The integration of advanced technologies, such as Artificial Intelligence (AI), into manufacturing processes is attracting significant attention, paving the way for the development of intelligent systems that enhance efficiency and automation. This paper uses the term ”Gentelligent system” to refer to systems that incorporate inherent component information (akin to genes in bioinformatics—where manufacturing operations are likened to chromosomes in this study) and automated mechanisms. By implementing reliable fault detection methods, manufacturers can achieve several benefits, including improved product quality, increased yield, and reduced production costs. To support these objectives, we propose a hybrid framework with a dominance-based multi-objective evolutionary algorithm. This mechanism enables simultaneous optimization of feature selection and classification performance by exploring Pareto-optimal solutions in a single run. This solution helps monitor various manufacturing operations, addressing a range of conflicting objectives that need to be minimized together. Manufacturers can leverage such predictive methods and better adapt to emerging trends. To strengthen the validation of our model, we incorporate two real-world datasets from different industrial domains. The results on both datasets demonstrate the generalizability and effectiveness of our approach.

Traditional artificial neural networks often face stability challenges in noisy Internet of Things (IoT) environments, owing to their dependence on static weights and deterministic computational frameworks. Spiking neural networks (SNNs), which mimic the spike-based coding mechanisms of biological neurons, present a promising alternative for edge devices due to their inherent temporal dynamics and energy efficiency. However, most existing Spiking neural network (SNN) models utilize fixed neuronal parameters, restricting their ability to adapt to diverse noise conditions commonly encountered in IoT applications. To address this issue, we propose a fractional-order plasticity approach integrated with spatio-temporal backpropagation algorithm. It enables simultaneous updating of synaptic weights and fractional-order parameters during training, allowing neurons to adjust their intrinsic dynamics. By incorporating fractional-order calculus, neurons can effectively integrate historical input information, enhancing their ability to filter out noise and extract relevant features effectively. Experimental results demonstrate that our proposed approach significantly improves classification performance, achieving accuracies of 99.67% on MNIST, 94.50% on Fashion-MNIST, and 93.43% on CIFAR-10, outperforming existing leading methods on all evaluated static datasets, and reaching a competitive accuracy of 97.25% on the DVS-Gesture dataset. The framework also exhibits superior resistance to noise compared to weight-training-only methods. This adaptive mechanism empowers neurons across different network layers to develop optimal fractional-order configurations, thereby enhancing networks’ temporal information processing ability and overall robustness in noisy environments. The proposed framework considerably advances the state of the art in efficient and robust spiking neural network learning for IoT applications.

The deployment of billions of Internet of Things (IoT) devices is driving unprecedented data generation at the network edge, demanding high computational power for real-time deep learning (DL) while raising serious concerns about energy consumption. While edge computing offers a viable paradigm for decentralized DL by preserving data privacy and reducing latency, the substantial energy costs of DL training and inference pose a major challenge for resource-constrained edge devices. This work provides a comprehensive review of state-of-the-art studies that address energy efficiency at the intersection of DL and edge computing. Moving beyond isolated solutions, we analyze the critical need for a co-design approach integrating hardware and software with adaptive resource management to build sustainable systems. The paper systematically examines hardware-level optimizations and software-level techniques for reducing energy consumption while maintaining model accuracy. Furthermore, it investigates how adaptive management of compute, memory, and communication resources is key to dynamic energy savings. Finally, the paper synthesizes recent trends, identifies emerging opportunities, and discusses open challenges, positioning hardware-software co-design as the most promising approach for achieving scalable and energy-efficient deep learning in edge computing.

Serverless computing offers fine-grained billing and elastic scalability, making it appealing for workflow execution. However, it also suffers from cold-start latency and stricter execution constraints. In contrast, traditional serverful cloud resources, such as virtual machines, provide coarser provisioning granularity but benefit from relatively lower unit cost. This work explores the potential of blending serverful and serverless resources to harness their complementary strengths for cost-effective workflow execution. We propose a hybrid resource management framework that dynamically allocates workflow tasks across both types of resources. A Budget-constrained Workflow scheduling algorithm for Blended cloud (BWB) is developed to minimize makespan while respecting user-specified budget. Evaluation experiments are conducted under real-world cloud settings by using realistic workflow applications. BWB is compared against state-of-the-art approaches for serverful, serverless, and blended clouds. Experimental results show that BWB consistently outperforms its state-of-the-art peers, achieving makespan reductions ranging from 10.6% to 37.6%, thereby demonstrating the cost-effectiveness of blending cloud resources for workflow execution.

A Fuzzy$C$-Means (FCM) algorithm has been widely applied to image segmentation due to its simplicity and effectiveness. However, conventional FCM and its variants often struggle to maintain robustness and accuracy when dealing with complex noise environments, particularly Poisson and mixed Poisson-Gaussian noise. To address this shortcoming, this work proposes a novel Poisson-specific Residual-driven FCM (PRFCM) algorithm for robust image segmentation, which is the first work to develop a dedicated residual regularization mechanism that effectively realizes the robust estimation of Poisson noise (regarded as residual between noisy and noise-free images). It incorporates a weighted$\ell _{2}$-norm regularization term with respect to Poisson noise distribution into FCM. An iterative residual approximation method is introduced to solve the minimization problem about residual, thus simplifying PRFCM's optimization procedure and enhancing its computational efficiency. PRFCM is also extended to cope with mixed Poisson-Gaussian noise scenarios without compromising performance. Experimental results on both simulated and real-scene images demonstrate that the proposed approach outperforms other FCM-related methods in terms of segmentation accuracy, noise resilience, and structural preservation, especially in challenging noise conditions.

Ethereum 2.0 (ETH2) marks a pivotal shift in blockchain technology, transitioning from a Proof-of-Work (PoW) to a Proof-of-Stake (PoS) consensus mechanism, with Gasper at its core. While this evolution promises enhanced scalability and energy efficiency, the performance of its block proposal stage is highly sensitive to network latency and system parameters, such as slot length. This sensitivity introduces a critical trade-off between throughput and security, measured by the probability of blockchain forking. This paper reveals that network latency is not just a passive risk but an exploitable attack surface. We introduce the "adaptive latency-driven equivocation attack", a novel adversarial strategy where an attacker deliberately creates forks while mimicking the behavior of a high-latency node, thus achieving plausible deniability. To formally analyze and quantify the impact of this threat, we develop a comprehensive theoretical model by using Markov chains to analyze the fork probability and throughput of the Gasper's block proposal mechanism under both honest and adversarial conditions. Through extensive simulations, we validate the accuracy of our model in both normal and bursty traffic conditions. Our findings provide a systematic methodology for optimizing system parameters to achieve a robust balance between performance and security, offering a foundational guide for configuring ETH2 networks against sophisticated, latency-based threats.

Robust object detection under varying weather conditions (e.g., rain, fog, and snow) presents significant challenges for industrial vision systems due to inherent visual degradations in manufacturing sites and outdoor facilities. While knowledge distillation offers promising potential by feature imitating and logit mimicking, existing methods face two critical limitations: first, inadequate mechanisms for effectively transferring localization capabilities in the presence of severe image degradation, and second, suboptimal strategies for identifying optimal distillation regions. To address these issues, we present a symmetric localization distillation loss based on the Jensen–Shannon divergence. Its mathematical characteristics, e.g., boundedness, symmetry, and gradient smoothness, enable robust preservation of spatial relationships and stabilize training processes. In addition, we present an adaptive label assignment strategy to select distillation regions, thus reducing the sparsity of positive samples during our knowledge distillation process. This work is the first one to apply knowledge distillation to object detection in inclement weather conditions. Extensive experiments on three challenging datasets show that our method improves the student model's object detection accuracy while maintaining its inference speed.

Vehicle-to-Infrastructure (V2I) collaboration constitutes an emerging paradigm for advancing autonomous driving. However, the integrated collaboration of perception, communication, and computation within V2I system remains a critical challenge. To address it, we propose a Software-Defined Network (SDN)-based collaborative approach for Autonomous Vehicle and Road Infrastructure Network (AVRIN). The architecture designates road infrastructures as road nodes and autonomous vehicles as dynamic vehicle nodes, establishing AVRIN through SDN. The control plane dynamically maintains global network topology and distributed flow tables by continuously evaluating node accessibility, while the forwarding plane is responsible for packet transmission via the OpenFlow protocol. In the perception module, road nodes divide the perception range into spatial units, whereas vehicle nodes dynamically align these units with their drivable areas across temporal sequences. Through coordinated communication and computation modules, road nodes strategically allocate dedicated bandwidth and computational resources. Building on this approach, we develop a particle swarm-based multi-objective optimization algorithm to achieve balanced co-optimization across perception, communication, and computation. Experimental validation demonstrates its superior collaborative Bird's Eye View (BEV) detection performance on the V2X-Sim 2.0 dataset, outperforming existing approaches by 10.37% in mean Average Precision. Furthermore, evaluations on the newly collected Jiading dataset, from a real-world urban roadway, confirm the approach's robustness with 1.823-second computation time under dynamic network conditions.

Vision Transformers (ViTs) demonstrate significant promise in medical image analysis but face two critical challenges: 1) their limited ability to capture local features in data-scarce scenarios, leading to data inefficiency, and 2) their high computational and storage demands of the full fine-tuning process in transfer learning, resulting in parameter inefficiency. To achieve efficient and accurate medical image analysis, we propose Shared Attention Vision Transformer (SAViT) that comprises three innovative modules: i) Shared Prior Attention (SPA) that enhances data efficiency by innovatively employing a visual prompt to sequentially share consistent attention weights across local image regions, thereby enabling the learning of translational invariance to capture locality; ii) MixPool that preserves global modeling ability by aggregating local features after SPA through a multi-pooling mechanism, thus effectively facilitating long-range dependency across local image regions; and iii) Low-rank Multi-head Self-Attention (Lr-MSA) that improves parameter efficiency by using low-rank weights of multi-head self-attention, hence reducing computational complexity while maintaining accuracy in medical image analysis. SAViT demonstrates strong generalization across multiple medical imaging modalities, including retinopathy, dermoscopy, and radiography. Extensive experiments are conducted. The results indicate its high data efficiency and outstanding performance in comparison with more than 20 medical-specific and ViT-based models when all of them are trained from scratch. It excels in parameter-efficient tuning by surpassing 17 models across 6 datasets in transfer learning, with only ${0}.{17}$ M/ ${0}.{23}$ M trainable parameters on ViT-B/SwinViT-B backbones requiring ${86}.{60}$ M/ ${88}.{00}$ M parameters. Source code can be found at: https://github.com/LYH-hh/SAViT.

Accurate fault detection in high-dimensional industrial environments remains a major challenge due to the inherent complexity, noise, and redundancy in sensor data. This article introduces compressed latent autoencoder for industrial representation and evaluation (CLAIRE), that is, a hybrid end-to-end learning framework that integrates unsupervised deep representation learning with supervised classification for intelligent quality control in smart manufacturing systems. It employs an optimized deep autoencoder to transform raw input into a compact latent space, effectively capturing the intrinsic data structure while suppressing irrelevant or noisy features. The learned representations are then fed into a downstream classifier to perform binary fault prediction. Experimental results on a high-dimensional dataset demonstrate that CLAIRE significantly outperforms conventional classifiers trained directly on raw features. Moreover, it incorporates a post hoc phase, using a game-theory-based interpretability technique, to analyze the latent space and identify the most informative input features contributing to fault predictions. The proposed framework highlights the potential of integrating explainable artificial intelligence with feature-aware regularization for robust fault detection. The modular and interpretable nature of the proposed framework makes it highly adaptable, offering promising applications in other domains characterized by complex, high-dimensional data, e.g., healthcare, finance, and environmental monitoring.

Fraud detection on dynamic graph (FDDG) is in high demand across many real-world applications, such as financial transaction networks and social networks. A graph neural network (GNN) is an advanced methodology for learning graph data and has been widely adopted in fraud detection tasks. Despite the promising results, research on GNN-based fraud detection faces two critical challenges. First, due to the lack of temporal consistency mining in existing methods, camouflaged fraudsters can easily evade detection by imitating the behavioral patterns of benign users. Second, with constantly evolving fraud patterns and a scarcity of fraud samples, existing methods can easily overfit to current fraud patterns and struggle to identify the evolved and new ones. To address these challenges, we propose CamFD, a semi-supervised camouflage-aware fraud detection model on dynamic graphs. To detect the temporal inconsistency in the behavioral patterns of camouflaged fraudsters, CamFD is equipped with a novel consistency-sensitive contrastive learning (CSCL) module. CSCL discriminatively learns the consistency and inconsistency in users’ behavioral patterns by constructing temporal and structural contrastive pairs. Since modeling the evolving fraud patterns with sparse fraud samples is almost impractical, CamFD concentrates on modeling the relatively stable patterns of extensive benign users with a multivariate Gaussian distribution modeling (MGDM) module. We conduct extensive experiments on a private credit card fraud dataset as well as three public datasets. The experimental results demonstrate that CamFD outperforms ten state-of-the-art (SOTA) baselines across all datasets, with the area under ROC curve (AUC) improvements ranging from 1.03% to 5.32%.

Topological semantic maps serve as an effective tool for the partially sighted or visually impaired (PSVI) in indoor navigation. Essential to their construction is the precise segmentation of floor plans. However, current techniques fall short in segmentation performance. To overcome this, we introduce a gated-dual-attention-based full-resolution network (GFNet) for floor plan segmentation. We leverage the inherent low-stage detailed features and intraclass-and-interclass contextual dependencies within floor plans to maximize segmentation effectiveness. Our proposed network incorporates modified residual blocks in early stages to maintain full-resolution capture with a low parameter count. We design and apply a novel gated dual attention (GDA) module that efficiently integrates channel and spatial contextual information to enhance local feature representation. This module improves the overall performance of our network while ensuring minimal parameter fluctuation. We also propose a 2-D deep supervision (TDDS) method to merge features from all stages, further enhancing the multilevel feature representation ability. Finally, a practical topological semantic mapping method for PSVI indoor navigation is introduced. All models are evaluated on the rasterized floor plan datasets R2V and R3D. Experimental results show that the proposed network achieves 2.21% and 2.74% mIoU improvements over the previous state-of-the-art method on R2V and R3D, respectively, for the challenging categories of walls and doors. This contributes to more accurate topological semantic mapping.

This study investigates a scheduling problem involving dual-arm cluster tools (CTs) that simultaneously handle two types of wafers, considering both wafer priority and residency time constraints. The two types of wafers have their own processing routes and processing times at each step. To fully utilize the resources of the CTs, we use the fewest processing modules (PMs) to produce one type of wafers with maximum productivity, and use the available PMs to produce the other type of wafers. Based on this, we introduce a swap sequence for scheduling a dual-arm robot, which is simple to implement and supports periodic operations. Without affecting the priority wafer production, we provide the necessary and sufficient conditions for scheduling a CT that processes two types of wafers, and present the optimal PM configuration. A high-performance algorithm is developed to determine an optimal periodic schedule, with its practicality and feasibility illustrated through several examples.

Neural architecture search (NAS) has emerged as a powerful paradigm that enables researchers to automatically explore vast search spaces and discover efficient neural networks. However, NAS suffers from a critical bottleneck, i.e., the evaluation of numerous architectures during the search process demands substantial computing resources and time. In order to improve the efficiency of NAS, a series of methods have been proposed to reduce the evaluation time of neural architectures. However, they are not efficient enough and still only focus on the accuracy of architectures. Beyond classification accuracy, real-world applications increasingly demand more efficient and compact network architectures that balance multiple performance criteria. To address these challenges, we propose the SMEMNAS, a pairwise comparison relation-assisted multiobjective evolutionary algorithm (EA) based on a multipopulation (MP) mechanism. In the SMEMNAS, a surrogate model is constructed based on pairwise comparison relations to predict the accuracy ranking of architectures, rather than the absolute accuracy. Moreover, two populations cooperate with each other in the search process, i.e., a main population that guides the evolutionary process and a vice population that enhances search diversity. Our method aims to discover high-performance models that simultaneously optimize multiple objectives. We conduct comprehensive experiments on CIFAR-10, CIFAR-100, and ImageNet datasets to validate the effectiveness of our approach. With only a single GPU searching for 0.17 days, competitive architectures can be found by SMEMNAS, which achieves 78.91% accuracy with the MAdds of 570 M on the ImageNet. This work makes a significant advancement in the field of NAS.

The performance of models trained by Multi-Agent Reinforcement Learning (MARL) is sensitive to perturbations in observations, lowering their trustworthiness in complex environments. Adversarial training is a valuable approach to enhance their performance robustness. However, existing methods often overfit to adversarial perturbations of observations and fail to incorporate prior information about the policy adopted by their protagonist agent, i.e., the primary one being trained. To address this important issue, this paper introduces Adversarial Training with Stochastic Adversary (ATSA), where the proposed adversary is trained online alongside the protagonist agent. The former consists of Stochastic Director (SDor) and SDor-guided generaTor (STor). SDor performs policy perturbations by minimizing the expected team reward of protagonists and maximizing the entropy of its policy, while STor generates adversarial perturbations of observations by following SDor's guidance. We prove that SDor's soft policy converges to a global optimum according to factorized maximum-entropy MARL and leads to the optimal adversary. This paper also introduces an SDor-STor loss function to quantify the difference between a) perturbations in the agent's policy and b) those advised by SDor. We evaluate our ATSA on StarCraft II tasks and autonomous driving scenarios, demonstrating that a) it is robust against diverse perturbations of observations while maintaining outstanding performance in perturbation-free environments, and b) it outperforms the state-of-the-art methods.

Achieving accurate segmentation in low-light scenes is challenging due to 1) severe domain shift encountered when models trained on daylight data are applied to such scenes and 2) lack of large-scale fine-grained labels in low-light conditions. A good idea is to use the generalization capabilities of segmentation foundation models like Segment Anything Model (SAM) to address the scarcity of annotated data. However, applying SAM to low-light scenes faces a severe domain shift issue due to the lack of inductive bias in effectively transforming low-light features into natural-light ones. To address this issue, we propose to adapt SAM for low-light scenes. To reduce the reliance on labels of low-light data, we develop a self-training method that makes SAM generate source-free predictions. To reduce the domain gap between low-light target data and SAM's natural-light trained data, we design a transformation head that enhances low-light features prior to the application of SAM. We further propose a domain shift compensation loss that trains our model to select a domain-adaptation-optimal illumination-enhanced feature map. Experimental results demonstrate that our method well outperforms the state of the art on the Dark Zurich and Nighttime Driving datasets. Code is available at https://github.com/HongminMu/SALS.

Large-scale outdoor navigation is essential for unmanned ground vehicles (UGVs), but despite significant advancements, they still face two key challenges in practical applications. The first one is how to ensure safe navigation in environments with dynamic and low-lying obstacles that LiDAR cannot detect. The second one is how to conduct the adaptive re-planning of target points while some of them are blocked by temporary obstacles. To address these challenges, this work proposes a Dynamic and Low-lying-obstacle Avoidance Navigation (DLAN) system to conduct perception, planning, and point correction for UGVs. To efficiently and accurately detect dynamic obstacles, it designs a lightweight Ensemble3D framework that integrates three fast but low-accuracy detection methods. A multi-criteria waypoint optimizer is used to assist UGVs in path planning. It ensures a balance between obstacle avoidance and path following. To adjust blocked target points through local re-planning, this work designs a checkpoint correction method. Extensive simulations and real-world experiments demonstrate that DLAN enables reliable navigation with high efficiency and robust obstacle avoidance in complex environments. More details can be found on our project homepage and video1.

Small object detection (SOD) given aerial images suffers from an information imbalance across different feature scales. This makes it extremely challenging to perform accurate SOD. Existing methods, e.g., Feature Pyramid Network (FPN)-based algorithms, focus on extracting high-resolution and low-resolution semantic features from different convolution layers. However, in deeper convolution layers, semantic feature misalignment and the loss of key information are inevitable. To tackle such issues, this work proposes a new encoder-decoder-based SOD framework with a Diffusion Model and Swin Transformer given aerial images. First, we reformulate an SOD task as a Noise-to-Box process. We then construct an encoder-decoder-based framework by using a diffusion model and Swin Transformer for dynamic bounding box generation. We introduce a decoupling training and inferencing strategy to recognize and locate small objects accurately. We finally evaluate the proposed framework on several public benchmarks. The experimental results well show its better SOD performance than the state of the art. Code is available at https://github.com/BrainPotter/EDSOD.

In order to provide natural and immersive interactive experience for teleoperation in the context of human-robot collaboration and interaction, this work introduces a natural human-robot interaction system for teleoperation based on ultrasonic haptic feedback. Specifically, our system can accurately capture an operator's hand movements and replicate these actions on the remote robot with low latency and high fidelity. It utilizes an ultrasonic phased array to achieve non-contact haptic feedback. We propose a dynamic ultrasonic array acoustic field customization method based on interactive feature information image. This method can dynamically adjust the acoustic field according to the operator's hand characteristics, focus on multiple target points in real time, and project them onto an operator's fingertips, thereby providing force-controllable non-contact haptic feedback to the operator. The operator is integrated into the feedback loop of our system, controlling the system through multimodal feedback to form a high-quality human-in-the-loop closed control system. The system's performance is validated in two classic robotic tasks: block pick-and-place and nut-tightening. The experimental results show that the system exhibits excellent accuracy and dexterity, and can efficiently complete tasks with high accuracy while providing great interactive experience for operators.

Semi-Supervised Learning (SSL) can leverage abundant unlabeled data to boost model performance. However, the class-imbalanced data distribution in real-world scenarios poses great challenges to SSL, resulting in performance degradation. Existing class-imbalanced semi-supervised learning (CISSL) methods mainly focus on rebalancing datasets but ignore the potential of using hard examples to enhance performance, making it difficult to fully harness the power of unlabeled data even with sophisticated algorithms. To address this issue, we propose a method that enhances the performance of Imbalanced Semi-Supervised Learning by Mining Hard Examples (SeMi). This method distinguishes the entropy differences among logits of hard and easy examples, thereby identifying hard examples and increasing the utility of unlabeled data, better addressing the imbalance problem in CISSL. In addition, we maintain a class-balanced memory bank with confidence decay for storing high-confidence embeddings to enhance the pseudo-labels' reliability. Although our method is simple, it is effective and seamlessly integrates with existing approaches. We perform comprehensive experiments on standard CISSL benchmarks and experimentally demonstrate that our proposed SeMi outperforms existing state-of-the-art methods on multiple benchmarks, especially in reversed scenarios, where our best result shows approximately a 54.8% improvement over the baseline methods. Our code is available at https://github.com/pywin/SeMi.

Accurate fault detection in industrial environments with high-dimensional sensor data presents significant challenges. This paper presents an explainable AI framework that combines unsupervised deep representation learning with supervised classification for enhanced quality control in smart manufacturing systems. It utilizes a finely tuned deep autoencoder to convert raw data into a compressed latent representation, effectively capturing the underlying structure while removing irrelevant or noisy features. These representations are then used by a downstream classifier to predict faults. Experimental results on a high-dimensional dataset show that the proposed solution outperforms traditional classifiers that process raw features directly. In addition, the framework incorporates an interpretability phase. It adopts a game-theory-based technique to analyze latent space and identify the most influential features that contribute to accurate faulty predictions.

Reliability assessment is essential for ensuring the stable operation of industrial systems. This study presents a hybrid reliability assessment framework that integrates mechanism-based and data-driven approaches for online degradation evaluation. A representative degradation mode— impact fatigue—is investigated to demonstrate the method. The mechanistic component is formulated by using crack propagation theory and Miner’s cumulative damage rule; while the data-driven one employs a generalized power-law Wiener process with an additive nonlinear drift. Both sources of information are fused through an evolving unscented Kalman filter that performs optimal and unbiased online state estimation via prediction and observation updates. In the proposed framework, the Wiener process serves as the state equation, and the mechanistic model functions as the measurement equation. The reliability of impact fatigue is further evaluated by using the inverse Gaussian distribution, which is inherently associated with the Wiener process. Application to a thrust bearing in the main pump of a pressurized water reactor demonstrates the effectiveness and superiority of the proposed method over existing ones.

Water quality prediction methods forecast the short- or long-term trends of its changes, providing proactive advice for preventing and controlling water pollution. Existing water quality prediction methods typically fail to capture water quality’s nonlinear characteristics accurately and only consider historical time series data. However, meteorology and other factors also significantly impact water quality indicators. Therefore, considering only historical data of water quality time series is not feasible. To solve this problem, this work proposes a hybrid water quality prediction model called CMLIP, which integrates convNeXt V2, multimodal bottleneck transformer, low-rank multimodal fusion, iTransformer, and PatchTST. CMLIP inputs water quality time series and meteorological remotely sensed rainfall images into a multimodal fusion module before prediction. Specifically, CMLIP integrates the model of ConvNeXt V2 to extract image features. Its multimodal fusion module combines a multimodal bottleneck transformer and the low-rank multimodal fusion to fuse the time series and images. Furthermore, CMLIP combines iTransformer and PatchTST to form an improved prediction module that realizes the prediction of fused features. Experimental results with real-life water quality time series and remotely sensed rainfall images demonstrate that CMLIP when fusing meteorological data, achieves an average improvement of 17% in water quality forecasting accuracy compared to forecasts using only water quality time series. Moreover, CMLIP outperforms other state-of-the-art algorithms in both data fusion and prediction, with an average enhancement of 6% in fusion effectiveness and an average improvement of 22% in prediction accuracy.

Given the forthcoming emergence of 6G communication models, the integration of terrestrial and nonterrestrial infrastructures is receiving increasing attention due to its widespread reach and broadcasting/multicast functions. The utilization of edge computing in space-related applications is appealing. However, the issue of positioning satellite edge servers and deploying services has yet to be resolved. Besides, existing studies mainly concentrate on energy consumption and latency problems, often neglecting the user mobility and potential privacy leakage issues in a mobile edge computing (MEC) environment. Yet it is crucial to optimize computation offloading and resource allocation for satellite-terrestrial edge computing networks. This work designs an innovative architecture for collaborative computation among multiple mobile devices and MEC servers deployed in ground stations and satellites. Based on this architecture, we formulate a nonlinear integer optimization problem to minimize the total system energy consumption. The model integrates several complex real-life nonlinear constraints, including operator cost, edge servers’ computing capacity, storage capacity, resource and latency, and privacy ones. To tackle the problem, this work proposes an advanced hybrid algorithm named a slime mold algorithm with genetic operations and individual updates of grey wolf optimizer (SMG2). SMG2 optimizes user mobility and privacy protection while optimizing server and service placement to minimize total energy consumption. Simulation experiments demonstrate that SMG2 reduces energy consumption drastically over the state of the art.

The advent of smart manufacturing in Industry 4.0 signifies the era of connections. As a communication protocol, Object linking and embedding for Process Control Unified Architecture (OPC UA) can address most semantic heterogeneity issues. However, its semantics are not formally defined at the application layer. To address the information silo problem caused by semantic heterogeneity, an integration framework named Querying of Ontology Mapping-based OPC UA (QOMOU) is proposed. QOMOU extracts information models of OPC UA servers into resource description framework triples and utilizes web ontology language for semantic enrichment and inference. Then, an Event Class Semantic Similarity Calculation (ECSSC) method is proposed for device type identification, enabling the classification of semantically heterogeneous OPC UA devices. The effectiveness of ECSSC is validated through queries with the RDF query language (SPARQL) protocol in Apache Jena. Experimental results demonstrate that ECSSC improves the accuracy of device identification by approximately 7% compared to benchmark device identification models. Specifically, compared with graph embedding-based methods, QOMOU’s query performance is approximately 13% higher, and its query efficiency is 5% higher on average compared to both structured query and extensible markup languages. Moreover, by employing a keyword-matching algorithm, the query accuracy of the existing heterogeneous data integration scheme is improved by 4% on average. This enhancement can boost the operational efficiency of Internet of Things systems based on the OPC UA architecture.

Industrial Internet of Things (IIoT) plays a crucial role in advancing smart manufacturing by connecting numerous devices, enabling data exchanges, and supporting industrial applications. Yet, the timely and proper scheduling of asynchronously generated Heterogeneous Computing Tasks (HCTs) in IIoT environments remains a significant challenge. In this article, we first introduce the representation and notation of such HCTs and define a computing network structure. We then propose an initial mathematical programming-based scheduling model aimed at minimizing HCT completion time. To make this model easy to solve, we reformulate it by using logical constraints and derive a constraint programming-based model, for which a feasibility-guaranteed solution algorithm is developed. This algorithm leverages two easily-verified propositions to either identify feasible solutions or demonstrate the infeasibility of the problem.Furthermore, we have proven a critical proposition that facilitates the development of a hybrid programming-based scheduling approach, effectively combining the strengths of both mathematical and constraint programming models. As demonstrated through extensive computational experiments, our proposed approach achieves an average reduction of 20% in HCT completion time in comparison with its existing peers. It consistently and timely provides the high-quality solutions that meet the required deadlines.

In many fields, time series prediction is gaining more and more attention, e.g., air pollution, geological hazards, and network traffic prediction. Water quality prediction uses historical data to predict future water quality. However, it is difficult to learn a representation map from a time series that captures the trends and fluctuations to effectively remove noise from the time series data and investigate complex nonlinear relationships. To solve these problems, this work proposes a time series prediction model, called DPSGT for short, which integrates Dual Patch Savitsky–Golay filtering and Transformer. First, DPSGT adopts the SG filtering to decompose the time series data and reduce the noise interference to improve long–term prediction capabilities. Second, to tackle the limitation of temporal representation capability, DPSGT adopts dual patches to ravel temporal series into local and global patches, which can tackle local semantic information and enlarge the receptive field. Third, it utilizes a transformer mechanism to address the nonlinear problem of the water quality time series and improve the accuracy of the prediction. Two real-world datasets are utilized to evaluate the proposed DPSGT, and experiments prove that DPSGT improves root mean-square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and R2 by 6%, 5%, 8%, and 7%, respectively, compared with other benchmark models.

Internet of Things (IoT) devices are widely distributed with large and scattered data, risk of privacy leakage and high bandwidth cost make data transmission and aggregation impractical. Federated Learning (FL) is an important privacy-preserving multi-party learning paradigm, involving collaborative learning with others and local updating based on private data. Data heterogeneity is one of the main challenges in FL scenarios, which can greatly limit FLs applicability and performance. Most existing approaches tackle the heterogeneity challenge during local training or model aggregation while ignoring the performance drop caused by direct model aggregation and catastrophic forgetting in the global model. This work proposes a novel FL approach, called Federated Learning with Dataset Condensation (FedDC). It aims to relieve the issue of knowledge discrepancy among local models and catastrophic forgetting in the global model. Specifically, this work proposes a differential distribution matching method to summarize local data for each client without compromising on data privacy. It introduces collaborative knowledge distillation to encourage local models to learn from others. To mitigate interround forgetting, this work stabilizes the averaged global models training by leveraging auxiliary information from its immediately past one. Extensive experiments are conducted. The results show that FedDC significantly outperforms the state-ofthe-art methods on various vision tasks.

A collaborative system that includes mobile devices (MDs), edge nodes (ENs), and the cloud is needed where ENs at the network edge can run offloaded tasks of MDs with limited resources and energy for timely processing for latency-sensitive applications. Unlike existing studies, we formulate a total cost minimization problem for the system for applications, which can be divided into several interdependent subtasks. Each subtask can be executed in MDs, ENs, and the cloud. This work formulates a mixed-integer nonlinear program to minimize the total system cost. To address it, a novel meta-heuristic optimization algorithm calledGeneticSimulated-annealing-basedParticle swarm optimization withAuto-Encoder (GSPAE) is proposed, which innovatively combines feature extraction of deep learning and global search of meta-heuristic optimization. Genetic operations provide diverse solutions, the Metropolis acceptance of annealing offers a robust global search, and autoencoders (AEs) extract distribution characteristics of particles toward high-quality regions for fast convergence. Thus, GSPAE optimizes the associations between ENs and MDs and the scheduling of subtasks among MDs, ENs, and the cloud. Experiments with large-scale Google cluster datasets show that compared to state-of-the-art benchmark methods, GSPAE reduces the total cost by at least 17% while strictly meeting limits of application latency, available energy, computing, and communication resources of ENs and MDs.

In recent years, cloud computing has witnessed widespread applications across numerous organizations. Predicting workload and computing resource data can facilitate proactive service operation management, leading to substantial improvements in quality of service and cost efficiency. However, these data often exhibit non-linearity, high volatility, and interdependencies across different categories, presenting challenges for accurate forecasting. Consequently, there is a critical need to develop a method that thoroughly and comprehensively analyzes all available data to forecast future trends effectively. This work proposes a novel integrated data-enhanced prediction model named SVAPI for achieving high-accuracy workload prediction in cloud computing systems. SVAPI employs the Savitzky-Golay filter, Variational mode decomposition, and the mode selection based on Amplitude-aware Permutation entropy for feature processing, whose features are subsequently utilized by Informer for multivariate joint analysis of the enhanced data, achieving high-precision prediction. Ablation and comparative experiments with advanced prediction models are conducted on the Google cluster trace and other typical datasets. Realistic data-driven results indicate that SVAPI improves the prediction accuracy by 37.7% compared to the original Informer, with each module contributing to the performance enhancement. Furthermore, compared with Autoformer, SVAPI enhances the prediction accuracy of workload, CPU, and memory by 65.6%, 66.9%, and 70.8%, respectively, demonstrating that SVAPI owns strong abilities in noise filtering, feature processing, and multivariate joint analysis for achieving higher prediction accuracy.

The widespread adoption of WiFi has made throughput efficiency a critical concern in wireless networks. While Full-Duplex (FD) technology promises to double network capacity by enabling simultaneous transmission and reception, existing FD-WiFi designs primarily focus on the data transmission phase, leaving the fundamental inefficiencies in channel contention unaddressed. This paper presents CollFree, a novel WiFi protocol that exploits FD capabilities during both contention and data transmission phases. At its core, CollFree introduces a Slotwise Arbitration (SA) mechanism that enables each node to simultaneously transmit contention signals and sense channel status in each contention slot. This dual-mode operation significantly reduces contention time and facilitates collision-free data transmissions through a unique winner-determination process. We then develop theoretical models to analyze CollFree’s contention performance and throughput efficiency under both perfect and imperfect Clear Channel Assessment (CCA) conditions, providing guidelines for parameter optimization in practical deployments. Extensive simulations demonstrate that CollFree enhances throughput efficiency by over 20% compared to state-of-the-art FD-WiFi systems while maintaining distributed control and compatibility with current WiFi standards. These results suggest that CollFree represents a significant step toward realizing the full potential of FD technology in next-generation WiFi networks.

ABSTRACT Background In recent years, Large Models (LMs) have been rapidly developed, including large language models, visual foundation models, and multimodal LMs. They are updated and iterated at a very fast pace. These LMs can accomplish many tasks, e.g., daily work assistant, intelligent customer service, and intelligent factory scheduling. Their development has contributed to various industries in human society. Aims The architectural flaws of LMs lead to several problems, including illusions and difficulty in locating errors, limiting their performance. Solving these problems properly can facilitate their further development. Methods This work first introduces the development of LMs and identifies their current problems, including data and energy consumption, catastrophic forgetting, reasoning ability, localization fault, and ethical problems. Then, potential solutions to these problems are provided, including increase data and computation capability, neural‐symbolic synergy, and data orientation to human pattern. Discussion This work discusses developing vertical domain LMs on top of some base LMs. In addition, this work introduces three typical real‐world applications of LMs, including autonomous driving, smart industrial productions, and intelligent medical assistance. Conclusion By embracing the advantages of LMs and solving their fundamental problems, many industries are expected to achieve promising prospects in the future.

Over the past decades of rapid development, the global water pollution problem became prominent. Accurate water quality prediction can detect the trend and anomaly of water quality changes in advance, thereby taking timely measures to avoid water quality problems. Traditional statistical methods for water quality prediction tend to fail to capture the complex relationship among multiple water quality variables. Deep learning models face a challenge to capture both temporal dependence and spatial correlation of the water quality series data. To solve the above problems, this work proposes an adaptive and dynamic graph fusion water quality prediction model based on a spatiotemporal attention mechanism namedSpatial-TemporalGraphFusionTransformer (STGFT). It integrates a spatial attention encoder, a temporal attention encoder, an adaptive dynamic adjacency matrix generator, and a multi-graph fusion layer. Among them, the first two are adopted to capture the spatial correlations and temporal characteristics among different water quality monitoring stations, respectively. The generator can produce adaptive and dynamic adjacency matrices to reflect potential spatial relationships in a river network. Experimental results with real-life water quality datasets reveal that the prediction accuracy of STGFT outperforms the existing state-of-the-art models. Note to Practitioners—This paper is motivated by the problem of long-term water quality prediction. The highly volatile water quality data and the nonlinear characteristics of the time series greatly affect the accuracy of the forecasting task. Existing approaches fail to simultaneously capture spatial correlations and temporal characteristics among different water quality monitoring stations, affecting the accuracy of water quality predictions. This work proposes a water quality prediction method that captures the spatial correlations and temporal characteristics among different water quality monitoring stations. Moreover, it produces adaptive and dynamic adjacency matrices to reflect potential spatial relationships in a river network. Experimental results from three real-world datasets show that this approach is feasible and obtains more accurate prediction results. Furthermore, this method can also be applied to other areas of time series prediction, including finance, traffic, and smart manufacturing.

Water quality prediction provides timely insights for addressing potential water environmental issues. Transformer-based models have been widely used in water quality prediction. However, the following challenges exist: 1) Noise in the time series of water quality causes nonlinear models to be overfit; 2) It is difficult to identify temporal correlations in complex time series data; and 3) Information utilization is limited in long-term prediction. This work introduces a large-scale water quality prediction model named SVD-Autoformer to address them. SVD-Autoformer combines aSavitzky-Golay (SG) filter,variational modedecomposition (VMD), anauto-correlation mechanism, and a deep decomposition architecture, which is achieved in the renovation of the transformer. First, the SG filter removes noise while retaining valuable data features. SVD-Autoformer employs the SG filter as a data preprocessing tool to reduce noise and prevent nonlinear models from overfitting. Second, VMD extracts major modes of the signals and their respective center frequencies, thus providing richer features for the prediction. Third, the deep decomposition architecture with embedded decomposition modules allows for gradual decomposition during the prediction process. SVD-Autoformer employs the architecture to extract more predictable components from complicated water quality time series for long-term forecasting. Finally, SVD-Autoformer applies the auto-correlation mechanism to capture the temporal dependence and enhance information utilization. Numerous experiments are conducted and the results demonstrate that SVD-Autoformer provides superior prediction accuracy over other advanced prediction methods with real-world datasets. Note to Practitioners—This paper explores the critical aspects of time series water quality prediction, aiming to provide valuable insights for engineers and decision-makers. Traditional water quality prediction methods primarily rely on linear time series approaches and suffer from high computational complexity when dealing with large-scale data. This study is motivated by the transformer architecture with highly parallel computing capability and innovatively proposes deep decomposition architecture to extract more predictable components. In practice, to handle massive data with low time complexity, we introduce an auto-correlation mechanism. We conduct experiments using real-world datasets to demonstrate that this method achieves superior water quality prediction accuracy. Additionally, the method has been deployed in a real-world water quality prediction platform. Our future work includes its applications to different real-world datasets arising from electric power, intelligent transportation, and meteorological rainfall prediction.

Many deadlock control methods rely on reachability graphs of Petri nets (PN), thus suffering from state-space explosion issues. This paper proposes a novel approach to designing liveness-enforcing supervisors for PN by constructing its maximally good step graphs (MGSG), a class of partial order techniques in mitigating the aforementioned issues. Specifically, we first categorize the MGSG markings into allowed and unallowed ones. Then, we define good and risky transitions at allowed markings. Through the execution of risky transitions, the initial marking cannot be reachable from the allowed ones. Next, we design a maximal number of risky transitions (MNRT) problem to compute control places. In MNRT, all allowed markings and the firing of all good transitions are permitted, while risky transitions are forbidden. The objective is to maximize the prevention of risky transitions by using a single control place, which can be achieved by solving an integer linear programming problem. MNRT problems are recursively solved for unforbidden risky transitions until their resulting markings are prohibited. Finally, a controlled PN is generated and has been demonstrated to retain liveness. The experimental results show that our approach effectively reduces the number of control places and mitigates state-space explosion issues over its state-of-the-art peers. Note to Practitioners—Raw materials are handled in manufacturing systems through a series of well-coordinated processes. However, multiple processes competing for the same resources (like machines or robots) may result in an undesirable deadlock situation, where production slows down or entirely halts. A liveness-enforcing supervisor is designed to counter such a situation. As a mathematical tool, PN can be applied to model manufacturing systems. The supervisor design involves adding control places and directed arcs into PN to ensure its liveness, which implies that the systems can execute their tasks and never enter into deadlocks. In this work, we develop a deadlock control approach, which can generate a live and controlled PN without exploring a complete state space of PN. Experimental results show the effectiveness of our approach in reducing the number of control places and its practicability applied to large-scale systems, thus surpassing state-of-the-art methods.

As massive data is generated by Internet of Things (IoT) devices, user-end devices are required to implement computation-intensive functionalities, including multi-sensory data processing and analysis, sophisticated system control schemes, and artificial intelligence. Mobile Edge Computing (MEC) is a significant technology that has the potential to extend the computation and storage capacities of user-end devices by the decentralization of required resources and contents near users and at the edge. A crucial challenge in this direction is the development of a smart mechanism to effectively cache contents upon Edge Servers (ESs) near users for high effectiveness and low latency of content delivery with the constraints on computational and storage capacities of ESs. This study employs a queuing model for analyzing total request delay and interprets the content caching problem as an adversarial semi-bandits problem. We propose an Online Self-feedback Adversarial Semi-bandits Learning (OSAL) algorithm that incorporates a dual-layer learning architecture for dynamically generating caching strategies and maximizes the long-term reward. Experimental results demonstrate that the proposed method significantly outperforms the state-of-the-art methods across various performance metrics in a real-world multi mobile-user content caching case.

Job-dependent tool switching is necessary in many batch processing systems (BPSs). Heterogeneous tool demand and extra time consumption for tool switch bring great challenge for high-performance production scheduling in BPSs. In this study, we present a novel batch scheduling problem derived from a satellite vibration test system. Different from basic scheduling problems in BPSs with tool switching, it has complications caused by many-to-many task-tool matching and job release time constraints. Specifically, each job has predetermined candidate tools. One of them should be selected and equipped to process a given job. Unlike sequence-dependent setup time that is only related to two adjacent jobs, tool switching time is additionally dependent on the tools selected by both jobs from their respective candidate tools. Both job sequence and tool selections need to be determined to minimize makespan. Mixed-integer linear programming (MILP) and constraint programming (CP) models are developed for formulating the problem. We prove that the problem isNP-hard. To obtain exact solutions, we propose a novel branch-andbound algorithm (B&B) based on the analysis and extraction of problem properties. It integrates a domain reduction mechanism, branching strategies, and dominance rules. Among them, the domain reduction mechanism reduces the number of jobs to be scheduled; the branching strategies accelerate solving speed; and the dominance rules save the memory consumption. Ablation studies and key influencing factor experiments are performed. The results show that the proposed B&B can optimally solve much larger instances than state-of-the-art methods (including some heuristics). It can optimally solve industrial-size problems in a short time, implying that it can well satisfy the performance demand of a satellite vibration test system and thus advancing the filed of optimal scheduling for BPSs.

Accurate prediction of grasp stability is crucial for reliable and precise operations with multi-fingered robotic hands. Traditional methods tend to oversimplify tactile information and pay equal attention to all regions of the data. This can obscure subtle yet critical variations and introduce noise, increasing the risk of stability assessment errors. To address these challenges, a novel self-supervised method, Adversarial Subgraph Contrastive Learning (ASCL), is proposed. It constructs an instance graph from the spatial distribution and features of perceptual nodes. It employs a bi-level adversarial strategy to enhance latent data representations by maximizing the mutual information between the instance graph and its semantic subgraphs, while minimizing it with its noisy subgraphs. To prevent trivial solutions and continuous relaxation of semantic subgraphs, node confidence and edge connection terms are incorporated to ensure stabilization. From an information-theoretic perspective, ASCL exhibits notable advantages on unlabeled or sparsely labeled data, well outperforming existing methods in empirical tests with robotic hands.

This work considers a task scheduling problem with deadline constraints in human-cyber-physical systems. To find its energy-efficient schedules in a short time, an autoencoder-embedded iterated local search algorithm is proposed to solve it. Iterated local search is selected as a main scheduler. In order to handle real-time requirements and high computational load involved in the problem solution, a Long Short-Term Memory-based AutoEncoder model (LSTM-AE) is constructed to capture the relevant and complementary features of the considered problem. The model is used to find low-order and high-fit sub-solutions via unsupervised end-to-end learning, and generates promising solutions in an informative low-dimensional solution space. To further reduce computational burden, a two-stage optimization framework is constructed, which includes an off-line training phase and an online optimization one. The former trains LSTM-AE by using expert knowledge and historical data. The latter designs optimal resource allocation strategies to build a high-quality initial solution. Then, LSTM-AE-assisted local search operators are proposed and used to reform the initial solution and generate better ones. Various numerical experiments are performed to compare the proposed method with several classic heuristics and some recently-developed methods. The results show its superiority over them.Note to Practitioners—In human-cyber-physical systems, a task scheduling problem is usually solved by using heuristics due to the limited computational resources. Nevertheless, fast dispatching rules tend to perform poorly. Meta-heuristics can find a relatively high-quality schedule but are time-consuming, especially for a population-based algorithm that requires to evaluate a fitness function for many candidate solutions at each iteration. To balance computational burden and solution quality, our idea is to combine machine-learning methods with meta-heuristics. Specially, we integrate a long short-term memory-based autoencoder model into iterated local search to improve the latter’s optimization ability. The combination of meta-heuristics and machine learning techniques makes it possible to obtain a high-quality schedule for the concerned problems in a short time. Theoretic analysis and experimental results show that the proposed method well outperforms its competitive peers.

Autonomous vehicles (AVs) are considered as next-generation delivery vehicles for logistics systems. This study proposes a dynamic Meal Delivery Routing Problem with a hybrid Rider-AV fleet (MDRP-RA). The hybrid fleet consists of riders and AVs. Each order must be fulfilled by an AV or a rider. Some orders can be delivered by riders or AVs only, while others can be delivered by both. The food of each order is one of three product segments (regular, frozen, and hot food), and each segment has a particular temperature need. An AV has multiple compartments, each of which needs to be cooled (heated) if it contains frozen (hot) food. Thus, AVs can deliver all kinds of food, while riders can deliver regular food only. A mathematical programming model is established for MDRP-RA, with the objective of minimizing the total cost, including the vehicle fixed cost, delivery fee to riders, energy consumption cost, and penalty cost for delay. An Adaptive Large Neighborhood Search based Approach (ALNS-A) is proposed to solve MDRP-RA. It involves a local search procedure with removal and insertion operators, where five operators are specifically devised for the problem. Experiments show that it can effectively solve MDRP-RA and outperforms comparative approaches.Note to Practitioners—AVs have great application potential in meal delivery since they have the advantages of saving labor costs, large capacity, and maintaining food temperature. The hybrid fleet of AVs and riders can meet diversified customer needs but brings challenges to the meal delivery route problem under a dynamic environment. This paper proposes a dynamic meal delivery routing problem with a hybrid rider-AV fleet, with the objective of minimizing the total cost. Some orders must be delivered by riders, some by AVs, and some by both. The food falls into three product segments, i.e., regular, frozen, and hot food. AVs can provide cooling or auxiliary heating to maintain the meal’s temperature. An adaptive large neighborhood search-based approach is proposed, which can provide high-quality solutions for problem instances. The approach can be embedded in takeout information platforms to realize intelligent scheduling of meal delivery.

Mobile edge computing (MEC) is an evolving paradigm for rendering services through network-accessible resources deployed over Internet of Things (IoT) nodes at the edge. Nevertheless, an MEC environment usually employs thousands of physical machines connected via hundreds of switches/routers that communicate and coordinate to deliver computing service. In such complicated systems, faults caused by software, human errors, and hardware are often unavoidable. The edge of network presents a dynamic environment with great quantities of terminals, high mobility of mobile devices, heterogeneous applications, and intermittent traffic. In such an environment, MEC can suffer from unbalanced resource provisioning and interruptions of faults occurring at different levels, which further causes task faults and affects service quality. To address this challenge, this work proposes a novel fault-tolerant offloading method for handling faults by leveraging a reinforcement-learning-based service offloading decision model. The model synthesizes a Dueling Deep Q Network (DQN)-based algorithm for deciding user offloading behaviors and an adaptive checkpointing method for improving task execution reliability. For the purpose of model validation and comparison, extensive simulations are conducted. Numerical results clearly demonstrate that the proposed method is highly effective and outperforms existing methods.

Industry 5.0 heralds a new wave of the industrial revolution, placing a spotlight on human-centric intelligent manufacturing. At the core of Industry 5.0 lies the human-cyber-physical system (HCPS), a composite intelligent system where interactions among humans, cyberspace, and physical assets are orchestrated across diverse manufacturing levels and phases. Understanding the pivotal roles played by humans in these advanced systems is of paramount importance. Nonetheless, the exploration of HCPS within the context of Industry 5.0 remains in its infancy. This paper presents a holistic literature review of industrial HCPS from a human-centric perspective. A united architecture is employed to encompass the aspects of cognitive-to-technology integration and human-to-human interaction in HCPS, highlighting human-in-the-loop, human-on-the-loop, and human-in-the-society paradigms. The mechanisms of these paradigms and their effects on design, production, and service are investigated to expand the research landscape of intelligent manufacturing in Industry 5.0. Key enabling technologies that facilitate harmonious tri-space integration are introduced, and the future challenges of industrial HCPS are discussed. This work is expected to attract more open discussions and in-depth research on HCPS in the new industrial revolution era.Note to Practitioners—This paper is motivated by the emergence of Industry 5.0 that integrates humans into cyber-physical systems to offset drawbacks on both sides. It presents an overview of HCPS-related works to identify the state-of-the-art and open problems in the Industry 5.0 era. The review of HCPS applications in the design, production, and service phases can benefit engineers in the intelligent manufacturing area. Key enabling technologies on human ability augmentation, human-robot interaction, digital twin, human-cyber-physical data fusion, crowdsourcing, and system modeling, are analyzed to attract researchers in broader research fields to join in the development of industrial HCPS.

In many fields, spatiotemporal prediction is gaining more and more attention,e.g., air pollution, weather forecasting, and traffic forecasting. Water quality prediction is a spatiotemporal prediction task. However, there are several challenges in water quality prediction: 1) Water quality time series has a complex nonlinear relationship, making it difficult to predict; 2) Water quality sensors are distributed on the river networks and have a strong spatial dependence on water quality prediction; and 3) Poor long-term forecast accuracy. To solve these problems, this work proposes a spatiotemporal prediction model called a Fusion Spatio-temporal Graph Convolution Neural network (FSGCN). First, This work uses a temporal attention mechanism to solve the nonlinear problem of water quality time series. Second, It adopts a graph convolution to extract spatial dependencies of river networks, and the fusion of spatiotemporal can more easily capture spatiotemporal features. Third, it adopts a temporal convolution residual mechanism, improving long-term series prediction accuracy. This work adopts two real-world datasets to evaluate the proposed FSGCN, and experiments demonstrate that FSGCN outperforms several state-of-the-art methods in terms of prediction accuracy.Note to Practitioners—This work considers the critical problem of spatiotemporal water quality prediction. Accurate water quality prediction can effectively prevent environmental pollution. Traditional water quality prediction only focuses on time series features without considering spatial features. In this work, a novel spatiotemporal prediction approach is proposed that combines spatial-temporal graph fusion construction networks for water quality time series prediction in a real-time manner. This work shows that this approach can achieve longer forecasting sequences and more accurate results than traditional forecasting methods. As a practical consequence of this research, spatiotemporal graph fusion convolution networks for water quality prediction can effectively integrate multi-dimensional data and improve the accuracy of long-term water quality prediction. This approach can also be applied to other fields, including intelligent transportation, smart manufacturing, finance, the Internet of Things, and urban computing.

Due to their graphical representation and capability for property analysis, Petri nets (PNs) have been widely used in developing automated manufacturing systems (AMS). When designing them, it is important to perform their state reachability analysis and verify their functionality. The most common approach is to traverse a reachability tree of their PN models. However, it has the problem of state space explosion. A way to determine a state’s reachability is to find a firing sequence (FS) that corresponds to nonnegative integer solutions (NISs) of a state equation. Our prior work has given an algorithm to decide the existence of FS corresponding to an NIS in polynomial time. Yet it is impossible to decide the reachability of a marking given an infinite number of NISs of a state equation. This work studies the relationship between the PN model and NIS count of any state equation. An innovative method is proposed to modify a PN structure such that, for any given initial state and destination one, its state equation has no more than one NIS. Given an initial PN model of an AMS, by analyzing the relationship between initial PN properties and the modified one’s, we have that the proposed method can maintain the functionality of the modeled AMS. As a result, by using the modified PN as the AMS’s final model, the reachability of any marking can be determined in polynomial time, which can be viewed as a breakthrough result in the field of PN analysis. The proposed method is illustrated via case studies.

Detecting critical nodes in sparse graphs is important in a variety of application domains, such as network vulnerability assessment, epidemic control, and drug design. The critical node problem (CNP) aims to find a set of critical nodes from a network whose deletion maximally degrades the pairwise connectivity of the residual network. Due to its general NP-hard nature, state-of-the-art CNP solutions are based on heuristic approaches. Domain knowledge and trial-and-error are usually required when designing such approaches, thus consuming considerable effort and time. This work proposes a feature importance-aware graph attention network for node representation and combines it with dueling double deep Q-network to create an end-to-end algorithm to solve CNP for the first time. It does not need any problem-specific knowledge or labeled datasets as required by most of existing methods. Once the model is trained, it can be generalized to cope with various types of CNPs (with different sizes and topological structures) without re-training. Computational experiments on 28 real-world networks show that the proposed method is highly comparable to state-of-the-art methods. It does not require any problem-specific knowledge and, hence, can be applicable to many applications including those impossible ones by using the existing approaches. It can be combined with some local search methods to further improve its solution quality. Extensive comparison results are given to show its effectiveness in solving CNP.Note to Practitioners—This work is motivated by the problems of identifying influential nodes from a sparse graph or network. Various practical applications can be naturally modeled as critical node problems, e.g., finding the most influential stations or airports within a transportation network, identifying a specific number of people to be vaccinated in order to reduce the overall transmissibility of a virus, and reinforcing the protection over some most important nodes to make the electric network more stable. It proposes an effective end-to-end deep learning algorithm to solve the critical node problem. The proposed approach combines feature importance-aware graph attention network with dueling double deep Q-network. Extensive numerical experiments and comparisons show that our proposed algorithm is highly comparable to state-of-the-art algorithms and can help decision-makers to discover valuable knowledge or influential nodes in a real-world network.

Scheduling single-robotic-arm cluster tools subject to wafer residency time constraints has received much attention. Compared to some scheduling strategies that use all processing modules (PMs) to process wafers, it is much more challenging to schedule a more general case whose optimal scheduling strategy is not limited to the case of using all PMs. The strategy of only adjusting the robot’s waiting time may fail to produce a desired schedule. When a tool using all PMs is not schedulable, it may become schedulable if only some PMs of a type are used. Therefore, it is very important to select an appropriate number of PMs to process wafers. This work studies the cyclic scheduling problem of wafer-residency-time-constrained single-robotic-arm cluster tools by simultaneously adjusting the number of PMs and robot waiting time. We build a virtual cell that includes an appropriate number of PMs to process wafers with the maximal productivity. We establish sufficient and necessary conditions under which the system is schedulable. The schedulability conditions are less conservative than the state-of-the-art one. A polynomial algorithm is developed to find the optimal cyclic schedule, a virtual cell’s configuration, and robot waiting time. We illustrate the practicability of the proposed algorithm via several examples, and its superiority over the existing one.Note to Practitioners—This paper addresses the optimal cyclic scheduling problem of wafer-residency-time-constrained single-robotic-arm cluster tools that are used in every wafer fabrication factory. This work for the first time simultaneously adjusts the number of PMs and the robot waiting time given such a cluster tool such that it can be scheduled with the highest productivity. It presents the optimal scheduling method with polynomial complexity. We confirm that the proposed approach can improve schedulability of single-robotic-arm cluster tools over existing methods by using multiple examples. It can thus be readily applied to industrial wafer fabrication.

Slab yards serve as temporary slab storage between a continuous casting stage and a rolling stage. Considering non-crossing and safe clearance constraints of slab yard cranes, this work studies a multi-crane assignment and scheduling problem in the slab yard. An mixed-integer linear programming (MILP) is formulated to minimize the slab completion time. Due to its NP-hardness, the problem for large-sized instances is computationally intractable. Thus, we develop a logic-based benders decomposition algorithm (LBBD) to solve it. First, we exploit a generalized decomposition of this problem into a relaxed main problem (RMP) and a sub-problem (SP). Solving the former allocates slabs to each crane. Then, the sequence of the assigned slabs can be found by solving its corresponding sub-problem. Finally, to verify the effectiveness of LBBD, we identify a lower bound (LB) of the optimal objective function. The problem instances on real data from an iron and steel plant are created. The result of LBBD is close to such lower bound and can be found efficiently. Note to Practitioners—This work deals with a crane assignment problem with multiple cranes for handling input slabs in a slab yard. This problem is formulated as an MILP model to minimize the completion time. Its time complexity grows exponentially with the problem size. Thus, we develop a LBBD to solve it. The numerical results reveal that LBBD can find the optimal or near-optimal solution for all realistic instances in affordable computational time. Its use can ensure the high utilization of cranes and efficient service in iron and steel plants.

As a computer-integrated manufacturing system, cluster tools are widely used for semiconductor manufacturing. To tackle their scheduling problems with parallel processing chambers, existing studies assume that the parallel processing chambers at each step are identical. However, due to machine aging, some differences could appear among them in reality. This work represents the first one to report the scheduling problems of cluster tools with non-identical parallel processing chambers. By focusing on time-constrained single and dual-armed cluster tools with non-identical parallel processing chambers, it derives the sufficient conditions under which a system is schedulable. Based on them, it develops two algorithms to obtain the feasible and optimal schedules. It also develops two linear programming models to find the feasible and optimal ones if the sufficient conditions are violated. The experimental results based on 5400 random instances show that the proposed scheduling method can be readily used to generate schedules for both time-constrained single and dual-armed cluster tools.

Rational allocation of resources can improve the profit margin of a steel enterprise. This paper deals with a multi-product multi-stage multi-period resource allocation problem. In it, product manufacturing involves multiple continuous production stages, each of which has parallel machines. According to process requirements, the tasks assigned to a machine need to be produced in batches. The process route of a product is a sequential combination of machines each of which is to be selected from a stage. The process route for each product and the batching rules of each machine are known in advance. Multi-period production means that the tasks released before a planning period can be processed in any of its periods. The demand for each product type in each period and the capacity of each machine are predetermined. Considering a customer’s demand, we optimally allocate machines for products in each planning period to achieve their efficient utilization. The objective is to minimize the sum of various costs related to transportation, resources, unmet demand, and product inventory. A mixed integer linear program is developed for the concerned problem. A fix-and-optimize heuristic with variable neighborhood size is newly designed to obtain high-quality solutions. Its solutions are compared with those of CPLEX (a commercial software) given a fixed solution time. Experimental results show that it can accurately solve small-scale instances and find better solutions than CPLEX for most large-scale instances. Comparison experiments are conducted and the results show that the proposed algorithm has excellent accuracy, speed, and stability in addressing the concerned problem. Note to Practitioners—As demand for steel products gradually shows a trend towards multiple varieties, small batches, and personalized customization, it increases the difficulty for practitioners to rationally allocate resources for their production in a steel enterprise. It is hard to achieve rational material and machine resource allocation subject to complex constraints for processing multiple products in multiple production stages and periods. To deal with a multi-product multi-stage multi-period resource allocation problem, it is essential to design efficient and stable algorithms. A fix-and-optimize heuristic with variable neighborhood size is thus proposed for addressing it. The method can decompose the problem into a series of subproblems according to a decomposition scheme. They are iteratively solved. In this work, our goal is to help practitioners to deal with the challenging resource allocation problem in a short time. The effectiveness of the proposed algorithm is validated and tested by comparing its results with those of a commercially available exact solver called CPLEX on various problem instances. Extensive experimental results demonstrate its effectiveness. It can quickly solve small-scale instances with no statistically significant difference from the optimal solutions obtained by CPLEX. When addressing large-scale instances, the proposed algorithm shows better solution performance than CPLEX in a given running time. The algorithm is flexible, accurate, and fast, which implies its great application potential for resource allocation in steel enterprises.

Path finding in multi-agent systems aims to identify collision-free and cost-optimized paths for all agents with distinct start and goal positions. It poses challenging optimization problems. Existing research typically treats all agents equally, overlooking their differences in practical scenarios where they undertake the tasks of varying importance. In many application scenarios, agents must be differentiated into critical (c-agents) and acritical ones (a-agents) due to premium/general service, no-loaded/full-loaded states, and urgent/non-urgent tasks. Facing this practical need, this work focuses on multi-agent systems in which the different importance of agents must be considered; and tackles a lexicographic dual-objective variant of path-finding problem. The consideration of c-agents makes the concerned problem more useful yet more challenging than basic multi-agent path-finding problems. Different from existing multi-agent path planning methods that minimize the sum-of-costs of all agents, we optimize two objectives with preferences to emphasize the influence of c-agents on the system. The primary one is to minimize the sum-of-costs of c-agents and the secondary one is to minimize that of a-agents. As existing methods are inadequate for this unique challenge, we adapt a conflict-based search framework and design new two-level lexicographic dual-objective optimization methods to deal with it. A high level is responsible for iteratively expanding a search tree and adding constraints to resolve conflicts among agents. A low level is responsible for finding the path of each agent for the node newly expanded in the high level. By conducting numerous computational experiments, we verify the great performance of the presented methods in solving the concerned problem. We further develop a prototype system incorporating our methods and make it public to promote their practical application. This research contributes valuable insights and solutions to pathfinding challenges in multi-agent systems with critical and acritical agents. Note to Practitioners—This work addresses a multi-agent path finding problem in multi-agent systems involving both critical and acritical agents. The former are more important than the latter since they are assigned to perform more important tasks. This is a common scenario in manufacturing and service environments. We propose a two-level lexicographic dual-objective optimization framework to deal with the problem and underscore the importance of c-agents in the path finding problem. Three solution approaches are designed for practitioners to select based on their practical application needs. The computational experimental results and statistical analyses highlight the exceptional performance of our proposed approaches. In order to facilitate the practical application, we further provide a prototype system with our proposed approaches embedded, which is openly accessible for practitioners to realize their specific applications.

A production workshop with mobile robots can be considered as a hybrid system consisting of a production system and a transportation one. Mobile robots are responsible for transferring production tasks among the machines of a production system and constitute a multi-robot transport system. It is highly coupled with a production system because of the interdependency that exists between production scheduling and mobile robot assignment. In this work, we study their integrated optimization problem for a mobile robot-based job shop with blocking properties. Its aim is to minimize total completion time as an objective function to improve overall operational efficiency. We consider the speed of a robot that varies according to whether it is loaded or not. We formulate this new problem into a mixed integer linear program to provide an algebraic description. Then, we propose a constraint programming method to solve it with high efficiency. The superiority of constraint programming over mixed integer linear programming in terms of the number of variables and constraints is analyzed. Numerous experiments on benchmark examples show that constraint programming can well handle the concerned problem. Under a one-hour time limit, it can exactly solve its instances while mixed integer linear programming cannot. Under a one-minute time limit, it obtains much better solutions than mixed integer linear programming and heuristic strategies, thus implying its high potential to be put into industrial applications. Note to Practitioners—The integration of mobile robots into production workshops has emerged as a pivotal strategy to enhance the operational efficiency of an advanced manufacturing system. This integration transforms the traditional job shop into a hybrid system, where mobile robots play a crucial role in transferring production tasks among machines. This brings a unique challenge to practitioners due to the intricate interdependencies between production scheduling and mobile robot assignment. The focus of our study is on the optimization of a mobile robot-based job shop to improve its overall operational efficiency. Our approach involves formulating this complex problem as a mixed-integer linear program, thereby providing a concise mathematical representation. We propose a constraint programming method to solve the problem efficiently. Through numerous experiments on benchmark examples, our findings indicate that the proposed constraint programming method can well solve the concerned problem given long or short solution time. This underscores its high potential for practical implementation in industrial scenarios.

Cleaning operations are required to clear particles and residual chemical gases in processing chambers of cluster tools in semiconductor manufacturing. To clean the chambers, condition-based chamber cleaning operations considering the actual state of chambers have been developed in practice. For single-armed cluster tools with such operations and wafer residency time constraints, the scheduling problem has been tackled. However, for dual-armed ones, this problem remains open and it is difficult to solve since their scheduling strategies are different from and more complex than single clusters’. To address this challenging problem, this work presents an efficient scheduling strategy to find the schedules that meet wafer residency time constraints and maximize their productivity. Two examples are given to show the effectiveness of the developed strategies. Note to Practitioners—Dual-armed cluster tools are extensively employed in semiconductor manufacturing processes. For residency-time-constrained dual-armed cluster tools with condition-based chamber cleaning operations, this work derives efficient scheduling strategies and algorithms for different cleaning cases. They can be used to realize feasible schedules that meet the wafer residency time constraints and restore a cluster tool to its steady-state as soon as possible. Hence, this work is significant and promising in helping semiconductor manufacturers enhance their fab’s throughput.

Cross-technology communication (CTC) enables direct communications among devices with heterogeneous wireless technologies, e.g., Bluetooth, WiFi, and ZigBee, thereby reducing the cost and complexity of their interconnections. Yet CTC is unreliable due to the technology heterogeneity, and most existing CTC designs do not provide acknowledgment (ACK) feedback to ensure reliable data transmission. Few ACK designs are only applicable to feedback for ZigBee-WiFi pair and vulnerable to sampling offsets that inherently exist in CTC. In this work, we propose a General and Offset-resistant Physical-layer ACK approach, called GOP-ACK, to support reliable communications. Its core idea lies in encoding ACK messages with offset-resistant signal that has two benefits: 1) it can be adapted to a wide range of CTC scenarios with minimal adjustment, and 2) it can be effortlessly and robustly detected even in the presence of sampling offsets. We offer practical guidelines to tackle key deployment challenges related to signal construction, efficient and robust transmission, and effective firmware module reuse, enabling the application of GOP-ACK to specific CTC scenarios. Based on them, we implement two designs: ZigBee-to-BLE and ZigBee-to-WiFi feedback, and propose a theoretical model to analyze their performance. We then conduct experiments and simulations to verify GOP-ACK’s feasibility and superiority over the state of the art, thereby enhancing the practicality of CTC greatly.

In the realm of computational social systems, the ability to edit facial attributes accurately plays a crucial role in enhancing user experience on social media platforms and virtual environments. However, we face significant challenges in isolated attribute manipulation and balancing the tradeoff between editing fidelity and facial identity preservation. Here, this article presents a novel approach to constructing an orthogonal decomposition subspace, enabling precise editing control over individual attributes with minimal impact on others and maintaining identity consistency. We introduce an adaptive weight modulation (AWM) method and a maximum slope truncation (MST) formula. The AWM method, founded on a sufficient convergent criterion, performs singular value decomposition to yield subspace parameters that preserve rich facial knowledge within the generative model, facilitating high-quality facial generation with reduced parameterization. This empowers meaningful semantic interpretation of attributes, supporting diverse editing tasks such as pose, age, and eyewear adjustments. The MST formula rigorously defines the editing bounds to effectively navigate the tradeoff between editing depth and identity retention. We also propose a guideline for deciphering the specific meanings of unsupervised semantics, potentially advancing interpretability in social behavioral studies. An accompanying web application, available athttps://github.com/mickoluan/GreenLimeSia, has been developed, granting users the freedom to perform tailored facial edits. Extensive experimental results show we pave the way for more personalized and authentic interactions within computational social platforms.

Real-time accurate visual object tracking (VOT) for quadrupedal robots is a great challenge when the scale or aspect ratio of moving objects vary. To overcome this challenge, existing methods apply anchor-based schemes that search a handcrafted space to locate moving objects. However, their performances are limited given complicated environments, especially when the speed of quadrupedal robots is relatively high. In this work, a newly designed VOT algorithm for a quadrupedal robot based on a Siamese network is introduced. First, a one-stage detector for locating moving objects is designed and applied. Then, position information of moving objects is fed into a newly designed Siamese adaptive network to estimate their scale and aspect ratio. For regressing bounding boxes of a target object, a box adaptive head with an asymmetric convolution (ACM) layer is newly proposed. The proposed approach is successfully used on a quadrupedal robot, which can accurately track a specific moving object in real-world complicated scenes.

Differential evolution (DE) is a widely recognized method to solve complex optimization problems as shown by many researchers. Yet, non-adaptive versions of DE suffer from insufficient exploration ability and uses no historical information for its performance enhancement. This work proposes Fractional Order Differential Evolution (FODE) to enhance DE performance from two aspects. Firstly, a bi-strategy co-deployment framework is proposed. The population-based and parameter-based strategies are combined to leverage their respective advantages. Secondly, the fractional order calculus is first applied to the differential vector to enhance DE’s exploration ability by using the historical information of populations, and ensures the diversity of population in an evolutionary process. We use the 2017 IEEE Congress on Evolutionary Computation (CEC) test functions, and CEC2011 real-world problems to evaluate FODE’s performance. Its sensitivity to parameter changes is discussed and an ablation study of multi-strategies is systematically performed. Furthermore, the variations of exploration and exploitation in FODE are visualized and analyzed. Experimental results show that FODE is superior to other state-of-the-art DE variants, the winners of CEC competitions, other fractional order calculus-based algorithms, and some powerful variants of classic algorithms.

Multi-source location is a significant application in the field of robot swarm and is required to find all sources whose number and distribution are unknown in advance. With few parameters and fast search, Particle Swarm Optimizer (PSO) variants that have certain grouping capability have been applied to address multi-source location problems by dividing a swarm such that every source has robots to locate. However, they are difficult to predetermine the exact number of groups, require a big number of robots, and are easily trapped in the no-signal areas when the proportion of no-signal areas is high. This work proposes a Virtual-source and Virtual-swarm-based PSO (VVPSO) to divide a search area into multiple cells equally, each of which has a virtual source in its center. Then, instead of robots grouping, only one group of robots is employed to traverse all virtual sources, and search their corresponding cells to locate real sources by a new PSO called Real-virtual mapping PSO (RMPSO). RMPSO asymmetrically maps a robot into a particle swarm with multiple virtual particles to perform PSO, which greatly reduces the requirements for the number of robots. Experimental results show that VVPSO has great search scalability and can solve large-scale multi-source location problems than two state-of-the-art grouping methods and three representative multimodal PSO variants, even with only one robot. Hence, this work greatly advances the field of multi-source location by using mobile robot swarm.

Ecological environment and natural resource issues are becoming more and more prominent, which promotes the recycling of waste products for green economy. Disassembly plays a key role in the remanufacturing and reuse of waste products. However, with the rapid development of production automation, designers tend to ignore the fact that manual operation is more flexible. It is of great importance to consider human factors in a disassembly process. This work considers two human disassembly postures, namely standing and sitting. The multiobjective disassembly line balancing problem considering human posture changes is studied. A mathematical model with the objective functions of maximizing profit, minimizing the number of posture changes at a workstation, and minimizing the difference of maximum posture changes between any two workstations is established. The model is solved through a newly proposed Pareto-based discrete harmony search algorithm. Three neighborhood structures are designed to enlarge the search space for better solutions. Furthermore, an elite reserve strategy is used to improve the global optimization ability of the proposed algorithm. Finally, the proposed model and algorithm are applied to cases of different scales of complexities, and the effectiveness of the proposed model and algorithm is verified in comparison with four competitive algorithms.

The automatic detection of defects in industrial photovoltaic film is crucial for ensuring the reliability of photovoltaic modules. Key challenges include limited defect samples, interclass feature similarities, and interference from complex backgrounds. Existing deep learning-based methods require large-scale datasets and focus solely on visual data, which limits their effectiveness in few-shot defect detection (FSDD). To address these challenges, we propose Contextual Ensemble Language-Image multimodal Network (CELIN), which enhances FSDD in photovoltaic films by incorporating textual information through prompt tuning. Unlike traditional language-image models that rely on single fixed text prompts, CELIN employs a position-aware context ensemble strategy to integrate position-specific prompt vectors, enabling the model to capture global contextual information and reduce background interference. In addition, a cross class mask method is introduced to differentiate between similar defect categories by blocking interclass interactions during attention computation, thereby minimizing misclassification. Extensive experiments on our own few-shot photovoltaic film defect dataset and various public benchmarks demonstrates that CELIN significantly outperforms existing methods.

Weather image translation technologies aim to convert sunny images into various weather scenes, addressing the challenge of the costly acquisition of highly-demanded diverse weather samples. However, existing weather translation methods based on generative adversarial networks (GANs) have limited generalization capability, resulting in translated images that lack authenticity and diversity. In contrast, the emerging image generation technologies based on diffusion models have greatly surpassed GAN-based ones in performance, thus becoming the dominant paradigm in various visual tasks. This work pioneers the application of diffusion models to weather translation and presents a novel Instruction-driven Multi-Weather Translation (InstructWT) method. InstructWT is built on the large image editing model, InstructPix2Pix, and leverages the latter’s zero-shot generalization capacities. We develop a user-friendly translation instruction set through prompt engineering and introduce a weather intensity factor for precise control of weather effects, thereby well enhancing the authenticity and diversity of weather images translated. A weather correlation-based blended editing technique is employed to maintain the layout and structure of the original image content. Additionally, a physical rendering approach of rain and snow is incorporated to further improve the translations’ realism. The results of comparative experiments on a public dataset, Cityscapes, demonstrate that InstructWT outperforms existing methods in terms of authenticity and fidelity. Specifically, InstructWT achieves Contrastive Language-Image Pre-Training (CLIP) image embedding cosine similarity and directional CLIP similarity scores of 0.8302 and 0.1598, respectively. Furthermore, several semantic segmentation algorithms fine-tuned using the multi-weather scene dataset augmented by InsturctWT show significant improvement in the segmentation effect on all complex weather scenarios.

Deep graph convolutional networks (GCNs) have shown promising performance in traffic prediction tasks, but their practical deployment on resource-constrained devices faces challenges. First, few models consider the potential influence of historical and future auxiliary information, such as weather and holidays, on complex traffic patterns. Second, the computational complexity of dynamic graph convolution operations grows quadratically with the number of traffic nodes, limiting model scalability. To address these challenges, this study proposes a deep encoder-decoder model named AIMSAN, which comprises an auxiliary information-aware module (AIM) and a sparse cross-attention-based graph convolutional network (SAN). From historical or future perspectives, AIM prunes multi-attribute auxiliary data into diverse time frames, and embeds them into one tensor. SAN employs a cross-attention mechanism to merge traffic data with historical embedded data in each encoder layer, forming dynamic adjacency matrices. Subsequently, it applies diffusion GCN to capture rich spatial-temporal dynamics from the traffic data. Additionally, AIMSAN utilizes the spatial sparsity of traffic nodes as a mask to mitigate the quadratic computational complexity of SAN, thereby improving overall computational efficiency. In the decoder layer, future embedded data are fused with feed-forward traffic data to generate prediction results. Experimental evaluations on three public traffic datasets demonstrate that AIMSAN achieves competitive performance compared to state-of-the-art algorithms, while reducing GPU memory consumption by 41.24%, training time by 62.09%, and validation time by 65.17% on average.

Aviation accidents are frequently related to pilots’ operations, especially during a landing phase. Therefore, accurately evaluating a pilot’s performance during this phase is crucial for minimizing landing risks. Traditional assessment methods, however, primarily focus on discrete monitoring points, failing to capture the continuous and dynamic nature of a pilot’s performance throughout the entire landing phase. To address this issue, we propose a Flight Process Importance (FPI) assessment framework that precisely determines accurate landing timing and captures the diverse operational characteristics of pilots. It consists of two components: Time-varying Importance Coefficient (TIC) and Pilot Characteristics Matrix (PCM). TIC develops a Spatio-Temporally Consistent Attention Network (STCAN) to classify Quick Access Recorder (QAR) data for anomalous event detection. It then determines the importance of different periods during the landing process by analyzing the STCAN model’s response to the data in an interpretable manner. PCM generates a parameter matrix for each flight by deriving the ideal intervals of various parameters through the interquartile range. This matrix is used to identify the duration and intensity of anomalies in operations across different pilots. By integrating TIC and PCM, our framework computes an evaluation matrix for each flight, quantifying the operational risk factors associated with pilots. Experimental results indicate that STCAN significantly surpasses other algorithms on QAR data. FPI provides a more precise and comprehensive assessment of a pilot’s performance. In particular, our findings highlight that the 10 seconds before landing to the touchdown are the most critical period of airplane landing.

In narrow environments with dynamic dense obstacles, it is difficult to find feasible paths for autonomous ground vehicles (AGVs) by using existing methods due to the strict security constraints on AGV movements. To overcome such difficulty, this work proposes an improved local path tracking algorithm based on model predictive control and a dynamic control barrier function with slack variable (MPC-DS). In this algorithm, slack variable is integrated with the control barrier function to convert the strict constraints to soft ones. To regulate the values of slack variable, a suitable penalty coefficient selected by using a series of comparative simulations is incorporated into MPC’s cost function. To test effectiveness of the proposed method, it is compared with three mainstream methods in environments with dense and sparse dynamic obstacles. Results of simulations and physical experiments show that AGV controlled by the proposed algorithm can avoid obstacles safely and efficiently in complex environments. It is worth noting that its use reduces energy consumption by 21.1% in comparison with the existing ones. To facilitate further research, our code is publicly available at https://github.com/BUCT-RobotLab417/MPC-DS

In the digital era, effective Transaction Fraud Detection (TFD) is essential to ensuring financial security. The considerable class imbalance, with legitimate transactions vastly outnumbering fraudulent ones, presents a significant challenge for TFD models to accurately identify fraudulent patterns. While existing sample-balancing strategies address class imbalance effectively in many contexts, they often fall short in TFD due to fraudsters’ sophisticated concealment tactics, which lead to pronounced behavioral overlap between fraudulent and legitimate transactions. In this paper, we introduce a novel Generative Adversarial Network-based Hybrid Sampling method (GANHS) to effectively address the class imbalance issue. GANHS employs a dual-discriminator generative adversarial network to generate synthetic samples that accurately reflect the characteristics of fraudulent activity, while an adaptive neighborhood-based undersampling technique refines these samples to minimize overlap with legitimate ones. This hybrid approach not only enhances the model’s ability to learn fraud patterns by generating high-quality samples but also improves its resilience against highly concealed fraudulent activities. Experiments on real-world and public datasets demonstrate that GANHS outperforms its competitive peers, with gains of 0.5%–8.7% in average$F_{1}$-Score and 1.0%–7.0% in G-mean, highlighting its strong potential for improving the reliability and effectiveness of TFD systems in complex, high-risk financial scenarios.

Accuracy and diversity represent two critical quantifiable performance metrics in the generation of natural and semantically accurate captions. While efforts are made to enhance one of them, the other suffers due to the inherent conflicting and complex relationship between them. In this study, we demonstrate that the suboptimal accuracy levels derived from human annotations are unsuitable for machine-generated captions. To boost diversity while maintaining high accuracy, we propose an innovative variational transformer (VaT) framework. By integrating "invisible information prior (IIP)" and "auto-selectable Gaussian mixture model (AGMM)," we enable its encoder to learn precise linguistic information and object relationships in various scenes, thus ensuring high accuracy. By incorporating the "range-median reward (RMR)" baseline into it, we preserve a wider range of candidates with higher rewards during the reinforcement-learning-based training process, thereby guaranteeing outstanding diversity. Experimental results indicate that our method achieves simultaneous improvements in accuracy and diversity by up to 1.1% and 4.8%, respectively, over the state-of-the-art. Furthermore, our approach demonstrates its performance that is the closest to human annotations in semantic retrieval, with its score of 50.3 versus the human score of 50.6. Thus, the method can be readily put into industrial use.

Ultrasound detection is a potent tool for the clinical diagnosis of various diseases due to its real-time, convenient, and noninvasive qualities. Yet, existing ultrasound beamforming and related methods face a big challenge to improve both the quality and speed of imaging for the required clinical applications. The most notable characteristic of ultrasound signal data is its spatial and temporal features. Because most signals are complex-valued, directly processing them by using real-valued networks leads to phase distortion and inaccurate output. In this study, for the first time, we propose a complex-valued convolutional gated recurrent (CCGR) neural network to handle ultrasound analytic signals with the aforementioned properties. The complex-valued network operations proposed in this study improve the beamforming accuracy of complex-valued ultrasound signals over traditional real-valued methods. Further, the proposed deep integration of convolution and recurrent neural networks makes a great contribution to extracting rich and informative ultrasound signal features. Our experimental results reveal its outstanding imaging quality over existing state-of-the-art methods. More significantly, its ultrafast processing speed of only 0.07 s per image promises considerable clinical application potential. The code is available at https://github.com/zhangzm0128/CCGR.

Social relationships offer crucial supplementary information for recommendations by leveraging users' social connections to gain insights into their preferences. However, prevalent social recommendation methods often grapple with the issues of sparsity and noise, which curtail their effectiveness. In addition, these methods overlook the intricacies of user interactions within social networks, which could provide invaluable information. Addressing their deficiencies, this article introduces a novel sociological-theory-based multitopic self-supervised recommendation method (SMSR). This method integrates user attitude information into the construction of social relationships and utilizes dynamic routing to identify and categorize topics, thereby mitigating the impact of social noise on recommendation accuracy. Furthermore, we reveal sophisticated higher order user relations within these topics by using motifs. By combining the light graph convolutional network with balance theory, SMSR efficiently aggregates information from diverse social relations to gain its outstanding performance. Moreover, we have devised and integrated four self-supervised signals, inspired by social theory and derived from heterogeneous graph analysis, to more effectively exploit the rich structural and semantic information inherent in social relationship graphs. Empirical results from extensive experiments on publicly available datasets underscore SMSR's superiority over the state of the art.

This paper proposes a hybrid long short-term motor (HLSM) optimization and control approach for a walking exoskeleton. It consists of long-term global optimization, short-term local optimization, human-in-the-loop trajectory adaptation, and hybrid cerebellar model articulation controller (HCMAC). In the long-term global optimization, a graphic Spiking Neural Network (SNN) is utilized for an optimal global path. Along the path, the short-term motor optimization includes footstep optimization and obtains a sequence of footsteps. While in response to the unexpected obstacles along the footstep sequence, a human-in-the-loop planning strategy is designed by a virtual impedance model between the Centers of Mass (COMs) of the human and the exoskeleton, regulating the COM of the exoskeleton and generating footstep adaptation of the exoskeleton such that the exoskeleton can avoid obstacles and maintain its original global trajectory. Moreover, considering the unmodeled dynamics, we propose an HCMAC based on an integral Lyapunov function, which is exploited to counteract the system's nonlinear uncertainties, external disturbances, and reduces a relatively high computational cost. We validate the effectiveness of the HLSM planner and controller in a practical indoor setting. The results demonstrate the effectiveness of HLSM planning and control in a real scenario for a walking exoskeleton.

Heterogeneous computing systems are extensively utilized to execute a wide range of time-critical services, which encompass numerous interdependent tasks organized in the form of workflows. In practice, the dynamic arrival of workflows often interleaves with their execution, leading to resource contention among multiple workflows and potentially causing QoS (Quality of Service) degradation. However, compared to the extensive research on single workflow scheduling, interleaved workflow scheduling has received relatively less attention. Moreover, the challenge of effectively scheduling limited computing resources to promptly complete consecutively arriving workflows remains underexplored, despite its practical importance. To fill this gap, this work proposes a method called Urgency-based List Scheduling (ULS) for dynamically scheduling deadline-constrained interleaved workflows. In ULS, a novel task property called urgency is introduced to prioritize tasks from multiple workflows by capturing real-time execution information, and each newly arrived workflow is scheduled with the outstanding tasks of prior workflows based on a list-based strategy to make more informed decisions. Extensive evaluation experiments are performed and the findings illustrate that ULS can achieve a reduction of at least 68% in deadline miss rates and 77% in overall tardiness compared to existing methods.

Large-scale many-objective optimization problems (LSMaOPs) pose great difficulties for traditional evolutionary algorithms due to their slow search for Pareto-optimal solutions in huge decision space and struggle to balance diversity and convergence among numerous locally optimal solutions. An objective space linear inverse mapping method has successfully achieved great saving in execution time in solving LSMaOPs. Linear mapping is a fast and straightforward way, but fails to characterize a complex functional relationship. If we can enhance the expressive capacity of a mapping model, and further obtain a more general function approximator, can the evolutionary search based on objective space mapping be more efficient? To answer this interesting question, this work proposes to employ nonlinear activation functions widely used in neural networks so as to enhance the efficiency of objective space inverse mapping, thus efficiently generating excellent offspring population. A new evolutionary optimization framework based on decision variable analysis is proposed to solve LSMaOPs. In order to demonstrate its performance, this work carries out empirical experiments involving massive decision variables and many objectives. Experimental results prove its superiority over some representative and updated ones.

This work studies the problem of block validation in a blockchain network where a block manager acting as a task publisher sends a task (block validation) to all the workers (miners) within the network. The latter carries out the block validation and finally returns the final results to the former. The goal of this work is to maximize the block manager’s profit by jointly optimizing the latency of the block verification process and the reward offered by the manager to miners. Note that the latency and reward are closely coupled. Therefore, in this case, if it is solved directly, they are offered separately, and their dependency is not well considered, leading to overall poor performance. This work formalizes it as an optimization problem considering both delay and reward, and proposes an evolutionary approach, namely, the reborn dandelion algorithm (RDA), to solve it. Specifically, in the proposed algorithm, each individual contains both delays and rewards for different types of miners. A reborn strategy is designed to reborn an individual to replace the worst one in the current population, with the aim of enhancing its exploration ability. A greedy selection strategy is proposed to enhance its exploitation ability. The experimental results on CEC2013 functions and blockchain network instances indicate that our proposed approach is significantly superior to other evolutionary-based ones.

During the past two decades or so, many researchers devoted themselves to avoiding the deadlocks of automated manufacturing systems (AMSs). Many deadlock control policies have been developed under the assumption that AMSs do not contain unreliable resources and assembly operations. This work focuses on the deadlock control of failure-prone AMS with assembly operations and a single unreliable resource. At first, an automata model is developed to characterize the studied AMS. Then, four properties that a robust deadlock control policy for the studied AMS should satisfy are proposed and rigorously formulated for the first time. Finally, a new robust deadlock control policy based on the modified Banker’s algorithms is proposed. Compared with existing control policies, only the proposed one can well satisfy the newly proposed properties, thus representing a significant contribution to the field of robust control of failure-prone manufacturing systems.

Petri nets (PNs) are graphical and mathematical tools used to model a variety of discrete event systems and analyze their properties. Reachability is their fundamental property that is undecidable for PNs in general and has exponential complexity with respect to net size. One of the possible computation techniques of reachable states (markings) is based on the determination of legal firing sequences (LFSs) that correspond to a non-negative integer solutions (NISs) of a state equation. Our previous work has proposed an algorithm to decide the existence of LFSs corresponding to an NIS in polynomial time. However, it is impossible to check all NISs to decide a marking’s reachability in the case of an infinite number of NISs in a state equation. Hence, this work studies the relationship between the structure of a PN and the number of NISs of its state equation. Furthermore, an innovative method is proposed to modify its structure such that its state equation has no more than one NIS. The relationship between the properties of a PN and its modified one is studied in this article. We conclude that we can modify its structure while maintaining the function of the modeled system such that the reachability of the model can be decided in polynomial time. This can be viewed as a breakthrough result in the area of PN theory and applications. The correctness and time efficiency of the algorithm are verified by case studies and experiments. This work is important in putting PNs into the industrial use.

In the periodic scheduling problems of a dual-arm wafer-residency-time-constrained (W-R-T-C) cluster tool (CT) with multivariety wafers, it is important to balance workload at each step. This work proposes a method of processing module (PM) two-layer configuration. First, one PM configuration is made for each wafer type, i.e., the suitable number of PMs is selected for processing wafers, ensuring that the natural workload is balanced at each step for a type of wafers. Then, the other PM configuration is made, i.e., adopt a virtual module method for balancing the natural workload at bottleneck processing steps of different types of wafers. Based on two-layer PM configuration, this work derives necessary and sufficient conditions for a CT’s schedulability, which are less restrictive than the current state-of-the-art ones. It proposes a polynomial-time algorithm for computing its minimum periodic schedule when it is schedulable. Several examples are given to show our algorithm’s superiority over existing ones.

While a deadlock control problem in complex resource allocation systems (RASs) has been extensively studied in the literature, the corresponding results that are applicable to assembly systems are quite limited, both, in terms of structural analysis of deadlocks and deadlock resolution. Taking Petri nets (PNs) as a modeling and analysis tool, this article focuses on the deadlock control problem for a forward-conflict free net (FCFN), which allows for batch assembly and multiple resource allocations. First, a new structural characterization of deadlocks in FCFN is proposed through two structural objects: 1) circuit and 2)$\omega $-structure. The starting point for this is motivated by the fact that deadlocks in assembly systems stem not only from the circular wait of resources but also the parts waiting for their assembly with other parts. Subsequently, based on these two objects, a necessary and sufficient condition about FCFNs liveness is obtained: 1) an FCFN is live if and only if no circuit and 2)$\omega $-structure are saturated at any reachable marking. Finally, in order to prevent each such object from inducing deadlocks, a hierarchical search algorithm appropriate for real-time implementation is developed to avoid its saturation. The proposed algorithm is proven to be capable of ensuring the deadlock-free operation of FCFNs. Moreover, several examples are provided to demonstrate its effectiveness.

In the era of booming e-commerce and Internet of Things technology, robotic mobile fulfillment systems (RMFSs) have gained more and more use in logistics industry. While these systems enhance labor efficiency, they introduce numerous optimization challenges. Order picking is a human–robot collaborative process in RMFS. It involves three critical and interrelated operational optimization issues: 1) PS; 2) resource scheduling; and 3) manual picking. Each of them is an NP-hard combinatorial optimization problem. Their integration is a significant challenge for operational optimization in RMFS with multiple picking stations and represents a novel problem that was not studied before to our best knowledge. To tackle this complex problem and fill the research gap, we first model it as a MIP to derive exact solutions for small-scale and illustrative cases. For industrial-scale cases that cannot be solved by exact methods given limited time, we propose a tailored learning-strategies-enhanced local search algorithm. It integrates a 3-D bi-section encoding strategy, a three-stage decoding policy, a learning-based ANS method, and two learning-based tabu mechanisms. Experimental results demonstrate the effectiveness of our proposed method, achieving 2.9% to 33.2% performance improvement over six competitive peers. This highlights its superiority in solving the concerned problem, providing significant potential for addressing practical order picking optimization challenges in RMFS.

In semiconductor manufacturing, chamber cleaning operations (CCOs) are performed from time to time to clear particles and chemical gases that remain in the processing chambers of cluster tools (CTs) to ensure wafer quality. For single-armed CTs with just one CCO that could be completed in one system cycle time, researchers have proposed several scheduling approaches. In practice, more than one CCO may be required concurrently, thus failing the existing ones under such cases. Furthermore, the duration of a cleaning operation may be larger than one system cycle time, which highly complicates the scheduling of such tools. This work addresses this challenge by proposing several scheduling approaches and algorithms for single-armed CTs with several CCOs, each of which exceeds the duration of one system cycle. Suggestions are given for engineers in semiconductor fabrication plants to take the most suitable scheduling approach and algorithm for enhancing the productivity of CTs given their specific scenarios. Finally, two illustrative examples are provided to showcase the power of the proposed concepts and techniques.

With the rapid development of the Internet, various network invasive behaviors are increasing rapidly. This seriously threatens the economic development of individuals, enterprises, and society. Network intrusion detection is important in network security systems, which can be regarded as a classification problem. It aims to distinguish between the specific categories of various network behaviors and determine whether the behavior belongs to network intrusion. However, network intrusions present a diverse and fast-changing trend, making categorizing difficult. Due to feature redundancy, uneven distribution of sample numbers, and inefficient parameter optimization, traditional rule-based approaches fail to achieve satisfying classification accuracy. This work proposes a multi-classification intrusion detection model based on Stacked Sparse Shrink AutoEncoder (SSSAE), Genetic Simulated annealing-based particle swarm optimization optimized Tabnet classifier (GS-Tabnet), and Decision Fusion (DF), called for SGTD short. Among them, SSSAE extracts multiple feature sets from the input data. Then GS-Tabnet trains a classifier for each feature set. Finally, the decision fusion fuses the results from these classifiers to obtain the final classification result. SGTD is compared with eight multi-classification benchmark models, and its intrusion detection accuracy is superior to its peers.

In the fields of robotic perception and computer vision, achieving accurate semantic segmentation of low-light or nighttime scenes is challenging. This is primarily due to the limited visibility of objects and the reduced texture and color contrasts among them. To address the issue of limited visibility, we propose a hierarchical gated convolution unit, which simultaneously expands the receptive field and restores edge texture. To address the issue of reduced texture among objects, we propose a dual closed-loop bipartite matching algorithm to establish a total loss function consisting of the unsupervised illumination enhancement loss and supervised intersection-over-union loss, thus enabling the joint minimization of both losses via the Hungarian algorithm. We thus achieve end-to-end training for a semantic segmentation network especially suitable for handling low-light scenes. Experimental results demonstrate that the proposed network surpasses existing methods on the Cityscapes dataset and notably outperforms state-of-the-art methods on both Dark Zurich and Nighttime Driving datasets.

Although remarkable advances have been achieved in generic object detection, small object detection (SOD) remains challenging owing to small objects’ information loss and noisy representation caused by their non-uniform distribution. Their limited width and height, scale variations, and redundant computation make SOD hard. To overcome them, this work proposes a new SOD method based on sparse convolutional network (SCNet) and Query Mechanism called QuerySOD. First, an extended feature pyramid network is constructed for extracting feature maps of small objects with more regional details. Then, a Sparse Head is neatly designed by using SCNet for accelerating the interfering speed and obtaining weights of each layer. After that, a Query Mechanism is innovatively introduced for harvesting the benefit of sparse value feature maps from the Sparse Head. QuerySOD is evaluated on public benchmarks including COCO and VisDrone. Finally, we apply it on ‘Jinghai’ unmanned survey vehicles and receive excellent SOD performance from this real-world application.

Topological semantic maps provide a practical solution to enhance indoor navigation for the Partially Sighted or Visually Impaired (PSVI). Segmenting indoor floor plans and extracting boundaries are key to constructing these maps. The existing methods exhibit low accuracy in segmentation. To achieve desired high segmentation accuracy, we introduce a Context-Enhanced Full-Resolution Network (CEFRN) for floor plan segmentation. It is designed to harness the shallow detailed features and inter-category contextual dependencies inherent in floor plans. CEFRN integrates modified residual blocks to capture the low-stage full-resolution features while maintaining its compactness. A position attention module is employed to refine the deep-stage contextual information. We also propose a two-dimensional deep supervision method to merge features from both stages, which significantly boosts the feature representation ability of CEFRN. Finally, a practical topological semantic mapping method for PSVI indoor navigation is introduced. Experimental results demonstrate that CEFRN’s segmentation accuracy well exceeds the state-of-the-art methods’. It can be used to well support accurate topological semantic mapping.

Image captioning is a multi-modal task that enables the transformation from scene images to natural language, providing valuable insights for visually impaired individuals to understand their environment. Therefore, its application to wearable navigation devices for visually impaired individuals holds immense potential. However, in practical applications, confusion between scene visuals and semantics, coupled with model complexity, often leads to performance degradation, resulting in inaccurate environmental interpretation. In light of this, we introduce a Lightweight De-confounding Transformer Network (LDTNet) for image captioning equipped with a Causal Adjustment module to eliminate confounders. Moreover, we design a Suppression Gate Unit that efficiently integrates fine-grained information from shallow features, while reducing the number of network layers to have a lightweight model. Experimental results demonstrate that our approach not only addresses the visual-semantic confusion issue effectively but also improves the response speed of wearable devices in comparison with the state of the art. Twenty volunteers are recruited to evaluate LDTNet’s efficacy in real-world settings in terms of both response speed and generated outputs by wearing the resulting assistive navigation devices. The outcomes well show its outstanding performance and great potential for visualy impaired individuals to use.

Remote photoplethysmography (rPPG) is a promising technique for non-contact physiological signal measurement. It has great potential applications in human health monitoring and emotion analysis. However, existing methods for the rPPG task ignore the long-tail phenomenon of physiological signal data, especially on multi-domain joint training. In addition, we find that the long-tail problem of the physiological label (phys-label) exists in different datasets, and the long-tail problem of some domain exists under the same phys-label. To tackle these problems, we propose a hierarchical balanced framework, to mitigate the bias caused by domain and phys-label imbalance. Specifically, we propose anti-spurious domain center learning tailored to learning domain-balanced embeddings space. Then, we adopt compact-aware continuity regularization to estimate phys-label-wise imbalances and construct continuity between embeddings. Extensive experiments demonstrate that our method outperforms the state-of-the-art in cross-dataset and intra-dataset settings. Our code is available at https://github.com/pywin/HiBa.

In recent years, distributed computing has wit-nessed widespread applications across numerous organizations. Predicting workload and computing resource data can facilitate proactive service operation management, leading to substantial improvements in quality of service and cost efficiency. However, these data often exhibit non-linearity, high volatility, and inter-dependencies across different categories, presenting challenges for accurate forecasting. Consequently, there is a critical need to develop a method that thoroughly and comprehensively analyzes all available data to forecast future trends effectively. This work proposes a novel integrated data-enhanced prediction model named SVI for achieving high-accuracy workload prediction in distributed computing systems. SVI employs the Savitzky-Golay filter and variational mode decomposition for feature processing, whose features are subsequently utilized by Informer for multivariate joint analysis of the enhanced data, achieving high-precision prediction. Ablation and comparative experiments with advanced prediction models are conducted on the Google cluster trace and other typical datasets. Realistic data-driven results indicate that SVI improves the prediction accuracy by 35.4% compared to the original Informer, with each module contributing to the performance enhancement. Furthermore, compared with Autoformer, SVI enhances the prediction accuracy of workload, CPU, and memory by 62.5%, 65.6%, and 69.1 %, respectively.

Water quality prediction methods forecast the future short or long-term trends of its changes, providing proactive advice for water pollution prevention and control. Existing water quality prediction methods only consider the historical data of single-type or multi-type water quality. However, meteorology and other factors also have a significant impact on water quality indicators. Therefore, only considering the historical data of water quality is not feasible. Unlike existing studies, this work proposes a hybrid water quality prediction model called CMI to solve the above problem. Before prediction, CMI incorporates a multimodal fusion mechanism of water quality time series and remote sensing images of meteorological rainfall. Moreover, CMI integrates the model of ConvNeXt V2 and a multimodal bottleneck transformer to extract image features for fusing the time series and images. Furthermore, it utilizes an emerging model of iTransformer to realize prediction with the fused features. Experimental results with real-life water quality time series and remotely sensed rainfall images demonstrate that CMI outperforms other state-of-the-art fusion algorithms, and the water quality prediction accuracy with fused meteorological data is 13% higher on average than that with only water quality time series.

Many new IoT applications have emerged with the fast evolution of 5G and the Internet of Things (IoT). These applications place higher demands on network energy consumption and processing capabilities. Mobile edge computing (MEC) significantly enhances execution efficiency, while energy harvesting (EH) modules further augment the operational features of IoT devices. However, existing studies mainly concentrate on energy consumption and latency problems, often neglecting issues about user mobility and potential privacy leakage within the MEC environment. Therefore, optimizing computation offloading and resource allocation for MEC-enabled IoT networks is essential. This work proposes an innovative architecture with EH for collaborative computing between multiple mobile devices (MDs) and MEC servers. To tackle the problem, this work also proposes an advanced hybrid algorithm named Self-adaptive Bat Optimizer with Genetic operations and individual update of Grey wolf optimizer (SBG2). With SBG2, this work aims to minimize the energy consumption of MDs while providing user mobility and privacy protection. Simulation experiments show that SBG2 reduces energy consumption by 79.15%, 93.20%, and 89.58%, respectively, compared to the other three typical algorithms.

Mobile edge computing (MEC) emerges as a vital paradigm to support the increasing use of mobile users (MUs) with capabilities similar to cloud computing. While most research concentrates on MEC facilitated by terrestrial base stations (BSs), its applicability in scenarios such as disaster rescue and field operations is limited. Efforts have been made to explore MEC assisted by unmanned aerial vehicles (UAVs) with efficient scheduling algorithms. However, relying solely on UAVs for MEC has limitations, particularly for computation-intensive applications. This work proposes a hybrid MEC system lever-aging UAVs and BS. Multiple UAVs and a BS are deployed to provide MEC services directly from UAVs or indirectly from the BS. We formulate an energy-efficient scheduling problem to minimize energy consumption by jointly optimizing UAV trajectories, task associations, and allocation of computing and transmitting resources. To solve it, this work designs a hybrid algorithm named _S_uccess History-based parameter Adaptation for Differential volution with a Niching-based population size reduction strategy and an efficient nsemble sinusoidal scheme (SHADE-NE). Experimental results validate the superiority of SHADE-NE over its benchmark peers, thus proving that SHADE-NE greatly enhances the performance of the system.

The advent of smart manufacturing in Industry 4.0 signifies the arrival of the era of connections. As an excellent communication protocol, Object linking and embedding for Process Control Unified Architecture (OPC UA) can address most semantic heterogeneity issues. However, its semantics are not formally defined at the application layer. To address the information silo problem caused by semantic heterogeneity, a method named Querying of Ontology Mapping-based OPC UA (QOMOU) is proposed. It extracts the information models of OPC UA servers into resource description framework triples, utilizes web ontology language for semantic enrichment and inference, and employs a semantic similarity model for event ontology mapping to improve query efficiency. The method's effectiveness is validated through functional queries using the SPARQL protocol in Apache Jena. The query efficiency is 5% higher on average compared to both structured query and extensible markup languages. Moreover, by employing a keyword-matching algorithm, the query accuracy of the existing heterogeneous data integration scheme is improved by 4% on average. This enhancement can boost the operational efficiency of Internet of Things systems based on the OPC UA architecture.

Multi-objective optimization algorithms are essential for addressing real-world challenges characterized by conflicting objectives. Although conventional algorithms are effective in exploring solution spaces and generating non-dominated solutions, solution quality and dynamic adaptability of true Pareto fronts need to be improved. This work proposes a multi-objective algorithm that integrates Non-dominated sorting genetic algorithm II (NSGA-II) and Multi-Objective Reinforcement Learning (N-MORL). N-MORL consists of two parts including upstream and downstream components. In the upstream component, this work improves the Variance-stabilized Multi-objective Proximal Policy Optimization (VMPPO) for enhanced convergence stability by adjusting its iteration mechanism. Additionally, this work optimizes variance networks and action sampling to balance exploration and exploitation, which improves experience sampling efficiency. This work adopts high-quality solution sets yielded by MORL as the initial solution set for downstream NSGA-II, guiding the exploration space and increasing the solution number. High-quality initial solutions significantly accelerate the iterative convergence speed of N-MORL. N-MORL provides the quality and the number of solutions, better covering or approaching the true Pareto front. Experimental results with five benchmark multi-objective functions demonstrate that N-MORL outperforms the other three multi-objective evolutionary algorithms regarding high-quality solutions with the same iterations.

Recent years have seen a surge in Internet of Things (IoT) technologies, with billions of mobile devices (MDs) straining limited computing and networking resources. Mobile edge computing offloads tasks from MDs to edge servers, saving energy and reducing network pressure. Edge servers provide closer services yet have fewer resources than cloud servers. A new heterogeneous edge and cloud computing paradigm combines the benefits of both. Edge servers provide close proximity services to MDs, while the cloud owns enough resources. The existence of mobile IoT devices makes it more practical to consider mobility when allocating resources of edge servers to decrease the energy consumption of the heterogeneous edge and cloud while meeting the latency needs of tasks. This work formulate a constrained energy consumption optimization problem and design a hybrid algorithm named Genetic Simulated-annealing-based particle swarm optimization (PSO) to yield a near-optimal solution. Simulation results prove that compared to genetic algorithm, PSO, simulated-annealing-based PSO, and Trex, GSPSO reduces the total energy consumption by 38.64%, 54.63%, 45.94%, and 36.21%, respectively.

In a cloud-edge system, data analysis, processing, and storage can be performed in edge servers, avoiding transferring data to more distant cloud servers. This greatly improves the efficiency of data processing, saves network bandwidth and cloud resources, and reduces operating and maintenance costs. However, it is a challenge of how to perform task scheduling. It is difficult to schedule tasks for joint optimization of the total energy consumption and completion time of a task sequence within a limited time in a resource-constrained cloud-edge system. The work proposes an improved Simulated-Annealing-based Firefly Algorithm with Linear position update, called SAFAL for short. SAFAL incorporates a simulated annealing mechanism and an efficient position update strategy into the firefly algorithm, enabling fireflies to find the optimal solution more quickly and avoid getting trapped in local optima. SAFAL adopts a probabilistic mapping operator to map the position of each firefly to a task scheduling sequence, thus linking the firefly space and the task space. Several test instances in cloud-edge systems are designed to validate the superiority of SAFAL over the firefly algorithm, simulated annealing, and firefly algorithm with a self-adaptive strategy. Results show that the weighted cost of total energy consumption and completion time of SAFAL is reduced by 16.32%, 17.62%, and 14.21%, respectively, with 20 tasks.

With the increasing amount of data and the need for real-time processing, Mobile Edge Computing (MEC) is growing rapidly, driving the shift from traditional cloud computing to distributed edge architectures. When offloading these applications with large amounts of data on mobile devices, a lot of computing and storage resources and high energy consumption are required. Yet, mobile devices' computing power, resource storage, and battery power are often limited and cannot meet these needs. To solve a computation offloading problem for joint optimization of time, cost, and energy, this work proposes an improved hybrid algorithm called Chaos and Lévy flights-based Whale Optimization Algorithm (CLWOA) to solve the multi-user offloading problem in an MEC-Cloud system. Each task is offloaded to local processors of mobile devices, edge servers, and cloud servers in proportion to jointly minimize the completion time, energy consumption, and total cost. Finally, compared with the whale optimization algorithm, lévy flight whale optimization algorithm, refined whale optimization algorithm, and chaos-based whale optimization algorithm, CLWOA reduces the weighted cost by 1.89%, 0.31%, 0.19%, and 0.42%, respectively.

In many fields, time series prediction is gaining more and more attention, e.g., air pollution, geological hazards, and network traffic prediction. Water quality prediction is based on historical data to predict future water quality. However, it is difficult to learn a representation map from a time series that captures the trends and fluctuations to effectively remove noise from time series data and capture complex nonlinear relationships. To solve these problems, this work proposes a time series prediction model, called PSGT for short, which integrates Patch Savitsky-Golay filtering and Transformer. First, this work adopts a Patching method to embed sub-time series data and obtains the trends and semantic information of the time series. Second, it uses the Savitsky-Golay filtering to effectively remove the noise data in the patch and improve the prediction accuracy. Third, it uses a Transformer mechanism to address the nonlinear problem of water quality time series and improve long-term prediction capability. Two real-world datasets are utilized to evaluate the proposed PSGT, and experiments prove that PSGT performs better than other benchmark models by at least 6%.

Currently, a cloud-edge collaborative system combines almost unlimited storage and computing resources where tasks can be migrated to high-performance servers in edge servers or the cloud. However, resource allocation and task offloading present big challenges due to the competition among mobile devices (MDs) for communication and computing resources of edge servers. Therefore, it is significant to properly offload MDs' tasks to edge servers or the cloud. This work proposes a collaborative edge-cloud architecture, including a centralized cloud, edge servers, and MDs. Then, this work jointly considers computing power, task sizes, computing resources, transmission power of MDs, transmission rates, computing power, transmission power, computing resource of edge servers, and computing resource of the cloud. Considering the abovementioned factors, this work designs a mixed-integer non-linear programming problem. To solve it, a Genetic Simulated annealing-based Particle Swarm Optimization (GSPSO) algorithm is proposed to obtain the best solution. Building upon it, this work proposes an energy-minimized task offloading and resource allocation strategy, thereby minimizing the system's energy consumption while ensuring strict task response time limits. Experimental results show that GSPSO reduces the system's energy by 66.34%, 34.65%, and 4.95% more than particle swarm optimization (PSO), self-adaptive PSO, and Tyrannosaurus optimization.

The unprecedented prosperity of the industrial Internet of Things has thoroughly facilitated the transition from traditional manufacturing towards intelligent manufacturing. In industrial environments, resource-constrained industrial equipments (IEs) often fail to meet the diverse demands of numerous compute-intensive and latency-sensitive tasks. Mobile edge computing has emerged as an innovative paradigm for lower latency and energy consumption for IEs. However, computational offloading and coordinating of multiple IEs with diverse task types and multiple edge nodes in industrial environments poses challenges. To address this challenge, we propose a multi-task approach encompassing scientific and concurrent workflow tasks to achieve energy-efficient and latency-optimized computation offloading. Furthermore, this work designs an improved Quantum Multi-objective Grey wolf optimizer with Manta ray foraging and Associative learning (QMGMA) to optimize multi-task computation offloading. Comprehensive experiments demonstrate the superior efficiency and stability of QMAGA compared to state-of-the-art algorithms in balancing latency and energy consumption. QMAGA improves average inverse generation distance and average spacing by 37% and 31% on average than multi-objective grey wolf optimizer, non-dominated sorting genetic algorithm II, and multi-objective multi-verse optimization, proving the convergence and diversity of its non-dominated solutions.

Most of current anomaly detection models assume that the normal pattern remains the same all the time. However, the normal patterns of web services can change dramatically and frequently over time. The model trained on old-distribution data becomes outdated and ineffective after such changes. Retraining the whole model whenever the pattern is changed is computationally expensive. Further, at the beginning of normal pattern changes, there is not enough observation data from the new distribution. Retraining a large neural network model with limited data is vulnerable to overfitting. Thus, we propose a Light Anti-overfitting Retraining Approach (LARA) based on deep variational auto-encoders for time series anomaly detection. In LARA we make the following three major contributions: 1) the retraining process is designed as a convex problem such that overfitting is prevented and the retraining process can converge fast; 2) a novel ruminate block is introduced, which can leverage the historical data without the need to store them; 3) we mathematically and experimentally prove that when fine-tuning the latent vector and reconstructed data, the linear formations can achieve the least adjusting errors between the ground truths and the fine-tuned ones. Moreover, we have performed many experiments to verify that retraining LARA with even a limited amount of data from new distribution can achieve competitive performance in comparison with the state-of-the-art anomaly detection models trained with sufficient data. Besides, we verify its light computational overhead.

Model-based Test Case Prioritization utilizing similarity metrics has proved effective in software testing. However, the utility of similarity metrics in it varies with test scenarios, hindering its universal effectiveness and performance optimization . To tackle this problem, we propose a Diversity-driven Learn-to-rank model-based TCP approach, named DLTCP, for optimizing early fault detection performance. Our method first employs the whale optimization algorithm to search for a suitable set of similarity metrics from a pool of existing candidates. This search process determines which metrics should be used. According to each selected metric, test cases are then prioritized. The resulting test case rankings are used as the training data for DLTCP. Finally, the proposed method incorporates random forest to train a ranking model for test case prioritization. As such, it can fuse multiple similarity metrics to improve the TCP performance. We conduct extensive experiments to evaluate our method’s performance using the average percentage fault detected (APFD) as metric. The experimental results show that DLTCP achieve an average APFD value of 0.953 for seven classic benchmark models , which is 11.37% higher than that of the state-of-the-art algorithms. It can well select a set of similarity metrics for effective fusion, demonstrating competitive performance in early fault detection.

Resource usage prediction in cloud data centers is critically important. It can improve providers’ service quality and avoid resource wastage and insufficiency. However, the time series of resource usage in cloud environments is characterized by multidimensional, nonlinear, and high-volatility characteristics. Achieving high-accuracy prediction for time series with such characteristics is necessary but difficult. Traditional prediction methods based on regression algorithms and recurrent neural networks cannot effectively extract nonlinear features from data sets. Besides, many deep learning models suffer from gradient explosion or gradient vanishing during the training stage. Current commonly used prediction methods fail to uncover some vital information about the frequency domain features in the time series. To resolve these challenges, we design a Forecasting method based on the Integration of a Savitzky–Golay (SG) filter, a frequency enhanced decomposed transformer (FEDformer) model, and a frequency-enhanced channel attention mechanism (FECAM), named FISFA. It adopts the SG filter to reduce noise and smooth sequences in the raw sequences of resources. Then, we develop a hybrid transformer-based model integrating FEDformer and the FECAM, effectively capturing the frequency domain patterns. Besides, a meta-heuristic optimization algorithm, i.e., genetic simulated annealing-based particle swarm optimizer, is proposed to optimize key hyperparameters of FISFA. Then, FISFA predicts the future needs for multidimensional resources in highly fluctuating traces in real-life cloud environments. Experimental results demonstrate that FISFA achieves higher accuracy and performs more efficient prediction than several benchmark forecasting methods with realistic data sets collected from Alibaba and Google cluster traces. FISFA improves the prediction accuracy on average by 32.14%, 25.49%, and 27.71% over vanilla long short-term memory, transformer, and Informer methods, respectively.

Smart mobile devices (SMDs) are integral for running advanced applications that demand significant computing resources and quick response time, e.g., immersive gaming and advanced image editing. However, SMDs often face constraints in computational capacity and battery duration, restricting their ability to process these tasks instantaneously. Cloud computing can circumvent these limitations by computation offloading, but cloud data centers (CDCs) are often deployed at long distances from users, which results in longer computational latency. To address the latency issue, the incorporation of small base stations (SBSs) in the vicinity of the user provides services with high bandwidth and low latency. The primary challenge lies in balancing the economics of the system consisting of different SMDs, SBSs, and a CDC, i.e., minimizing cost while still meeting the latency requirements of applications. In this work, a cost-minimized computation offloading framework is formulated and solved by a two-stage optimization algorithm named Lévy flight and Simulated Annealing-based Grey wolf optimizer (LSAG). The optimal edge selection strategy is defined in the first stage for dealing with the case of several available SBSs. The second stage coordinates task scheduling and optimizes the allocation of resources among SMDs, SBSs, and CDC. LSAG integrates the extended search property of Lévy flight and the individual selection strategy of simulated annealing in the grey wolf optimizer, which reduces the risk of falling into local optima and finds the global optimum. Experimental results of executing real-life tasks show that LSAG outperforms its state-of-the-art peers in terms of cost and speed of convergence.

Precise and real-time prediction of future network attacks can not only prompt cloud infrastructures to fast respond and protect network security, but also prevents economic and business losses. In recent years, neural networks, e.g., Bi-directional Gated Recurrent Unit network and Temporal Convolutional Network (TCN), have been proven to be suitable for predicting time series data. Attention mechanisms are also widely used for the prediction of the time series of network attacks. This work proposes a hybrid deep learning prediction method that combines capabilities of Savitzky-Golay filter, TCN, Multi-head self attention, and Bi-directional Gated Recurrent Unit (STMB) for the prediction of network attacks. This work first adopts a Savitzky-Golay filter to smooth possible outliers and noise in network attack traffic data. It applies TCN to extract abstract features from one-dimensional time series to make full use of data. It then adopts multi-head self-attention to capture internal correlations among multi-dimensional features, by increasing weights of key features and reducing those weight of non-key features, making that SMTB captures important features adaptively. Finally, this work adopts Bi-directional Gated Recurrent Unit to extract bi-directional and long-term correlations in the time series to imporve the prediction accuracy. This work also utilizes a hybrid algorithm named genetic simulated-annealing-based particle swarm optimizer to determine the hyperparameter setting of STMB. Experimental results with real-life datasets show that STMB outperforms several commonly-used algorithms in terms of prediction accuracy.

In recent years, ongoing advancements in the Industrial Internet of Things (IIoT) have yielded massive volumes of data, taxing the capabilities of cloud computing infrastructure. Allocating limited computing resources to numerous incoming requests is one of the difficulties of cloud computing, which is typically referred to as a Task-Scheduling-in-Cloud-Computing (TSCC) problem. In order to ameliorate the performance of a particle swarm optimizer (PSO) and broaden its application to TSCC, this paper introduces an Opposition-based Simulated Annealing Particle Swarm Optimizer (OSAPSO) to address PSO’s premature convergence issue, particularly when tackling high-dimensional complex problems like TSCC. OSAPSO is a novel combination of opposition-based learning (OBL), evolution strategy, simulated annealing (SA), and swarm intelligence. At its initial stage, the swarm is formed at random by using OBL to guarantee swarm diversity with a light computational burden. A multi-way tournament selection approach is then utilized to pick parents to produce a new offspring swarm by using two novel evolutionary operators, namely, damping-based mutation and inversion–scrambling-based crossover. OSAPSO is given a powerful exploration capacity by adopting the survivor probabilistic selection of SA, which accepts subpar solutions with a certain probability. Finally, PSO itself kicks in, making a good trade-off between solution diversity and convergence speed of the proposed method. Due to the non-convex discontinuous nature of TSCC, OSAPSO is modified to clone it into a discrete optimization problem. Within a heterogeneous cloud computing environment, OSAPSO and eight well-regarded competitors are examined on a set of multi-scale IIoT heterogeneous task groups (realistic and synthetic). In terms of power consumption, monetary cost, service makespan, and system throughput, experimental results reveal that OSAPSO has a winning performance and is statistically more significant in handling the challenge of IIoT task scheduling on two different replicated scenarios of cloud systems.

Cloud computing is a powerful paradigm that can provide high-quality computation services to customers. Because its energy consumption has a large effect on its service price, this study investigates how to minimize the energy consumption while achieving adequate response times for requested computations. Such a problem is formulated as a nonlinear integer program. By deriving a state transition equation, this problem is transformed into an unconventional 0–1 knapsack problem, and dynamic programming is then used to solve it. In addition to this solution, we develop an energy-efficient job accommodation scheme that can manage dynamic jobs with varying frequencies throughout a day. Unlike existing studies that abruptly switch off old virtual machines and create new ones for upcoming jobs, this scheme tries to accommodate them with current virtual machines, and new virtual machines are not created unless necessary. Conversely, when the workload declines, jobs on energy-inefficient servers are moved to other servers, such that some energy-inefficient servers can be switched off to save energy. This scheme adjusts the computing power adaptively and smoothly without lowering the system’s quality of service. Experimental results demonstrate that the proposed solution outperforms a particle swarm optimizer and two other heuristics in terms of accommodating jobs and saving energy consumption.

Multirobot patrolling systems with various sensing and communications devices are deployed to guarantee maritime safety. Patrolling path planning for multiple robots can be modeled as a multiobjective optimization problem. The positions of patrolling nodes impact the length of patrolling paths and execution efficiency of robots. To compute them, a huge solution space is encountered. Besides, multiple patrolling nodes on the same line lead to the same patrolling scheme. Thus, how to promote solution (population) diversity becomes a new challenge. To tackle it, this work proposes a learning-inspired immune algorithm. It uses the historical information in the previous generations during iterations to realize a learning process. Unlike saving all the individuals themselves and training a model for them, the useful historical information is extracted by using upper confidence bound-based and actor–critic-inspired methods. Both time consumption and storage space can be dramatically saved. The experimental results indicate that the proposed algorithm can generate multiple patrolling schemes for the decision makers and outperforms the state-of-the-art.

In emerging applications of the Internet of Things, wireless sensor networks (WSNs) are often utilized to gather, track, and monitor data in remote areas with limited communication infrastructure. Since the majority of WSNs employ sensors powered by batteries, maintaining energy efficiency and conservation is crucial for ensuring their sustained operations over time. This work designs an Energy-efficient Unmanned aerial vehicle (UAV)-assisted hierarchical architecture of WSNs (EUW). EUW supports fast transmission of data collected from WSNs to a cloud server. Based on this architecture, this work first formulates a joint optimization problem for cluster head selection, time slot allocation, and UAV path planning to minimize the weighted sum of energy consumption of WSNs and that of a UAV. Then, a hybrid meta-heuristic algorithm named knowledge transfer-based particle swarm optimization (KTPSO) is designed, which utilizes previous optimization results to increase the convergence speed and find better results. Finally, numerical analysis and evaluation results are shown to demonstrate the efficiency of KTPSO and the proposed UAV-assisted architecture of hierarchical WSNs.

With the rapid development of ubiquitous networks and smart devices, artificial intelligence-based unmanned aerial vehicles (UAVs) are drawing more and more attention. The rise in popularity of deep neural networks (DNNs) has spawned a research effort to deploy various kinds of DNN models on vehicles. They have been used to accomplish complicated vehicular tasks and enable the construction of intelligent vehicular networks. Despite the promising prospects, how to train and run them on resource-limited and hardware-constrained UAVs faces huge challenges. Furthermore, the tradeoff between accuracy and latency needs to be considered while reducing the computational cost of DNN training.

Node-assisted WiFi backscatter communication (NWB) is a promising technology that allows backscatter tags to communicate over long distances and achieve high throughput by using WiFi nodes as relays and enabling concurrent transmissions. However, NWB lacks an accurate theoretical model to evaluate and optimize its network performance, which is challenging to develop due to the location and contention randomness of both WiFi nodes and backscatter tags. Existing backscatter models that only account for one type of randomness are not suitable for NWB. To address this issue, we propose a novel stochastic geometry-based model that captures Location and Contention Randomness as well as the involved dependency and interference (named LoCoR). We use the Matérn hard-core point process and Matérn cluster process to model the repulsive and clustering attributes of the locations of WiFi nodes and backscatter tags, respectively. We also introduce a unified time unit to analyze the randomness and dependency of WiFi and backscatter contentions. Our model factors in various design parameters (e.g., the density and transmission power of tags) and can be used to evaluate their impacts on system throughput. We conduct extensive simulations to validate the accuracy of our model. With our accurate model, one can easily configure the optimal design parameters to maximize system throughput.

Accurate workload and resource prediction are critical to realizing proactive, dynamic, and self-adaptive resource allocation for building cost-effective, energy-efficient, and green cloud data centers (CDCs), providing satisfactory quality services to users and high revenue to cloud providers. However, it is challenging because patterns of dramatically increasing and large-scale workload and resource usage in CDCs vary significantly with time. Current prediction methods often fail to handle implicit noise data and capture nonlinear, long and short-term, and spatial characteristics in workload and resource time series, thus leading to limited prediction accuracy. To tackle these issues, this work designs a novel prediction approach named VSBG that seamlessly and innovatively combines Variational mode decomposition, Savitzky Golay, Bi-directional long short-term memory (LSTM), and Grid LSTM to predict workload and resource usage in CDCs accurately. VSBG innovatively integrates variational mode decomposition (VMD) and a Savitzky Golay (SG) filter in a four-step manner before exploring its prediction. VSBG leverages VMD to divide non-stationary workload and resource time series into multiple mode functions. Then, VSBG designs a quadratic penalty, solves it with a Lagrangian multiplier, and adopts a logarithmic operation and the SG filter to smooth the first mode function to eliminate noise interference. Finally, VSBG, for the first time, systematically and simultaneously captures depth and temporal characteristics of fluctuating and complex time series data with two BiLSTM layers, between which a GridLSTM layer lies, thereby accurately predicting workload and resources in CDCs. Extensive experiments with different real-world datasets prove that VSBG outperforms a holistic set of state-of-the-art algorithms on prediction accuracy and convergence speed.

Current Industrial Internet supports the sharing of information on heterogeneous resources and elements in a process of industrial production. It enables intelligent production processes and supports cost-effective scheduling. However, collaborative manufacturing and scheduling planning for enterprises with multiple plants cause several major challenges because of a large number of decision variables and constraints of manufacturing abilities of plants, resources of production, etc. Existing methods cannot comprehensively optimize the cost of multiple products in different plants, and fail to consider machine-level optimization of tasks of manufacturing. We propose a comprehensive machine-level architecture for enterprises with multiple plants. Based on this architecture, we formulate a limited non-linear integer optimization problem to decrease the total cost of transportation, production, and sales. In it, several real-life complicated nonlinear constraints are jointly considered, and they include constraints of storage space, replacement times, pairing production, substitution, and order fulfillment rates. To solve this optimization problem, we design a hybrid meta-heuristic optimization algorithm named Genetic Simulated annealing-based Particle Swarm Optimizer with Auto-Encoders (GSPAE). Extensive experiments with real-life data show that GSPAE decreases the total cost by 25% than other state-of-the-art methods.

Mobile edge computing (MEC) paradigm supports cloud-like computing capabilities at the edge of the network and offers low-latency services. Proxy servers of MEC with mobility and limited computing, e.g., flying unmanned aerial vehicles (UAVs) have emerged as competitors in providing services. This work considers a task offloading problem for an UAV-assisted MEC system and designs an integrated cloud-edge network with multiple mobile users (MUs) and layered UAVs to improve MEC with a network of UAVs. In our system, edge UAVs (EUAVs) and the cloud collaborate to provide caching and computing services for MUs. We consider static and dynamic applications that support task offloading. Our proposed approach minimizes the weighted cost of latency and energy consumption by jointly optimizing caching and offloading, deployment of EUAVs, and allocation of computation resources. Simultaneously, this work also considers UAVs’ caching and computation capacities while meeting MUs’ latency and energy constraints. Thus, a constrained mixed integer nonlinear program for a layered UAV-assisted hybrid cloud-edge system is formulated. To solve it, this work designs a hybrid metaheuristic algorithm named adaptive and genetic simulated annealing (SA)-based particle swarm optimization (AGSP). Experimental results with a real-life dataset verify that the AGSP’s system energy consumption and task latency are reduced by at least 7.4% and 8.46%, respectively, compared with the state-of-the-art algorithms, thus proving that AGSP greatly enhances the energy and latency of the system.

The rapid expansion of Internet users results in an immense influx of network traffic within extensive cloud data centers. Accurate and instantaneous identification and forecasting of network traffic aid system managers in efficiently distributing resources, assessing network performance based on specific service demands and scrutinizing the health of network status. However, sources and distributions of traffic are different, which makes accurate warnings of cyberattack traffic difficult. Recently, emerging neural networks have demonstrated their efficacy in forecasting time series data of network cyberattacks. The time series has temporal and spatial features, which can be efficiently captured with Informer and convolutional neural networks (CNNs). To realize high-performance spatiotemporal detection of cyberattacks, this work for the first time designs a hybrid and spatiotemporal prediction framework, which integrates CNNs, Informer, and a Softmax classifier to realize high-classification accuracy of normal and abnormal cyberattacks. Real-life data are adopted to evaluate the proposed method, which yields significant improvement in classification accuracy over typical benchmark classification models.

Hybrid cloud-edge systems combine the advantages of cloud computing and mobile edge computing (MEC) to achieve flexible integration and fluidity of data between the cloud and the edge. To address dynamic and stochastic loads caused by mobile users (MUs) and time-varying tasks, MEC network operators need to continuously migrate installed services among edge servers, significantly increasing network maintenance costs. Existing studies often overlook the service migration cost resulting from MU mobility. Therefore, we present a joint optimization scheme focusing on minimizing the operational cost of hybrid cloud-edge systems while considering the dynamic service migration cost induced by MUs. With the rapid development of 5G/6G technologies, many MUs require connectivity to edge nodes (ENs) or cloud data centers (CDCs) for processing. Minimizing the operational cost of hybrid cloud-edge systems while considering many heterogeneous decision variables is a challenge. To solve this complex high-dimensional mixed-integer nonlinear problem, we develop a novel deep learning-based evolutionary algorithm called autoencoder-based multiswarm gray wolf optimizer based on genetic learning (AMGG). Experimental results with real data demonstrate that AMGG achieves lower system cost by 49.69% while strictly meeting task latency requirements of MUs compared with state-of-the-art algorithms.

Cooperative vehicle safety systems rely on periodic broadcasts of beacons to track positions and movements of concerned vehicles. In vehicular networking, vehicle driving environment is changing rapidly. This unique characteristic can cause dynamic network topology and heavy traffic conditions. In scenarios where traffic density is high, a large number of beacons could cause channel congestion, and the tracking performance of safety applications can thus be seriously impacted. To maintain high tracking accuracy for each node under varying traffic situations, this paper presents a distributed and coordinated channel access control strategy based on Multi-agent Model Predictive Control theory. First, we propose a multi-dimensional and hybrid Petri net model to characterize the interactions among multiple vehicles. The interaction model describes the possibility of collisions among vehicles. We then propose an application-dependent utility function that incorporates inter-vehicle collision behaviour. A model predictive control problem for beaconing rate adaption is formulated based on the function. Next, a distributed and coordinated decision-making scheme is designed. In this scheme, each node is treated as an agent. Each agent uses a model predictive control controller and coordinates with its neighboring agents to take channel access control actions. Simulation results validate that it improves channel resource utilization and tracking accuracy under dynamic driving situations.

With the development of robotics and Internet of Things, robot-assisted goods-to-person order picking systems become popular in smart warehouses. Order picking in such systems is a human-robot collaborative process, where robots carry pods to a picking station with human pickers who pick the demanded goods from them to fulfill orders. In it, pod selection, robot scheduling, and manual picking are highly coupled and together influence the efficiency of order picking. Their joint optimization is the key to enhancing operational efficiency but rarely studied in existing work. In order to fill such a research gap and meet high market demand, this work focuses on a novel human-robot collaborative order picking optimization problem. A mixed integer program is formulated to model it and provide an exact solution method for small-scale instances. To provide large-scale problems with efficient solutions in practical application scenarios, we propose an adaptive large-neighborhood-based tabu search algorithm. Specifically, an adaptive large neighborhood search method is designed and embedded into a tabu search algorithm with two tabu mechanisms. Experimental results indicate that the presented algorithm has significant advantages in solving the newly proposed problem. It substantially outperforms: 1) the independent use of adaptive large neighborhood search or tabu search, 2) Gurobi subject to an hour execution time, and 3) several competitive benchmark and newest well-performing algorithms. Its high performance implies its great potential in solving practical order picking optimization problems for Internet-of-Things-enabled robot-assisted smart warehouses.

Existing studies on knowledge distillation typically focus on teacher-centered methods, in which the teacher network is trained according to its own standards before transferring the learned knowledge to a student one. However, due to differences in network structure between the teacher and the student, the knowledge learned by the former may not be desired by the latter. Inspired by human educational wisdom, this paper proposes a Student-Centered Distillation (SCD) method that enables the teacher network to adjust its knowledge transfer according to the student network's needs. We implemented SCD based on various human educational wisdom, e.g., the teacher network identified and learned the knowledge desired by the student network on the validation set, and then transferred it to the latter through the training set. To address the problems of current deficiency knowledge, hard sample learning and knowledge forgetting faced by a student network in the learning process, we introduce and improve Proportional-Integral-Derivative (PID) algorithms from automation fields to make them effective in identifying the current knowledge required by the student network. Furthermore, we propose a curriculum learning-based fuzzy strategy and apply it to the proposed PID control algorithm, such that the student network in SCD can actively pay attention to the learning of challenging samples after with certain knowledge. The overall performance of SCD is verified in multiple tasks by comparing it with state-of-the-art ones. Experimental results show that our student-centered distillation method outperforms existing teacher-centered ones.

Additive manufacturing (AM) can build up complex parts in a layer-by-layer manner, which is a kind of novel and flexible production technology. The special manufacturing capability of AM shows great application potential in various fields. However, an open-loop control method cannot guarantee the reliability and repeatability of an AM process. Defects often occur to deteriorate product quality and lead to material and time waste, which hinders the development of AM industry. In this regard, a lot of efforts have been made to make an AM process more controllable. This work proposes an AM control framework that divides the related studies into three feedback loops, including the in-situ monitoring of process defects, fault diagnosis of 3-D printers, and closed-loop control of an AM process. These three loops constitute the inspection and control of AM from the machine level to product level. Specifically, the measurement requirements for monitoring techniques, defect detection, fault diagnosis, and closed-loop control are summarized. The challenges and future trends in realizing a more reliable and repeatable AM process are discussed. Note to Practitioners—This survey is motivated by urgent need to solve product quality problems in additive manufacturing (AM) caused by open-loop control. Three feedback loops can be established to solve them. The first one is defect detection that inspects part quality during fabrication. The second one is the fault diagnosis of a 3-D printer that monitors the health and operation conditions of its actuators. The last one is closed-loop control that improves AM process reliability and repeatability by regulating process variables in real time. These three loops are all based on the feedback signals of in-situ monitoring systems. This paper reviews the related studies and provides guidance for establishing the monitoring systems, performing defect detection and fault diagnosis, and designing closed-loop control systems, which helps realize more reliable and repeatable AM.

Product disassembly is critically important in recycling end-of-life products, reducing their negative impact on environmental pollution and minimizing resource waste. Disassembly line balancing problems have attracted much attention from researchers and industrial practitioners. Most of the existing studies, however, consider only human disassembly or robot disassembly alone. This work considers human-robot collaboration. It proposes an human-robot collaborative disassembly line balancing model considering stochastic task time, where an AND/OR graph is adopted to describe a product’s disassembly process. The objectives are to maximize the total profit and minimize energy consumption. A Pareto improved multi-objective shuffled frog leaping algorithm with a stochastic simulation strategy is proposed to solve the model. In addition, an elite strategy is introduced in global search to enhance the algorithm’s optimization capability. Through experiments on disassembling products of different sizes, the feasibility and effectiveness of this algorithm are demonstrated. Its comparison with some most popular state-of-the-art methods is performed.Note to Practitioners—This paper is motivated by the benefits of human-robot collaboration in the disassembly systems. The presented approach is suitable for disassembly lines with multiple objectives, and the weight of each objective cannot be accurately grasped. Most of the existing operation allocation methods are based on the correlation between humans and robots and the factors affecting disassembly. This paper suggests the selection of humans and robots is completely random and decided by an optimization algorithm. This paper designs an improved multi-objective shuffled frog leaping algorithm based on Pareto’s rule. Experimental results show that this algorithm can be applied to solve practical disassembly line balancing problems.

This work addresses the scheduling problem of deadlock-prone automated manufacturing systems (AMSs) modeled by a class of Petri nets called systems of simple sequential processes with resources (S$^{3}$PRs), and proposes a novel hybrid heuristic search (HHS) algorithm to minimize the makespan. First, based on place-timed Petri net models of AMSs, a timed state space (TSS) composed of timed states is defined. TSS-based breadth-first search and A$^{\ast}$algorithms are developed, through which the optimal solutions for small-scale problems can be obtained. Then, in order to effectively solve the scheduling problems of AMSs with different scales, TSS is condensed, and HHS is designed to search the condensed state space (CSS). In HHS, a duplicate detection policy is proposed for ensuring that only one transition path is reserved for each state in CSS. A pruning policy is proposed for pruning unpromising states to achieve the goal of searching a small part of CSS only, and three different cost estimation functions are developed for evaluating states. A hybrid search strategy is defined to achieve high search efficiency and find better solutions. The integration of the proposed duplicate detection policy, pruning policy, and hybrid search strategy ensures the high optimization performance and computational efficiency of HHS. Experimental results on benchmark AMS instances demonstrate the superiority of the proposed algorithm over the existing ones. The applicability of HHS in solving different industrial problems and manufacturing scheduling problems is also verified.Note to Practitioners—Automated manufacturing systems (AMSs) are computer-controlled manufacturing systems. They exhibit a high degree of resources sharing and route flexibility and can be highly adaptable to various production plans. Solving their scheduling problems is of great significance to manufacturers. When designing a scheduling algorithm for such systems, engineers face two challenges: how to deal with deadlocks caused by jobs competing for limited resources, and how to maintain high solution ability when solving large-scale problems. Existing algorithms for scheduling deadlock-prone AMSs have to rely on specific deadlock handling strategies to ensure their feasibility, and are unable to find high-quality solutions for large-scale problems. This article presents a hybrid heuristic search (HHS) algorithm for minimizing the makespan of deadlock-prone AMSs, in which a duplicate detection policy, a pruning policy, and a hybrid search strategy are specially designed. The combination of these policies ensure that HHS can find a high-quality solution in a short computation time, even if no deadlock handling strategy is used in the algorithm. Experimental tests and comparisons show that HHS significantly outperforms existing algorithms. The proposed HHS can be applied to other AMS scheduling problems, and can be used as an online or real-time scheduling method due to its high computational efficiency.

Community is a fundamental and highly desired pattern in a Large-scale Undirected Network (LUN). Community detection is a vital issue when LUN representation learning is performed. Owing to its good scalability and interpretability, a Symmetric and Non-negative Matrix Factorization model is frequently utilized to tackle this issue. It adopts a unique Latent Factor (LF) matrix for precisely representing LUN’s symmetry, which, unfortunately, leads to a reduced LF space that decreases its representation learning ability to a target LUN. Motivated by this discovery, this study proposes a Symmetry and Graph Bi-regularized Non-negative Matrix Factorization (B-NMF) method that: a) leverages multiple LF matrices when representing LUN, thereby boosting the representation learning ability; b) constructs a symmetry regularization term that implies the equality constraint among its multiple LF matrices, thereby illustrating LUN’s intrinsic symmetry; and c) incorporates graph regularization into its learning objective, thereby illustrating LUN’s local geometry. A theoretical proof is given to theoretically validate B-NMF’s convergence ability. The regularization hyperparameters are selected by validating model modularity, thereby guaranteeing B-NMF’s practicability in addressing real application issues. Extensive experimental results on ten LUNs from real applications demonstrate that the proposed B-NMF-based community detector significantly outperforms several baseline and state-of-the-art models in achieving highly-accurate community detection results. Note to Practitioners—LUNs are very-commonly seen in real applications like a social network system. Communities in LUNs are vital for various knowledge discovery-related applications. For accurately detecting them, a detector should guarantee its high representation learning ability to a target LUN. To do so, this paper presents a B-NMF model that is able to perform precise representation learning to LUNs, thereby achieving accurate community detection results. In comparison with conventional Symmetric and Non-negative Matrix Factorization-based community detectors, a B-NMF-based community detector enjoys its enlarged latent feature space, which ensures its higher representation ability to a target LUN. It depends on two regularization hyperparameters, which can be selected by performing grid-search on the target LUN via its modularity evaluation. This paper gives the empirical values of B-NMF’s regularization hyperparameters based on the parametersensitivity tests on the involved experimental datasets. The proposed B-NMF model is shown to be highly suitable for addressing community detection and clustering tasks on LUNs from real applications.

Latent factor analysis (LFA) is efficient in knowledge discovery from a high-dimensional and incomplete (HDI) matrix frequently encountered in industrial big data-related applications. A stochastic gradient descent (SGD) algorithm is commonly adopted as a learning algorithm for LFA owing to its high efficiency. However, its sequential nature makes it less scalable when processing large-scale data. Although alternating SGD decouples an LFA process to achieve parallelization, its performance relies on its hyper-parameters that are highly expensive to tune. To address this issue, this paper presents three extended alternating SGD algorithms whose hyper-parameters are made adaptive through particle swarm optimization. Correspondingly, three Parallel Adaptive LFA (PAL) models are proposed and achieve highly efficient latent factor acquisition from an HDI matrix. Experiments have been conducted on four HDI matrices collected from industrial applications, and the benchmark models are LFA models based on state-of-the-art parallel SGD algorithms including the alternative SGD, Hogwild!, distributed gradient descent, and sparse matrix factorization parallelization. The results demonstrate that compared with the benchmarks, with 32 threads, the proposed PAL models achieve much speedup gain. They achieve the highest prediction accuracy for missing data on most cases.Note to Practitioners—HDI data are commonly encountered in many industrial big data-related applications, where rich knowledge and patterns can be extracted efficiently. An SGD based-LFA model is popular in addressing HDI data due to its efficiency. Yet when dealing with large-scale HDI data, its serial nature greatly reduces its scalability. Although alternating SGD can decouple an LFA process to implement parallelization, its performance depends on its hyper-parameter whose tuning is tedious. To address this vital issue, this study proposes three extended alternating SGD algorithms whose hyper-parameters are made via through a particle swarm optimizer. Based on them, three models are realized, which are able to efficiently obtain latent factors from HDI matrices. Compared with the existing and state-of-the-art models, they enjoy their hyper-parameter-adaptive learning process, as well as highly competitive computational efficiency and representation learning ability. Hence, they provide practitioners with more scalable solutions when addressing large HDI data from industrial applications.

With the rapid changes and diversity of market demand, fabs have to produce wafers in many varieties and small batches. This brings a great challenge to the scheduling of wafer-residency-time-constrained cluster tools that concurrently process multiple types of wafers. Existing practice tends to employ all processing modules (PMs) of a required type to process wafers, and adopt a virtual module technology to balance PM workload for different types of wafers. However, since the virtual module technology may fail to well balance the natural workload of each processing step of a type of wafers, the corresponding scheduling results can be too conservative for the cluster tools to reach their optimal productivity. Can we improve their schedulability by choosing the most appropriate number of PMs to process wafers instead of all PMs? Under what conditions, may we use fewer PMs in a cluster tool to reach the same cycle time reached by using all PMs, thus saving resources and reducing energy consumption? This work presents a PM dual-layer configuration method. The first configuration is performed to decide the proper number of PMs, and the second one is done for so-called virtual PMs that are not physically existing PMs. It derives necessary and sufficient conditions for the schedulability of a cluster tool, which are more relaxed than all the existing ones. Furthermore, this work presents a polynomial-complexity algorithm to find an optimal cyclic schedule, and its superiority over existing ones is shown via several examples.Note to Practitioners—This paper addresses the cyclic scheduling problem of wafer-residency-time-constrained single-arm cluster tools handling multiple wafer types, and gives an efficient method to configure PMs for processing wafers. Through the method, we can balance not only the PM workload for different types of wafers, but also the PM workload for each step of a type of wafers. Based on it, a cyclic scheduling approach is presented, which is readily applied to industrial wafer manufacturing. By using our method, the schedulability of the tool can be greatly improved, and some energy and resource consumption can be eliminated to make wafer fabrication greener than the existing methods.

Credit card fraud detection (CCFD) is an important issue concerned by financial institutions. Existing methods generally employ aggregated or raw features as their representations to train their detection models. Yet such features tend to fall short of effectively exposing the characteristics of various frauds. In this work, we propose a spatial-temporal gated network (STGN) to automatically learn new informative transactional representations containing users’ transactional behavioral information for CCFD. A gated recurrent neural net unit is specifically constructed with a time-aware gate and location-aware gate to extract users’ spatial and temporal transactional behaviors. A spatial-temporal attention module is designed to expose the transaction motive of users in their historical transactional behaviors, which allows the proposed model to better extract the fraudulent characteristics from successive transactions with time and location information. A representation interaction module is offered to make rational decisions and learn compositive transactional representations. A real-world transaction dataset is used in experiments to verify the efficacy of the learned new representations. The results demonstrate that our proposed model outperforms the state-of-the-art ones, thus greatly advancing the field of CCFD.Note to Practitioners—The features of transaction records reflect the characteristics of users’ transactional behaviors. Therefore, effective features are critical for accurate CCFD. However, fraudsters often pretend to be legitimate users during transactions to deceive the CCFD system. As a result, fraudulent behaviors become concealed within legitimate ones, signifying that original features are inadequate for accurate CCFD. Thus, it is imperative for researchers and practitioners to extract new features that can well expose fraud characteristics. While existing methods employing some transaction aggregation strategies can spot certain fraudulent behaviors, they fail to clearly cluster all the anomalous behaviors and distinguish them from legitimate behaviors. Therefore, this work is driven by the urgent demand to extract new informative features for CCFD. Its primary focus is to unveil the aggregation of fraudulent transactional behaviors from both temporal and spatial perspectives, enabling more accurate CCFD. Specifically, this work introduces a new STGN model that automatically learns new transactional representations incorporating users’ transactional behavioral information for CCFD. By comprehensively considering the time interval and location interval of consecutive user transactions, we thoroughly reveal the temporal and spatial aggregation of fraudulent behavior, which provides valuable insights for CCFD practitioners: 1) employing features that integrate the behavioral characteristics of fraudsters instead of the original features can enhance the model’s capability to identify frauds, and 2) taking into account the time and location intervals of users’ consecutive historical transactions can better uncover the behavioral characteristics of fraudsters.

With the rapid development of service computing, the demand for service recommendation is increasing. Quality of Service (QoS) prediction has been one of the key challenges for service recommendation. Existing deep learning-based methods have been proposed for QoS prediction, but further improvement of their neural network structures is still needed to improve the prediction accuracy. This work introduces multi-stage multi-scale feature fusion with individual evaluations to a deep learning model for accurate QoS prediction. In our model, non-negative matrix factorization is used to extract three-scale (i.e., global, local, and individual) features; distance similarity is exploited to find similar users and services; a multi-stage deep neural network is designed to fuse multi-scale features, where individual evaluations are input to each stage to correct QoS features. Finally, our model is compared with often-cited prediction methods, and the experimental results show that it can more accurately predict QoS than its peers.Note to Practitioners—Accurate QoS prediction is very helpful to recommend the most suitable services to users among many similar services. Affected by the sparsity of historical data, the accuracy of existing QoS prediction methods is often limited. The multi-scale features of users and services can be used to improve prediction accuracy. This work proposes a new QoS prediction method to do so. Specifically, it first extracts global and individual features through non-negative matrix factorization and uses distance similarity to obtain local features. Then, it proposes a new deep neural network that fuses the extracted multi-scale features in each learning stage, thereby improving QoS prediction for services recommendation.

In cloud computing, private cloud tends to exhibit high controllability but lack scalability, whereas public cloud is just the opposite. The hybrid cloud formed by combining them can effectively balance controllability and scalability, and has been widely adopted in industry for executing workflows. The network connecting public and private clouds goes through Internet and its bandwidth is rather limited in comparison with that inside clouds. Hence, cross-cloud data transmission can become a bottleneck of executing workflows in such environment. Moreover, multiple data communications may contend for bandwidth resources, thereby incurring delay. To address these concerns, this work establishes a scheduling model for workflow execution across public and private clouds, where a queueing mode is innovatively employed to address potential network communication contention. A contention-cognizant list scheduling (CCLS) heuristic equipped with task duplication is devised to minimize workflow makespan. It adopts a novel task sorting attribute to schedule tasks and cross-cloud data communications by using available computation and communication resources, and employs task duplication to obviate the needs for certain data communications. Experiments are conducted with realistic workflows and diverse settings, and the results verify the superiority of CCLS over the existing ones as it can always achieve the best makespanNote to Practitioners—A crucial challenge for workflow scheduling across public and private clouds is that the cross-cloud bandwidth resource is relatively limited and multiple data communications may contend for it in practice. However, this communication contention issue has largely been neglected in existing investigations. To advance the state of the art, this paper proposes a communication contention-cognizant scheduling approach based on a queueing mode to minimize the makespan for workflow execution across public and private clouds. The proposed approach can be readily put into use and experimental results show that it performs better than traditional scheduling approaches that fail to consider communication contention.

A bottleneck-minimized colored traveling salesman problem is an important variant of colored traveling salesman problems. It is useful in handling the planning problems with partially overlapped workspace such as the scheduling transportation resources for timely delivery of goods. In this work, we propose an efficient bi-trajectory hybrid search method for it. The proposed method integrates a route-based crossover operator to generate promising offspring solutions, a multi-neighborhood simulated annealing to perform local optimization, and a stagnation-detect-escape mechanism to help the search escape from local optima. We also propose a bidirectional adjacency solution representation method to encode a solution, which extends the traditional adjacency representation method by additionally employing an array to represent its reverse counterpart. Extensive evaluations on two sets of 58 widely used benchmark instances demonstrate that the proposed method significantly outperforms state-of-the-art algorithms. Investigations on key algorithm modules are performed to confirm the novelty and effectiveness of the proposed ideas and strategies. Finally, we verify the generalization of the proposed method via its use to solve a colored traveling salesman problem with its aim to balance workload among salesmen. Note to Practitioners—This work is motivated by the problems of scheduling transportation resources for timely delivery of goods. It proposes an effective bi-trajectory hybrid search to tackle bottleneck-minimized colored traveling salesman problem. The proposed approach performs bi-trajectory hybrid evolutionary search by maintaining only two individuals during the whole search process. Extensive numerical experiments and comparisons show that our proposed algorithm significantly outperforms state-of-the-art algorithms and can help decision-makers in designing best-routing solutions.

High-dimensional and incomplete (HDI) interactions among numerous nodes are commonly encountered in a big data-related application, like user-item interactions in a recommender system. Owing to its high efficiency and flexibility, a stochastic gradient descent (SGD) algorithm can enable efficient latent feature analysis (LFA) of HDI data for its precise representation, thereby enabling efficient solutions to knowledge acquisition issues like missing data estimation. However, LFA on HDI data involves a bilinear issue, making SGD-based LFA a sequential process, i.e., the update on a feature can impact the results on the others. Intervening the sequence of SGD-based LFA on HDI data can affect the training results. Therefore, a parallel SGD algorithm to LFA should be designed with care. Existing parallel SGD-based LFA models suffer from a) low parallelization degree, and b) slow convergence, which significantly restrict their scalability. Aiming at addressing these vital issues, this paper presents an Adaptively-accelerated Parallel Stochastic Gradient Descent (AP-SGD) algorithm to LFA by: a) establishing a novel local minimum-based data splitting and scheduling scheme to reduce the scheduling cost among threads, thereby achieving high parallelization degree; and b) incorporating the adaptive momentum method into the learning scheme, thereby accelerating the convergence rate by making the learning rate and acceleration coefficient self-adaptive. The convergence of the achieved AP-SGD-based LFA model is theoretically proved. Experimental results on three HDI matrices generated by real industrial applications demonstrate that the AP-SGD-based LFA model outperforms state-of-the-art parallel SGD-based LFA models in both estimation accuracy for missing data and parallelization degree. Hence, it has the potential for efficient representation of HDI data in industrial scenes. The future study will focus on the GPU-deployed extensions of the proposed AP-SGD algorithm and an AP-SGD-based LFA model.

The innovations in technologies have led to the emergence of digital finance such as online payment, online insurance, online lending, and supply chain finance. Digital finance has greatly facilitated people’s lives, accelerated the circulation of capital in various fields, and enhanced the vitality of financial markets. However, it exposes many increasing risks and hidden dangers such as stock volatility, trading fraud, credit card fraud, and privacy leakage[1],[2],[3],[4],[5],[6],[7]. How to effectively calculate, control, manage, and utilize financial big data and make full use of artificial intelligence technology to ensure financial security is an important research question. Solving it faces many challenges. These challenges not only include the complexity of data and computation but also the effectiveness of intelligent optimization algorithms and ways to deal with human behaviors and social environments[8],[9].

Many machine learning methods have been proposed to achieve accurate transaction fraud detection, which is essential to the financial security of individuals and banks. However, most existing methods either leverage original features only or require manual feature engineering so that they show a weak ability to learn discriminative representations from transaction data. Moreover, criminals often commit fraud by imitating cardholders’; behaviors, which causes the poor performance of existing detection models. In this article, we propose an adaptive sampling and aggregation-based graph neural network (ASA-GNN) that learns discriminative representations to improve the performance of transaction fraud detection. A neighbor sampling strategy is performed to filter noisy nodes and supplement information for fraudulent nodes. Specifically, we use cosine similarity and edge weights to adaptively select neighbors with similar behavior patterns for target nodes and then find multihop neighbors for fraudulent nodes. A neighbor diversity metric is designed by calculating the entropy of neighbors to tackle the camouflage issue of fraudsters and explicitly alleviate the oversmoothing phenomena. Extensive experiments on three real financial datasets demonstrate that ASA-GNN outperforms state-of-the-art ones.

To study the asymmetric jump behaviors of the stock markets, we propose a novel autoregressive conditional jump intensity (ARJI)—generalized autoregressive conditional heteroskedasticity (GARCH) model with a Markov chain. Compared with the existing models, it considers the asymmetric effects of the positive and negative shocks on jump volatilities. It is proposed to estimate the asymmetric jump volatilities of the stock markets in mainland China and Hong Kong under different volatility regimes. Multiple time-varying copula models are used to analyze the dynamic dependences of the jump risks between the two markets. Furthermore, we construct dynamic hedging portfolios for their spot and futures markets, estimate the minimum risk hedging ratios, and measure the hedging performance. Compared with other benchmark models, the results show that the proposed one has the best fitting effect for the Chinese stock markets. The correlations between the Chinese mainland and Hong Kong markets are always positive. When constructing hedging portfolios, the proposed model is superior to other models, which means that introducing asymmetric shocks on both normal and jump volatilities into a Markovian ARJI-GARCH model can effectively improve the performance of hedging portfolios. In addition, the results of the robustness test indicates that our proposed model performs well and is robust.

Thek-vertex cut (k-VC) problem belongs to the family of the critical node detection problems, which aims to find a minimum subset of vertices whose removal decomposes a graph into at leastkconnected components. It is an important NP-hard problem with various real-world applications, e.g., vulnerability assessment, carbon emissions tracking, epidemic control, drug design, emergency response, network security, and social network analysis. In this article, we propose a fast local search (FLS) approach to solve it. It integrates a two-stage vertex exchange strategy based on neighborhood decomposition and cut vertex, and iteratively executes operations of addition and removal during the search. Extensive experiments on both intersection graphs of linear systems and coloring/DIMACS graphs are conducted to evaluate its performance. Empirical results show that it significantly outperforms the state-of-the-art (SOTA) algorithms in terms of both solution quality and computation time in most of the instances. To evaluate its generalization ability, we simply extend it to solve the weighted version of thek-VC problem. FLS also demonstrates its excellent performance.

Since minority samples are substantially less common than majority samples, many industrial applications, such as credit card fraud detection (CCFD) and defective part identification, call for imbalanced classification. The performance of a classifier tends to suffer from the noisy samples in majority or minority classes. This work proposes a new undersampling scheme, called a clustering-based noisy-sample-removed undersampling scheme (NUS) for imbalanced classification. The majority class samples are first clustered. The distance of the majority class sample from the cluster center that is furthest away is used as the radius to build a hypersphere, with each cluster’s center assumed to be a spherical center. We determine the Euclidean distance between the center of a cluster and each minority sample to find whether they are in the hypersphere or not. Afterward, we exclude noisy samples from the minority class. The noisy samples of majority classes are removed by using the same procedure. Second, we propose an NUS, which combines noisy sample removal with undersampling techniques. Finally, to prove the effectiveness of NUS, we integrate NUS with the basic classifiers random forest (RF), decision tree (DT), and logistics regression (LR). We conduct their comparison with seven undersampling, oversampling, and noisy-sample-removed methods. This work performs experiments on 13 public and three real transaction datasets related to e-commerce. The results show that NUS plays a positive role in promoting existing classifiers’ performance.

Lane detection, one of the crucial foundations of the autonomous driving of Rubber-Tired Gantries (RTGs), plays a vital role in automating manual container terminals. Deep-learning-based lane detection methods have robust and generalized global feature extraction capabilities to deal with complex scenarios well. However, the high preparation cost of large-scale labeled data has limited their application in RTG lane detection. Therefore, this paper presents a cost-effective, scalable incremental learning-based detection method. Specifically, some lane images are collected online, with reliable segmentation labels generated by an image-processing-based lane detection method. Next, a semi-supervised clustering approach is employed to construct a dynamically expanding sample pool, ensuring that samples are representative and diverse. Finally, a lane detection network model is self-trained by using all labeled and unlabeled samples. Extensive experimental results show that our proposed method outperforms existing methods and can achieve a lane detection accuracy of 94.87% and a detection success rate of 99.06%, with the potential for further performance improvement as data size increases..

This work considers an extended flexible job-shop scheduling problem from a semiconductor manufacturing environment. To find its high-quality solution in a reasonable time, a learning-based genetic algorithm (LGA) that incorporates a parallel long short-term memory network-embedded autoencoder model is proposed. In it, genetic algorithm is selected as a main optimizer. A novel autoencoder model is trained offline via end-to-end unsupervised learning without relying on labeled data. This model captures the major linkages among decision variables and generates promising solutions in an informative low-dimensional space, striking a balance between computational efficiency and solution quality. To further improve its search ability, a co-evolving framework is designed, which includes both a network-embedded subpopulation and a regular one. The former focuses on its global search while the latter ensures LGA's convergence. An information exchange method between the two subpopulations balances global and local search, improving its overall optimization ability. This work conducts various numerical experiments to compare LGA with the CPLEX optimizer, several classical heuristics, and some popular methods. Results show that LGA outperforms its peers in finding high-quality solutions in a reasonable time.

Wearable walking exoskeletons show great potentials in helping patients with neuro musculoskeletal stroke. Key to the successful applications is the design of effective walking trajectories that enable smooth walking for exoskeletons. This work proposes a walking planning method based on the divergent component of motion to obtain a stable joint angle trajectory. Since periodic and nonperiodic disturbances are ubiquitous in the repeating walking motion of an exoskeleton system, a major challenge in the walking control of wearable exoskeleton is the joint angle drift problem, that is, the joint angle motion trajectories are not necessarily periodic due to the presence of disturbance. To address this challenge, this work develops an adaptive repetitive control strategy to guarantee that the motion trajectories of joint angle are repetitive. In particular, by treating the disturbance as system uncertainties, an adaptive controller is designed to compensate for the uncertainties based on an integral-type Lyapunov function. A fully saturated learning approach is then developed to achieve asymptotic tracking of repetitive walking trajectories. Extensive experiments are carried out to demonstrate the effectiveness of the tracking performance.

Human-centered environments provide affordance for the use of two-handed mobile manipulators. Yet robots designed to function in and physically interact with such environments are not yet capable of meeting human users' requirements. This work proposes a whole body control framework of a two-handed mobile manipulator driven by series elastic actuators (SEAs) for cart pushing tasks. A whole body dynamic model for an integrated mobile platform and on-board arms is revealed, which takes into account the interaction forces with the cart. Then, the explicit force/position control of the mobile manipulator is performed. It enables the robot to interact dynamically with the environment while providing motion, i.e., the manipulators provide both output force control and motion control for pushing a cart. To cope with the highly nonlinear system dynamics and parameter variation of a SEA-driven mobile manipulator, this work proposes an adaptive robust controller based on a novel integral barrier Lyapunov function for cart pushing tasks by considering model uncertainty. The proposed controller enables the mobile manipulator to complete cart pushing tasks by regulating the position and output force of the mobile base and arms. The experimental results show the effectiveness of this approach in cart pushing tasks.

Since a noisy image has inferior characteristics, the direct use of Fuzzy C -Means (FCM) to segment it often produces poor image segmentation results. Intuitively, using its ideal value (noise-free image) benefits FCM's robustness enhancement. Therefore, the realization of accurate noise estimation in FCM is a new and important task. To date, only two noise-estimation-based FCM algorithms have been proposed for image segmentation, that is: 1) deviation-sparse FCM (DSFCM) and 2) our earlier proposed residual-driven FCM (RFCM). In this article, we make a thorough comparative study of DSFCM and RFCM. We demonstrate that an RFCM framework can realize more accurate noise estimation than DSFCM when different types of noise are involved. It is mainly thanks to its utilization of noise distribution characteristics instead of noise sparsity used in DSFCM. We show that DSFCM is a particular case of RFCM, thus signifying that they are the same when only impulse noise is involved. With a spatial information constraint, we demonstrate RFCM's superior effectiveness and efficiency over DSFCM in terms of supporting experiments with different levels of single, mixed, and unknown noise.

Coverage path planning (CPP) is essential for robotic tasks, such as environmental monitoring and terrain surveying, which require covering all surface areas of interest. As the pioneering approach to CPP, inspired by the concept of predation risk in predator-prey relations, the predator-prey CPP (PPCPP) has the benefit of adaptively covering arbitrary bent 2-D manifolds and can handle unexpected changes in an environment, such as the sudden introduction of dynamic obstacles. However, it can only work in bounded environment and cannot handle tasks in unbounded one, e.g., search and rescue tasks where the search boundary is unknown. Sometimes, robots are required to handle both bounded and unbounded environments, i.e., dual environments, such as capturing criminals in a city. Once encountering a building, the robot enters it to cover the bounded environment, then continues to cover the unbounded one when leaving the building. Therefore, the capability of swarm robots for the coverage tasks both in bounded and unbounded environments is important. In nature, herbivores live in groups to find more food and reduce the risk of predation. Especially the juvenile ones prefer to forage near the herd to protect themselves. Inspired by the foraging behavior of animals in a herd, this article proposes an online adaptive CPP approach that enables swarm robots to handle both bounded and unbounded environments without knowing the environmental information in advance, called dual-environmental herd-foraging-based CPP (DH-CPP). It's performance is evaluated in dual environments with stationary and dynamic obstacles of different shapes and quantity, and compared with three state-of-the-art approaches. Simulation results demonstrate that it is highly effective to handle dual environments.

Myocardial scar regions appear in cardiac magnetic resonance images of patients with myocardial infarction. Implantable cardioverter defibrillators (ICDs) can be used to effectively prevent arrhythmias and even death caused by myocardial infarction. Whether or not to implant an ICD and deciding the precise location of implantation are huge clinical challenges. This work proposes a noise-estimation-dominated fuzzy segmentation strategy for ICD implantation. It achieves accurate noise estimation in cardiac magnetic resonance image segmentation by weighting mixed noise distributions and adding a spatial information constraint. To be specific, a weightedl2-norm regularization term is proposed to form a universal noiseestimation-based FuzzyC-Means algorithm that can perform accurate segmentation of images subject to mixed or unknown noise. Through region growth and flood fill in order, the region and volume of myocardial scars are precisely obtained. Thus, the ICD implantation is accurately estimated. Finally, a criterion for ICD implantation estimation is reported. Experimental results on different myocardial infarction datasets show that the proposed strategy is more effective and efficient than its peers.

Human–robot hybrid disassembly cells provide a shared workspace that synergizes the strength of both humans and robots. These cells are characterized by adaptability and reconfigurability to accommodate the frequent changes stemming from diverse products, championing the mass personalization paradigm in Industry 5.0. Disassembly sequence planning assumes paramount importance within hybrid cells but proves to be a formidable challenge. Conventional methods prioritize the fulfillment of product-related constraints while neglecting the ergonomics considerations of operators. This oversight runs counter to the human-centric ethos central to Industry 5.0. This article proposes a personalized disassembly sequence planning approach for a human–robot hybrid disassembly cell. It presents a biobjective disassembly sequence planning model that concurrently addresses sequence scheduling and task allocation. Personal ergonomics are automatically assessed by analyzing joint angles within the operator's body skeleton. To yield disassembly plans that optimize both benefit and efficiency, a hybrid multiobjective ant lion optimizer is proposed featuring improved encoding/decoding mechanisms, updating strategies, and constraint satisfaction strategies. It adeptly addresses the discrete nature of disassembly sequences and the binary attributes associated with task execution and assignment. Personalized disassembly experiments are carried out to illustrate the feasibility and practicability of the proposed approach.

Batch scheduling problems are NP-hard and often coupled with logistics optimization problems in industrial manufacturing scenarios, further increasing the challenge of decision-making. This work focuses on a stainless steel hot-rolling production process, where it is essential to maintain the temperature of the products from the previous process beyond a certain threshold. Our work diverges from typical hot-rolling production processes by necessitating the utilization of alternative thermal devices, thereby intricately linking batch scheduling with logistics optimization and resulting in a novel lexicographical dual-objective optimization problem. To address this complex problem, we first introduce a mathematical model formulated as a mixed integer program, providing exact solutions of the concerned problem but taking much computation time. To provide effective and efficient solutions, we then propose an enhanced simulated annealing algorithm, which integrates destruction and construction methods inspired by iterated greedy algorithms. This algorithm is tailored to the specific characteristics of the problem, incorporating specialized encoding-decoding mechanisms, neighborhood search operators, and a Metropolis acceptance criterion. Our experimental results highlight the effectiveness of proposed approaches, demonstrating their superiority over competitive peers. Thus, this research contributes valuable insights and innovative solutions to the scheduling and optimization challenges inherent in batch manufacturing processes with temperature constraints and thermal devices.

When we recognize images with the help of Artificial Neural Networks (ANNs), we often wonder how they make decisions. A widely accepted solution is to point out local features as decisive evidence. A question then arises: Can local features in the latent space of an ANN explain the model output to some extent? In this work, we propose a modularized framework named MemeNet that can construct a reliable surrogate from a Convolutional Neural Network (CNN) without changing its perception. Inspired by the idea of time series classification, this framework recognizes images in two steps. First, local representations named memes are extracted from the activation map of a CNN model. Then an image is transformed into a series of understandable features. Experimental results show that MemeNet can achieve accuracy comparable to most models' through a set of reliable features and a simple classifier. Thus, it is a promising interface to use the internal dynamics of CNN, which represents a novel approach to constructing reliable models.

In an intelligent transportation system, accurate traffic flow prediction can provide significant help for travel planning. Even though some methods are proposed to do so, they focus on either algorithm or data level studies. This work focuses on both by proposing a Community-based dandelion algorithm-enabled Feature selection and Broad learning system (CFB). Specifically, a feature selection method is adopted to choose suitable features aiming to avoid redundant ones affecting prediction accuracy, and a neural network-based learning algorithm, namely a Broad Learning System (BLS), is used to predict traffic flow. In order to further boost its prediction performance, a Community-based Dandelion Algorithm (CDA) is proposed by considering an individual and its multiple offspring as a community and adopting a learning strategy for different communities. The proposed CDA is used to a) choose the suitable features as a feature selection method; and b) optimize the parameters and network structure of BLS. CDA’s superiority over its competitive peers is first verified on CEC2013’s benchmark functions, and then the proposed CFB is applied to handle the traffic flow prediction problems. The results indicate that it can improve the prediction accuracy by 5%-16% compared to the updated traffic flow prediction methods.

To assist the Partially Sighted and Visually Impaired (PSVI), a variety of Obstacle Avoidance (OA) methods have been developed. These methods mostly use depth cameras for distance measurement in terms of perception, and communicate to users through voice broadcasts. However, due to insufficient detection accuracy and slow system response, they are difficult to apply to narrow and multi-pedestrian areas. To overcome this difficulty, this work aims to develop a dynamic OA system using an improved instance segmentation network for high-precision detection. To improve the segmentation accuracy of accessible paths for PSVI users, it proposes a new 2D convolution unit that couples multi-scale receptive fields of deep features. This unit focuses on the global context of an input image by constructing a hierarchical residual-like structure. To improve the efficiency of exploration, this work adopts a bidirectional A* algorithm with safety distance constraints to plan optimal paths for PSVI users, thus avoiding their trial-and-error path finding. To ensure safety, it proposes a collision avoidance algorithm based on regional safety analysis, which can generate and transmit timely vibration response to users. Experimental results demonstrate that our developed system can help PSVI users to pass through those challenging areas safely and effectively.

Bus scheduling is vital for public transportation to ensure high transit service quality. In actual bus operation, buses’ travel time may change due to some uncertain factors, which makes the planned scheduling scheme fail to meet users’ actual requirements. This work proposes a Controller-based Bus Scheduling Approach (CBSA). In this approach, each departure time in a timetable is regarded as a decision point, and a controller is devised to select a bus in real-time to depart from the departure time. The controller makes a decision at each departure time to cover all departure times in a given timetable. The controller consists of a Duty Type Converter (DTC) and a Bus Selector (BS). DTC determines bus duty types to improve bus utilization, while BS selects a bus to cover the departure time. Since the controller makes decisions in a real-time manner, it can effectively handle uncertain events and factors (such as uncertain travel time). Some key parameters of the controller are optimized by a particle swarm optimizer (PSO) to improve its performance. CBSA is applied to real-world problem instances. Experimental results show that it outperforms the compared algorithms and a manual scheduling scheme. It can schedule buses in real-time to generate a high-quality scheduling solution under uncertain environments.

A transit route network design problem is a vitally important problem in the area of public transit systems. Most of studies on this problem aim to design a new transit network, which is often an infeasible option in practice since it is highly challenging to replace an existing network with a completely new one. In this paper, we propose a Multi-objective Ant Colony System-based Approach (MACSA) to adjust routes of bus lines for an existing transit network, such that transit service quality is improved while making the smallest deviation of the adjusted network from the existing one. First, all the bus lines in a network are sorted according to their performance. Then, a multi-objective ant colony system is adapted to adjust the sorted bus lines one by one. Besides traditional optimization objectives to maximize direct passenger flow and minimize line repetition coefficient, a new optimization objective (metric), termed adjustment degree, is proposed to measure the difference between adjusted bus lines and existing ones. Needleman-Wunsch algorithm is introduced to calculate the adjustment degree. A multi-pheromone updating mechanism is suggested to guide ants to search for better bus lines for each objective. MACSA is applied to benchmark problem instances and a real-world problem and compared with six approaches. Experiments show that MACSA can achieve an adjusted network with higher direct passenger flow, lower repetition coefficient and smaller adjustment degree. The adjustment degree achieved by MACSA is 1.61-53.82% smaller than that of other comparative approaches.

Road network persistent surveillance requires a swarm of unmanned ground vehicles (UGVs) to repeatedly surveil a sequence of places called viewpoints on a road network and detect randomly occurring events and problems at viewpoints. In most existing work, UGVs are responsible for detection but not handling emergencies like extinguishing fires, capturing intruders, or managing pollution. Hence, existing methods fail to perform both persistent surveillance and emergency handling. This work copes with both and proposes a cooperative dual-task path planning method for multiple UGVs. UGVs first treat all the viewpoints as candidates and compute their rewards to select their targets. Emergencies have a higher reward, allowing UGVs to solve dual tasks and prioritize emergencies. A heuristic protocol is then employed to choose appropriate UGVs for emergencies. Local path planning is finally considered when UGVs plan paths to targets to avoid deadlock and backtracking. Thus, a necessary and appropriate subswarm of UGVs with the nearest distance to emergencies are selected to handle them, while others continue surveilling viewpoints for potential emergencies, avoiding any long delayed handling of emergencies. The performance of the proposed method is evaluated in three road networks. The simulation and analysis results demonstrate its superiority over the state-of-the-art surveillance methods.

This work investigates how multiple unmanned aerial vehicles (UAVs) assist the large-scale IoT devices (its count$\geq$100) in the edge computing system in accomplishing their tasks. The UAVs serve the latter as edge servers, and fly to footholds to collect task data from the latter, execute tasks locally and return results to the latter. The goal of this work is to minimize overall energy consumption by jointly optimizing the association between each UAV and ground-based IoT devices, deployments of UAVs, and their flight trajectories. To achieve this, this work proposes a joint optimization approach (JOA). It has three parts: 1) an improved k-means method is designed to handle the association between each UAV and ground-based IoT devices, where the number of clusters is equal to that of UAVs, which means that each UAV is responsible for the IoT devices within a cluster; 2) for the deployments of UAVs, an improved fireworks algorithm (IFWA) with variable-length encoding strategy and population size update strategy is proposed to optimize the number and locations of footholds of each UAV, where each member of the population symbolizes a UAV foothold, and each firework and its offspring are considered as the deployment of UAV. Also, the population size update strategy is employed to dynamically change the number of footholds; and 3) regarding UAV flight trajectory, a pre-computed greedy algorithm based on the footholds of UAVs obtained by IFWA is proposed to minimize the total UAV distance. The proposed approach is verified on ten large-scale instances, and the results demonstrate its effectiveness in achieving minimal energy consumption when compared to other state-of-the-art methods.

Ultrasound imaging is widely used in medical diagnosis. It has the advantages of being performed in real time, cost-efficient, noninvasive, and nonionizing. The traditional delay-and-sum (DAS) beamformer has low resolution and contrast. Several adaptive beamformers (ABFs) have been proposed to improve them. Although they improve image quality, they incur high computation cost because of the dependence on data at the expense of real-time performance. Deep-learning methods have been successful in many areas. They train an ultrasound imaging model that can be used to quickly handle ultrasound signals and construct images. Real-valued radio-frequency signals are typically used to train a model, whereas complex-valued ultrasound signals with complex weights enable the fine-tuning of time delay for enhancing image quality. This work, for the first time, proposes a fully complex-valued gated recurrent neural network to train an ultrasound imaging model for improving ultrasound image quality. The model considers the time attributes of ultrasound signals and uses complete complex-number calculation. The model parameter and architecture are analyzed to select the best setup. The effectiveness of complex batch normalization is evaluated in training the model. The effect of analytic signals and complex weights is analyzed, and the results verify that analytic signals with complex weights enhance the model performance to reconstruct high-quality ultrasound images. The proposed model is finally compared with seven state-of-the-art methods. Experimental results reveal its great performance.

To construct a strong classifier ensemble, base classifiers should be accurate and diverse. However, there is no uniform standard for the definition and measurement of diversity. This work proposes a learners' interpretability diversity (LID) to measure the diversity of interpretable machine learners. It then proposes a LID-based classifier ensemble. Such an ensemble concept is novel because: 1) interpretability is used as an important basis for diversity measurement and 2) before its training, the difference between two interpretable base learners can be measured. To verify the proposed method's effectiveness, we choose a decision-tree-initialized dendritic neuron model (DDNM) as a base learner for ensemble design. We apply it to seven benchmark datasets. The results show that the DDNM ensemble combined with LID obtains superior performance in terms of accuracy and computational efficiency compared to some popular classifier ensembles. A random-forest-initialized dendritic neuron model (RDNM) combined with LID is an outstanding representative of the DDNM ensemble.

Performing highly accurate representation learning on a high-dimensional and sparse (HiDS) matrix is of great significance in a big data-related application such as a recommender system. A latent factor (LF) model is one of the most efficient approaches to the HiDS matrix representation. However, an LF model's representation learning ability relies heavily on an HiDS matrix's known data density, which is extremely low due to numerous missing data entities. To address this issue, this work proposes a prediction-sampling-based multilayer-structured LF (PMLF) model with twofold ideas: 1) constructing a loosely connected multilayered LF architecture to increase the known data density of an input HiDS matrix by generating synthetic data layer by layer and 2) constraining this synthetic data generating process through a random prediction-sampling strategy and nonlinear activations to avoid overfitting. In the experiments, PMLF is compared with six state-of-the-art LF-and deep neural network (DNN)-based models on four HiDS matrices from industrial applications. The results demonstrate that PMLF outperforms its peers in well-balancing prediction accuracy and computational efficiency.

Credit card fraud detection is a challenging task since fraudulent actions are hidden in massive legitimate behaviors. This work aims to learn a new representation for each transaction record based on the historical transactions of users in order to capture fraudulent patterns accurately and, thus, automatically detect a fraudulent transaction. We propose a novel model by improving long short-term memory with a time-aware gate that can capture the behavioral changes caused by consecutive transactions of users. A current-historical attention module is designed to build up connections between current and historical transactional behaviors, which enables the model to capture behavioral periodicity. An interaction module is designed to learn comprehensive and rational behavioral representations. To validate the effectiveness of the learned behavioral representations, experiments are conducted on a large real-world transaction dataset provided to us by a financial company in China, as well as a public dataset. Experimental results and the visualization of the learned representations illustrate that our method delivers a clear distinction between legitimate behaviors and fraudulent ones, and achieves better fraud detection performance compared with the state-of-the-art methods.

Multiagent deep reinforcement learning (DRL) makes optimal decisions dependent on system states observed by agents, but any uncertainty on the observations may mislead agents to take wrong actions. The mean-field actor-critic (MFAC) reinforcement learning is well-known in the multiagent field since it can effectively handle a scalability problem. However, it is sensitive to state perturbations that can significantly degrade the team rewards. This work proposes a Robust MFAC (RoMFAC) reinforcement learning that has two innovations: 1) a new objective function of training actors, composed of a policy gradient function that is related to the expected cumulative discount reward on sampled clean states and an action loss function that represents the difference between actions taken on clean and adversarial states and 2) a repetitive regularization of the action loss, ensuring the trained actors to obtain excellent performance. Furthermore, this work proposes a game model named a state-adversarial stochastic game (SASG). Despite the Nash equilibrium of SASG may not exist, adversarial perturbations to states in the RoMFAC are proven to be defensible based on SASG. Experimental results show that RoMFAC is robust against adversarial perturbations while maintaining its competitive performance in environments without perturbations.

Owing to real-world demands for global collaboration and increasing volumes of data to be analyzed, many data-intensive workflow applications are deployed in geographically distributed (geo-distributed) datacenters (DCs). In such an environment, inter-DC bandwidths are much slower than intra-DC ones, and how to effectively schedule inter-DC data communication without contention is crucial to a workflow's execution time. Meanwhile, the diversity of data privacy requirements in geo-distributed DCs causes an additional challenge. This paper introduces a workflow scheduling model for geo-distributed DCs where inter-DC communications are explicitly considered and data privacy is protected. A Communication-contention-Aware Privacy-preserving Scheduling (CAPS) method is proposed to solve it for the first time. CAPS distributes workflow tasks to DCs via a simulated annealing method such that privacy constraints are respected and the overall inter-DC data transmission time is minimized. It adopts a list scheduling heuristic to schedule tasks and data communications to computation and network resources, respectively. In experiments, CAPS is compared against leading-edge methods with realistic workflows and network settings. The results reveal that it can reduce workflow makespan by 7.08-87.53% in comparison with its peers, while guaranteeing data privacy and resolving all the communication contention issues, which has not been seen in the existing work.

Current cloud data centers (CDCs) provide highly scalable, flexible, and cost-effective services to meet the performance needs of emerging applications. It is critical for CDC providers to predict future incoming workloads such that they can perform accurate resource provisioning in CDCs. Prediction accuracy is important and its improvement has been pursued in much existing work. This work adopts two different real-life Google data traces, based on which such prediction is conducted. Specifically, this work first gives a novel prediction mechanism that integrates wavelet decomposition, Savitzky–Golay (SG) filter, and autoregressive integrated moving average (ARIMA) to realize workload prediction in each time interval. The time series of the workload is smoothed with an SG filter and further divided into several components with wavelet decomposition. Then, an integrated approach is developed to predict statistical trends and their detail components. Real-life trace-driven experiments are done and the results suggest that the proposed method provides higher accuracy of prediction than its existing peers.

Differential evolution (DE) has been developed as a state-of-the-art population-based stochastic optimizer for continuous nonconvex search space in recent decades. It uses difference vectors among individuals to adaptively control the balance between exploration and exploitation for achieving an excellent search performance. Specifically, one-to-one selection in DE guarantees that difference vectors vary from large for exploration to small for exploitation. However, the selection may generate inappropriate scale of difference vectors, which causes insufficient exploitation when facing complex problems with, e.g., multimodal with a huge number of local optima or plain regions. In this article, we propose a multiselection-based DE (MSDE) to address such limitation. Specifically, three different types of selection strategies are adopted to deal with the different situation during a search process. The proposed MSDE is evaluated on the 2013, 2014, and 2017 IEEE congress on evolutionary computation real parameter optimization competitions. Experimental results demonstrate that the proposed MSDE significantly outperforms the winners of those competitions. We further show that, when integrating the proposed multiselection-based strategy with the original DE or other advanced DE variants, it can also improve their search performance on most of test cases. This work makes a significant contribution to advance the state of the art in the area of population-based stochastic optimizers.

In most interactive image generation tasks, given regions of interest (ROI) by users, the generated results are expected to have adequate diversities in appearance while maintaining correct and reasonable structures in original images. Such tasks become more challenging if only limited data is available. Recently proposed generative models complete training based on only one image. They pay much attention to the monolithic feature of the sample while ignoring the actual semantic information of different objects inside the sample. As a result, for ROI-based generation tasks, they may produce inappropriate samples with excessive randomicity and without maintaining the related objects’ correct structures. To address this issue, this work introduces a morphologic-structure-aware generative adversarial network named MOGAN that produces random samples with diverse appearances and reliable structures based on only one image. For training for ROI, we propose to utilize the data coming from the original image being augmented and bring in a novel module to transform such augmented data into knowledge containing both structures and appearances, thus enhancing the model’s comprehension of the sample. To learn the rest areas other than ROI, we employ binary masks to ensure the generation isolated from ROI. Finally, we set parallel and hierarchical branches of the mentioned learning process. Compared with other single image generative adversarial network schemes, our approach focuses on internal features, including the maintenance of rational structures and variation on appearance. Experiments confirm a better capacity of our model on ROI-based image generation tasks than its competitive peers.

Liveness is among the most significant properties when Petri net (PN) models of automated systems are analyzed, which ensures systems’ deadlock-freeness. Traditionally, the liveness analysis methods based on reachability graphs (RGs) of PNs often suffer from state-space explosion problems. In this article, we propose a novel liveness-analysis method for PN based on maximally good-step graphs (MGs), namely, the reduced form of RGs, which can effectively alleviate such problems in liveness analysis. First, we introduce the concept of sound steps and establish an algorithm for assessing the soundness of an enabled step at the current marking from a practice point of view. Second, we propose a definition of maximal sound steps and construct an algorithm for calculating a maximal-sound-step set at each marking whose computational complexity grows polynomial with the number of places and transitions. Then, we introduce a definition for good steps and an algorithm for generating maximally good step graphs of PN; and discuss its computational complexity with respect to the net size and initial marking. Next, we for the first time answer how to evaluate the liveness of PN by using MGs. Experiments in diverse large-scale automated manufacturing systems demonstrate that the proposed method significantly reduces state space and time consumption in the liveness analysis of network systems.

Recently, energy conservation in manufacturing industry, particular in energy-intensive industries, receives much attention in order to meet the environmental protection and sustainable development needs. Optimal job scheduling is of great importance in reducing unnecessary energy consumption. To this end, both energy and time-related criteria need to be taken into consideration to achieve an efficient and sustainable production process. Generally, it is difficult to obtain the accurate processing time of jobs in advance due to various uncertainties in open shop scheduling problems arising from manufacturing and service systems. This work formulates a stochastic multiobjective hybrid open shop scheduling problem that consists of open shop and parallel-machine models. First, a multiobjective chance-constrained program is established to minimize total tardiness and energy consumption while meeting makespan requirements. Second, we newly develop a multiobjective framework integrating a brain storm optimizer and a simulation system to solve this problem. We combine population evolution to enhance exploration and external archive evolution to strengthen exploitation into the brain storm optimizer to seek for promising solutions. A simulation system is accordingly designed by using stochastic simulation and discrete-event simulation to assess the searched solutions. Finally, by conducting experiments and comparing the proposed method with several existing algorithms and an exact solver, our results confirm that it significantly outperforms its peers in tackling the considered problem.

An oversampling technique balances a dataset by increasing the number of minority samples. It is a common and effective method in imbalanced learning. However, most oversampling methods have randomness in generating minority samples, which would have negative impacts on the prediction performance of subsequent classifiers. This study treats the prediction made by classifiers as a black-box optimization problem. The optimization objective is to improve the classification accuracy of subsequent classifiers for minority samples. The solution of this optimization problem can be regarded as a minority sample that can be and added to the imbalanced dataset. The minority samples are iteratively generated by Bayesian optimization (BO). We determine two valuable intervals for each 1-D continuous variable feature. One is the interval with the densest minority samples. The other is that with the sparsest majority samples distributed among the minority samples. By adjusting the proportion of samples generated in the two areas, the presented algorithm can be flexibly applied to different datasets. In order to reduce the noise that may be caused by the exploration phase of BO, a sample selection procedure is carried out to eliminate the samples that are worse than those generated at the previous iteration. The samples generated in this way are based on the principle of improving the performance of the classifier, thus avoiding the negative effects of randomness. Experimental results via twenty open imbalanced datasets show that the proposed method obtains better results than existing state-of-the-art oversampling models, thus well advancing the important field of imbalanced learning.

This work addresses multiobjective dynamic surgery scheduling problems with considering uncertain setup time and processing time. When dealing with them, researchers have to consider rescheduling due to the arrivals of urgent patients. The goals are to minimize the fuzzy total medical cost, fuzzy maximum completion time, and maximize average patient satisfaction. First, we develop a mathematical model for describing the addressed problems. The uncertain time is expressed by triangular fuzzy numbers. Then, four meta-heuristics are improved, and eight variants are developed, including artificial bee colony, genetic algorithm, teaching-learning-base optimization, and imperialist competitive algorithm. For improving initial solutions’ quality, two initialization strategies are developed. Six local search strategies are proposed for fine exploitation and a$Q$-learning algorithm is used to choose the suitable strategies among them in the iterative process of the meta-heuristics. The states and actions of$Q$-learning are defined according to the characteristic of the addressed problems. Finally, the proposed algorithms are tested for 57 instances with different scales. The analysis and discussions verify that the improved artificial bee colony with$Q$-learning is the most competitive one for scheduling the dynamic surgery problems among all compared algorithms.

As a powerful reinforcement learning method, the learning automaton (LA) has been studied, analyzed, and applied to various engineering systems for decades. However, the state-of-the-art LA-based methods can select only the optimal action or optimal subset and cannot select an arbitrary target subset like selecting the best and worst actions or the ones in a given rank range. In order to solve the problem of selecting a given arbitrary subset of actions, this work proposes a novel pursuit learning scheme, called a discretized equal pursuit reward-inaction algorithm for arbitrary subset selection (${\text {DEP}_{RI}}$-AS). The proposed scheme pursues the currently estimated arbitrary action subset and makes the probabilities of selecting each action in the subset equal, so as to increase the convergence speed. The proof of its$\epsilon $-optimality property is presented. Simulation results of comparison experiments, parameter analysis, and a real-world application demonstrate its power in selecting a given subset of user-desired actions.

Multiagent deep reinforcement learning (MADRL) has been recently applied in many fields, including industry 5.0, but it is sensitive to adversarial attacks. Although adversarial attacks can be detrimental, they are crucial for testing and assisting in enhancing the robustness of models. Existing attacks on MADRL-based models are not sufficient since these attacks involve fixed perturbed agents, without taking into account cases where perturbed agents change. In this article, we present a novel adversarial attack framework. In this framework, we define critical agents that change over time, i.e., when they are perturbed a little, the whole multiagent system is perturbed greatly. Then, we identify critical agents through their worst-case joint actions. In this identifying process, we use gradient information, differential evolution, and SARSA to deal with the challenge caused by changes in the perturbed agents and to compute the worst-case joint actions. After identifying them, we use the target attack method to perturb them. We apply our method to attack the models trained by two state-of-the-art MADRL algorithms under three environments, including two industry-related ones. The experimental results demonstrate our method has a stronger perturbing ability than the existing methods.

Evolutionary algorithms are commonly used to solve many complex optimization problems in such fields as robotics, industrial automation, and complex system design. Yet, their performance is limited when dealing with high-dimensional complex problems because they often require enormous computational resources to yield desired solutions, and they may easily trap into local optima. To solve this problem, this work proposes a Self-adaptive Teaching-learning-based Optimizer with an improved Radial basis function model and a sparse Autoencoder (STORA). In STORA, a Self-adaptive Teaching-learning-based Optimizer is designed to dynamically adjust parameters for balancing exploration and exploitation during its solution process. Then, a sparse autoencoder (SAE) is adopted as a dimension reduction method to compress search space into lower-dimensional one for more efficiently guiding population to converge towards global optima. Besides, an Improved Radial Basis Function model (IRBF) is designed as a surrogate model to balance training time and prediction accuracy. It is adopted to save computational resources for improving overall performance. In addition, a dynamic population allocation strategy is adopted to well integrate SAE and IRBF in STORA. We evaluate it by comparing it with several state-of-the-art algorithms through six benchmark functions. We further test it by applying it to solve a real-world computational offloading problem.

High-dimensional expensive problems are often encountered in the design and optimization of complex robotic and automated systems and distributed computing systems, and they suffer from a time-consuming fitness evaluation process. It is extremely challenging and difficult to produce promising solutions in a high-dimensional search space. This work proposes an evolutionary optimization framework with embedded autoencoders that effectively solve optimization problems with high-dimensional search space. Autoencoders provide strong dimension reduction and feature extraction abilities that compress a high-dimensional space to an informative low-dimensional one. Search operations are performed in a low-dimensional space, thereby guiding whole population to converge to the optimal solution more efficiently. Multiple subpopulations coevolve iteratively in a distributed manner. One subpopulation is embedded by an autoencoder, and the other one is guided by a newly proposed Multi-swarm Gray-wolf-optimizer based on Genetic-learning (MGG). Thus, the proposed multi-swarm framework is named Autoencoder-based MGG (AMGG). AMGG consists of three proposed strategies that balance exploration and exploitation abilities, i.e., a dynamic subgroup number strategy for reducing the number of subpopulations, a subpopulation reorganization strategy for sharing useful information about each subpopulation, and a purposeful detection strategy for escaping from local optima and improving exploration ability. AMGG is compared with several widely used algorithms by solving benchmark problems and a real-life optimization one. The results well verify that AMGG outperforms its peers in terms of search accuracy and convergence efficiency.

In semiconductor manufacturing, multicluster tools are widely employed for many wafer fabrication processes. With the demand for high-mix integrated circuit chips and shrinkage of circuit width, a scheme in which multiple wafer types are fabricated inside multicluster tools is adopted by wafer foundries to make more profits. Multiple wafer types, multiple robots, and wafer residency time constraints make the resulting scheduling problems challenging. This work focuses on scheduling a single-arm multicluster tool to process two wafer types concurrently subject to wafer residency time constraints in which a conventional one-wafer cycle and backward strategy are not efficient. With such properties, several necessary and sufficient conditions are presented to check the feasibility of a periodic schedule. Polynomial-time-complex algorithms are proposed to examine a tool's schedulability and coordinate multiple robots to handle wafers for schedulable scenarios. The cycle time of an obtained schedule can reach the lower bound. A practical example is used to show the effectiveness of the proposed algorithm.

This work deals with the verification of a decision-making system for additive manufacturing (AM) technology adoption specified by a Petri net. An innovative verification technique of a Petri net-based system is oriented toward practical applications, and can detect errors at the early design and modelling stage. The idea is illustrated by a real-life case study of supporting decision making in AM technology adoption affecting supply chain management (SCM). Two main issues are addressed. Firstly, making optimal decisions about AM technology requires models rarely possessed within a company. Therefore, this work proposes a model supporting the decision making related to the implementation of AM technology, based on a Petri net, by utilizing its main advantages: graphical modelling, and strong mathematical support of formal verification techniques. Contrary to the most popular analysis methods (which are bounded exponentially in a general case), it is proved that the presented method is bounded by a cubic polynomial with net size. Secondly, an investment in AM technology is often financially assessed and does not affect other processes, as in our SCM case. Hence, strong connectivity within the proposed Petri net-based model is examined.

By using an autoencoder as a dimension reduction tool, an Autoencoder-embedded Teaching-Learning Based Optimization (ATLBO) has been proved to be effective in solving high-dimensional computationally expensive problems through several widely used function problems. However, the following two crucial issues have not been resolved, 1) ATLBO should be verified by solving real-life optimization problems; and 2) how autoencoder parameters and structures impact AEO's performance. In this work, ATLBO is verified by an energy consumption minimization problem (ECM) in mobile edge computing systems. To design an effective autoencoder for ATLBO, this work proposes a parameter tuning optimization strategy for autoencoders. By using the proposed Autoencoder Parameter Tuning (APT) strategy, ATLBO can enjoy higher robustness than those without it. The experimental results show that it is three to six times better than state-of-the-art methods in solving ECM. We consider the strategy-induced overhead and take the execution time as the primary criterion to evaluate them. In addition, the experimental results show that, against the conventional wisdom that higher-accuracy auto encoders bring higher system performance, lower-accuracy ones can actually assist ATLBO in locating the best solutions. This work promotes a novel application of autoencoders in optimization theory and practice.

Social networks are an essential component of the Internet of People (IoP) and play an important role in stimulating interactive communication among people. Graph convolutional networks provide methods for social network analysis with its impressive performance in semi-supervised node classification. However, the existing methods are based on the assumption of balanced data distribution and ignore the imbalanced problem of social networks. In order to extract the valuable information from imbalanced data for decision making, a novel method named minority-weighted graph neural network (mGNN) is presented in this article. It extends imbalanced classification ideas in the traditional machine learning field to graph-structured data to improve the classification performance of graph neural networks. In a node feature aggregation stage, the node membership values among nodes are calculated for minority nodes’ feature aggregation enhancement. In an oversampling stage, the cost-sensitive learning is used to improve edge prediction results of synthetic minority nodes, and further raise their importance. In addition, a Gumbel distribution is adopted as an activation function. The proposed mGNN is evaluated on six social network data sets. Experimental results show that it yields promising results for imbalanced node classification.

With the development of multiaccess edge computing (also called mobile-edge computing, MEC), more and more service-based applications are deployed to edge servers in order to ensure desired Quality of Service (QoS). In edge environment, how to reasonably deploy application services emerges as a challenging problem due to limited resources, heterogeneous servers, and different geographical locations of users. Benefiting from its reusability, a single service can be used by multiple applications. Yet only a few studies of the deployment problem in edge environment consider such property. This work considers the redundant deployment of reused services by different applications, so as to achieve high QoS. Due to the importance of cost for providers, it aims to minimize transmission cost and network latency under the constraint of deployment budget. This work first builds a redundant service deployment model under a heterogeneous edge environment and defines it as a multiobjective optimization problem under a given budget constraint. Then, service priority is calculated to determine redundancy, and the K-medoids clustering algorithm based on request frequency filtering is used to conduct edge server selection. It next proposes a genetic algorithm based on priority to obtain an optimized plan. Finally, this work conducts experiments on real-world datasets to prove the superiority of the proposed method over existing ones.

Cloud-edge hybrid systems are known to support delay-sensitive applications of contemporary industrial Internet of Things (IoT). While edge nodes (ENs) provide IoT users with real-time computing/network services in a pay-as-you-go manner, their resources incur cost. Thus, their profit maximization remains a core objective. With the rapid development of 5G network technologies, an enormous number of mobile devices (MDs) have been connected to ENs. As a result, how to maximize the profit of ENs has become increasingly more challenging since it involves massive heterogeneous decision variables about task allocation among MDs, ENs, and a cloud data center (CDC), as well as associations of MDs to proper ENs dynamically. To tackle such a challenge, this work adopts a divide-and-conquer strategy that models applications as multiple subtasks, each of which can be independently completed in MDs, ENs, and a CDC. A joint optimization problem is formulated on task offloading, task partitioning, and associations of users to ENs to maximize the profit of ENs. To solve this high-dimensional mixed-integer nonlinear program, a novel deep-learning algorithm is developed and named as a Genetic Simulated-annealing-based Particle-swarm-optimizer with Stacked Autoencoders (GSPSA). Real-life data-based experimental results demonstrate that GSPSA offers higher profit of ENs while strictly meeting latency needs of user tasks than state-of-the-art algorithms.

Blockchain-enabled Internet of Things (IoT) has been receiving growing attention. However, IoT nodes are usually resources heterogeneous and subject to malicious attacks, such as the intentional delay of block verification and the transmission of invalid blocks. As a result, a random node-selection model for block transmission may lead to a low transmission rate and serious security risk. To solve the problem, this article proposes a rate-and-trust-based node selection model for block transmission. In our model, we calculate the block transmission rate of a node by the latency and connectivity among nodes and its trust value by its historical transmission and verification behaviors. On this basis, we propose a PageRank-based optimization algorithm for node selection that makes a tradeoff between the transmission rate and the security risk. Extensive experimental results show that the proposed model can achieve better performance than the state-of-the-art methods, including Bitcoin network, Ethereum network, and BlockP2P-EP protocol.

In wireless visual sensor networks, the generation and transmission of huge amounts of image data consume much energy of sensor nodes (SNs), and their routing and processing take quite a long time. It is of great importance to shorten event reporting delay (ERD) and prolong network lifetime, which can be achieved by the appropriate deployment of edge nodes (ENs) that can not only collect but also process data. This work investigates how to jointly optimize SN deployment, EN deployment, data routing, and data offloading to minimize the number of deployed SNs, the number of deployed ENs, and ERD and maximize network lifetime. We formulate this problem as a mixed-integer nonlinear program and propose a multiobjective differential evolution algorithm to solve it. A large number of simulation results demonstrate that it can deliver a more accurate Pareto set than the nondominated sorting genetic algorithm III.

This paper considers a stochastic parallel machine scheduling problem in a just-in-time manufacturing context, in which its processing time can be described by a gamma or log-normal distribution. In order to obtain a high-performance schedule in a reasonable time, this work proposes a two-stage genetic algorithm with optimal computing budget allocation (OCBA) and improved Monte-Carlo Policy Evaluation (MCPE). In it, a genetic algorithm is selected as a main optimizer. An OCBA-based approach is developed to improve search efficiency, which is designed for two scenarios in a just-in-time manufacturing context. Different from most prior OCBA studies, this work considers that the stochastic processing time of jobs does not obey normal distribution. It extends the application area of OCBA by laying a theoretical foundation. A parameter control scheme based on MCPE is proposed, which aims to balance the global and local search in GA. To further enhance the efficiency and effectiveness of the proposed method, a two-stage framework is constructed. In the first stage, the performance is estimated roughly aiming at locating satisfactory solution regions. In the second stage, OCBA is incorporated to provide the reliable evaluation of excellent individuals. The theoretic interpretation of the proposed OCBA, and the convergence analysis results of the proposed method are presented. Various simulation results with benchmark and randomly generated cases validate that the proposed algorithm is more efficient and effective than several existing optimization algorithms. Note to Practitioners—A parallel machine scheduling problem under stochastic processing time is usually solved via meta-heuristic algorithms. However, their computational efficiency requires substantial improvement, especially for a stochastic optimization case that requires Monte Carlo sampling to estimate the actual objective function values in a precise manner. Most of them are parameter-sensitive, and choosing their proper parameters is highly challenging. For the first thorny issue, we develop an OCBA-based approach for determining the optimal numbers of simulations according to both prior knowledge and simulation results. In order to select proper control parameters of the proposed algorithm iteratively, we introduce a parameter control scheme based on MCPE. The combination of a meta-heuristic algorithm, OCBA and MCPE makes it possible to find high-quality solutions for the concerned scheduling problems in a short time. Theoretic analysis and numerical simulation results suggest that the proposed framework is valid and efficient. Hence, it can be readily applicable to practical systems, e.g., semiconductor manufacturing.

Recounts the career and contributions of Peter Luh.

In mobile interactive systems, there exist several components that can move and interact with each other. The current methods fail to give a comprehensive description of their properties and have inadequate modeling and analysis capacity for them. This paper concludes three system properties called system connectivity, interaction soundness and data validity, and focuses on their modeling and analysis based on Variable Petri Nets (VPNs), which have recently been proposed and are able to describe their dynamic interactions. A VPN- based model including component nets and interaction structure nets is constructed given a mobile interactive system. It depicts its structure and event-driven dynamics. Three properties and their related analysis methods of a VPN-based model are presented. An example is given to demonstrate the proposed concepts and methods. Note to Practitioners—Due to the mobility and frequent disconnections, the correctness of mobile interaction systems, such as mobile robot systems and mobile payment systems, are often difficult to analyze. This paper introduces three critical properties of systems, called system connectivity, interaction soundness and data validity, and presents a related modeling and analysis method, based on a kind of Petri net called VPN. For a given system, a model including component nets and interaction structure nets is constructed by using VPNs. The component net describes the internal process of each component, while the interaction structure net reflects the dynamic interaction between components. Based on this model, three properties are defined and analyzed. The case study of a practical mobile payment system shows the effectiveness of the proposed method.

In this paper, we propose and evaluate the idea that data need to be preconditioned prior to compression, such that they can better match the design philosophies of lossy compressors for HPC scientific data. In particular, we aim to identify a reduced model that can be utilized to transform the original data into a more compressible form. We begin with two PDE applications as a proof of concept, in which we demonstrate that a reduced model can indeed reside in the full model output, and can be utilized to improve compression ratios. A mathematical proof is also presented to show how the compression ratio is improved by the reduced model. We further explore more general dimension reduction techniques to extract the reduced model, including principal component analysis, singular value decomposition, and discrete wavelet transform. After preconditioning, the reduced model in conjunction with difference between the reduced model and full model is stored, which results in higher compression ratios. We evaluate the reduced models on ten scientific datasets, and the results show the effectiveness of our approaches. Given that there is no single method that consistently achieves the best performance, we further propose a selection strategy that guides users to select the best reduced model prior to data reduction.

As vehicles integrate more and more autonomous driving functionalities, it becomes more and more important to use vehicular networks to fully guarantee the safety and real-time performance of on-board computing tasks. Current studies on vehicular networks pay much attention to the performance improvement of network communication and resource allocation, while ignoring the fact that automotive on-board computing tasks play a significant role in vehicle safety and need to be elegantly handled in vehicular networks. In this paper, we propose a hybrid scheduling framework for meeting all hard real-time task deadlines while minimizing soft real-time task deadline misses. In particular, the proposed scheduler is composed of some local schedulers and a global scheduler, where the former guarantees the schedulability of all hard real-time tasks, and the latter decides the assignment of soft real-time jobs dynamically online. Depending on the remaining processing capability of a vehicular network, arrival jobs are either assigned for further processing or discarded. An approach combining the utilization-based schedulability test and demand-supply analysis is proposed to effectively assign tasks to processors offline. To meet as many soft real-time task deadlines as possible, tasks' demand and supply bounds are computed at runtime, such that online execution information can be used to compensate for the scheduling loss with the worst-case assumption made by the offline test. Experimental results demonstrate that, compared to a genetic algorithm, our proposed approach needs far less computation to assign more tasks offline. The online scheduling also saves many soft real-time jobs that were to be dropped by the offline algorithm.

As an increasingly mature business model, crowdsourcing, especially spatial crowdsourcing, has played an important role in data collection, disaster response, urban planning and other fields. However, the rapid growth of user scale and massive data collected inevitably brings serious challenges to computing and storage resources. The emergence of cloud computing provides an opportunity to handle such challenges. Its nearly unlimited resource provision capability can provide reliable services for different crowdsourcing applications. Nevertheless, considering the risks of privacy leakage and vendor lock-in using only a single cloud, as well as the additional restrictions caused by the wide geographical distribution of data and associations among workers, the use of multi-cloud seems to be a better choice. In this article, we define a problem to find an effective data placement scheme for spatial crowdsourcing in multi-cloud environment to achieve the cost-effectiveness and minimal latency. We take full account of the interval pricing strategy. Then we analyze the geographical distribution characteristics of data centers through a clustering algorithm, and propose an effective data initialization strategy. Finally, we use a genetic algorithm to further optimize the results. Through experiments on real-world data from cloud providers, the efficiency and effectiveness of our proposed method is verified. Compared with some existing algorithms, the proposed method can significantly reduce the system cost and latency, among which the cost reduction is up to 150 times and the latency reduction is up to twice.

Computation tree logic of knowledge (CTLK) can be used to specify many properties related to privacy of multiagent systems (MASs). Our previous work defined knowledge-oriented Petri nets (KPNs) to formally describe both the interacting/collaborating process of multiagents and their epistemic evolutions. Our KPN-based verification of CTLK required to obtain all reachable states, the transition relation of all states, and the equivalence relations of all states with respect to knowledge, which resulted in a serious state explosion problem and thus only fit some small-scale systems. This article adopts reduced ordered binary decision diagram (ROBDD) to deal with this problem. Especially, the ROBDD technique is used to encode and store all reachable states but not to encode and store any transition relation or equivalence relation. However, when verifying a CTLK formula, the transition relation and equivalence relation of some states must be known. To solve this problem, we design the related algorithms to compute only those required relations and prove their correctness. We design the related model checking algorithms and develop a tool. A number of experiments are done by using a famous benchmark about the privacy problem of MAS: the dining cryptographers protocol (DCP), and the results illustrate the advantages of our methods. For example, our tool running with a general PC spends less than 14 h to verify the DCP with 1200 cryptographers, which involves about 101080 states and two CTLK formulas with more than 6000 atomic propositions and more than 3600 operators. This represents a significant advance in the field of model checking.

From the aspect of behavioral finance, which is an emerging area integrating human behavior into finance, this work studies a robust portfolio problem for loss-averse investors under distribution and mean return ambiguity. A loss-aversion distributionally-robust optimization model is constructed if the return distribution of risky assets is unknown. Then, under the premise that the mean returns of risky assets belong to an ellipsoidal uncertainty set, a model under joint ambiguity in distribution and mean returns is constructed. This study solves both robust models and derives their analytical solutions, respectively. Moreover, the effect of ambiguity aversion and loss aversion on robust optimal portfolio returns is studied. The results show that ambiguity-neutral investors who do not know the return distribution obtain more robust optimal portfolio returns than ambiguity-averse investors who are unaware of both the distribution and mean return. The difference between them decreases with the increase of loss aversion coefficients and increases with ambiguity aversion coefficients. Both loss aversion and ambiguity aversion play important roles in investors’ behavioral portfolio selection.

With the popularity of credit cards worldwide, timely and accurate fraud detection has become critically important to ensure the safety of their user accounts. Existing models generally utilize original features or manually aggregated features as their transactional representations, while they fail to reveal the hidden fraudulent behaviors. In this work, we propose a novel model to extract transactional behaviors of users and learn new transactional behavioral representations for credit card fraud detection. Considering the characteristics of transactional behaviors, two time-aware gates are designed in a recurrent neural net unit to learn long- and short-term transactional habits of users, respectively, and to capture behavioral changes of users caused by different time intervals between their consecutive transactions. A time-aware-attention module is proposed and employed to extract the behavioral information from their consecutive historical transactions with time intervals, which enables the proposed model to capture behavioral motive and periodicity inside their historical transactional behaviors. An interaction module is designed to learn more comprehensive and rational representations. To prove the effectiveness of the learned transactional behavioral representations, experiments are conducted on a large real-world transaction dataset and a public one. The results show that the learned representation can well distinguish fraudulent behaviors from legitimate ones, and the proposed method can improve the performance of credit card fraud detection in terms of various evaluation criteria over the state-of-the-art methods.

Convolutional neural networks (CNNs) have demonstrated excellent capability in various visual recognition tasks but impose an excessive computational burden. The latter problem is commonly solved by utilizing lightweight sparse networks. However, such networks have a limited receptive field in a few layers, and the majority of these networks face a severe information barrage due to their sparse structures. Spurred by these deficiencies, this work proposes a Squeeze Convolution block with Information Flow Enhancement (SCIFE), comprising a Divide-and-Squeeze Convolution and an Information Flow Enhancement scheme. The former module constructs a multi-layer structure through multiple squeeze operations to increase the receptive field and reduce computation. The latter replaces the affine transformation with the point convolution and dynamically adjusts the activation function’s threshold, enhancing information flow in both channels and layers. Moreover, we reveal that the original affine transformation may harm the network’s generalization capability. To overcome this issue, we utilize a point convolution with a zero initial mean. SCIFE can serve as a plug-and-play replacement for vanilla convolution blocks in mainstream CNNs, while extensive experimental results demonstrate that CNNs equipped with SCIFE compress benchmark structures without sacrificing performance, outperforming their competitors.

Human mesh recovery from one single image has achieved rapid progress recently, but many methods suffer from the image appearance overfitting since the training data are collected along with accurate 3D annotations in controlled settings of monotonous backgrounds or simple clothes. Some methods regress human mesh vertices from poses to tackle the above problem. However the mesh topologies have not been well exploited, and artifacts are often generated. In this paper, we aim to find an efficient low-cost solution to human mesh reconstruction. To this end, we propose a Progressive Quadric Graph Convolutional Network (PQ-GCN), and design a simple and fast method for 3D human mesh recovery from a single image in the wild. Specifically, we apply quadric-based surface simplification to human meshes and design a progressive graph convolution network, accompanied by mesh feature up-sampling, to deal with the mesh topologies. We carry out a series of studies to validate our method. The results prove that our method achieves superior performance on a challenging in-the-wild dataset, while using 66% fewer parameters than the existing method, Pose2Mesh. Artifacts have also been eliminated and better visual quality has been obtained without any further post-processing and model fitting. Besides, the recovery can be stopped at an earlier stage by adding a decoder head. Consequently, the computational complexity can be reduced greatly.

Knowledge distillation is a deep learning method that mimics the way that humans teach, i.e., a teacher network is used to guide the training of a student one. Knowledge distillation can generate an efficient student network to facilitate deployment in resource-constrained edge computing devices. Existing studies have typically mined knowledge from a teacher network and transferred it to a student one. The latter can only passively receive knowledge but cannot understand how the former acquires the knowledge, thus limiting the latter’s performance improvement. Inspired by the old Chinese saying “Give a man a fish and you feed him for a day; teach a man how to fish and you feed him for a lifetime,” this work proposes a Skill-transferring Knowledge Distillation (SKD) method to boost a student network’s ability to create new valuable knowledge. SKD consists of two main meta-learning networks: Teacher Behavior Teaching and Teacher Experience Teaching. The former captures the process of a teacher network’s learning behavior in the hidden layers and can predict the teacher network’s subsequent behavior based on previous ones. The latter models the optimal empirical knowledge of a teacher network’s output layer at each learning stage. With their help, a teacher network can provide its actions to a student one in the subsequent behavior and its optimal empirical knowledge in the current stage. SKD’s performance is verified through its application to multiple object recognition tasks and comparison with the state of the art.

As a novel neural network-based learning framework, a broad learning system (BLS) has attracted much attention due to its excellent performance on regression and balanced classification problems. However, it is found to be unsuitable for imbalanced data classification problems because it treats each class in an imbalanced dataset equally. To address this issue, this work proposes a weighted BLS (WBLS) in which the weight assigned to each class depends on the number of samples in it. In order to further boost its classification performance, an improved differential evolution algorithm is proposed to automatically optimize its parameters, including the ones in BLS and newly generated weights. We first optimize the parameters with a training dataset, and then apply them to WBLS on a test dataset. The experiments on 20 imbalanced classification problems have shown that our proposed method can achieve higher classification accuracy than the other methods in terms of several widely used performance metrics. Finally, it is applied to fault diagnosis in self-organizing cellular networks to further show its applicability to industrial application problems.

In this article, a distributed and time-delayed k-winner-take-all (DT-kWTA) network is established and analyzed for competitively coordinated task assignment of a multirobot system. It is considered and designed from the following three aspects. First, a network is built based on a k-winner-take-all (kWTA) competitive algorithm that selects k maximum values from the inputs. Second, a distributed control strategy is used to improve the network in terms of communication load and computational burden. Third, the time-delayed problem prevalent in arbitrary causal systems (especially, in networks) is taken into account in the proposed network. This work combines distributed kWTA competition network with time delay for the first time, thus enabling it to better handle realistic applications than previous work. In addition, it theoretically derives the maximum delay allowed by the network and proves the convergence and robustness of the network. The results are applied to a multirobot system to conduct its robots' competitive coordination to complete the given task.

This work considers an extended version of flexible job-shop problem from a postprinting or semiconductor manufacturing environment, which needs a directed acyclic graph rather than a linear order to describe the precedences among operations. To obtain its reliable and high-quality schedule in a reasonable time, a learning-based cuckoo search (LCS) algorithm is presented. In it, cuckoo search is selected as an optimizer. To produce promising solutions in a high-dimensional solution space, a sparse autoencoder is introduced to compress a high-dimensional solution into an informative low-dimensional one. It extends the application area of autoencoder-embedded evolutionary optimization methods into combinational optimization by developing an improved one-hot encoding method. Then, in order to reveal the linkages among decision variables and enhance the explore ability of the proposed method, a factorization machine (FM) is used, for the first time, to capture the relevant and complementary features of population. Hence, a parallel framework involving three co-evolved subpopulations is constructed. The first one is an autoencoder embedded subpopulation, the second one is assisted by an FM, and the last one undergoes a regular iteration process. To balance the exploration and exploitation of the proposed framework and avoid unnecessary computation, a reinforcement learning algorithm is used to adaptively adjust the proportion of subpopulations and tune parameters of each subpopulation iteratively. Numerical simulations with benchmarks are performed to compare it with CPLEX, some classical heuristics, and several recently developed methods. The results shows that it well outperforms them.

Feature selection (FS) has received significant attention since the use of a well-selected subset of features may achieve better classification performance than that of full features in many real-world applications. It can be considered as a multiobjective optimization consisting of two objectives: 1) minimizing the number of selected features and 2) maximizing classification performance. Ant colony optimization (ACO) has shown its effectiveness in FS due to its problem-guided search operator and flexible graph representation. However, there lacks an effective ACO-based approach for multiobjective FS to handle the problematic characteristics originated from the feature interactions and highly discontinuous Pareto fronts. This article presents an Information-theory-based Nondominated Sorting ACO (called INSA) to solve the aforementioned difficulties. First, the probabilistic function in ACO is modified based on the information theory to identify the importance of features; second, a new ACO strategy is designed to construct solutions; and third, a novel pheromone updating strategy is devised to ensure the high diversity of tradeoff solutions. INSA's performance is compared with four machine-learning-based methods, four representative single-objective evolutionary algorithms, and six state-of-the-art multiobjective ones on 13 benchmark classification datasets, which consist of both low and high-dimensional samples. The empirical results verify that INSA is able to obtain solutions with better classification performance using features whose count is similar to or less than those obtained by its peers.

Multitask optimization (MTO) is a new optimization paradigm that leverages useful information contained in multiple tasks to help solve each other. It attracts increasing attention in recent years and gains significant performance improvements. However, the solutions of distinct tasks usually obey different distributions. To avoid that individuals after intertask learning are not suitable for the original task due to the distribution differences and even impede overall solution efficiency, we propose a novel multitask evolutionary framework that enables knowledge aggregation and online learning among distinct tasks to solve MTO problems. Our proposal designs a domain adaptation-based mapping strategy to reduce the difference across solution domains and find more genetic traits to improve the effectiveness of information interactions. To further improve the algorithm performance, we propose a smart way to divide initial population into different subpopulations and choose suitable individuals to learn. By ranking individuals in target subpopulation, worse-performing individuals can learn from other tasks. The significant advantage of our proposed paradigm over the state of the art is verified via a series of MTO benchmark studies.

Stochastic point location (SPL) involves a learning mechanism (LM) determining an optimal point on the line when the only inputs LM receives are stochastic information about the direction in which LM should move. The complexity of SPL comes from the stochastic responses of the environment, which may lead LM completely astray. SPL is a fundamental problem in optimization and was studied by many researchers during the last two decades, including improvement of its solution and all-pervasive applications. However, all existing SPL studies assume that the whole search space contains only one optimal point. Since a multimodal optimization problem (MMOP) contains multiple optimal solutions, it is significant to develop SPL's multimodal version. This article extends it from a unimodal problem to a multimodal one and proposes a parallel partition search (PPS) solution to address this issue. The heart of the proposed solution involves extracting the feature of the historical sampling information to distinguish the subintervals that contain the optimal points or not. Specifically, it divides the whole search space into multiple subintervals and samples them parallelly, then utilizes the feature of the historical sampling information to adjust the subintervals adaptively and to find the subintervals containing the optimal points. Finally, the optimal points are located within these subintervals according to their respective sampling statistics. The proof of the ϵ -optimal property for the proposed solution is presented. The numerical testing results demonstrate the power of the scheme.

Feature selection (FS) is an essential technique widely applied in data mining. Recent studies have shown that evolutionary computing (EC) is very promising for FS due to its powerful search capability. However, most existing EC-based FS methods use a length-fixed encoding to represent feature subsets. This inflexible encoding turns ineffective when high-dimension data are handled, because it results in a huge search space, as well as a large amount of training time and memory overhead. In this article, we propose a length-adaptive genetic algorithm with Markov blanket (LAGAM), which adopts a length-variable individual encoding and enables individuals to evolve in their own search space. In LAGAM, features are rearranged decreasingly based on their relevance, and an adaptive length changing operator is introduced, which extends or shortens an individual to guide it to explore in a better search space. Local search based on Markov blanket (MB) is embedded to further improve individuals. Experiments are conducted on 12 high-dimensional datasets and results reveal that LAGAM performs better than existing methods. Specifically, it achieves a higher classification accuracy by using fewer features.

The generation and updating of solutions, e.g., crossover and mutation, of many existing evolutionary algorithms directly operate on decision variables. The operators are very time consuming for large-scale and many-objective optimization problems. Different from them, this work proposes an objective space-based population generation method to obtain new individuals in the objective space and then map them to decision variable space and synthesize new solutions. It introduces three new objective vector generation methods and uses a linear mapping method to tightly connect objective space and decision one to jointly determine new-generation solutions. A loop can be formed directly between two spaces, which can generate new solutions faster and use more feedback information in the objective space. In order to demonstrate the performance of the proposed algorithm, this work performs a series of empirical experiments involving both large-scale decision variables and many objectives. Compared with the state-of-the-art traditional and large-scale algorithms, the proposed method exceeds or at least reaches its peers’ best level in overall performance while achieving great saving in execution time.

An attack–defense confrontation problem arises from a robot swarm attacking a territory protected by another one. In denied environments, global positioning and communication are hardly available. It becomes difficult for a swarm to realize collaboration and handle confrontation against another. Commonly used deep reinforcement learning (DRL) relies on pretraining, which is time consuming and has strong environmental dependence, especially in denied environments. To study attack strategies in denied environments, this work proposes a novel evolutionary algorithm (EA)-based attack strategy with Swarm Robots for the first time. Each robot obtains its situation information by perceiving its nearby peers and enemies. Such information is utilized to evaluate the benefits or threats of a robot’s next perceptible attack positions. Then, each robot uses EA to optimize its fitness function and searches for its optimal position. A collision-avoidance strategy is integrated into the algorithm. Hence, a robot swarm realizes collaboration and handles confrontation as long as each robot can sense its surroundings. They utilize their own sensors to detect others locally without using global positioning and communication devices. The experimental result analyses show that the EA-based attack strategy has better scalability and more potential in solving large-scale confrontational problems than the DRL-based algorithms. Rationales of the proposed method are presented to show the great capability of the proposed method.

This article presents a framework to enable a human and robot to perform collaborative tasks safely and efficiently. It consists of three functions. First, human motion is predicted by utilizing Gaussian mixture regression. Second, the motion planning of an online robot is performed such that the robot can appropriately react to a human coworker while executing a task. In our proposed framework, the predicted human motion is transferred to a virtual force acting on a robot's end effector. Its initial trajectory is modified so as to avoid any collisions with the human. To obtain a smooth, collision-free, and energy-minimized trajectory, a constrained optimization problem is formulated. A neural dynamics optimization algorithm is then adopted to solve it. Third, an adaptive fuzzy controller is proposed to track the robot's desired trajectory with uncertain dynamics parameters. We provide the rigorous proof of stability for the proposed methods. The physical experiments are conducted to demonstrate the effectiveness of the proposed collaborative strategy.

In order to help robots understand and perceive an object's properties during noncontact robot-object interaction, this article proposes a deeply supervised subspace learning method. In contrast to previous work, it takes the advantages of low noise and fast response of noncontact sensors and extracts novel contactless feature information to retrieve cross-modal information, so as to estimate and infer material properties of known as well as unknown objects. Specifically, a depth-supervised subspace cross-modal material retrieval model is trained to learn a common low-dimensional feature representation to capture the clustering structure among different modal features of the same class of objects. Meanwhile, all of unknown objects are accurately perceived by an energy-based model, which forces an unlabeled novel object's features to be mapped beyond the common low-dimensional features. The experimental results show that our approach is effective in comparison with other advanced methods.

For tracking suspicious objects using intelligent robots, Multiple Object Tracking (MOT) has gained great attention. MOT is easily affected by long-term severe occlusion. This work proposes a joint MOT algorithm to handle such occlusion. Pairs of frames in complicated environments are taken as input. A center-based feature extraction framework is designed for precisely detecting objects and extracting their feature maps. A ConvGRU module is applied to learn permanent representations by using historical spatio-temporal information of objects. A Hungarian matching method is applied to match the detected objects and predicted predictions. The proposed algorithm is compared with several representative methods on two public multi-object tracking benchmarks. Furthermore, this work constructs a database with videos captured from street scenarios and uses it to test the proposed algorithm and its peers. Experimental results demonstrate that the proposed algorithm outperforms its peers, especially under long-term severe occlusion, thus advancing the field of MOT.

This work studies perishable products’ distribution using electric commercial vehicles (ECVs). Extra energy is consumed for refrigeration to keep such products from deteriorating, which shortens the limited driving range of ECVs. Besides charging at public recharging stations, their travel speed can be adjusted to improve their driving range and decrease distribution cost. We propose an Electric Vehicle Routing problem with variable vehicle speed and soft time windows for perishable products (EVRP-VS). An energy consumption rate function of refrigerated ECVs during driving is introduced into the problem. The function considers refrigeration and has a nonlinear relationship with ECV speed and weight. As long as the carriage is not empty, the vehicle needs refrigeration during driving and during its stay at customers and stations. A mathematical programming model is developed for EVRP-VS, to minimize total distribution cost, including vehicle cost, power cost, refrigeration cost, and penalty cost due to delayed delivery. An adaptive hybrid ant colony optimization (AHACO) with a two-stage speed optimization strategy is proposed to solve EVRP-VS. In the first stage, local speed optimization is used to optimize the speed of each ant (ECV) in each transfer step. In the second one, global speed optimization further optimizes the speed for each fixed route constructed by ants. AHACO is applied to many problem instances. Experimental results show that it can effectively solve EVRP-VS in comparison with CPLEX and other meta-heuristics.

Formulating a cooperative autonomous vehicle group is challenging in an urban scene that has complex road networks and diverse disturbance. Existing methods of vehicle cluster cooperation in a vehicular ad-hoc network cannot be applied to autonomous vehicles because the latter have different requirements for a vehicle group structure and communication quality. Existing studies focus on autonomous vehicle group cooperation in closed and highway scenes only. Their outcomes cannot be directly applied to an urban scene because of its complex road conditions, incomplete cooperation properties, and lack of a vehicle group size control strategy. In this work, we formulate a cooperation model for autonomous vehicle groups in such scene. First, we analyze cooperation criteria based on the non-colliding aggregate motion of flocks and deduce the connectivity, coupling, timeliness, evolvability, and adaptivity of a vehicle group, based on which we propose a cooperation model. Next, we solve our model by using a modified distributed evolutionary multi-objective optimization method, prove its convergence, and analyze its computational complexity. Finally, we conduct simulations on synthetic and real roads to show its performance in terms of average connectivity, coupling, timeliness, evolvability, and adaptivity of vehicle groups.

Locating multiple sources in an unknown environment based on their signal strength is called a multi-source location problem. In recent years, there has been great interest in deploying autonomous devices to solve it. A particle swarm optimizer (PSO) is a widely employed source location method. Yet most work in this field focuses on a flat search space while ignoring height information. An unmanned aerial vehicle (UAV) has a coarser but wider view as it flies higher. Inspired by such facts, this paper focuses on improving the efficiency of locating sources by utilizing height information through UAVs. A novel source location model is designed where their sensing range gradually increases as their flying height rises, but their obtained signal strength fades away. It can be directly deployed to existing PSO-based multi-source location methods and improve their performance, especially in a large-scale environment with sparse sources. UAVs can spontaneously switch their search schemes between a rough search at a higher height and a fine one at a lower height. Experimental results of three PSO-based methods show their significant improvement after deploying our model. Given the same computation resources, its deployment leads to over 30% hike in both location accuracy and speed. This represents a great advance to the field of source location.

Recently, using a particle swarm optimizer (PSO) to guide robots in a source location problem has attracted widespread interest. While being navigated by PSO, robots are easily trapped into U-shape-like concave obstacles such that they move back and forth cyclically and fail to locate a correct source. Existing obstacle avoidance strategies perform well when robots have information about all obstacles. Yet in many real scenes, robots have no prior information. This work proposes a novel PSO based on Tabu Search (PSO-TS) for robots to locate multiple sources. Instead of traditionally setting obstacles as tabu objects, PSO-TS innovatively sets trapping areas as tabu objects such that robots do not need prior knowledge or expensive hardware and much time to obtain obstacle information. The weighted average velocity of a robot is employed to determine if it is stuck inside an obstacle-induced area. If so, a rectangular tabu area is set to push robots out of the area and prevents robots from searching the same area again. The proposed method can be embedded into various source location algorithms to improve their performance. Its obstacle avoidance capability is proved. Finally, experimental results show the algorithmic compatibility, environmental adaptability and obstacle avoidance performance of the proposed method.

Lane reservation optimization is important in intelligent transportation systems. Most existing studies are carried out under deterministic road conditions by assuming constant road travel time. However, road conditions vary due to various factors, resulting in uncertain road travel time. This work addresses a new reliability-based lane reservation and route design problem by considering uncertain road travel time with its known mean and covariance matrix. It aims to decide which road segments in a network should implement reserved lanes and to design routes for special time-crucial transportation tasks. The objective is to maximize transportation service reliability (i.e., the probability of completing the special tasks on time). For this problem, a novel distributionally robust optimization model is first established. To solve it, this work proposes i) an adapted sample average approximation-based approach and ii) a two-stage hierarchical heuristic algorithm based on second-order cone programming. Experimental results on an illustrative example and a real-world case demonstrate that the latter is more effective and efficient than the former. In addition, we conduct a series of parameter sensitivity analysis experiments to reveal the factors affecting lane reservation and provide optimal solutions given different parameter settings.

The$\alpha $-separator problem ($\alpha $-SP) consists of finding the minimum set of vertices whose removal separates the network into multiple different connected components with fewer than a limited number of vertices in each component, which belongs to the family of critical node detection problems. The$\alpha $-SP problem is an important NP-hard problem with various real-world applications. In this paper, we propose a frequent itemset-driven search (FIS) algorithm to solve$\alpha $-SP, which integrates the concept of frequent itemset into the well-known memetic search framework. Starting from a high-quality population built by population construction and population repair, FIS then iteratively employs a frequent itemset recombination operator (to generate promising offspring solution), a tabu-based simulated annealing (to find local optima), a population repair procedure, and a population management strategy (to guarantee healthy/diverse population). Extensive evaluations on 50 benchmark instances show that FIS significantly outperforms the state-of-the-art algorithms. In particular, it discovers 29 new upper bounds and matches 18 previous best-known bounds. Finally, we experimentally analyze the importance of each key algorithmic component, and perform a case study on an air transportation network for understanding its network structure and identifying its influential airports.

Due to the generation and transmission of video data, the high energy consumption rates of wireless nodes result in short lifetime of a visual sensor network, which has been commonly deployed in 3D space. In many applications, lots of targets need to be covered, but the budgets of both deployment and maintenance are limited. Therefore, how to maximize total weighted target coverage while prolonging network lifetime is of great significance. This work investigates the maximum weighted target coverage deployment of a rechargeable visual sensor network charged by an unmanned aerial vehicle (UAV) in 3D space, which should ensure network permanence and satisfy the budgets of both deployment and maintenance. In this problem, since UAV has a limited energy capacity, it is allowed to return to the depot for energy replenishment more than once in each charging period of wireless nodes so as to increase target coverage. After formulating this problem as a mixed integer nonlinear program, this work proposes a greedy heuristic and a particle swarm optimizer to approximately solve this NP-hard problem. Extensive simulation results reveal that the latter can deliver a better solution despite consuming more time than the former.

Feature selection can be used to solve imbalanced classification problems encountered in big data projects. There often exist multiple feature subsets achieving the same accuracy. These subsets tend to exhibit different acquisition difficulty and reliability, thus offering decision-makers with multiple choices if they can be well-identified. This work formulates feature selection as a Multimodal Multiobjective Problem (MMOP), where a point on Pareto front in objective space has multiple equivalent feature subsets in decision space. To seek more equivalent feature subsets, this work proposes a new multiobjective fireworks algorithm. It extends a latest single-objective fireworks algorithm to a multiobjective version such that it becomes suitable for solving MMOP. An adaptive strategy and special archive guidance are newly designed to improve its performance. A weighted extreme learning machine is chosen to classify datasets and return classification accuracy due to its fast learning speed. Experimental results show that the proposed algorithm outperforms its compared ones on 15 imbalanced classification datasets including 5 low-dimensional, 5 high-dimensional feature selection problems and 5 large-scale problems with larger imbalanced ratio, and its runtime is the least among them. Also, fault diagnosis in self-organizing cellular networks, as an important imbalance classification problem, is performed by the proposed algorithm and the results show that it can perform fault diagnosis well.

A fast non-negative latent factor (FNLF) model for a high-dimensional and sparse (HiDS) matrix adopts a Single Latent Factor-dependent, Non-negative, Multiplicative and Momentum-incorporated Update (SLF-NM2U) algorithm, which enables its fast convergence. It is crucial to achieve a rigorously theoretical proof regarding its fast convergence, which has not been provided in prior research. Aiming at addressing this critical issue, this work theoretically proves that with an appropriately chosen momentum coefficient, SLF-NM2U enables the fast convergence of an FNLF model in both continuous and discrete time cases. Empirical analysis of HiDS matrices generated by representative industrial applications provides empirical evidences for the theoretical proof. Hence, this study represents an important milestone in the field of HiDS matrix analysis.

Graphs are widely used to model various practical applications. In recent years, graph convolution networks (GCNs) have attracted increasing attention due to the extension of convolution operation from traditional grid data to graph one. However, the representation ability of current GCNs is undoubtedly limited because existing work fails to consider feature interactions. Toward this end, we propose a Dual Feature Interaction-based GCN. Specifically, it models feature interaction in the aspects of 1) node features where we use Newton's identity to extract different-order cross features implicit in the original features and design an attention mechanism to fuse them; and 2) graph convolution where we capture the pairwise interactions among nodes in the neighborhood to expand a weighted sum operation. We evaluate the proposed model with graph data from different fields, and the experimental results on semi-supervised node classification and link prediction demonstrate the effectiveness of the proposed GCN. The data and source codes of this work are available athttps://github.com/ZZY-GraphMiningLab/DFI-GCN.

Typical adversarial-training-based unsupervised domain adaptation (UDA) methods are vulnerable when the source and target datasets are highly complex or exhibit a large discrepancy between their data distributions. Recently, several Lipschitz-constraint-based methods have been explored. The satisfaction of Lipschitz continuity guarantees a remarkable performance on a target domain. However, they lack a mathematical analysis of why a Lipschitz constraint is beneficial to UDA and usually perform poorly on large-scale datasets. In this article, we take the principle of utilizing a Lipschitz constraint further by discussing how it affects the error bound of UDA. A connection between them is built, and an illustration of how Lipschitzness reduces the error bound is presented. A local smooth discrepancy is defined to measure the Lipschitzness of a target distribution in a pointwise way. When constructing a deep end-to-end model, to ensure the effectiveness and stability of UDA, three critical factors are considered in our proposed optimization strategy, i.e., the sample amount of a target domain, dimension, and batchsize of samples. Experimental results demonstrate that our model performs well on several standard benchmarks. Our ablation study shows that the sample amount of a target domain, the dimension, and batchsize of samples, indeed, greatly impact Lipschitz-constraint-based methods' ability to handle large-scale datasets. Code is available at https://github.com/CuthbertCai/SRDA.

A single dendritic neuron model (DNM) that owns the nonlinear information processing ability of dendrites has been widely used for classification and prediction. Complex-valued neural networks that consist of a number of multiple/deep-layer McCulloch-Pitts neurons have achieved great successes so far since neural computing was utilized for signal processing. Yet no complex value representations appear in single neuron architectures. In this article, we first extend DNM from a real-value domain to a complex-valued one. Performance of complex-valued DNM (CDNM) is evaluated through a complex XOR problem, a non-minimum phase equalization problem, and a real-world wind prediction task. Also, a comparative analysis on a set of elementary transcendental functions as an activation function is implemented and preparatory experiments are carried out for determining hyperparameters. The experimental results indicate that the proposed CDNM significantly outperforms real-valued DNM, complex-valued multi-layer perceptron, and other complex-valued neuron models.

A traveling salesman problem (TSP) is a well-known NP-complete problem. Traditional TSP presumes that the locations of customers and the traveling time among customers are fixed and constant. In real-life cases, however, the traffic conditions and customer requests may change over time. To find the most economic route, the decisions can be made constantly upon the time-point when the salesman completes his service of each customer. This brings in a dynamic version of the traveling salesman problem (DTSP), which takes into account the information of real-time traffic and customer requests. DTSP can be extended to a dynamic pickup and delivery problem (DPDP). In this article, we ameliorate the attention model to make it possible to perceive environmental changes. A deep reinforcement learning algorithm is proposed to solve DTSP and DPDP instances with a size of up to 40 customers in 100 locations. Experiments show that our method can capture the dynamic changes and produce a highly satisfactory solution within a very short time. Compared with other baseline approaches, more than 5% improvements can be observed in many cases.

Sharding technology is a promising solution for improving the scalability of blockchain systems. However, it faces the problem of allocating suitable trusted nodes into separate shards to satisfy security and efficiency requirements. Existing blockchain sharding methods fail to consider shard trust difference, communication latency difference, and node count difference among shards. This tends to increase the risk of a blockchain failure. This work proposes a novel blockchain sharding model for node allocation by considering shard trust difference. Its key idea is to allocate nodes of different trust levels to suitable shards to make shards have almost the same trust, such that shards’ reliability increases and blockchain failure probability decreases. To reduce the communication delay among shards, this work considers the communication latency difference and node count difference among shards. It proposes a sharding algorithm to iteratively adjust node allocations such that an optimal or near-optimal node allocation set is obtained. Simulation results show that the proposed method can effectively improve shard security and the performance of blockchain sharding compared with two state-of-the-art methods, i.e., Monoxide and Rapidchain, in terms of throughput, latency, and blockchain failure probability.

With the continuous development of consortium blockchains, the storage overhead of blocks and storage load of nodes are increasing. The existing storage models consider either low-reliability issues or high storage overhead but not both. This work proposes a storage model that combines group-based block storage with block splitting and unit encoding. The former is used to improve the reliability of a block, while the latter is adopted to reduce the storage overhead and increase the reliability of a block in a consortium blockchain. Based on this model, this work designs a storage method to minimize the storage overhead of blocks and storage load of nodes, and maximize reliability of a consortium blockchain. Experimental results show that the proposed method outperforms Multi-layer Practical Byzantine Fault Tolerance, RapidChain, and Byzantine Fault Tolerance-Store in terms of storage overhead and reliability, thus greatly advancing the field of consortium blockchains.

Service-oriented applications commonly involve high-dimensional and sparse (HiDS) interactions among users and service-related entities, e.g., user-item interactions from a personalized recommendation services system. How to perform precise and efficient representation learning on such HiDS interactions data is a hot yet thorny issue. An efficient approach to it is latent factor analysis (LFA), which commonly depends on large-scale non-convex optimization. Hence, it is vital to implement an LFA model able to approximate second-order stationary points efficiently for enhancing its representation learning ability. However, existing second-order LFA models suffer from high computational cost, which significantly reduces its practicability. To address this issue, this paper presents a Momentum-accelerated Hessian-vector algorithm (MH) for precise and efficient LFA on HiDS data. Its main ideas are two-fold: a) adopting the principle of a Hessian-vector-product-based method to utilize the second-order information without manipulating a Hessian matrix directly, and b) incorporating a generalized momentum method into its parameter learning scheme for accelerating its convergence rate to a stationary point. Experimental results on nine industrial datasets demonstrate that compared with state-of-the-art LFA models, an MH-based LFA model achieves gains in both accuracy and convergence rate. These positive outcomes also indicate that a generalized momentum method is compatible with the algorithms, e.g., a second-order algorithm, which implicitly rely on gradients.

Accurately predicting Quality of Service (QoS) is one of the main challenges in the area of service recommendation, and it has thus attracted much attention in recent years. This field exists many methods, most of which are inspired by collaborative filtering in service recommendation. They predict the missing QoS values of services by collecting the historical information of similar users/services, but their prediction accuracy needs further improvements. This work proposes user and service graphs are proposed for the first time in the field of QoS prediction by exploiting deep relationships among users and services. Based on the graphs, user/service feature vector sets are found via similar users/services. A two-stream deep learning-based prediction model is proposed for service QoS prediction. It has a deep convolutional neural network with two efficient deep convolutional units to deal with user/service feature vectors parallelly. Experiments are carried out to show that the proposed method can achieve better QoS prediction accuracy of services than the existing approaches such as non-negative matrix factorization, probabilistic matrix factorization, covering-based neighbor-hood-aware matrix factorization, neighbor integration deep matrix factorization, and several traditional methods.

This study designs a joint decision model to solve cell formation and product scheduling problems together in multifactory cellular manufacturing systems. If a product is processed in a factory with a small logistics cost of its raw material, the logistics cost of the product delivered to the distributor may be large. If managers prefer the pair of equipment unit and worker with a high production rate, operational error rate may be large. Meanwhile, it influences the labor cost, raw material cost, and completion time of each product. Managers can assign the products of a type to several cells for processing. If a cell receives fewer products of a type, it can complete their processing and becomes available to process the products of other types. However, some cells receiving more products of this type may be delayed to handle the products of other types. If a type of product in a cell has a priority to be processed, its backorder cost may be none or reduced, but this schedule may delay other product types with lower priority in the cell, and give rise to their backorder cost. For solving the intertwined optimization problem, an improved bacterial foraging algorithm with a priority rule-based heuristic (IBFAP) is developed to maximize profit. Experiments are conducted to show that IBFAP outperforms the improved bacterial foraging algorithm with no heuristic, the classical bacterial foraging algorithm, genetic algorithm, and simulated annealing.

In multirobot patrolling problems, various dynamic situations require a higher level of cooperation among robots. The dynamic problems caused by moving objects are rarely studied yet. This work proposes a distributed event-driven cooperative strategy for multirobot systems to patrol moving objects autonomously. First, forward and backward utility functions are defined as criteria for robots to conduct two-way evaluation when they choose their targets to patrol. Then, three event types and a cooperative action considering energy consumption and visiting frequency comprehensively are proposed to improve coordination among robots during their execution processes. In simulation experiments, the proposed strategy shows significant advantages on decreasing the average and maximum unvisited time of moving objects compared with the state-of-the-art. A marine pollution monitoring case is simulated to demonstrate the practicability of this strategy.

Particle swarm optimizer (PSO) is an optimization technique that has been applied to solve various problems. In its variants, hierarchical learning and variable population are two commonly used learning strategies. The former is used to employ more potentially good particles to lead the swarm, which is very effective in the early search phase. However, in the later search phase, such mechanism impedes PSO’s convergence. This work proposes an adaptive particle swarm optimizer combining hierarchical learning with variable population (PSO-HV), in which a heap-based hierarchy is first proposed to organize particles to hierarchically learn from the ones with better fitness in the same and upper levels. The levels of particles are determined and updated according to their current fitness in each iteration. Meanwhile, an adaptive variable population strategy is introduced and eliminates redundant particles based on the population’s evolution state. In this way, the swarm is more explorative upon the hierarchical structure and improves its exploitation capability due to the variable population mechanism. Ten state-of-the-art PSO contenders, including two hierarchical ones and two variable population-based ones, are compared with the proposed method on 57 benchmark functions and the experimental results verify its effectiveness and efficiency.

Reachability is the basis for studying other dynamic properties of Petri nets (PNs). When a state equation is used to determine the reachability of a marking, we need to judge whether there is a corresponding legal firing sequence (LFS) for a non-negative integer solution (NIS), i.e., a firing count vector, of the state equation. The search for an LFS is an NP-hard problem, and previous work cannot always find an LFS for any NISs. This article proposes that transition-dependent circuits or firing-dependent circuits are the root cause that a state equation has an NIS but the marking is nonreachable, i.e., there is no LFS corresponding to an NIS in PNs. Based on this, we propose a state-equation-based backward algorithm (SBA) to determine whether there is an LFS corresponding to an NIS of the state equation in a PN. The correctness and effectiveness of SBA are verified by a case study on a PN-based flexible manufacturing system and through simulation on an S4PR net. The experimental results show that the time required for SBA to determine the existence of an LFS increases linearly with the transition firing count in NISs. When the number of NISs of a state equation is finite, we can efficiently determine the reachability of a marking. This represents an important result in theory and applications of PNs.

Resource allocation systems (RASs) exist in various fields of modern society. The deadlock control problem is a crucial issue in control theory of RAS. This work is concentrated on a special class of shared resource and process-oriented Petri nets whose initial marking can have only a token in every resource place. Using mixed-integer programming (MIP) and iterative siphon control, we present a two-stage deadlock prevention policy. In particular, a modified MIP technique is developed for the first stage to compute a specific type of emptiable siphons and a siphon control method introducing monitors with related arcs whose weights all equal to one is established in the second stage. This policy leads to a maximally permissive liveness-enforcing supervisor and such an obtained controlled net is ordinary. Moreover, it avoids the exhaustive enumeration of siphons and the reachability analysis. Examples are provided to explain the policy.

Domain adaptation (DA) aims to accomplish tasks on unlabeled target data by learning and transferring knowledge from related source domains. In order to learn a discriminative and domain-invariant model, a critical step is to align source and target data well and thus reduce their distribution divergence. But existing DA methods mainly align the global feature distributions in distorted original space, which neglects their fine-grained local information and intrinsic geometrical structures. Moreover, some methods rely heavily on pseudo-labels to align features, which may undermine adaptation performance and lead to negative transfer. We propose an efficient discriminative manifold distribution alignment (DMDA) approach, which improves feature transferability by aligning both global and local distributions and refines a discriminative model by learning geometrical structures in manifold space. In addition, when learning geometrical structures, DMDA is exempt from the uncertainty and error brought by pseudo-labels of a target domain. It is very concise and efficient to be implemented by integrating learning steps and obtaining solutions directly. Extensive experiments on 68 DA tasks from seven benchmarks and subsequent analyses show that DMDA outperforms the compared methods in both classification accuracy and time efficiency, thus representing a significant advance in the DA field.

Accurately processing dynamic evolution events is extremely challenging for autonomous vehicle groups in an urban scene, which can be disturbed by manned vehicles, roadside obstacles, traffic lights, and pedestrians. Existing work focuses on a dynamic evolution method for such groups in a highway scene only. Its outcomes cannot be directly used to an urban scene due to different environmental factors, incomplete dynamic evolution events, and lack of simulation evaluation with real road networks. In this work, we present a dynamic evolution method for such groups in an urban scene. First, we analyze their dynamic evolution reasons. Then, we abstract five dynamic evolution events, i.e., joining, leaving, merging, splitting, and disappearing, and introduce a dynamic evolution method to process them. Finally, we deduce the evolvability that can reflect dynamic evolution states of a vehicle group. The simulation results in synthetic and real urban scenes show that the connectivity, coupling, timeliness, and evolvability of vehicle groups using the proposed dynamic evolution method are higher than those of using a dynamic evolution method for a highway scene.

In today’s semiconductor manufacturing industry, wafer foundries often face the challenge of producing a variety of integrated circuit chip products using a single manufacturing line. To address this, multicluster tools have become a popular choice for processing multiple wafer types simultaneously. Operating such tools involves coordinating the robots in adjacent individual tools to transport multitype wafers through a shared buffer. This study aims to develop a scheduling method for the concurrent fabrication processes of two wafer types, performed by a multicluster tool with wafer residency time constraints. The proposed approach presents a two-backward sequence, based on a backward strategy of a single wafer type, to convert a one-wafer cyclic schedule into a one-wafer-per-type cyclic schedule while revealing its temporal properties. To ensure a smooth operation of a single-arm multicluster tool system and synchronize multiple robots, several necessary and sufficient conditions are derived for the first time. Two efficient algorithms are then proposed to determine the feasibility of a periodic schedule and obtain a schedule that achieves the lower-bound cycle time under a two-backward strategy, maximizing the productivity of such a multicluster tool. Finally, numerical simulations and two practical examples are presented to demonstrate the applications and performance of the proposed approach.

Real-time and precise prediction for traffic of networks is critically important for allocating the optimal computing/network resources based on users' business requirements, analyzing the network performance, and realizing intelligent congestion control and high-accuracy anomaly detection. The dramatic growth of users' applications significantly increases the volume, uncertainty, and complexity of workload, thereby making it highly challenging to precisely predict future net-work traffic. Temporal Convolutional Networks (TCNs) and Long Short-Term Memory (LSTM) can be effectively used to analyze and predict time series. This work designs an improved prediction approach for the prediction of network traffic, which combines a Savitzky-Golay filter, TCN, and LSTM, called ST-LSTM for short. It first removes the noise of data with the filter of Savitzky-Golay. It then investigates temporal characteristics of data by using TCN. At last, it investigates the long-term dependency in the time series by using LSTM. Experimental results on a real-life website dataset show the prediction accuracy of ST-LSTM is higher than autoregressive integrated moving average, support vector regression, eXtreme Gradient Boosting, backpropagation, TCN, and LSTM, in terms of several commonly used performance indicators.

An environment where a robot swarm attacks a territory protected by another one leads to an attack-defense confrontation problem. Commonly-used deep reinforcement learning-based methods rely on pre-training and become intractable due to the curse of dimensionality. To develop effective attack strategies, inspired by a particle swarm optimizer (PSO), this work proposes a PSO-based strategy for a robot swarm for the first time. During the moving of a robot swarm, each robot obtains situation information through perceiving its nearby peers and enemies and uses such information to construct its fitness function. Then, each robot uses PSO to optimize its fitness function and searches for its optimal attack position, which guides it to move in the next time slot. The experimental analyses show that the PSO-based attack strategy has more potential in solving large-scale confrontational problems than the deep reinforcement learning-based algorithms.

Currently, cloud computing service providers face big challenges in predicting large-scale workload and resource usage time series. Due to the difficulty in capturing nonlinear features, traditional forecasting methods usually fail to achieve high performance in predicting resource usage and workload sequences. Much noise implicit in the original sequences of resources and workloads is another reason for their low performance. To address these problems, this work proposes a hybrid prediction model named SABG that integrates an adaptive Savitzky-Golay (SG) filter, Attention mechanism, Bidirectional and Grid versions of Long and Short Term Memory (LSTM) networks. SABG adopts an adaptive SG filter in the data pre-processing to eliminate noise and extreme points in the original time series. It uses bidirectional and grid LSTM networks to capture bidirectional features and dimension ones, respectively. Then, it utilizes an attention mechanism to explore importance of different data dimensions. SABG aims to predict resource usage and workloads in highly variable traces in cloud computing systems. Extensive experimental results demonstrate that SABG achieves higher-accuracy prediction than several benchmark prediction approaches with datasets from Google cluster traces.

Precise real-time prediction of the number of future network attacks cannot only prompt cloud infrastructures to fast respond to them and protect network security, but also prevents economic and business losses. In recent years, neural networks, e.g., Bi-direction Long and Short Term Memory (LSTM) and Temporal Convolutional Network (TCN), have been proven to be suitable for predicting time series data. Attention mechanisms are also widely used for the time series prediction. In this work, we propose a novel hybrid deep learning prediction method by combining the capabilities of a Savitzky-Golay (SG) filter, TCN, Multi-head self attention, and BiLSTM for the prediction of network attacks. This work first adopts a SG filter to eliminate noise in the raw data. It applies TCN to extract short-term features from the sequences. It then adopts multi-head self attention to capture intrinsic connections among features. Finally, this work adopts Bi-LSTM to extract bi-directional and long-term correlations in the sequences. Experimental results with a real-life dataset show that the proposed method outperforms several typical algorithms in terms of prediction accuracy.

With the rapid development of recycling and remanufacturing technologies, disassembly line balancing problems (DLBP) have drawn great attention. Considering the limitation of disassembly by humans or robots alone, this paper focuses on human-robot collaborative disassembly lines. Specifically, this work proposes a multi-product human-robot collaborative disassembly line balancing model to tackle the inflexibility of single product disassembly and inconsistency in recovery values of different product components. Its objective is to maximize disassembly profit. As a commercially available solver, IBM’s CPLEX is used to obtain the exact solution of DLBP and verify the correctness of the proposed mathematical model. A discrete shuffled frog leading algorithm is newly designed to solve the sizable problems. Experimental results show that the proposed algorithm has a fast convergence rate and can find solutions consistent with those with CPLEX but requires much less time than the latter, thus advancing the field of disassembly automation.

Multi-Agent Path Finding (MAPF) is an important problem with a variety of applications. Its aim is to find collision-free paths for agents having separate start and goal positions. This work proposes a new lexicographic bi-objective MAPF considering different task types, where agents are divided into two kinds to perform critical and acritical tasks. This is common in practical intelligent warehousing scenarios where a critical/acritical-task-performing agent (called c-agent and a-agent, respectively) may represent a full-load/no-load one or the one conducting urgent/non-urgent tasks. The primary objective is to minimize the sum-of-costs of c-agents, while the secondary objective is to minimize the sum-of-costs of a-agents. Two MAPF algorithms are modified to fit and solve the concerned problem for the first time. Moreover, four improvement strategies are embedded to the proposed algorithms and proved to be effective in solving MAPF problems with different task types.

Spatio-temporal prediction has a wide range of applications in many fields, e.g., air pollution, weather forecasting, and traffic forecasting. Water quality prediction is also one of spatio-temporal prediction tasks. However, it faces the following challenges: 1) Water quality in river networks has complex spatial dependencies; 2) There are complex nonlinear relations in water quality time series; and 3) It is difficult to realize long-term forecasting. To address these challenges, this work proposes a spatio-temporal prediction model called a Graph Attention-based Spatio-Temporal (GAST) neural network. GAST investigates spatial and temporal dependencies of water quality time series. First, we introduce a temporal attention mechanism to capture time series dependencies, which can effectively handle nonlinear relationships in time series. Second, we adopt a spatial attention mechanism to extract spatial dependencies of river networks and fuse temporal features of spatial nodes. Third, we adopt a temporal convolution residual mechanism based on the spatio-temporal fusion, which improves the accuracy of long-term series prediction. This work adopts two real-world datasets to evaluate the proposed GAST and experiments demonstrate that GAST outperforms several state-of-the-art methods in terms of prediction accuracy.

If people simply trash their used products, they would face many issues such as pollution to environment and resource waste. Recycling and remanufacturing used products are thus necessary, which makes the study of disassembly line balancing problems important. At present, manual disassembly is popular and it does not guarantee personal safety in the event of dangerous disassembly parts. Targeting at this problem, a mixed human-robot disassembly method is proposed. An improved Q-learning algorithm based on reinforcement learning is used to solve the two-sided disassembly line balancing problem with the objective of minimizing total disassembly time. The improved algorithm is compared with the SARSA algorithm. The results show that it can find better solutions than SARSA, and outperforms SARSA particularly in large-scale cases.

Protecting the high voltage transmission lines has been one of the most significant problems in the power systems. Precise and timely detection, identification, and location estimation of line-to-ground, line-to-line, line-to-line-to-ground, and line-to-line-to-line faults can considerably enhance the speed of a recovery process of transmission lines and hence reduce the costs associated with the downtime of a power system. Consequently, having a robust, affordable, and accurate fault diagnosis system is crucial to perform these tasks within an acceptable time window after a fault occurs in the presence of system uncertainties. Mistakenly detected or undetected faults can be expensive in the conventional techniques and this fact has motivated us to present a robust detection, identification, and location estimation system by using a machine learning method called generalized regression neural networks. The robustness of this technique is tested with respect to the variations of fault resistance, phase difference between two connected buses, fault inception angle, local bus voltage fluctuations, source inductance fluctuations, and measurement noise. Besides, the effect of noise on the GRNN method is revealed in this paper. Its comparison with the existing state-of-the-art methods shows its outstanding performance in the accurate fault classification and location estimation for transmission lines.

The storage overhead of blocks and storage load of nodes are increasing as consortium blockchains grow. The existing storage models consider either low-reliability issues or high storage overhead but not both. This work proposes a new storage model that combines group-based block storage with block splitting and unit encoding for the first time. The former is used to improve the reliability of a block, while the latter is adopted to reduce the storage overhead and increase the reliability of a block in a consortium blockchain. This work analyzes the storage overhead of blocks, storage load of nodes, and reliability of a consortium blockchain and designs a storage optimization method to reduce the storage overhead and improve the reliability of a blockchain system. Experimental results show that the proposed method outperforms Multi-layer Practical Byzantine Fault Tolerance, RapidChain, and Byzantine Fault Tolerance-Store in terms of storage overhead and reliability.

Coverage path planning is important in helping us perform such tasks as map construction and criminal capture. As an outstanding method, predator-prey coverage path planning employs a predator-prey mechanism to enable a robot to adaptively cover an arbitrary 2-D surface with dynamic obastacles. However, it is designed for bounded environments only and cannot work in unbounded environments. Inspired by the foraging behavior of herds in nature, this work proposes an adaptive coverage path planning algorithm suitable for unbounded environments, called herd foraging-based coverage path planning. It employs a virtual herd to control the overall coverage direction of a robot, allowing it to be applied in unbounded environments. The experimental results demonstrate its effectiveness in unbounded environments.

Vehicle groups that are composed of autonomous vehicles can increase the perception range of vehicles, and their dynamic evolution can provide guidance for the operation of autonomous vehicles. Most existing studies on vehicle group formation neither propose a standard vehicle group model, nor consider vehicle mobility and dynamic topology of vehicle groups. Instead, they focus on detecting dynamic evolution without predicting it. This work proposes a dynamic evolution method for autonomous vehicle groups. It first defines five vehicle states and their transitions. Then, it proposes an autonomous vehicle group formation method based on vehicle states and formulates an autonomous vehicle group model. Next, it uses meta vehicle group sequences to manage vehicle groups at different times. Finally, it gives detection and prediction methods of vehicle group dynamic evolution. Extensive simulation results show that the proposed method can be used to establish interconnection among autonomous vehicle nodes, detect dynamic evolution characteristics inside a vehicle group precisely, and predict dynamic evolution trends of vehicle groups effectively.

The Internet of Things (IoT) is a paradigm characterized by a network of embedded sensors and services. These sensors are incorporated to collect various information, track physical conditions, e.g., waste bins’ status, and exchange data with different centralized platforms. The need for such sensors is increasing; however, the proliferation of technologies comes with various challenges. For example, how can IoT and its associated data be used to enhance waste management? In smart cities, an efficient waste management system is crucial. Artificial intelligence (AI) and IoT-enabled approaches can empower cities to manage the waste collection. This work proposes an intelligent approach to route recommendation in an IoT-enabled waste management system given spatial constraints. It performs a thorough analysis based on AI-based methods and compare their corresponding results. Our solution is based on a multiple-level decision-making process in which bins’ status and coordinates are taken into account to address the routing problem. Such AI-based models can help engineers design a sustainable infrastructure system.

Spatiotemporal analysis ranges from simple univariate descriptive statistics to more complex multivariate analyses. Such an analysis can be used to explore spatial and temporal patterns in different domains, i.e., spatial and temporal information of subscribers in Internet of Things networks. Most spatial and temporal analysis techniques are based on conventional quantitative and traditional data mining approaches, such as the$k$-means algorithm. Clustering approaches based on artificial neural networks can be more efficient since they can reveal nonlinear patterns. Hence, in this work, we tailor an AI-based spatiotemporal unsupervised model such that the underlying pattern structure of a mobile phone network can be revealed, relative similarity among interactions extracted, and the associated patterns analyzed. The proposed approach is based on an optimized self-organizing feature map. It deals with high-dimensionality concerns and preserves inherent data structures. By identifying the spatial and temporal associations, decision makers can explore dominant interactions that can be used for resource optimization in network planning, content distribution, and urban planning.

In a distributed heterogeneous fog environment, fog nodes may change their state at any time. Their reliability changes accordingly. A dynamic analysis of state changes can help one detect fault-tolerant fog nodes, which is conducive to promoting the reliability of fog services. This article proposes a fault-tolerant model based on a Markov chain for a fog system’s performance optimization. The real-time reliability of fog nodes is analyzed by using dynamic distributed parameters. Thus, the state transition process of fog nodes is modeled with a continuous-time Markov chain. The steady-state probability of a fog system is analyzed. Then, a fault-tolerant strategy and its algorithms are designed to select nodes with the minimum cost based on their steady-state probabilities. The proposed method can predict the number of faulty ones of a fog system via the steady-state probability. An intelligent optimization method called simulated annealing (ISA) is designed and used to select the most appropriate fog nodes to substitute faulty ones. The experimental results show that the method is feasible and effective for selecting the right fault-tolerant nodes according to different performance requirements. ISA can well outperform such methods as random selection, discrete differential evolution, and simulated annealing in terms of cost and time.

With the rapid development of Internet of Things (IoT), much data has been produced, and new requirements have been posed for data mining. Clustering plays an essential role in discovering the underlying patterns of IoT data. It is widely used in health prognoses, pattern recognition, information retrieval, and computer vision. Density clustering is crucial to find arbitrary-shaped clusters and noise points without knowing the number of clusters in advance. However, its efficiency and applicability are reduced sharply when there exists mutual interference among parameters. In this article, a new algorithm called density-center-based automatic clustering (DAC) is proposed. First, this work presents a nonparametric density computing method. Second, it proposes to use an adaptive neighborhood whose radius is automatically calculated based on all the points in a data set. Finally, it selects appropriate density centers from a decision graph, which merge their surrounding points into the same groups. Experiments are conducted to show that DAC has higher accuracy than six classic and updated algorithms. Its effectiveness is shown via data from photovoltaic power and oil extraction systems. As an outstanding feature that its compared peers lack, it can determine parameters automatically. Thus this work greatly advances the state-of-the-art of clustering algorithms in the field of IoT data analysis.

Full-view coverage realized by camera sensor networks (CSNs) is highly demanded for monitoring and recognizing objects appearing at target points. However, it aggravates the energy shortage in CSNs as caused by the need to generate and process much sensed data. Undoubtedly, enabling CSN nodes to be rechargeable and harvest energy from their surroundings is an effective method to overcome the energy limitation of a CSN and ensuring its perpetual operation. Moreover, using rechargeable nodes can avoid the replacement of batteries, and thus can reduce network maintenance cost. In this article, we investigate how to design and deploy a rechargeable CSN with the fewest nodes to achieve full-view coverage of all target points while guaranteeing its connectivity and perpetual operation. We first formulate the problem as an integer linear program and prove its NP-hardness, and then propose a greedy heuristic and a differential evolution algorithm to solve it. Extensive simulation results reveal that the latter is able to achieve a larger success rate and higher solution quality but spends more time than the former.

Accurate and real-time prediction of network traffic can not only help system operators allocate resources rationally according to their actual business needs but also help them assess the performance of a network and analyze its health status. In recent years, neural networks have been proved suitable to predict time series data, represented by the model of a long short-term memory (LSTM) neural network and a temporal convolutional network (TCN). This article proposes a novel hybrid prediction method named SG and TCN-based LSTM (ST-LSTM) for such network traffic prediction, which synergistically combines the power of the Savitzky–Golay (SG) filter, the TCN, as well as the LSTM. ST-LSTM employs a three-phase end-to-end methodology serving time series prediction. It first eliminates noise in raw data using the SG filter, then extracts short-term features from sequences applying the TCN, and then captures the long-term dependence in the data exploiting the LSTM. Experimental results over real-world datasets demonstrate that the proposed ST-LSTM outperforms state-of-the-art algorithms in terms of prediction accuracy.Note to Practitioners—This work considers real-time and high-accuracy prediction of network traffic. It is highly important to well predict network traffic by capturing long-term dependence and effectively extracting high- and low-frequency information from time series data. Yet, it is a big challenge to achieve it because there are unstable characteristics and strong nonlinear features in the network traffic due to continuous expansion of network scale and fast emergence of new services. Current prediction methods usually have oversimplified theoretical assumptions, need significant time and memory, or suffer problems of gradient disappearance or early convergence. Thus, they fail to effectively capture the nonlinear characteristics of large-scale network sequences. This work proposes a hybrid prediction method named SG and TCN-based LSTM (ST-LSTM), which integrates the merits of the Savitzky–Golay filter, the temporal convolutional network (TCN), and the long short-term memory (LSTM), serving as smoothing time series, capturing short-term local features, and capturing long-term dependence, respectively. Experimental results based on the real-life dataset demonstrate that it achieves better prediction accuracy than its state-of-the-art peers, including the TCN and the LSTM. It can be readily implemented and deployed in many real-life industrial areas including smart city, edge computing, cloud computing, and data centers.

This IEEE Transactions on Automation Science and Engineering (T-ASE) Special Section on New Frontiers in Smart Factories: Smart Automation and Human–Robot Interaction focuses on promising, innovative research outcomes and industrial applications of different key technologies for smart automation and human–robot interaction.

System scheduling is a decision-making process that plays an important role in improving the performance of robotic cellular manufacturing (RCM) systems. Timed Petri nets (PNs) are a formalism suitable for graphically and concisely modeling such systems and obtaining their reachable state graphs. Within their reachability graphs, timed PNs’ evolution and intelligent search algorithms can be combined to find an efficient operation sequence from an initial state to a goal one for the underlying systems of the nets. To schedule RCM systems, this work proposes an A* search with a new heuristic function based on timed PNs. When compared with related approaches, the proposed one can deal with token remaining time, weighted arcs, and multiple resource copies commonly seen in the PN models of RCM systems. The admissibility of the proposed heuristic function is proved. Finally, experimental results are given to show the effectiveness and efficiency of the proposed method and heuristic function.Note to Practitioners—Robotic cellular manufacturing (RCM) systems are among the most common and complicated discrete-event dynamic systems, which provide a great number of choices of resources and processing routes to allow high system productivity. Timed Petri nets (PNs) and intelligent search algorithms on their reachability graphs are ideal tools to handle the RCM scheduling problem. This work proposes an A* search method based on the evolutions of timed PNs to optimally schedule RCM systems. The proposed method can deal with token remaining time, weighted arcs, and multiple resource copies often encountered in the PN models of RCM systems.

Recently, deep neural network (DNN) models work incredibly well, and edge computing has achieved great success in real-world scenarios, such as fault diagnosis for large-scale rotational machinery. However, DNN training takes a long time due to its complex calculation, which makes it difficult to optimize and retrain models. To address such an issue, this work proposes a novel fault diagnosis model by combining binarized DNNs (BDNNs) with improved random forests (RFs). First, a BDNN-based feature extraction method with binary weights and activations in a training process is designed to reduce the model runtime without losing the accuracy of feature extraction. Its generated features are used to train an RF-based fault classifier to relieve the information loss caused by binarization. Second, considering the possible classification accuracy reduction resulting from those very similar binarized features of two instances with different classes, we replace a Gini index with ReliefF as the attribute evaluation measure in training RFs to further enhance the separability of fault features extracted by BDNN and accordingly improve the fault identification accuracy. Third, an edge computing-based fault diagnosis mode is proposed to increase diagnostic efficiency, where our diagnosis model is deployed distributedly on a number of edge nodes close to the end rotational machines in distinct locations. Extensive experiments are conducted to validate the proposed method on the data sets from rolling element bearings, and the results demonstrate that, in almost all cases, its diagnostic accuracy is competitive to the state-of-the-art DNNs and even higher due to a form of regularization in some cases. Benefited from the relatively lower computing and storage requirements of BDNNs, it is easy to be deployed on edge nodes to realize real-time fault diagnosis concurrently.Note to Practitioners—Rotating machines, such as engines and motors, are the cornerstones of the modern industry. Edge computing is an emerging computing paradigm where computation is performed on the edges of networks rather than on the central cloud, thereby reducing system response time, transmission overhead, storage space, and computation resources of the cloud. Motivated by the high demand on computation for deploying DNN models and lower computation complexity for running BDNN models and easiness for large-scale deployment of BDNNs, an edge computing-based method for real-time fault diagnosis of rotating machines is proposed. First, we design a BDNN-based feature extractor to decrease the amount of computation and speed up a diagnosis processes. Then, the resulting binary features are fed to train an RF-based classifier, where we use ReliefF instead of Gini index when training a random forest model to further improve the proposed method’s diagnostic accuracy. Finally, a novel cloud-edge collaborative computing-based fault diagnostic mode is presented, where the model trained from the central cloud is deployed on the edge computing devices distributed in large-scale scenarios to realize real-time fault diagnosis. Experiment results show that the proposed method can maintain the desired accuracy but greatly enhance the diagnosis speed when deployed on the edge nodes near end physical machines. It is easily extended and used for fault detection in many industrial sectors.

Cloud computing becomes a promising technology to reduce computation cost by providing users with elastic resources and application-deploying environments as a pay-per-use model. More scientific workflow applications have been moved or are being migrated to the cloud. Scheduling workflows turns to the main bottleneck for increasing resource utilization and quality of service (QoS) for users. This work formulates workflow scheduling as multiobjective optimization problems and proposes a Scoring and Dynamic Hierarchy-based NSGA-II (Nondominated Sorting Genetic Algorithm II), called SDHN for short, to minimize both makespan and cost of workflow execution. First, a scoring criterion is developed to calculate the total score for each individual during population updating, which is used as a quantitative index to evaluate the dominance degree of individuals among the whole population. Hence, SDHN can distinguish individuals within the same dominance level and target its search toward the directions of elite solutions as their different dominance degrees and accordingly improve search efficiency. Second, a population-based dynamic hierarchical structure (HS) and its evolutionary rules are presented to update HS by comparing each child with all parental individuals from bottom to up until finding a proper dominant level. Since traversing all HS levels is not needed in most cases, the number of individual comparisons is reduced and SDHN’s updating efficiency is greatly improved, especially for large-scale and complex applications. Third, to guarantee its converging to the near-optimal solutions, adaptive adjustment strategies (AASs) are designed to prevent the search from falling into local optima or diverging by checking the number of individuals at the highest HS level and then modifying the relevant genetic operations to guide the evolutionary process to approach the global Pareto Front. Extensive experiments are conducted to verify SDHN, and the results show that it outperforms the existing algorithms in the quality and diversity of resulting solutions as well as convergence time.Note to Practitioners—Most scientific applications are computation and/or data-intensive and need large-scale or high-performance resources for their execution. More and more scientists use workflows to manage their applications, but how to efficiently run them in the cloud is a big challenge due to their large scale as well as the dynamic characteristics of the elastic and heterogeneous cloud resources. In this article, we develop a novel multiobjective optimization technique for workflow scheduling such that the makespan and cost can be minimized simultaneously. A scoring criterion, dynamic hierarchical structure and its evolutionary rules, and adaptive adjustment strategies are designed to cooperate with each other and increase the search ability and efficiency of the original and widely used NSGA-II. Adequate experiments are conducted to verify the proposed method’s performance, and the experimental results show that it can provide more near-optimal solutions than the existing methods. It can be readily applied for implementing more efficient and effective cloud data centers to execute large-scale scientific workflows.

This work considers a stochastic flexible job shop scheduling with limited extra resources and machine-dependent setup time in a semiconductor manufacturing environment, which is an NP-hard problem. In order to obtain its reliable and high-performance schedule in a reasonable time, a learning-based grey wolf optimizer is proposed. In it, an optimal computing budget allocation-based approach, which is designed for two scenarios from real manufacturing environments, is proposed to intelligently allocate computing budget and improve search efficiency. It extends the application area of optimal computing budget allocation by laying a theoretic foundation. Besides, to obtain proper control parameters iteratively, a reinforcement learning algorithm with a newly designed delay update strategy is used to build a parameter tuning scheme of a grey wolf optimizer. The scheme acts as a guide for balancing global and local search, thereby enhancing effectiveness of the proposed algorithm. The theoretic interpretation of the developed optimal computing budget allocation-based approach and the convergence analysis results of the proposed algorithm are presented. Various experiments with benchmarks and randomly generated cases are performed to compare it with several updated algorithms. The results shows its superiority over them. Note to Practitioners—Meta-heuristic are often deployed to solve semiconductor manufacturing scheduling problems. However, they face to two thorny issues when they face stochastic manufacturing environments. 1) their computational efficiency is quite low, thus requiring substantial improvement, since a stochastic optimization problem requires Monte Carlo sampling to estimate the actual objective function values in a precise manner; and 2) most of them are parameter-sensitive, and choosing their proper parameters is highly challenging in such environments. To address the first issue, we develop an optimal computing budget allocation-based method for deciding the optimal numbers of sampling times based on both prior knowledge and simulation results. To address the second one, we propose a reinforcement learning algorithm to self-adjust the parameters of our proposed method called Learning-based Grey Wolf Optimizer. In addition, we design a delay update strategy to enhance its robustness, and thus, a feasible and high-quality schedule can be founded in a short time for real-time scheduling problems. Theoretic proofs and experimental results show that the proposed method is effective and efficient. Consequently, it can be readily applicable to practical semiconductor manufacturing systems.

Cloud platforms have recently become a popular target execution environment for numerous workflow applications. Hence, effective workflow scheduling strategies in cloud environments are in high demand. However, existing scheduling algorithms are grounded on an idealized target platform model where virtual machines are fully connected, and all communications can be performed concurrently. A significant aspect neglected by them is endpoint communication contention when executing workflows, which has a large impact on workflow makespan. This article investigates how to incorporate contention awareness into cloud workflow scheduling and proposes a new practical scheduling model. Endpoint communication contention-aware List Scheduling Heuristic (ELSH) is designed to minimize workflow makespan. It uses a novel task ranking property and schedules data communications to communication resources besides scheduling tasks to computing resources. Moreover, a rescheduling technique is employed to improve the schedule. In experiments, ELSH is evaluated against the traditional contention-oblivious list scheduling algorithm, which is adapted to address contention during execution in practice. The experimental results reveal that ELSH performs more efficaciously compared with the adapted traditional ones.Note to Practitioners—This article aims to advance the state of the art for workflow scheduling in clouds by taking into account endpoint communication contention that can occur in practice but has largely been neglected in existing investigations. A scheduling method called Endpoint communication contention-aware List Scheduling Heuristic (ELSH) is then proposed to optimize workflow makespan. Experimental results based on synthetic and realistic workflows show that ELSH performs better than traditional scheduling algorithms that fail to consider endpoint communication contention, especially for the workflow with a large communication-to-computation-cost ratio. The proposed approach can be readily put into use and help cloud service providers to offer their customers high-quality services when executing the latter’s workflows.

The enormous energy consumed by clouds becomes a significant challenge for cloud providers and smart grid operators. Due to performance concerns, applications typically run in different clouds located in multiple sites. In different clouds, many factors, including electricity prices, available servers, and task service rates, exhibit spatial variations. Therefore, it is important to manage and schedule tasks among multiple clouds in a high-quality-of-service and low-energy-cost manner. This work proposes a task scheduling method to jointly minimize energy cost and average task loss possibility (ATLP) of clouds. A problem is formulated and tackled with an adaptive biobjective differential evolution based on simulated annealing to determine a real-time and near-optimal set of solutions. A final knee solution is further chosen to specify suitable servers in clouds and task allocation among web portals. Simulation results based on realistic data prove that less average loss possibility of tasks, and smaller energy cost is obtained with it than its widely used peers.Note to Practitioners—This work considers joint optimization of both ATLP and average energy cost of all clouds. It is of great significance to execute tasks among multiple clouds by jointly allocating all tasks among multiple web portals and specifying suitable servers in different clouds. Yet, it is challenging to achieve joint optimization in a market where factors, including prices of electricity and available servers, show spatial variations. Current studies are coarse-grained and fail to jointly achieve average energy cost minimization and quality-of-service optimization of tasks. In this work, a novel algorithm named adaptive simulated-annealing-based biobjective differential evolution is proposed for an energy cost and quality-of-service-optimized task scheduling strategy in a real-time manner. Experiments prove that it realizes lower energy cost and ATLP compared with its typical widely used peers. It can also be applied to other industrial areas, including smart manufacturing, Internet of Things, and smart city.

Wire rod and bar rolling is an important batch production process in steel production systems. A scheduling problem originated from this process is studied in this work by considering the constraints on sequence-dependent family setup time and release time. For each serial batch to be scheduled, it contains several jobs and the number of late jobs within it varies with its start time. First, we model a rolling process using a Petri net (PN), where a so-called rolling transition describes a rolling operation of a batch. The objective of the concerned problem is to determine a firing sequence of all rolling transitions such that the total number of late jobs is minimal. Next, a mixed-integer linear program is formulated based on the PN model. Due to the NP-hardness of the concerned problem, iterated greedy algorithm (IGA)-based methods by using different neighborhood structures and integrating a variable neighborhood descent method are developed to obtain its near-optimal solutions. To test the accuracy, speed, and stability of the proposed algorithms, we compare their solutions of different-size instances with those of CPLEX (a commercial software) and four heuristic peers. The results indicate that the proposed algorithms outperform their peers and have great potential to be applied to industrial production process scheduling.Note to Practitioners—This work deals with a scheduling problem of a batch production process, i.e., wire rod and bar rolling, which is modeled by a Petri net (PN). Due to the NP-hardness of the concerned problem, four iterated greedy algorithm-based methods are developed to solve it. The proposed methods are validated and tested by comparing their solutions with those of four heuristic peers and the exact ones (when available via CPLEX). Extensive experimental results show that they can fast solve one-week-scale instances with better performance than their peers’, thereby proving the readiness to put them in industrial use. When solving a one-month-scale instance, the proposed methods show much better performance than others.

An iterated greedy algorithm (IGA) is a simple and powerful heuristic algorithm. It is widely used to solve flow-shop scheduling problems (FSPs), an important branch of production scheduling problems. IGA was first developed to solve an FSP in 2007. Since then, various FSPs have been tackled by using IGA-based methods, including basic IGA, its variants, and hybrid algorithms with IGA integrated. Up until now, over 100 articles related to this field have been published. However, to the best of our knowledge, there is no existing tutorial or review paper of IGA. Thus, we focus on FSPs and provide a tutorial and comprehensive literature review of IGA-based methods. First, we introduce a framework of basic IGA and give an example to clearly show its procedure. To help researchers and engineers learn and apply IGA to their FSPs, we provide an open platform to collect and share related materials. Then, we make classifications of the solved FSPs according to their scheduling scenarios, objective functions, and constraints. Next, we classify and introduce the specific methods and strategies used in each phase of IGA for FSPs. Besides, we summarize IGA variants and hybrid algorithms with IGA integrated, respectively. Finally, we discuss the current IGA-based methods and already-solved FSP instances, as well as some important future research directions according to their deficiency and open issues.Note to Practitioners—Many practical scheduling problems can be transformed into flow-shop scheduling problems (FSPs), most of which are NP-hard. In order to solve them in an industrial system setting, designing effective heuristics is important and practically useful and has, thus, attracted much attention from both researchers and engineers. As an easy and high-performance heuristic, an iterated greedy algorithm (IGA) is widely used and adapted to solve numerous FSPs. Its simple framework makes it easy to be implemented by practitioners, and its high performance implies its great potential to solve industrial scheduling problems. In this work, we aim to give practitioners a comprehensive overview of IGA and help them apply IGA to solve their particular industrial scheduling problems. We review the papers that solve FSPs with IGA-based methods, including basic IGA, its variants, and hybrid algorithms with IGA integrated. First, we provide practitioners with a tutorial on IGA, where an example for solving an FSP is introduced and an open platform is constructed. The platform collects and shares the related materials, e.g., open-source code, benchmarks, and website links of important papers. Then, we introduce various FSPs and specific designs of IGA-based methods. Finally, we discuss the current research and point out future research issues.

A Stochastic Gradient Descent (SGD)-based Latent Factor Analysis (LFA) model is highly efficient in representative learning on a High-Dimensional and Sparse (HiDS) matrix, where the learning rate adaptation is vital in its efficiency and practicability. The learning rate adaptation of an SGD-based LFA model can be achieved efficiently by learning rate evolution with an evolutionary computing algorithm. However, a resultant model commonly suffers from twofold premature convergence issues, i.e., a) the premature convergence of the learning rate swarm relying on an evolution algorithm, and b) the premature convergence of an LFA model relying on the compound effects of evolution-based learning rate adaptation and adopted optimization algorithm. Aiming at addressed such issues, this work proposes an Hierarchical Particle swarm optimization-incorporated Latent factor analysis (HPL) model with a two-layered structure. The first layer pre-trains desired latent factors with a position-transitional particle swarm optimization-based LFA model with learning rate adaptation; while the second layer performs latent factor refinement with a newly-proposed mini-batch particle swarm optimization algorithm. Experimental results on four HiDS matrices generated by industrial applications demonstrate that an HPL model can well handle the mentioned premature convergence issues, thereby achieving highly-accurate representation to HiDS matrices.

Attributed graphs are widely used to represent network data where the attribute information of nodes is available. To address the problem of identifying clusters in attributed graphs, most of existing solutions are developed simply based on certain particular assumptions related to the characteristics of clusters of interest. However, it is yet unknown whether such assumed characteristics are consistent with attributed graphs. To overcome this issue, we innovatively introduce an inductive clustering algorithm that tends to address the clustering problem for attributed graphs without any assumption made on the clusters. To do so, we first process the attribute information to obtain pairwise attribute values that significantly frequently co-occur in adjacent nodes as we believe that they have potential ability to represent the characteristics of a given attributed graph. For two adjacent nodes, their likelihood of being grouped in the same cluster can be weighted by their ability to characterize the graph. Then based on these verifed characteristics instead of assumed ones, a depth-first search strategy is applied to perform the clustering task. Moreover, we are able to classify clusters such that their significance can be indicated. The experimental results demonstrate the performance and usefulness of our algorithm.

Cloud computing has become an important paradigm in which scalable resources such as CPU, memory, disk and IO devices can be provided to users to remotely process their applications. In a cloud computing platform, energy consumption accounts for a significant cost portion. This article thus aims to present an energy-efficient scheduling algorithm for processing a user application with a real-time requirement. This problem is formulated as a non-linear mixed integer programming problem. We start with providing an optimal closed-form solution to its relaxation problem that aims to minimize the energy consumption without considering real-time requirements. To meet real-time requirements, we propose how to adjust task placement and resource allocation by making a good tradeoff between energy consumption and task execution time. Lastly, we find two equivalent optimal resource allocation strategies once task placement has been done. We then propose to adjust the start time of task execution such that an application’s completion time can be further shortened. Experimental results on two real-case enchmarks and extensive synthetic applications demonstrate that our proposed method finds a schedule that generally has 30 and 20 percent less energy consumption than enhancement heterogeneous earliest finish time (E-HEFT) and genetic algorithm, respectively. Besides, the proposed method has a higher rate to successfully find a feasible schedule than them, and its computation time is close to E-HEFT’s, but far less than the genetic algorithm's.

Infrastructure in Distributed Green Data Centers (DGDCs) is concurrently shared by multiple different applications to flexibly provide a growing number of services to global users in a cost-effective way. A highly challenging problem is how to maximize the total profit of the DGDC provider in a market where Internet Service Provider (ISP) bandwidth price, availability of green energy, price of power grid, and revenue brought by the execution of tasks all vary with geographical locations. Unlike existing studies, this article proposes a Geography-Aware Task Scheduling (GATS) approach by considering spatial variations in DGDCs to maximize the total profit of the DGDC provider by intelligently scheduling tasks of all applications. In each time slot, the formulated profit maximization problem is solved as a convex optimization one via the interior point method. Trace-driven simulations show that GATS achieves larger total profit and higher throughput than two typical task scheduling approaches.

The emerging applications for smart cities intend to promote the quality of citizens’ life. Among them, ubiquitous user connectivity and real-time computation offloading are significant for the ever-increasing requirements of delay-sensitive and mission-critical applications. By integrating human social behaviors (such as relationship, similarity, community, and social ties) with physical Internet of Things (IoT) systems, social IoT systems are promising to provide ubiquitous connectivity among users. As the applications of social IoT systems are transferring from information dissemination to user entertainment (such as image identification, online games, and augmented reality), computation offloading is significant to reduce the execution delay of applications.

This work investigates direction control and path following of a 3-D snake-like robot. In order to control such robots accurately, this work researches the relationships between its phase offsets of pitch joints and directions. A new direction control method is proposed for the robot based on these relationships. An adaptive path-following algorithm based on the line-of-sight guidance law is proposed and combined with the direction control method to steer the robot to move forward and along desired paths. Simulation and experimental results are presented to demonstrate the performances of the proposed 3-D model and control methods. They well outperform the classical and commonly used path-following method.

Visual relationship detection (VRD) is one newly developed computer vision task, aiming to recognize relations or interactions between objects in an image. It is a further learning task after object recognition, and is important for fully understanding images even the visual world. It has numerous applications, such as image retrieval, machine vision in robotics, visual question answer (VQA), and visual reasoning. However, this problem is difficult since relationships are not definite, and the number of possible relations is much larger than objects. So the complete annotation for visual relationships is much more difficult, making this task hard to learn. Many approaches have been proposed to tackle this problem especially with the development of deep neural networks in recent years. In this survey, we first introduce the background of visual relations. Then, we present categorization and frameworks of deep learning models for visual relationship detection. The high-level applications, benchmark datasets, as well as empirical analysis are also introduced for comprehensive understanding of this task.

To quantify user-item preferences, a recommender system (RS) commonly adopts a high-dimensional and sparse (HiDS) matrix. Such a matrix can be represented by a non-negative latent factor analysis model relying on a single latent factor (LF)-dependent, non-negative, and multiplicative update algorithm. However, existing models' representative abilities are limited due to their specialized learning objective. To address this issue, this study proposes an α- β -divergence-generalized model that enjoys fast convergence. Its ideas are three-fold: 1) generalizing its learning objective with α- β -divergence to achieve highly accurate representation of HiDS data; 2) incorporating a generalized momentum method into parameter learning for fast convergence; and 3) implementing self-adaptation of controllable hyperparameters for excellent practicability. Empirical studies on six HiDS matrices from real RSs demonstrate that compared with state-of-the-art LF models, the proposed one achieves significant accuracy and efficiency gain to estimate huge missing data in an HiDS matrix.

In this article, we elaborate on a Kullback-Leibler (KL) divergence-based Fuzzy C -Means (FCM) algorithm by incorporating a tight wavelet frame transform and morphological reconstruction (MR). To make membership degrees of each image pixel closer to those of its neighbors, a KL divergence term on the partition matrix is introduced as a part of FCM, thus resulting in KL divergence-based FCM. To make the proposed FCM robust, a filtered term is augmented in its objective function, where MR is used for image filtering. Since tight wavelet frames provide redundant representations of images, the proposed FCM is performed in a feature space constructed by tight wavelet frame decomposition. To further improve its segmentation accuracy (SA), a segmented feature set is reconstructed by minimizing the inverse process of its objective function. Each reconstructed feature is reassigned to the closest prototype, thus modifying abnormal features produced in the reconstruction process. Moreover, a segmented image is reconstructed by using tight wavelet frame reconstruction. Finally, supporting experiments coping with synthetic, medical, and real-world images are reported. The experimental results exhibit that the proposed algorithm works well and comes with better segmentation performance than other peers. In a quantitative fashion, its average SA improvements over its peers are 4.06%, 3.94%, and 4.41%, respectively, when segmenting synthetic, medical, and real-world images. Moreover, the proposed algorithm requires less time than most of the FCM-related algorithms.

This work proposes a scalable gamma non-negative matrix network (SGNMN), which uses a Poisson randomized Gamma factor analysis to obtain the neurons of the first layer of a network. These neurons obey Gamma distribution whose shape parameter infers the neurons of the next layer of the network and their related weights. Upsampling the connection weights follows a Dirichlet distribution. Downsampling hidden units obey Gamma distribution. This work performs up-down sampling on each layer to learn the parameters of SGNMN. Experimental results indicate that the width and depth of SGNMN are closely related, and a reasonable network structure for accurately detecting brain fatigue through functional near-infrared spectroscopy can be obtained by considering network width, depth, and parameters.

Current brain cognitive models are insufficient in handling outliers and dynamics of electroencephalogram (EEG) signals. This article presents a novel self-paced dynamic infinite mixture model to infer the dynamics of EEG fatigue signals. The instantaneous spectrum features provided by ensemble wavelet transform and Hilbert transform are extracted to form four fatigue indicators. The covariance of log likelihood of the complete data is proposed to accurately identify similar components and dynamics of the developed mixture model. Compared with its seven peers, the proposed model shows better performance in automatically identifying a pilot's brain workload.

A colored traveling salesman problem (CTSP) as a generalization of the well-known multiple traveling salesman problem utilizes colors to distinguish the accessibility of individual cities to salesmen. This work formulates a precedence-constrained CTSP (PCTSP) over hypergraphs with asymmetric city distances. It is capable of modeling the problems with operations or activities constrained to precedence relationships in many applications. Two types of precedence constraints are taken into account, i.e., 1) among individual cities and 2) among city clusters. An augmented variable neighborhood search (VNS) called POPMUSIC-based VNS (PVNS) is proposed as a main framework for solving PCTSP. It harnesses a partial optimization metaheuristic under special intensification conditions to prepare candidate sets. Moreover, a topological sort-based greedy algorithm is developed to obtain a feasible solution at the initialization phase. Next, mutation and multi-insertion of constraint-preserving exchanges are combined to produce different neighborhoods of the current solution. Two kinds of constraint-preserving k -exchange are adopted to serve as a strong local search means. Extensive experiments are conducted on 34 cases. For the sake of comparison, Lin-Kernighan heuristic, two genetic algorithms and three VNS methods are adapted to PCTSP and fine-tuned by using an automatic algorithm configurator-irace package. The experimental results show that PVNS outperforms them in terms of both search ability and convergence rate. In addition, the study of four PVNS variants each lacking an important operator reveals that all operators play significant roles in PVNS.

Particle swarm optimizer (PSO) and mobile robot swarm are two typical swarm techniques. Many applications emerge separately along both of them while the similarity between them is rarely considered. When a solution space is a certain region in reality, a robot swarm can replace a particle swarm to explore the optimal solution by performing PSO. In this way, a mobile robot swarm should be able to efficiently explore an area just like the particle swarm and uninterruptedly work even under the shortage of robots or in the case of unexpected failure of robots. Furthermore, the moving distances of robots are highly constrained because energy and time can be costly. Inspired by such requirements, this article proposes a moving-distance-minimized PSO (MPSO) for a mobile robot swarm to minimize the total moving distance of its robots while performing optimization. The distances between the current robot positions and the particle ones in the next generation are utilized to derive paths for robots such that the total distance that robots move is minimized, hence minimizing the energy and time for a robot swarm to locate the optima. Experiments on 28 CEC2013 benchmark functions show the advantage of the proposed method over the standard PSO. By adopting the given algorithm, the moving distance can be reduced by more than 66% and the makespan can be reduced by nearly 70% while offering the same optimization effects.

Stochastic point location deals with the problem of finding a target point on a real line through a learning mechanism (LM) with the stochastic environment (SE) offering directional information. The SE can be further categorized into an informative or deceptive one, according to whether p is above 0.5 or not, where p is the probability of providing a correct suggestion of a direction to LM. Several attempts have been made for LM to work in both types of environments, but none of them considers a dynamically changing environment where p varies with time. A dynamic dual environment involves fierce changes that frequently cause its environment to switch from an informative one to a deceptive one, or vice versa. This article presents a novel weak estimator-based adaptive step search solution, to enable LM to track the target in a dynamic dual environment, with the help of a weak estimator. The experimental results show that the proposed solution is feasible and efficient.

Surrogate-assisted evolutionary algorithms (EAs) have been intensively used to solve computationally expensive problems with some success. However, traditional EAs are not suitable to deal with high-dimensional expensive problems (HEPs) with high-dimensional search space even if their fitness evaluations are assisted by surrogate models. The recently proposed autoencoder-embedded evolutionary optimization (AEO) framework is highly appropriate to deal with high-dimensional problems. This work aims to incorporate surrogate models into it to further boost its performance, thus resulting in surrogate-assisted AEO (SAEO). It proposes a novel model management strategy that can guarantee reasonable amounts of re-evaluations; hence, the accuracy of surrogate models can be enhanced via being updated with new evaluated samples. Moreover, to ensure enough data samples before constructing surrogates, a problem-dimensionality-dependent activation condition is developed for incorporating surrogates into the SAEO framework. SAEO is tested on seven commonly used benchmark functions and compared with state-of-the-art algorithms for HEPs. The experimental results show that SAEO can further enhance the performance of AEO on most cases and SAEO performs significantly better than other algorithms. Therefore, SAEO has great potential to deal with HEPs.

Protein structure prediction (PSP) problems are a major biocomputing challenge, owing to its scientific intrinsic that assists researchers to understand the relationship between amino acid sequences and protein structures, and to study the function of proteins. Although computational resources increased substantially over the last decade, a complete solution to PSP problems by computational methods has not yet been obtained. Using only one energy function is insufficient to characterize proteins because of their complexity. Diverse protein energy functions and evolutionary computation algorithms have been extensively studied to assist in the prediction of protein structures in different ways. Such algorithms are able to provide a better protein with less computational resources requirement than deep learning methods. For the first time, this study proposes a many-objective PSP (MaOPSP) problem with four types of objectives to alleviate the impact of imprecise energy functions for predicting protein structures. A many-objective evolutionary algorithm (MaOEA) is utilized to solve MaOPSP. The proposed method is compared with existing methods by examining 34 proteins. An analysis of the objectives demonstrates that our generated conformations are more reasonable than those generated by single/multiobjective optimization methods. Experimental results indicate that solving a PSP problem as an MaOPSP problem with four objectives yields better PSPs, in terms of both accuracy and efficiency. The source code of the proposed method can be found athttps://toyamaailab.github.io/sourcedata.html.

The generation of motor adaptation in response to mechanical perturbation during human walking is seldom considered in an exoskeleton system. Reshaping step trajectory over consecutive gait cycles for a walking exoskeleton is investigated in this article. Step adjustment of a walking exoskeleton can adapt to human walking intention by shaping step trajectory. This work develops an admittance adaptive fuzzy control strategy for a walking exoskeleton robot to provide assistance for human lower limb movement. Considering human walking intention and utilizing an admittance model, it shapes a reference trajectory to ensure that the walking exoskeleton follows it according to the human–robot force produced by its wearer. Considering a nonlinear and dynamic model with uncertainties, this work designs an integral-type Lyapunov function controller to track a reference trajectory. A disturbance observer is integrated into the controller design to compensate for uncertain disturbance in order to achieve an effective tracking performance. Finally, this work conducts experiments on two healthy subjects with the proposed method on a walking exoskeleton to validate its effectiveness. The results show that it can be applied to walking exoskeletons to enhance human mobility.

Extreme learning machine (ELM) is suitable for nonlinear soft sensor development. Yet it faces an overfitting problem. To overcome it, this work integrates bound optimization theory with variational Bayesian (VB) inference to derive novel L1 norm-based ELMs. An L1 term is attached to the squared sum cost of prediction errors to formulate an objective function. Considering the nonconvexity and nonsmoothness of the objective function, this article uses bound optimization theory, and constructs a proper surrogate function to equivalently convert a challenging L1 norm-based optimization problem into easy one. Then, VB inference is adopted for optimizing the converted problem. Thus, an L1 norm-based ELM can be efficiently optimized by an alternating optimization algorithm with a proved convergence. Finally, a soft sensor is developed based on the proposed algorithm. An industrial case study is carried out to demonstrate that the proposed soft sensor is competitive against recent ones.

Intelligent video surveillance is an important computer vision application in natural environments. Since detected objects under surveillance are usually low-resolution and noisy, their accurate recognition represents a huge challenge. Knowledge distillation is an effective method to deal with it, but existing related work usually focuses on reducing the channel count of a student network, not feature map size. As a result, they cannot transfer "privilege information" hidden in feature maps of a wide and deep teacher network into a thin and shallow student one, leading to the latter's poor performance. To address this issue, we propose a Feature Map Distillation (FMD) framework under which the feature map size of teacher and student networks is different. FMD consists of two main components: Feature Decoder Distillation (FDD) and Feature Map Consistency-enforcement (FMC). FDD reconstructs the shallow texture features of a thin student network to approximate the corresponding samples in a teacher network, which allows the high-resolution ones to directly guide the learning of the shallow features of the student network. FMC makes the size and direction of each deep feature map consistent between student and teacher networks, which constrains each pair of feature maps to produce the same feature distribution. FDD and FMC allow a thin student network to learn rich "privilege information" in feature maps of a wide teacher network. The overall performance of FMD is verified in multiple recognition tasks by comparing it with state-of-the-art knowledge distillation methods on low-resolution and noisy objects.

Accurate estimation of network capacity is very important for Vehicular Infrastructure-based NETwork (VINET) in an urban scene that may involve greatly dynamic typology and complex driving conditions. The node mobility, network behavior, and network scale of a VINET are different from those of a wireless network, and, therefore, the existing capacity estimation methods of wireless networks cannot be used to estimate VINET capacity. In addition, most existing studies on VINET capacity only derive asymptotic descriptions when the number of nodes is large enough. In this work, a novel approach is proposed for the modeling and calculating VINET capacity. More specifically, we first analyze communication characteristics in a VINET, and introduce two transmission modes, i.e., a vehicle-based mode and a Road Side Unit (RSU)-based one. Then, we propose a probability-based transmission mode selecting strategy with which vehicle nodes can choose either transmission mode independently and such choice is probabilistic. Next, we analyze the characteristics of an RSU-based mode, divide a VINET into a number of communities according to the position and communication range of RSUs, and derive the capacity contributed by an RSU-based mode. Then, we calculate the capacity contributed by a vehicle-based mode based on fluid mechanics. Finally, the VINET capacity can be calculated. The proposed VINET capacity estimation approach is validated to be consistent with simulation results.

This work investigates an unmanned aerial vehicle (UAV) assisted IoT system, where a UAV flies to each foothold to collect data from IoT devices, and then return to its start point. For such a system, we aim to minimize the energy consumption by jointly optimizing the deployment and flight trajectory of UAV. It is a mixed-integer non-convex and NP-hard problem. In order to address it, a bilevel optimization approach is proposed, where an upper-level method aims to optimize the deployment of UAV and a lower-level one aims to plan UAV flight trajectory. Specifically, the former optimizes the number and locations of footholds of UAV. This work proposes an improved dandelion algorithm with a novel encoding strategy, in which each dandelion represents a foothold of UAV and the entire dandelion population is seen as an entire deployment. Then, two mutation strategies are designed to adjust the number and locations of footholds. Based on the footholds of the UAV provided by the former, the latter transforms flight trajectory planning into a traveling salesman problem (TSP). This work proposes an iterated greedy algorithm to solve it efficiently. The effectiveness of the proposed bilevel optimization approach is verified on ten instances, and the experimental results show that it significantly outperforms other benchmark approaches.

It is important to sufficiently guarantee an automotive system’s safety, because otherwise terrible consequences may happen. Generally the safety in an automotive system includes two aspects: reliability and timeliness. Previous studies have proposed many approaches to how to improve them. However, few of them consider the development cost along with their improvement. In this study, we aim to propose a method that can build a safety-guaranteed and development cost-minimized schedule for functionality modeled as a directed acyclic graph running on an automotive system. Unlike previous studies that tightly couple the development cost minimization with other requirements together, we start by building a schedule with the minimum development cost by ignoring safety requirement. Then, reliability and real-time requirements are subsequently taken into consideration. Together with automotive safety integrity level decomposition options provided by International Standard called ISO 26262, the decomposition is evaluated for each task to improve its safety, and tasks are then successively chosen to adjust the schedule, such that its safety can be maximized with incurring the least extra development cost. This procedure continues until a schedule that meets safety requirement is built. Experiments on a real-life automotive benchmark and extensive synthetic functionality demonstrate that our proposed heuristics outperform the state-of-the-art heuristic algorithm, and a typical intelligent optimization algorithm.

In modern logistics and smart warehouse, a terminal exists as a hub to connect multiple transportation modes and exchange goods. Solving a block relocation problem (BRP) arising from block retrieval processes in a terminal is fundamentally important to enhance the terminal’s overall efficiency and save its energy. In this paper, we improve the state-of-the-art mixed integer programming (MIP) formulation of an unrestricted BRP by extracting valid inequalities from structural properties and proposing a new variable fixation method to solve it. Computational results show that the improved model can be optimally solved much more easily than the original model. Among over five-hundred benchmark instances with height limits, the improved model can solve 12.3% more than the original one. For the instances that can be optimally solved by the latter, the improved model shows six times faster speed than the latter. This work represents a significant advance in this important area.

The real-time traffic control of railway networks imposes safety constraints and authorizes movements of trains. This work deals with it by focusing on conflict and deadlock avoidance problems. A place-timed Petri net model is developed for a railway network. Based on it, a conflict avoidance method is presented to ensure that safety principles in railway networks are respected. Also, a polynomial deadlock avoidance policy based on the Banker’s algorithm is proposed. It authorizes the movement of trains between block sections and stations. The proposed conflict avoidance method and deadlock avoidance policy together ensure a railway network to be conflict-free and deadlock-free. Experimental results show that they are effective and efficient.

A colored traveling salesman problem (CTSP) is a generalization of the well-known multiple traveling salesman problem. In it, each city has one to multiple colors and allows a salesman in the same color to visit exactly once. This work presents for the first time a CTSP whose edge weights among the cities change over time. It can be applied to dynamic routing problems arising in logistic distribution systems with various goods accessibilities to different types of vehicles. A non-linear integer mathematical program is constructed. A Variable Neighborhood Search (VNS) algorithm with a direct-route encoding and random initialization is then presented to solve it. To increase the convergence rate and population diversity of VNS, it is further combined with two-stage greedy initialization and an appropriate population immigrant scheme to perform the population search in a dynamic environment. Then, off-line performance evaluation and Wilcoxon test of the proposed algorithms are performed. The results show that their solution quality is improved by 30% ~ 60% over the basic VNS’s. They can track the environmental changes of CTSP-VEW more rapidly and effectively. In the end, a case study with a practical scenario is conducted.

It is critical to reduce the operating cost of shared buses for bus companies and improve the user experience of passengers. However, existing studies focus on either bus scheduling or route planning, which cannot accomplish the above mentioned goals concurrently. In this paper, we construct a joint bus scheduling and route planning framework to maximize the number of passengers, minimize the total length of routes and the number of required buses, as well as guarantee good user experience of passengers. First, we establish a system model based on a real-world scenario and formulate a multi-objective combinational optimization problem. Then, based on the extracted traffic topology of urban traffic networks and the generated candidate line set, we propose an offline algorithm to cope with the similar passenger flow distributions, e.g., morning or evening peak of every day. In order to cope with dynamic real-time passenger flows, an online algorithm is designed. Experiments are carried out based on real-word scenarios. The results show that the proposed algorithms can greatly reduce the operating cost of bus companies and guarantee good user experience based on real-world scheduling data in comparison with several existing methods.

We study a new competitive multi-facility location and quality design problem in a continuous space. The facility location and quality design are considered together because of their interdependence. Especially, new entrant facilities compete for customer demands with existing ones and the latter’s reactions are taken into account. The goal is to maximize the profit of all new entrant facilities by optimally determining their locations and qualities. For this problem, a probabilistic Huff-like gravity model is adopted to analyze the market share to be captured by new and existing facilities, and then a mathematical programming model is provided based on the market share analysis. Since it is shown to be strongly NP-hard, a new iterative solution framework is first proposed to solve it, where at each iteration, new configurations of facility locations are firstly generated, and then the quality decisions of all facilities are modelled as a competitive decision process by a non-cooperative game. The best qualities for new and existing facilities are determined by their Nash equilibrium. Finally, optimal or near-optimal solutions are calculated. Then based on the proposed solution framework, a particle swarm optimization-based approach is developed. Computational results for randomly generated instances indicate that the devised algorithm is able to find suitable locations and qualities of newly entering facilities in a competitive environment and outperforms favorably a genetic algorithm-based approach.

The relationship between flight performance and multi-physiological parameters under different flight operating patterns is unknown. This work proposes a Stacked Gaussian Process Network (SGPN) to reveal it. SGPN is a multi-layer network model formed by recursion from a regular Gaussian process and random disturbance. This work constructs an auxiliary variable strategy with the induced points to improve its learning efficiency, thus leading to a sparse SGPN model. In it, a Gaussian process acts as an activation function of each node, but the entire model is no longer a Gaussian process and thus very challenging to solve it. This work presents its solution via variational approximate inference. Experimental results of pilot flight performance evaluation show that the proposed model has stronger learning and generalization ability than its seven competitive peers. It is able to approximate non-linear coupling relationship between multi-physiological parameters and flight height differences.

A mobile edge computing (MEC) paradgim is evolving as an increasingly popular means for developing and deploying smart-city-oriented applications. MEC servers can receive a great deal of requests from devices of mobile users, especially in crowded scenes, e.g., a city’s central business district and school areas. It thus remains a great challenge for appropriate scheduling and managing strategies to avoid hotspots, guarantee load-fairness among MEC servers, and maintain high resource utilization at the same time. To address this challenge, we propose a coalitional-game-based and location-aware approach to MEC service migration for mobile user reallocation in crowded scenes. Our proposed method includes: 1) dividing MEC servers into multiple coalitions according to their inter-Euclidean distance by using a modified$k$-means clustering method; 2) discovering hotspots in every coalition area and scheduling services based on their corresponding cooperations; and 3) migrating services to appropriate edge servers to achieve high utilization and load-fairness among coalition members. Experimental results based on a real-world mobile trajectory dataset for crowded scenes, and an urban-edge-server-position dataset demonstrate that our method outperforms existing ones in terms of load fairness, number of migrations, and utilization rate of edge servers.

As a generalization of the well-known multiple traveling salesman problem, a colored traveling salesman problem (CTSP) utilizes colors to describe the accessibility of individual cities to salesmen. To expand its application scope, this work presents a bi-objective CTSP (BCTSP) over hypergraphs, taking into account the balance of workload among salesmen. To solve it, a bi-objective variable neighborhood search (BVNS) is proposed as a solution framework. In BVNS, we exploit a two-stage initialization and a probability-based insertion to produce feasible solutions and generate their neighborhood. Next, population-based multi-insertion, color-preserving exchange and 2-opt constitute a powerful local search procedure, where color-preserving exchange improves the solutions by modifying the route intersections among distinct salesmen. Besides, Delaunay triangulation is utilized to prepare candidates for multi-insertion, thereby increasing the possibility to find an optimal route by shortening links among vertices. Extensive experiments are conducted on 20 cases. To make a comprehensive comparison, four existing methods, i.e., two genetic algorithms and two variable neighborhood search methods, are adapted by exploiting an elitist non-dominated sorting for solving BCTSP instances. The experimental results show that BVNS is superior to its peers in achieving Pareto optima in terms of three popular performance metrics, i.e., hypervolume, inverted generational distance, and C-metric. In addition, the study of four BVNS variants reveals that probability-based insertion, population-based multi-insertion, and color-preserving exchange play significant roles in BVNS’s high performance.

A coloring traveling salesman problem (CTSP) generalizes the well-known multiple traveling salesman problem, where colors are used to differentiate salesmen’s the accessibility to individual cities to be visited. As a useful model for a variety of complex scheduling problems, CTSP is computationally challenging. In this paper, we propose a Multi-neighborhood Simulated Annealing-based Iterated Local Search (MSAILS) to solve it. Starting from an initial solution, it iterates through three sequential search procedures: a multi-neighborhood simulated annealing search to find a local optimum, a local search-enhanced edge assembly crossover to find nearby high-quality solutions around a local optimum, and a solution reconstruction procedure to move away from the current search region. Experimental results on two groups of 45 medium and large benchmark instances show that it significantly outperforms state-of-the-art algorithms. In particular, it is able to discover new upper bounds for 29 instances while matching 8 previous best-known upper bounds. Hence, this work greatly advances the field of CTSP.

Fuzzy Petri nets (FPNs) are a useful instrument for modelling expert systems to conduct knowledge representation and reasoning. Many studies have been carried out for improving the performance of FPNs in terms of their accurate representation of knowledge and power of approximate reasoning. Nevertheless, the current representation methods with FPNs are unable to handle the uncertain linguistic knowledge given by domain experts and the reliability of their judgments. In addition, the existing reasoning algorithms have no way to capture the interrelationship of the propositions with the same output transition. Therefore, we present a new type of FPNs, called 2-dimensional uncertain linguistic Petri nets (2DULPNs). The 2-dimensional uncertain linguistic variables (2DULVs) and Choquet integral are combined for knowledge representation and reasoning for the first time. The truth degrees of propositions, thresholds and certainty values of linguistic production rules are denoted as 2DULVs. Some new aggregated operators based on Choquet integral are proposed and used in the approximate reasoning to capture the interactions among antecedent propositions. Finally, an equipment fault diagnosis example is provided to illustrate the correctness and effectiveness of the proposed 2DULPN model.

Community detection is a popular yet thorny issue in social network analysis. A symmetric and nonnegative matrix factorization (SNMF) model based on a nonnegative multiplicative update (NMU) scheme is frequently adopted to address it. Current research mainly focuses on integrating additional information into it without considering the effects of a learning scheme. This study aims to implement highly accurate community detectors via the connections between an SNMF-based community detector's detection accuracy and an NMU scheme's scaling factor. The main idea is to adjust such scaling factor via a linear or nonlinear strategy, thereby innovatively implementing several scaling-factor-adjusted NMU schemes. They are applied to SNMF and graph-regularized SNMF models to achieve four novel SNMF-based community detectors. Theoretical studies indicate that with the proposed schemes and proper hyperparameter settings, each model can: 1) keep its loss function nonincreasing during its training process and 2) converge to a stationary point. Empirical studies on eight social networks show that they achieve significant accuracy gain in community detection over the state-of-the-art community detectors.

This work proposes a decision tree (DT)-based method for initializing a dendritic neuron model (DNM). Neural networks become larger and larger, thus consuming more and more computing resources. This calls for a strong need to prune neurons that do not contribute much to their network's output. Pruning those with low contribution may lead to a loss of accuracy of DNM. Our proposed method is novel because 1) it can reduce the number of dendrites in DNM while improving training efficiency without affecting accuracy and 2) it can select proper initialization weight and threshold of neurons. The Adam algorithm is used to train DNM after its initialization with our proposed DT-based method. To verify its effectiveness, we apply it to seven benchmark datasets. The results show that decision-tree-initialized DNM is significantly better than the original DNM, k-nearest neighbor, support vector machine, back-propagation neural network, and DT classification methods. It exhibits the lowest model complexity and highest training speed without losing any accuracy. The interactions among attributes can also be observed in its dendritic neurons.

Recent decades have witnessed a trend that control-theoretical techniques are widely leveraged in various areas, e.g., design and analysis of computational models. Computational methods can be modeled as a controller and searching the equilibrium point of a dynamical system is identical to solving an algebraic equation. Thus, absorbing mature technologies in control theory and integrating it with neural dynamics models can lead to new achievements. This work makes progress along this direction by applying control-theoretical techniques to construct new recurrent neural dynamics for manipulating a perturbed nonstationary quadratic program (QP) with time-varying parameters considered. Specifically, to break the limitations of existing continuous-time models in handling nonstationary problems, a discrete recurrent neural dynamics model is proposed to robustly deal with noise. This work shows how iterative computational methods for solving nonstationary QP can be revisited, designed, and analyzed in a control framework. A modified Newton iteration model and an improved gradient-based neural dynamics are established by referring to the superior structural technology of the presented recurrent neural dynamics, where the chief breakthrough is their excellent convergence and robustness over the traditional models. Numerical experiments are conducted to show the eminence of the proposed models in solving perturbed nonstationary QP.

Text classification is a fundamental and important area of natural language processing for assigning a text into at least one predefined tag or category according to its content. Most of the advanced systems are either too simple to get high accuracy or centered on using complex structures to capture the genuinely required category information, which requires long time to converge during their training stage. In order to address such challenging issues, we propose a dynamic embedding projection-gated convolutional neural network (DEP-CNN) for multi-class and multi-label text classification. Its dynamic embedding projection gate (DEPG) transforms and carries word information by using gating units and shortcut connections to control how much context information is incorporated into each specific position of a word-embedding matrix in a text. To our knowledge, we are the first to apply DEPG over a word-embedding matrix. The experimental results on four known benchmark datasets display that DEP-CNN outperforms its recent peers.

Many scientific applications can be well modelled as large-scale workflows. Cloud computing has become a suitable platform for hosting and executing them. Workflow scheduling has gained much attention in recent years. However, since cloud service providers must offer services for multiple users with various QoS demands, scheduling multiple applications with different QoS requirements is highly challenging. This work proposes a Multi-swarm Co-evolution-based Hybrid Intelligent Optimization (MCHO) algorithm for multiple-workflow scheduling to minimize total makespan and cost while meeting the deadline constraint of each workflow. First, we design a multi-swarm co-evolutionary mechanism where three swarms are adopted to sufficiently search for various elite solutions. Second, to improve global search and convergence performance, we embed local and global guiding information into the updating process of a Particle Swarm Optimizer, and develop a swarm cooperation technique. Third, we propose a Genetic Algorithm-based elite enhancement strategy to exploit more non-dominated individuals, and apply the Metropolis Acceptance rule of Simulated Annealing to update the local guiding solution for each swarm so as to prevent it from being stuck into a local optimum at an early stage. Extensive experimental results demonstrate that MCHO outperforms the state-of-art scheduling algorithms with better distributed non-dominated solutions.

Detecting and understanding the differences among process models is important for business improvement. Most of the existing work in analysing the differences between two process models employs an edit script approach, i.e., using a sequence of edit operations that transform one to another by applyingdeleteorinsertoperations. However, describing process differences this way is hard for users to understand and interpret. To overcome the problem, we propose a pattern-based method for process difference detection namedProDiff. We specify a set of process difference patterns as Single-Entry-Single-Exit (SESE) fragments of a process model. Process differences are detected by decomposing process models into different levels of SESE fragments, based on whichProDifflocates the positions of differences and provides assistance for users to carry out further analysis. A case study is provided to show the effectiveness and extensibility of the proposed method.

Today's large-scale parallel workflows are often processed on heterogeneous distributed computing platforms. From an economic perspective, computing resource providers should minimize the cost while offering high service quality. It has become well-recognized that energy consumption accounts for a large part of a computing system's total cost, and timeliness and reliability are two important service indicators. This work studies the problem of scheduling a parallel workflow that minimizes the system energy consumption under the constraints of response time and reliability. We first mathematically formulate this problem as a Non-linear Mixed Integer Programming problem. Since this problem is hard to solve directly, we present some highly-efficient heuristic solutions. Specifically, we first develop an algorithm that minimizes the schedule length while meeting reliability requirement, on top of which we propose a processor-merging algorithm and a slack time reclamation algorithm using a dynamic voltage frequency scaling (DVFS) technique to reduce energy consumption. The processor-merging algorithm tries to turn off some energy-inefficient processors such that energy consumption can be minimized. The DVFS technique is applied to scale down the processor frequency at both processor and task levels to reduce energy consumption. Experimental results on two real-life workflows and extensive synthetic parallel workflows demonstrate their effectiveness.

A non-negative latent factor (NLF) model with a single latent factor-dependent, non-negative and multiplicative update (SLF-NMU) algorithm is frequently adopted to extract useful knowledge from non-negative data represented by high-dimensional and sparse (HiDS) matrices arising from various service-oriented applications. However, its convergence rate is slow. To address this issue, this study proposes aGeneralized Nesterov's acceleration-incorporated,Non-negative andAdaptiveLatentFactor (GNALF) model. It results from a) incorporating a generalized Nesterov's accelerated gradient (NAG) method into an SLF-NMU algorithm, thereby achieving anNAG-incorporated andelement-orientednon-negative (NEN) algorithm to perform efficient parameter update; and b) making its regularization and acceleration parameters self-adaptive via incorporating the principle of a particle swarm optimization algorithm into the training process, thereby implementing a highly adaptive and practical model. Empirical studies on six large sparse matrices from different recommendation service applications show that a GNALF model achieves very high convergence rate without the need of hyper-parameter tuning, making its computational efficiency significantly higher than state-of-the-art models. Meanwhile, such efficiency gain does not result in accuracy loss, since its prediction accuracy is comparable with its peers. Hence, it can better serve practical service applications with real-time demands.

Dynamic relationships are frequently encountered in big data and services computing-related applications, like dynamic data of user-side QoS in Web services. They are modeled into a high-dimensional and sparse (HiDS) tensor, which contain rich knowledge regarding temporal patterns. A non-negative latent factorization of tensors (NLFT) model is very effective in extracting such patterns from an HiDS tensor. However, it commonly suffers from overfitting with improper regularization schemes. To address this issue, this article investigates NLFT models with diversified regularization schemes. Six regularized NLFT models, i.e.,$L_{2}, L_{1}$, elastic net, log, dropout, and swish-regularized ones, are proposed and carefully investigated. Moreover, owing to their diversified regularization designs, they possess strong model diversity to achieve an effective ensemble. Empirical studies on HiDS QoS tensors from real applications demonstrate that compared with state-of-the-art models, the proposed ones better describe the temporal patterns hidden in an HiDS tensor, thereby achieving significantly higher prediction accuracy for missing data. Moreover, their ensemble further outperforms each of them in terms of prediction accuracy for missing QoS data.

Mobile computing systems, service-based systems, and some other systems with mobile interacting components have recently received much attention. However, because of their characteristics, such as mobility and disconnection, it is difficult to model and analyze them by using a structure-fixed model. This work proposes a new Petri net model called variable Petri net (VPN) for modeling and analyzing these systems. The definition, firing rule, and related analysis technology of VPN are introduced in detail. In a VPN, the possible interaction interfaces are abstracted as a new kind of places called virtual places, and the occurrences of (dis)connections are described by new functions, which makes it appropriate to describe the component collaboration in systems and realize the scalability and pluggability of systems. Moreover, to overcome the shortcoming that markings cannot reflect the link capability of a system, VPNs add a constraint function along with a marking to represent a complete system configuration. Several examples are used to demonstrate the newly proposed model and method.

Disassembly is an essential step in a remanufacturing process via which valuable parts and material of end-of-life (EOL) products can be well reused and resource waste is reduced. Disassembly sequence planning focuses on finding the best disassembly sequence for a given EOL product by considering economic and environmental performance. In a practical disassembly process, one may face a disassembly operation failure risk due to the difficulty of knowing EOL products’ exact information in advance. Despite its importance in impacting disassembly outcomes, the existing work fails to consider it comprehensively. This work proposes a stochastic biobjective DSP problem with the objectives of maximizing disassembly profit and minimizing energy consumption by doing so. A chance-constrained programming model is established, where a chance constraint ensures a fixed confidence level of disassembly failure. To solve it efficiently, a multiobjective multiverse optimization algorithm with stochastic simulation is proposed. Experiments are carried out on four products. Results demonstrate that it outperforms some state-of-the-art algorithms in terms of solution performance.

Most of the techniques involved in customer clustering and segmentation are based on conventional methods of quantitative analysis or traditional data mining approaches such as the K-Means algorithm. However, clustering approaches based on artificial neural networks (ANNs), evolutionary algorithms, and fuzzy methods can be more efficient since they can reveal nonlinear patterns. They also seem to be more robust in coping with noise-related issues and relevant noise handling operations. They do not make any statistical distributional assumptions regarding the nature of the data. In this article, we develop a hybrid approach based on ANNs and swarm intelligence to reveal the underlying pattern structure of customers of an insurance company in the Republic of Ireland. This model is tailored to the scope of segmenting administrative districts, or “small areas,” given policyholders’ spatial characteristics. To that end, the geospatial features of customers are taken into account. Geodemographically speaking, by implementing such a hybrid model, the relative similarity among spatial objects (small areas in this work) are preserved. In this way, the similarity of each small area to all other small areas is characterized. Consequently, the pattern of customers is analyzed using an optimal and intelligent solution. We can also visualize the results of this study.

A Dandelion algorithm (DA) inspired by the seed dispersal process of dandelions has been proposed as a newly intelligent optimization algorithm. For improving its exploration ability as well as reducing the probability of its falling into a local optimum, this work proposes to add a novel competition mechanism with historical information feedback to current DA. Specifically, the fitness value of each dandelion in the next generation, which is calculated by linear prediction, is compared with the current best dandelion, and the loser is replaced by a new offspring. Current DA generates new offsprings without considering historical information. This work improves its offspring generation process by exploiting historical information with an estimation-of-distribution algorithm. Three historical information models are designed. They are best, worst, and hybrid historical information feedback models. The experimental results show that the proposed algorithms outperform DA and its variants, and the proposed algorithms are superior or competitive to nine participating algorithms benchmarked on 28 functions from CEC2013. Finally, the proposed algorithms demonstrate the effectiveness on four real-world problems, and the results indicate that the proposed algorithms have better performance than its peers.

Feature selection has been considered as an effective method to solve imbalanced classification problems. It can be formulated as a multiobjective optimization problem (MOP) aiming to find a small feature subset while achieving a high classification accuracy. With traditional MOP, the focus is on deriving an optimal solution (i.e., a feature subset), while ignoring the diversity in solution space (e.g., there could exist multiple feature subsets achieving the same accuracy). Providing more options for feature selection would be beneficial since some features can be more difficult to obtain than others. In this work, we treat feature selection as a multimodal MOP (MMOP) whose goals are to find an excellent Pareto front in objective space and as many equivalent Pareto optimal solutions (feature subsets) as possible in feature space. Note that though several multimodal multiobjective evolutionary algorithms (MMEAs) have been proposed, their use of a convergence-first selection criterion could cause the loss of solution diversity in an objective and feature space. To address the issue, a novel competition-driven mechanism is designed to assist the existing multimodal MMEAs in locating more equivalent feature subsets and a desired Pareto front. The effectiveness of the proposed mechanism is first verified on all 22 MMOPs from CEC2019. Then, the proposed method is applied to feature selection in imbalanced classification problems and a real-world application, i.e., credit card fraud detection. Experimental results show that the proposed mechanism can not only provide more equivalent feature subsets but also improve classification accuracy.

In recent decades, deadlock control for automated manufacturing systems has been an active area. Most researchers have assumed that allocated resources, such as sensors, actuators, and controllers never fail. However, this case is not prevalent in practice due to the unexpected failure of resources. Thus, the objective of robust control is presented in this article. Several methods have been developed along this direction, such as methods that combine neighborhood constraints and the modified Banker’s algorithm, as well as methods based on critical places. To explore their effectiveness and performance, we not only conduct a comparison investigation but also develop new theoretical results. According to the experimental results, critical place-based approaches are simpler, more efficient, and more comprehensive than the Banker’s algorithm-based approaches in response to resource failures. This article is motivated by the control of production Petri nets; however, the results are also applicable to other more complex systems.

Online streaming feature selection (OSFS) has attracted extensive attention during the past decades. Current approaches commonly assume that the feature space of fixed data instances dynamically increases without any missing data. However, this assumption does not always hold in many real applications. Motivated by this observation, this study aims to implement online feature selection from sparse streaming features, i.e., features flow in one by one with missing data as instance count remains fixed. To do so, this study proposes a latent-factor-analysis-based online sparse-streaming-feature selection algorithm (LOSSA). Its main idea is to apply latent factor analysis to pre-estimate missing data in sparse streaming features before conducting feature selection, thereby addressing the missing data issue effectively and efficiently. Theoretical and empirical studies indicate that LOSSA can significantly improve the quality of OSFS when missing data are encountered in target instances.

A growing number of organizations are hosting their software applications in distributed data centers (DCs) in the cloud, for faster response time and higher energy efficiency. The dramatic increase of user tasks, however, poses a significant challenge on DC providers to retain users’ expectations on both aspects. To tackle this challenge, this work first formulates the problem into a constrained biobjective optimization problem. A biobjective algorithm, named simulated-annealing-based adaptive differential evolution (SADE), is presented to simultaneously reduce both the response time of tasks and energy cost. Meanwhile, a method of minimal Manhattan distance is adopted to search for a final knee, for achieving a good balance between response time minimization and energy cost reduction. Experimental results on real-life datasets, i.e., the electricity prices and tasks collected from a Google cluster trace, have proved that SADE yields less task response time and lower energy cost compared with state-of-the-art algorithms.

Current large-scale green cloud data centers (GCDCs) tend to consume a huge amount of energy and generate enormous carbon emissions. Existing studies have tried to solve this problem by either realizing prediction of green energy, or optimizing task scheduling. In contrast, this work seamlessly combines green energy prediction and task scheduling to jointly optimize revenue and energy cost of GCDCs. Specifically, this work designs a prediction method, named Savitzky-Golay and Long Short-Term Memory network (SG-LSTM), to realize noise filtering and forecast green energy. Based on such prediction, a bi-objective optimization method, named Decomposition-based Multi-objective evolutionary algorithm with Gaussian mutation and Crowding distance (DMGC), is developed to optimize the revenue and energy cost of GCDCs. Its performance is demonstrated over real-life datasets including Google cluster traces, wind speeds, solar irradiance and prices of electricity. Experimental results show that SG-LSTM outperforms its two peers, back propagation neural network and gated recurrent unit, in terms of root mean square errors and mean absolute errors. In addition, DMGC surpasses its such peers as NSGA-II, SPEA2, and MOEA/D in terms of revenue, energy cost and average execution time. Particularly, DMGC's revenue is 18%, 20% and 13.1% higher, energy cost is 16%, 19.8% and 15.2% lower, and average execution time is 60.02%, 38.47% and 24.17% lower than those of NSGA-II, SPEA2, and MOEA/D, respectively.

Since a single private cloud cannot satisfy the increasing computational requirements for executing multiple workflows simultaneously, hybrid clouds are often adopted to perform such execution. Multi-workflow scheduling is challenging as users may request various applications with different QoS requirements. This work proposes a Chaotic-nondominated-sorting Owl Search Algorithm (COSA) by combining an Owl Search Algorithm (OSA) with a Nondominated Sorting Genetic Algorithm II (NSGA-II) to schedule resource-constrained multiple workflows in hybrid clouds with makespan, cost and energy consumption minimized under the given deadline and budget constraints. First, a hierarchical evolving mechanism is designed to update the better half and worse half of population by NSGA-II and OSA, respectively to guarantee a good trade-off between exploration and exploitation. Second, a chaotic sequence is introduced to adaptively adjust OSA's step size during population evolution for better exploration. Third, we adopt a chaotic operator for searching around the resulting Non-Dominated Solutions (NDS) to improve COSA's local search ability. Experiments are conducted to compare COSA with four peers and the results show its superiority in the number of obtained NDS, diversity preservation and convergence towards the near optimal Pareto set. In particular, it can find at least 19% more NDS than its peers.

User-side Quality-of-Service (QoS) data are vital for efficient cloud service selection, while available QoS data in real applications are commonly described by a high-dimensional and Sparse (HiDS) matrix due to a) increasing users and services, and b) the impossibility of observing the full invoking mapping among users and services. A latent factor analysis (LFA) model has proven to be efficient in performing representation learning on such an HiDS QoS matrix. However, existing LFA models commonly adopt a first-order optimizer that cannot makes it approach the second-order stationary point of the learning objective, thereby resulting in accuracy loss. Aiming at addressing this issue, this study proposes to incorporate a Generalized Nesterov's Acceleration (GNA) method in to a Hessian-vector algorithm for LFA, thereby establishing a GNA-incorporated Hessian-vector-based LFA (GNHL) model with two-fold ideas: a) adopting the principle of a Hessian-vector method to acquire a proper Newton step efficiently, and b) incorporating a GNA method into its linear search for accelerating its convergence rate. Experimental results on six real QoS datasets demonstrate that a GNHL model outperforms state-of-the-art LFA models in generating highly accurate predictions for missing QoS data with low computational burden.

A Particle Swarm Optimizer (PSO) and mobile robot swarm are two widely studied subjects. Many applications emerge separately while the similarity between them is rarely explored. When a solution space is a certain region in reality, a robot swarm can replace a particle swarm to explore the optimal solution by performing PSO. In this way, a mobile robot swarm should be able to efficiently explore an area just like a particle swarm and uninterruptedly work even under the shortage of robots or in the case of unexpected failure of robots. Furthermore, the moving distances of robots are highly constrained because energy and time of robots can be costly. Inspired by such requirements, this presentation discusses a Moving-distance-minimized PSO for a mobile robot swarm to minimize the total moving distance of its robots while performing optimization and collaboration. The distances between the current robot positions and the particle ones in the next generation are utilized to derive paths for robots such that the total distance that all robots move is minimized, hence minimizing the energy and time for a robot swarm to locate the optima. Experimental results on optimizing 28 CEC2013 benchmark functions show the advantage of the proposed method over the standard PSO. By adopting it, the moving distance of robots can be reduced by more than 40% while offering the same optimization effects. The implication is enormous since all population-based optimization algorithms can be potentially benefited from such replacement of their individuals with mobile robots, thus leading to their moving-distanceminimized variants.

A stochastic-gradient-descent-based Latent Factor Analysis (LFA) model is highly efficient in representative learning of a High-Dimensional and Sparse (HiDS) matrix. Its learning rate adaptation is vital in ensuring its efficiency. Such adaptation can be realized with an evolutionary computing algorithm. However, a resultant model tends to suffer from two issues: a) the pre-mature convergence of the swarm of learning rates as caused by an adopted evolution algorithm, and b) the pre-mature convergence of the LFA model as caused jointly by evolution-based learning rate adaptation and an optimization algorithm. This paper focuses on the methods to address such issues. A Hierarchical Particle-swarm-optimization-incorporated Latent factor analysis (HPL) model with a two-layered structure is proposed, where the first layer pre-trains desired latent factors with a position-transitional particle-swarm-optimization-based LFA model, and the second layer performs latent factor refining with a newly-proposed mini-batch particle swarm optimizer. With such design, an HPL model can well handle the pre-mature convergence, which is supported by the positive experimental results achieved on HiDS matrices from industrial applications.

Protein-protein interactions are of great significance for us to understand the functional mechanisms of proteins. With the rapid development of high-throughput genomic technology, the amount of protein-protein interaction data has become so big that most of existing prediction algorithms are no longer applicable. To address this problem, we develop a distributed framework by reimplementing one of state-of-the-art algorithms, i.e., CoFex, by using MapReduce. In particular, we adopt a novel tree-based data structure to reduce the heavy memory consumption cased by the huge sequence information of proteins. After that, the procedure of CoFex is modified by following the paradigm of MapReduce such that the prediction task can be completed in a distributed manner, thus fulfilling the demanding requirements of large-scale protein-protein interaction prediction. A series of experiments have been conducted to evaluate the performance of the proposed distributed framework in terms of both efficiency and effectiveness. Experimental results demonstrate that the proposed framework can considerably improve the efficiency of CoFex by achieving more than two-orders-of-magnitude improvement in computational efficiency while retaining a comparable level of accuracy.

Pruning is widely used for neural network model compression. It removes redundant links from a weight tensor to lead to smaller and more efficient neural networks for system implementation. A compressed neural network can enable faster run and reduced computational cost in network training. In this paper, a novel pruning method is proposed for a dendritic neuron model (DNM). It calculates the significance of each DNM dendrite. The calculated significance is expressed numerically and a dendrite whose significance is lower than a pre-set threshold is removed. Experimental results verify that it obtains superior performance over the existing one in terms of both accuracy and computational efficiency.

A big-data-related application like a Terminal Interaction Pattern Analysis System (TIPAS) commonly involves massive entities interacting with each other dynamically. Such interactions can be represented by a Dynamically Weighted Directed Network (DWDN). A large number of involved entities results in a high-dimensional and incomplete (HDI) network, which can be represented by an HDI tensor with numerous missing entries. In spite of its HDI nature, this tensor contains much useful knowledge regarding various desired patterns like unobserved links in DWDN. However, due to its extremely high dimensionality and low data density, it is very challenging to build a learning model that can precisely represent an HDI tensor. To address this issue, this work proposes a Neural Latent Factorization of Tensors (NeuLFoT) model with three interesting ideas: a) adopting the principle of density-oriented modeling and Canonical Polyadic tensor factorization to build rank-one tensor series relying on three-dimensional latent factors for precisely representing an HDI tensor’s known data; b) treating the obtained rank-one tensors as neurons to form a novel neural tensor network model; and c) proposing a novel Backward Propagation algorithm for Latent factorization of tensors (BPL) to ensure high training efficiency. Experimental results on two large-scale DWDNs generated by a real TIPAS demonstrate that compared with state-of-the-art models, the proposed model achieves significant gain in prediction accuracy for missing links of a DWDN and achieves highly competitive computational efficiency.

Transactions in blockchain networks can be transmitted through channels. Existing methods for channel transaction transmission depend on transaction routing information. If a node randomly chooses a channel to transmit a transaction, the transmission may be aborted due to insufficient funds (also called balance) or low transmission rate. To increase transmission success rate and reduce transmission delay across all transactions, this work proposes a transaction transmission model for blockchain channels based on non-cooperative game theory. The model considers the channel balance, states, and nodes’ channel preferences. Steady-state transmission rates of selected channels are analyzed. The ideal solution for the model is found.

Smart mobile devices (SMDs) can meet users' high expectations by executing computational intensive applications but they only have limited resources, including CPU, memory, battery power, and wireless medium. To tackle this limitation, partial computation offloading can be used as a promising method to schedule some tasks of applications from resource-limited SMDs to high-performance edge servers. However, it brings communication overhead issues caused by limited bandwidth and inevitably increases the latency of tasks offloaded to edge servers. Therefore, it is highly challenging to achieve a balance between high-resource consumption in SMDs and high communication cost for providing energy-efficient and latency-low services to users. This work proposes a partial computation offloading method to minimize the total energy consumed by SMDs and edge servers by jointly optimizing the offloading ratio of tasks, CPU speeds of SMDs, allocated bandwidth of available channels, and transmission power of each SMD in each time slot. It jointly considers the execution time of tasks performed in SMDs and edge servers, and transmission time of data. It also jointly considers latency limits, CPU speeds, transmission power limits, available energy of SMDs, and the maximum number of CPU cycles and memories in edge servers. Considering these factors, a nonlinear constrained optimization problem is formulated and solved by a novel hybrid metaheuristic algorithm named genetic simulated annealing-based particle swarm optimization (GSP) to produce a close-to-optimal solution. GSP achieves joint optimization of computation offloading between a cloud data center and the edge, and resource allocation in the data center. Real-life data-based experimental results prove that it achieves lower energy consumption in less convergence time than its three typical peers.

Existing work on the Internet-of-Vehicles (IoV) community mainly focuses on the detection of IoV community using static network detection and evolution methods of complex networks. These methods are prone to over-centralization, high computational complexity, and poor stability during the evolution of a community. In this work, we present an IoV community model and its evolution mechanism in an urban scene. More specifically, we first propose an IoV community detection model based on node similarity merging. Then, we use a network increment-based strategy to analyze node increment, edge increment, and weight increment. Finally, we give a dynamic evolution mechanism of an IoV community. Simulation-based experimental evaluation results show that the proposed mechanism achieves better real-time performance and accuracy than existing methods.

One of the important Internet-of-Things applications is to use image and video to realize automatic people monitoring, surveillance, tracking, and reidentification (Re-ID). Despite some recent advances, pedestrian Re-ID remains a challenging task. Existing algorithms based on fully supervised learning for it usually requires numerous labeled image and video data, while often ignoring the problem of data imbalance. This work proposes a method based on unlabeled samples generated by cycle generative adversarial networks. For a newly generated unlabeled sample, it learns its pseudorelationship between unlabeled samples and labeled ones in a low-dimensional space by using a self-paced learning approach. Then, these unlabeled ones having pseudo-relationship with labeled ones are added in a training set to better mine discriminative information between positive and negative samples, which is in turn used to learn a more effective metric. We name this method as a semi-supervised learning approach based on the built pseudopairwise relations between labeled data and unlabeled one. It can greatly enhance the performance of pedestrian Re-ID in case of insufficient labeled images. By using only about 10% labeled images in a given database, the proposed method obtains higher accuracy than state-of-the-art supervised learning methods using all labeled ones, e.g., deep-learning ones, thus greatly advancing the field of pedestrian Re-ID.

This work investigates the problem of undersampled face recognition (i.e., insufficient training data) encountered in practical Internet-of-Things (IoT) applications. Insufficient and uncertain samples captured by IoT devices may include background and facial disguise that makes face recognition more challenging than that with sufficient and reliable images. Many models work well in face recognition on a big data set, but when training data are insufficient, they achieve unsatisfactory performance. This work proposes a novel method named sparse common feature-based representation (SCFR) that provides a unique and stable result and completely avoids very time-consuming training required by a deep learning model. Specially, it constructs a common feature dictionary using both training and test images. Thereinto, a common feature is based on a discriminative common vector and learned by a Gaussian mixture model for both training and test images in a semisupervised learninig manner, which would reduce the difference among samples in each class. In the optimization, the latent indicator of test data is initialized by the estimated label. This can avoid learning invalid information and lead to good prototype images. A new variation dictionary characterizes variables that can be shared by different classes. Finally, this work adopts minimum reconstruction residuals to recognize test images, thus bringing about a substantial improvement in SCFR's performance. Extensive results on benchmark face databases demonstrate that the proposed method is better than the state-of-the-art methods handling undersampled face recognition.

The performance of face recognition has been greatly improved by deep neural network algorithms when a dataset is large. However, when face data are insufficient as in practical Internet of Things (IoT) applications and captured by IoT devices under the same intrasubject variation, both data quantity and quality bring big challenges to construct a model or representation, and most of the time it becomes infeasible to build a deep neural network model. This work proposes a sparse individual low-rank component-based representation (SILR) such that the representation of testing images can be based on individual subjects’ low-rank component. Theoretically, we put the$l_{2}$-norm constraint on intrasubject coefficients to represent testing images, thus making intrasubject coefficients dense. Hence, we alleviate the impact of an undersampled training dataset and its same intersubject variation on classification performance. We solve a convex minimization problem in polynomial time via an augmented lagrange multiplier scheme to get the solution of SILR. The scheme can reduce the influences from the same intersubject variation and contribute to an accurate recognition of the undersampled training dataset. We adopt sparse individual low-rank component representation and minimum reconstruction residual to recognize testing images. Extensive results on various databases show that SILR outperforms the other state-of-the-art methods for face recognition.

Mobile-edge computing provides an effective approach to reducing the workload of smart devices and the network delay induced by data transfer through deploying computational resources in the proximity of the devices. In a mobile-edge computing system, it is of great importance to improve the quality of experience of users and reduce the deployment cost for service providers. This article investigates a joint cloudlet deployment and task offloading problem with the objectives of minimizing energy consumption and task response delay of users and the number of deployed cloudlets. Since it is a multiobjective optimization problem, a set of tradeoff solutions ought to be found. After formulating this problem as a mixed-integer nonlinear program and proving its NP-completeness, we propose a modified guided population archive whale optimization algorithm to solve it. The superiority of our devised algorithm over other methods is confirmed through extensive simulations.

A style-based generative adversarial network (StyleGAN2) yields remarkable results in image-to-latent embedding. This work proposes a Detached Dual-channel Domain Encoder as an effective and robust method to embed an image to a latent code, i.e., GAN inversion. It infers a latent code from two aspects: a) a detached dual-channel design to support faithful image reconstruction; and b) a local skip connection that allows conveying pieces of information with image details. We further introduce a hierarchical progressive training strategy that allows the proposed encoder to separately capture different semantic features. The qualitative and quantitative experimental results show that the well-trained encoder can embed an image into a latent code in StyleGAN2 latent space with less time than its peers while preserving facial identity and image details well.

Scheduling of complex manufacturing systems entails complicated constraints such as the mating operational one. Focusing on the real settings, this article considers an extended version of a flexible job shop problem that allows the precedence between the operations to be given by an arbitrary directed acyclic graph instead of a linear order. In order to obtain its reliable and high-performance schedule in a reasonable time, this article contributes a knowledge-based cuckoo search algorithm (KCSA) to the scheduling field. The proposed knowledge base is initially trained off-line on models before operations based on reinforcement learning and hybrid heuristics to store scheduling information and appropriate parameters. In its off-line training phase, the algorithm SARSA is used, for the first time, to build a self-adaptive parameter control scheme of the CS algorithm. In each iteration, the proposed knowledge base selects suitable parameters to ensure the desired diversification and intensification of population. It is then used to generate new solutions by probability sampling in a designed mutation phase. Moreover, it is updated via feedback information from a search process. Its influence on KCSA's performance is investigated and the time complexity of the KCSA is analyzed. The KCSA is validated with the benchmark and randomly generated cases. Various simulation experiments and comparisons between it and several popular methods are performed to validate its effectiveness.

Industrial products' reuse, recovery, and recycling are very important due to the exhaustion of ecological resources. Effective product disassembly planning methods can improve the recovery efficiency and reduce harmful impact on the environment. However, the existing approaches pay little attention to disassembly resources, such as tools and operators that can significantly influence the optimal disassembly sequences. This article considers a multiobjective resource-constrained disassembly optimization problem modeled with timed Petri nets such that energy consumption is minimized, while disassembly profit is maximized. Since its solution complexity has exponential growth with the number of components in a product, a multiobjective genetic algorithm based on an external archive is used to solve it. Its effectiveness is verified by comparing it with nondominated sorting genetic algorithm II and a collaborative resource allocation strategy for a multiobjective evolutionary algorithm based on decomposition. Note to Practitioners-This article establishes a novel dual-objective optimization model for product disassembly subject to multiresource constraints. In an actual disassembly process, a decision-maker may want to minimize energy consumption and maximize disassembly profit. This article considers both objectives and proposes a multiobjective genetic algorithm based on an external archive to solve optimal disassembly problems. The experimental results show that the proposed approach can solve them effectively. The obtained solutions give decision-makers multiple choices to select the right disassembly process when an actual product is disassembled.

Workflow management technology has become a key means to improve enterprise productivity. More and more workflow systems are crossing organizational boundaries and may involve multiple interacting organizations. This article focuses on a type of loosely coupled workflow architecture with collaborative tasks, i.e., each business partner owns its private business process and is able to operate independently, and all involved organizations need to be synchronized at a certain point to complete certain public tasks. Because of each organization’s privacy consideration, they are unwilling to share the business details with others. In this way, traditional correctness verification approaches via reachability analysis are not practical as a global business process model is unavailable for privacy preservation. To ensure its globally correct execution, this work establishes a correctness verification approach for the cross-organizational workflow with task synchronization patterns. Its core idea is to use local correctness of each suborganizational workflow process to guarantee its global correctness. We prove that the proposed approach can be used to investigate the behavioral property preservation when synthesizing suborganizational workflows via collaborative tasks. A medical diagnosis running case is used to illustrate the applicability of the proposed approaches.Note to Practitioners—Cross-organizational workflow verification techniques play an increasingly important role in ensuring the correct execution of collaborative enterprise businesses. This work addresses the issue of correctness verification for loosely coupled interactive workflows with collaborative tasks. To ensure the globally correct execution, a behavioral correctness verification approach is established. All proposed concepts and techniques are supported by open-source tools, and evaluation over a medical diagnosis process case has shown their applicability. The proposed methodology is readily applicable to industrial-size workflow correctness verification problems.

A nonnegative latent factorization of tensors (NLFT) model precisely represents the temporal patterns hidden in multichannel data emerging from various applications. It often adopts a single latent factor-dependent, nonnegative and multiplicative update on tensor (SLF-NMUT) algorithm. However, learning depth in this algorithm is not adjustable, resulting in frequent training fluctuation or poor model convergence caused by overshooting. To address this issue, this study carefully investigates the connections between the performance of an NLFT model and its learning depth via SLF-NMUT to present a joint learning-depth-adjusting scheme for it. Based on this scheme, a Depth-adjusted Multiplicative Update on tensor algorithm is innovatively proposed, thereby achieving a novel depth-adjusted nonnegative latent-factorization-of-tensors (DNL) model. Empirical studies on two industrial data sets demonstrate that compared with the state-of-the-art NLFT models, a DNL model achieves significant accuracy gain when performing missing data estimation on a high-dimensional and incomplete tensor with high efficiency.Note to Practitioners—Multichannel data are often encountered in various big-data-related applications. It is vital for a data analyzer to correctly capture the temporal patterns hidden in them for efficient knowledge acquisition and representation. This article focuses on analyzing temporal QoS data, which is a representative kind of multichannel data. To correctly extract their temporal patterns, an analyzer should correctly describe their nonnegativity. Such a purpose can be achieved by building a nonnegative latent factorization of tensors (NLFT) model relying on a single latent factor-dependent, nonnegative and multiplicative update on tensor (SLF-NMUT) algorithm. But its learning depth is not adjustable, making an NLFT model frequently suffer from severe fluctuations in its training error or even fail to converge. To address this issue, this study carefully investigates the learning rules for an NLFT model’s decision parameters using an SLF-NMUT and proposes a joint learning-depth-adjusting scheme. This scheme manipulates the multiplicative terms in SLF-NMUT-based learning rules linearly and exponentially, thereby making the learning depth adjustable. Based on it, this study builds a novel depth-adjusted nonnegative latent-factorization-of-tensors (DNL) model. Compared with the existing NLFT models, a DNL model better represents multichannel data. It meets industrial needs well and can be used to achieve high performance in data analysis tasks like temporal-aware missing data estimation

This work investigates a scheduling problem of deadlock-prone flexible manufacturing systems modeled by place-timed Petri nets. It proposes an anytime branch and bound (AB&B) algorithm for it to minimize system makespan based on the branch tree of a net model and a highly permissive deadlock controller. The proposed algorithm searches a sequence of transitions in the branch tree that evolves the model from the initial marking to the final one. In order to prune the branch tree and increase search speed, this work develops two pruning rules, a lower bound of makespan, and a novel branching strategy. Their usage ensures AB&B’s high search efficiency. Experimental results demonstrate that the proposed algorithm surpasses the state-of-the-art ones.Note to Practitioners—In practice, scheduling is one of the most important issues for production managers to address. Existing approaches to deadlock-prone flexible manufacturing system (FMS) scheduling have multiparameters. Their settings can impact the scheduling results greatly. Since their turning is a challenging task, it is difficult to find the best parameter settings. This article proposes an anytime branch and bound (AB&B) algorithm for minimizing the makespan of deadlock-prone FMSs. It has only one parameter, i.e., maximal CPU time, which needs no turning. When it is highly tight, AB&B can output a feasible and decent schedule. If it is larger and larger, AB&B can output a better and better schedule. Comparison studies show that AB&B outclasses existing ones significantly. It is suitable for real-time scheduling cases in which satisfied schedules must be offered in a very short time.

An opportunistic link between two mobile devices or nodes can be constructed when they are within each other’s communication range. Typically, cyber–physical environments consist of a number of mobile devices that are potentially able to establish opportunistic contacts and serve mobile applications in a cost-effective way. Opportunistic mobile service computing is a promising paradigm capable of utilizing the pervasive mobile computational resources around the users. Mobile users are thus allowed to exploit nearby mobile services to boost their computing capabilities without investment in their resource pool. Nevertheless, various challenges, especially its quality-of-service and reliability-aware scheduling, are yet to be addressed. Existing studies and related scheduling strategies consider mobile users to be fully stable and available. In this article, we propose a novel method for reliability-aware and deadline-constrained service composition over opportunistic networks. We leverage the Krill–Herd-based algorithm to yield a deadline-constrained, reliability-aware, and well-executable service composition schedule based on the estimation of completion time and reliability of schedule candidates. We carry out extensive case studies based on some well-known mobile service composition templates and a real-world opportunistic contact data set. The comparison results suggest that the proposed approach outperforms existing ones in terms of success rate and completion time of composed services.Note to Practitioners—Recently, the rapid development of mobile devices and mobile communication leads to the prosperity of mobile service computing. Services running on mobile devices within a limited range are allowed to be composed to coordinate through wireless communication technologies and perform complex tasks and business processes. Despite its great potential, mobile service compositions remains a challenge since the mobility of users and devices imposes high unpredictability on the execution of tasks. A careful investigation into existing methods has found their various limitations, e.g., assuming time-invariant availability of mobile services. This article presents a novel reliability-aware and deadline-constrained service composition method for mobile opportunistic networks. Instead of assuming time-invariant availability of mobile nodes, the proposed method is capable of estimating service availability at run-time and leveraging a Krill–Herd-based algorithm to yield the deadline-constrained, reliability-aware, and well-executable service composition schedules. Case studies based on well-known service composition templates and real-world data sets suggest that it outperforms traditional ones in terms of success and completion time of composed services. It can thus aid the design and optimization of composite services as well as their smooth execution in a mobile environment. It can help practitioners better manage the reliability and performance of real-world applications built upon mobile services.

This article deals with improved formal verification of reconfigurable discrete-event systems (DESs) modeled by reconfigurable timed net condition event systems (R-TNCESs). An R-TNCES consists of a set of timed net condition event systems, each of which represents a particular behavior of a DES, and a reconfiguration scenario is a switching mode from a timed net condition event system to another. However, the verification with the classical computation tree logic (CTL) as well as the related extensions increases the number of properties for complete verification of a complex R-TNCES. We propose reconfigurable CTL as a new extension of CTL to reduce such a number. New connectors of reconfigurable CTL are proposed, with their formal syntax and semantics, and a set of new algorithms is proposed to control the complexity of model checking. We use a benchmark production system for the performance evaluation of the proposed approach. Reduction in the number of properties to be checked is shown, and consequently, the related validation time is reduced.Note to Practitioners—This research represents a new orientation for guiding efficiently the model checking of reconfigurable discrete-event systems. A classification of properties described in computation tree logic (CTL), according to their dominance and equivalence relations, allows one to conduct an efficient verification by avoiding inefficient calculation due to redundant properties. In this case, giving a verification order for these properties allows one to shorten their verification time. An extension named reconfigurable CTL describes the new syntax of the proposed classification. This approach can be applied in modeling and verification of advanced reconfigurable systems arising from smart grids, adaptive sensor networks, intelligent transportation, reconfigurable manufacturing, and embedded systems.

In a steel industry, a slab yard plays a role of a buffer between continuous casting stage and rolling mill. An effective assignment and scheduling of cranes can guarantee the operation efficiency in the slab yard. This work studies a multicrane scheduling problem with noncrossing constraints of slabs. A mixed-integer programming model is used to formulate the problem that minimizes the whole traveling distance of all the cranes and ensures the workload balance among cranes. As it is an NP-hard problem, classical programming mathematical methods are difficult to get an optimal solution for large-size instances. Thus, we develop a branch and price algorithm to solve this problem. First, we formulate the model as a generalized set covering problem and a set partition problem. Then, we solve them and combine the solutions to obtain the solution of the original problem. Finally, we conduct computational experiments based on real data from an iron–steel plant. The comparisons of proposed methods with an exact solution method show its effectiveness.Note to Practitioners—This work deals with a multicrane scheduling problem. Aiming to minimize the total traveling distance of all the cranes, it establishes a mixed-integer programming model with a workload balance constraint on cranes. It presents a branch and price algorithm to solve the problem whose solution complexity grows exponentially with problem size. The integration of crane assignment and scheduling enables the better utilization of cranes and faster service in iron–steel enterprises and, hence, improving customer satisfaction. The experimental results reveal the effectiveness of the proposed approach. It can readily be put into use in the steel industry.

The industry of data centers is the fifth largest energy consumer in the world. Distributed green data centers (DGDCs) consume 300 billion kWh per year to provide different types of heterogeneous services to global users. Users around the world bring revenue to DGDC providers according to actual quality of service (QoS) of their tasks. Their tasks are delivered to DGDCs through multiple Internet service providers (ISPs) with different bandwidth capacities and unit bandwidth price. In addition, prices of power grid, wind, and solar energy in different GDCs vary with their geographical locations. Therefore, it is highly challenging to schedule tasks among DGDCs in a high-profit and high-QoS way. This work designs a multiobjective optimization method for DGDCs to maximize the profit of DGDC providers and minimize the average task loss possibility of all applications by jointly determining the split of tasks among multiple ISPs and task service rates of each GDC. A problem is formulated and solved with a simulated-annealing-based biobjective differential evolution (SBDE) algorithm to obtain an approximate Pareto-optimal set. The method of minimum Manhattan distance is adopted to select a knee solution that specifies the Pareto-optimal task service rates and task split among ISPs for DGDCs in each time slot. Real-life data-based experiments demonstrate that the proposed method achieves lower task loss of all applications and larger profit than several existing scheduling algorithms. Note to Practitioners-This work aims to maximize the profit and minimize the task loss for DGDCs powered by renewable energy and smart grid by jointly determining the split of tasks among multiple ISPs. Existing task scheduling algorithms fail to jointly consider and optimize the profit of DGDC providers and QoS of tasks. Therefore, they fail to intelligently schedule tasks of heterogeneous applications and allocate infrastructure resources within their response time bounds. In this work, a new method that tackles drawbacks of existing algorithms is proposed. It is achieved by adopting the proposed SBDE algorithm that solves a multiobjective optimization problem. Simulation experiments demonstrate that compared with three typical task scheduling approaches, it increases profit and decreases task loss. It can be readily and easily integrated and implemented in real-life industrial DGDCs. The future work needs to investigate the real-time green energy prediction with historical data and further combine prediction and task scheduling together to achieve greener and even net-zero-energy data centers.

The significant growth in the number and types of tasks of heterogeneous applications in green cloud data centers (GCDCs) dramatically increases their providers' revenue from users as well as energy consumption. It is a big challenge to maximize such revenue, while minimizing energy cost in a market where prices of electricity, availability of renewable power generation, and behind-the-meter renewable generation contract models differ among the geographical sites of the GCDCs. A multiobjective optimization method that investigates such spatial differences in the GCDCs is for the first time proposed to trade off such two objectives by cost-effectively executing all tasks while meeting their delay constraints. In each time slot, a constrained biobjective optimization problem is formulated and solved by an improved multiobjective evolutionary algorithm based on decomposition. Realistic data-based simulations prove that the proposed method achieves a larger total profit in faster convergence speed than the two state-of-the-art algorithms. Note to Practitioners-This article considers the tradeoff between profit maximization and energy cost minimization for the green cloud data center (GCDC) providers while meeting the delay constraints of all tasks. Current task-scheduling methods fail to take the advantage of spatial variations in many factors, e.g., prices of electricity and availability of renewable power generation at geographically distributed GCDC locations. As a result, they fail to execute all tasks of heterogeneous applications within their delay bounds in a high-revenue and low-energy-cost manner. In this article, a multiobjective optimization method that addresses the disadvantages of the existing methods is proposed. It is realized by a proposed intelligent optimization algorithm. Simulations demonstrate that in comparison with the two state-of-the-art scheduling algorithms, the proposed one increases the profit and reduces the convergence time. It can be readily implemented and integrated into actual industrial GCDCs.

Edge computing is a new architecture to provide computing, storage, and networking resources for achieving the Internet of Things. It brings computation to the network edge in close proximity to users. However, nodes in the edge have limited energy and resources. Completely running tasks in the edge may cause poor performance. Cloud data centers (CDCs) have rich resources for executing tasks, but they are located in places far away from users. CDCs lead to long transmission delays and large financial costs for utilizing resources. Therefore, it is essential to smartly offload users’ tasks between a CDC layer and an edge computing layer. This work proposes a cloud and edge computing system, which has a terminal layer, edge computing layer, and CDC layer. Based on it, this work designs a profit-maximized collaborative computation offloading and resource allocation algorithm to maximize the profit of systems and guarantee that response time limits of tasks are strictly met. In each time slot, this work jointly considers CPU, memory, and bandwidth resources, load balance of all heterogeneous nodes in the edge layer, maximum amount of energy, maximum number of servers, and task queue stability in the CDC layer. Considering the abovementioned factors, a single-objective constrained optimization problem is formulated and solved by a proposed simulated-annealing-based migrating birds optimization procedure to obtain a close-to-optimal solution. The proposed method achieves joint optimization of computation offloading between CDC and edge, and resource allocation in CDC. Realistic data-based simulation results demonstrate that it realizes higher profit than its peers.Note to Practitioners—This work considers the joint optimization of computation offloading between Cloud data center (CDC) and edge computing layers, and resource allocation in CDC. It is important to maximize the profit of distributed cloud and edge computing systems by optimally scheduling all tasks between them given user-specific response time limits of tasks. It is challenging to execute them in nodes in the edge computing layer because their computation resources and battery capacities are often constrained and heterogeneous. Current offloading methods fail to jointly optimize computation offloading and resource allocation for nodes in the edge and servers in CDC. They are insufficient and coarse-grained to schedule arriving tasks. In this work, a novel algorithm is proposed to maximize the profit of distributed cloud and edge computing systems while meeting response time limits of tasks. It explicitly specifies the task service rate and the selected node for each task in each time slot by considering resource limits, load balance requirement, and processing capacities of nodes in the edge, and server and energy constraints in CDC. Real-life data-driven simulations show that the proposed method realizes a larger profit than several typical offloading strategies. It can be readily implemented and incorporated into large-scale industrial computing systems.

Non-negativity is vital for a latent factor (LF)-based model to preserve the important feature of a high-dimensional and sparse (HiDS) matrix in recommender systems, i.e., none of its entries is negative. Current non-negative models rely on constraints-combined training schemes. However, they lack flexibility, scalability, or compatibility with general training schemes. This work aims to perform unconstrained non-negative latent factor analysis (UNLFA) on HiDS matrices. To do so, we innovatively transfer the non-negativity constraints from the decision parameters to the output LFs, and connect them through a single-element-dependent mapping function. Then we theoretically prove that by making a mapping function fulfill specific conditions, the resultant model is able to represent the original one precisely. We subsequently design highly efficient UNLFA algorithms for recommender systems. Experimental results on four industrial-size HiDS matrices demonstrate that compared with four state-of-the-art non-negative models, a UNLFA-based model obtains advantage in prediction accuracy for missing data and computational efficiency. Moreover, such high performance is achieved through its unconstrained training process which is compatible with various general training schemes, on the premise of fulfilling non-negativity constraints. Hence, UNLFA algorithms are highly valuable for industrial applications with the need of performing non-negative latent factor analysis on HiDS matrices.

Different from multiobjective optimization problems (MOPs), multimodal MOPs (MMOPs) focus on both decision and objective spaces rather than only objective one. Thus, finding a good Pareto front approximation and finding the maximal number of equivalent Pareto optimal solutions for each objective vector in the Pareto front are two core tasks for them. Although some multimodal multiobjective evolutionary algorithms have been proposed to handle them, they can quickly converge to the easy-to-find equivalent Pareto optimal solutions, thereby losing their ability to improve solution diversity in decision space and performance in objective space. To address the above issues, this work proposes a new information utilization method. Its core idea is to randomly extract a certain amount of decision variable information from the current optimal solutions to construct an information vector, which is, in turn, used to assist the generation of elite solutions. The proposed method can assist any available intelligent optimizers to improve their performance in solving MMOPs. This is confirmed by experimental results obtained from solving 22 such problems from CEC2019 and 12 scalable imbalanced distance minimization problems through a number of optimizers. Finally, we apply the proposed method to credit card fraud detection problems to show its practical significance.

As big-data-driven complex systems, commercial recommendation systems (RSs) have been widely used in such companies as Amazon and Ebay. Their core aim is to maximize total profit, which relies on recommendation accuracy and profits from recommended items. It is also important for them to treat new items equally for a long-term run. However, traditional recommendation techniques mainly focus on recommendation accuracy and suffer from a cold-start problem (i.e., new items cannot be recommended). Differing from them, this work designs a multiobjective RS by considering item profit and novelty besides accuracy. Then, a hybrid probabilistic multiobjective evolutionary algorithm (MOEA) is proposed to optimize these conflicting metrics. In it, some specifically designed genetic operators are proposed, and two classical MOEA frameworks are adaptively combined such that it owns their complementary advantages. The experimental results reveal that it outperforms some state-of-the-art algorithms as it achieves a higher hypervolume value than them.

A model can be easily influenced by unseen factors in nonstationary environments and fail to fit dynamic data distribution. In a classification scenario, this is known as a concept drift. For instance, the shopping preference of customers may change after they move from one city to another. Therefore, a shopping website or application should alter recommendations based on its poorer predictions of such user patterns. In this article, we propose a novel approach called the multiscale drift detection test (MDDT) that efficiently localizes abrupt drift points when feature values fluctuate, meaning that the current model needs immediate adaption. MDDT is based on a resampling scheme and a paired student t -test. It applies a detection procedure on two different scales. Initially, the detection is performed on a broad scale to check if recently gathered drift indicators remain stationary. If a drift is claimed, a narrow scale detection is performed to trace the refined change time. This multiscale structure reduces the massive time of constantly checking and filters noises in drift indicators. Experiments are performed to compare the proposed method with several algorithms via synthetic and real-world datasets. The results indicate that it outperforms others when abrupt shift datasets are handled, and achieves the highest recall score in localizing drift points.

This paper investigates a finite-frequency H-/H∞fault detection method for discrete-time T-S fuzzy systems with unmeasurable premise variables. To minimize the effect of uncertainties on system performance and maximize that of actuator faults on the generated residual, both the H∞disturbance attenuation index and finite-frequency H-fault sensitivity index are utilized. Since the premised variables are unmeasurable, the existing generalized Kalman-Yakubovich-Popov lemma cannot be directly extended to these nonlinear systems. In this paper, the conditions of allowing one to design the proposed H-/H∞fault detection observer are established and transformed into linear matrix inequalities. Some scalars and slack matrices are introduced to bring extra degrees of freedom in observer design. Finally, a single-link robotic manipulator model is utilized to illustrate that the proposed technique can detect faults with smaller amplitude than that required by a normal H∞observer technique.

G-images refer to image data defined on irregular graph domains. This article elaborates on a similarity-preserving FuzzyC-Means (FCM) algorithm for G-image segmentation and aims to develop techniques and tools for segmenting G-images. To preserve the membership similarity between an arbitrary image pixel and its neighbors, a Kullback–Leibler divergence term on partition matrix is introduced as a part of FCM. As a result, similarity-preserving FCM is developed by considering spatial information of image pixels for its robustness enhancement. Due to superior characteristics of a wavelet space, the proposed FCM is performed in this space rather than the Euclidean one used in conventional FCM to secure its high robustness. Experiments on synthetic and real-world G-images demonstrate that it indeed achieves higher robustness and performance than the state-of-the-art segmentation algorithms. Moreover, it requires less computation than most of them.

In this article, we develop a residual-sparse FuzzyC-Means (FCM) algorithm for image segmentation, which furthers FCM's robustness by realizing the favorable estimation of the residual (e.g., unknown noise) between an observed image and its ideal version (noise-free image). To achieve a sound tradeoff between detail preservation and noise suppression, morphological reconstruction is used to filter the observed image. By combining the observed and filtered images, a weighted sum image is generated. Tight wavelet frame decomposition is used to transform the weighted sum image into its corresponding feature set. Taking such feature set as data for clustering, we impose an$\ell _0$regularization term on residual to FCM's objective function, thus resulting in residual-sparse FCM, where spatial information is introduced for improving its robustness and making residual estimation more reliable. To further enhance segmentation accuracy of the proposed FCM, we employ morphological reconstruction to smoothen the labels generated by clustering. Finally, based on the prototypes and smoothed labels, a segmented image is reconstructed by using tight wavelet frame reconstruction. Experimental results regarding synthetic, medical, and real-world images show that the proposed algorithm is effective and efficient, and outperforms its peers.

The conventional fuzzy C-means (FCM) algorithm is not robust to noise and its rate of convergence is generally impacted by data distribution. Consequently, it is challenging to develop FCM-related algorithms that have good performance and require less computing time. In this article, we elaborate on a comprehensive FCM-related algorithm for image segmentation. To make FCM robust, we first utilize a morphological grayscale reconstruction (MGR) operation to filter observed images before clustering, which guarantees noise-immunity and image detail-preservation. Since real images can generally be approximated by sparse coefficients in a tight wavelet frame system, feature spaces of observed and filtered images can be obtained. Taking such features to be clustered, we investigate an improved FCM model in which a sparse regularization term is introduced into the objective function of FCM. We design a three-step iterative algorithm to solve the sparse regularization-based FCM model, which is constructed by the Lagrangian multiplier method, hard-threshold operator, and normalization operator, respectively. Such an algorithm can not only perform well for image segmentation, but also come with high computational efficiency. To further enhance the segmentation accuracy, we use MGR to filter the label set generated by clustering. Finally, a large number of supporting experiments and comparative studies with other FCM-related algorithms available in the literature are provided. The obtained results for synthetic, medical and color images indicate that the proposed algorithm has good ability for multiphase image segmentation, and performs better than other alternative FCM-related algorithms. Moreover, the proposed algorithm requires less time than most of the existing algorithms.

In past decades, we witnessed many applications and fast development of cloud computing technologies. Cloud faults are encountered in a cloud computing environment. They badly impact users and cause serious economic losses in business. As a vital technology, fault detection can guarantee a high reliability cloud environment. However, fault detection with a fixed detection cycle has defects and shortcomings. On the one hand, for the service with good performance, if a small cycle is set, it may need a lot of system overhead due to unnecessary over detection; on the other hand, for the service with poor performance, if a large cycle is set, it may result in the omission of faults which should be detected. To address these issues, in this paper, a fault detection model is proposed to improve the detection accuracy based on support vector machine and a decision tree. For abnormal samples, their abnormality is calculated by using the model. We design algorithms to adaptively and dynamically adjust cycles for fault detection. The cycle is shortened if a system experiences many faults, thus increasing fault detection success rate; it is lengthened if the system runs without any problem, thereby reducing much computational overhead. Experimental results show that the proposed method outperforms two classical methods, i.e., one based on self-organizing competitive neutral network and the other based on a probabilistic neural network.

A style-based architecture (StyleGAN2) yields outstanding results in data-driven unconditional generative image modeling. This work proposes a Domain-guided Noise-optimization-based Inversion (DNI) method to perform facial image manipulation. It works based on an inverse code that includes: 1) a novel domain-guided encoder called Image2latent to project the image to StyleGAN2 latent space, which can reconstruct an input image with high-quality and maintain its semantic meaning well; 2) a noise optimization mechanism in which a set of noise vectors are used to capture the high-frequency details such as image edges, further improving image reconstruction quality; and 3) a mask for seamless image fusion and local style migration. We further propose a novel semantic alignment evaluation pipeline. It evaluates the semantic alignment with an inverse code by using different attribute boundaries. Extensive qualitative and quantitative comparisons show that DNI can capture rich semantic information and achieve a satisfactory image reconstruction. It can realize a variety of facial image manipulation tasks and outperform state of the art.

Achieving the high accuracy of blind roads and crosswalks recognition is important for blind guiding equipment to help blind people sense the surrounding environment. A lightweight semantic segmentation network is proposed to quickly and accurately segment blind roads and crosswalks in a complex road environment. Specifically, a lightweight network with depthwise separable convolution as a component is used as a basic module to reduce the number of parameters of the model and increase the speed of semantic segmentation. In order to ensure the segmentation accuracy of the network, we use a densely connected atrous spatial pyramid pooling module to extract feature information of different angles and context feature modules to enhance the effectiveness of different levels of feature information fusion. To verify the effectiveness of the proposed method, we collect and produce a data set from a real environment, which contains two objects of blind roads and crosswalks1. Experimental results demonstrate that, compared to some state-of-the-art approaches, the proposed approach greatly improves the segmentation speed, while achieving better or similar accuracy, which shows that the proposed approach provides a better basis for the application of devices for guiding the blind.1https://github.com/qweawq/Blind-road-and-crosswalk-dataset

Pedestrian re-identification (Re-ID) is a very challenging and unavoidable problem in the field of multi-camera surveillance in smart transportation. Among many ways to solve this problem, keep it simple and straightforward (KISS) metric learning (KISSME) stands out since it has unbeatable advantages in running time while maintaining highly acceptable matching rate. It can be used to realize effective pedestrian Re-ID in an open world. Although it has achieved highly acceptable performance in some applications, it encounters a small sample size (S3) problem that causes too small eigenvalues of its covariance matrix, thus resulting in an instability issue. Its large eigenvalues are overestimated; while its small ones are underestimated. In order to solve this problem, we use an orthogonal basis vector to generate virtual samples to overcome the S3problem. The resulting algorithm named KISS+ is experimentally shown to have the eigenvalues of its covariance matrix significantly larger than those of the original KISSME. In order to show its advantage in pedestrian Re-ID, this work uses multi-feature fusion to extract more discriminant features, and obtain a low-dimensional expression of features through dimension reduction. Experiments based on several well-known databases show that our method can improve the matching rate, while maintaining the advantage of fast computation. Compared with deep learning algorithms, our algorithm does not achieve their matching rate, but it is highly suitable for real-time pedestrian Re-ID of an open world due to its simplicity, easy operation and fast execution.

This paper presents an arc-routing problem with time-dependent penalty cost (ARPTPC), which arises from a practical application in garbage collection service. ARPTPC considers the minimization of service cost, traveling cost and penalty cost. While the first two parts are known as the traditional objectives of arc-routing problems, the third part is determined by the parking pattern and service period on each arc. We formulate the problem by using a mixed integer linear model. To solve it, we design a dynamic programming to determine the optimal service beginning time on each edge when a routing sequence is given. We then propose a problem-specific intelligent heuristic search approach involving six neighborhood operators, a priority maintenance mechanism and a perturbation process. Through numerical experiments, we demonstrate that the proposed approach is able to produce satisfactory solutions of ARPTPC. Additional experiments are also carried out to analyze the effects of operators and parameters on solution quality.

Location-Based Services (LBS), which answer users’ location-dependent queries to Points of Interest (POI), have become popular along with mobile devices’ widespread use. While benefiting from convenience, users may suffer privacy leak risk, due to publishing sensitive information, i.e., locations, to servers. Previous studies have proposed a number of methods for protecting LBS. Through providing provable privacy protection, Private Information Retrieval (PIR) becomes well-known for its high effectiveness in protecting privacy. However, the PIR-protected LBS must cautiously handle the consequent transmission/computation cost and service error brought by adopting PIR, as both of them impact user experience. Then an important question arises: Can the user experience of PIR-protected LBS be greatly improved? We answer it by presenting a novel interface from PIR to LBS, gaining benefits from road networks. The proposed interface is comprised of two road partition methods fully considering POI’s distribution along roads, and two response-calculating methods to satisfy users’ requirements on driving distance. Since Vehicular Location Based Services contain adequate road network knowledge, we accomplish our work on them. The experimental results on a real dataset validate that our interface can improve PIR-protected LBS greatly in user experience while protecting privacy.

This work investigates the problem of optimally locating an automotive service firm (ASF) subject to stochastic customer demands, varying setup cost and regional constraints. The goal is to minimize the transportation cost while maintaining the specified profit of the ASF. This work studies two variants of the problem: ASF location with known demand probability distributions and with partial demand information, i.e., only the support and mean of the customer demands are known. For the former, a chance-constrained program is formulated that improves an existing model, and then an equivalent deterministic nonlinear program is constructed based on our property analysis results. For the latter, a novel distribution-free model is developed. The proposed models are solved by solver LINGO. Computational results on the benchmark examples show that: i) for the first variant, the proposed approach outperforms the existing one; ii) for the second one, the proposed distribution-free model can effectively handle stochastic customer demands without complete probability distributions; and iii) the results of the distribution-free model are slightly worse than those of the deterministic nonlinear one, but the former is more cost-efficient for the practical ASF location as it is less expensive in obtaining demand information. Moreover, the proposed models and approaches are extended to address a multi-ASF location allocation under demand uncertainty.

A colored traveling salesman problem (CTSP) is a generalization of the well-known multiple traveling salesman problem. It utilizes colors to differentiate the accessibility of its cities to its salesmen. In our prior work, CTSPs are formulated over graphs associated with a city-color matrix. This work redefines a general colored traveling salesman problem (GCTSP) in the framework of hypergraphs and reveals several important properties of GCTSP. In GCTSP, the setting of city colors is richer than that in CTSPs. As results, it can be used to model and address various complex scheduling problems. Then, a Delaunay-triangulation-based Variable Neighborhood Search (DVNS) algorithm is developed to solve large-scale GCTSPs. At the beginning stage of DVNS, a divide and conquer algorithm is exploited to prepare a Delaunay candidate set for lean insertion. Next, the incumbent solution is perturbed by utilizing greedy multi-insertion and exchange mutation to obtain a variety of neighborhoods. Subsequently, 2-opt and 3-opt are used for local search in turn. Extensive experiments are conducted for many large scale GCTSP cases among which two maximal ones are up to 33000+ cities for 4 salesmen and 240 salesmen given 11000+ cities, respectively. The results show that the proposed method outperforms the existing four genetic algorithms and two VNS methods in terms of search ability and convergence rate.

Snake-like robots move flexibly in complex environments due to their multiple degrees of freedom and various gaits. However, their existing 3-D models are not accurate enough, and most gaits are applicable to special environments only. This work investigates a 3-D model and designs hybrid 3-D gaits. In the proposed 3-D model, a robot is considered as a continuous beam system. Its normal reaction forces are computed based on the mechanics of materials. To improve the applicability of such robots to different terrains or tasks, this work designs hybrid 3-D gaits by mixing basic gaits in different parts of their bodies. Performances of hybrid gaits are analyzed based on extensive simulations. These gaits are compared with traditional gaits including lateral undulation, rectilinear, and sidewinding ones. Results of simulations and physical experiments are presented to demonstrate the performances of the proposed model and hybrid gaits of snake-like robots.

In the field of computer vision, without sufficient labeled images, it is challenging to train an accurate model. However, through visual adaptation from source to target domains, a relevant labeled dataset can help solve such problem. Many methods apply adversarial learning to diminish cross-domain distribution difference. They are able to greatly enhance the performance on target classification tasks. Generative adversarial network (GAN) loss is widely used in adversarial adaptation learning methods to reduce an across-domain distribution difference. However, it becomes difficult to decline such distribution difference if generator or discriminator in GAN fails to work as expected and degrades its performance. To solve such cross-domain classification problems, we put forward a novel adaptation framework called generative adversarial distribution matching (GADM). In GADM, we improve the objective function by taking cross-domain discrepancy distance into consideration and further minimize the difference through the competition between a generator and discriminator, thereby greatly decreasing cross-domain distribution difference. Experimental results and comparison with several state-of-the-art methods verify GADM's superiority in image classification across domains.

A continuous stirred-tank reactor (CSTR) system is widely applied in wastewater treatment processes. Its control is a challenging industrial-process-control problem due to great difficulty to achieve accurate system identification. This work proposes a deep learning-based model predictive control (DeepMPC) to model and control the CSTR system. The proposed DeepMPC consists of a growing deep belief network (GDBN) and an optimal controller. First, GDBN can automatically determine its size with transfer learning to achieve high performance in system identification, and it serves just as a predictive model of a controlled system. The model can accurately approximate the dynamics of the controlled system with a uniformly ultimately bounded error. Second, quadratic optimization is conducted to obtain an optimal controller. This work analyzes the convergence and stability of DeepMPC. Finally, the DeepMPC is used to model and control a second-order CSTR system. In the experiments, DeepMPC shows a better performance in modeling, tracking, and antidisturbance than the other state-of-the-art methods.

A growing number of global companies select Green Data Centers (GDCs) to manage their delay-constrained applications. The fast growth of users’ tasks dramatically increases the energy consumed by GDCs owned by a company, e.g., Google and Amazon. The random nature of tasks brings a big challenge of scheduling tasks of each application with limited infrastructure resources of GDCs. Therefore, hybrid cloud is widely employed to smartly outsource some tasks to public clouds. However, the temporal variation in many factors including revenue, price of power grid, solar irradiance, wind speed, price of public clouds makes it challenging to schedule all tasks of each application in a cost-effective way while strictly meeting their expected delay constraints. This work proposes a temporal task scheduling algorithm investigating the temporal variation in green hybrid cloud to schedule all tasks within their delay constraints. Besides, it explicitly presents a mathematical equation of service rates and task refusal. The maximization problem is formulated and tackled by the proposed hybrid optimization algorithm called Genetic Simulated-annealing-based particle swarm optimization. Trace-driven experiments demonstrate that larger profit are achieved than several existing scheduling algorithms.

This article focuses on the problem of deadlock for sequential automated manufacturing systems (AMSs) that allow for the general resource allocation and flexible routings. A class of Petri nets, systems of sequential systems with shared resources (S4PR), are used to model these considered AMSs. Our previous work has showed that deadlocks in S4PR are characterized by saturated perfect activity-circuit (PA-circuit). In this article, we divide all saturable PA-circuits into two categories: 1) dependent and 2) independent. An algorithm is proposed to compute all independent saturable PA-circuits. We prove that by adding a monitor for each independent PA-circuit to ensure that it is not saturated, all dependent PA-circuits cannot be saturated either and deadlocks in S4PR are successfully prevented. The presented method simplifies the structure of the deadlock controller without imposing tight constraints on the system. Finally, the proposed controller is illustrated by some examples.

The Internet of Things (IoT) is the latest example of the System of Systems (SoS), demanding for both innovative and evolutionary approaches to tame its multifaceted aspects. Over the years, different IoT methodologies, frameworks, platforms, and tools have been proposed by industry and academia, but the jumbled abundance of such development products have resulted into a high (and disheartening) entry-barrier to IoT system engineering. In this survey, we steer IoT developers by: 1) providing baseline definitions to identify the most suitable class of development products-methodologies, frameworks, platforms, and tools-for their purposes and 2) reviewing seventy relevant products through a comparative and practical approach, based on general SoS engineering features revised in the light of main IoT systems desiderata (i.e., interoperability, scalability, smartness, and autonomy). Indeed, we aim to lessen the confusion related to IoT methodologies, frameworks, platforms, and tools as well as to freeze their current state, for eventually easing the approach towards IoT system engineering.

JADE is a differential evolution (DE) algorithm and has been shown to be very competitive in comparison with other evolutionary optimization algorithms. However, it suffers from the premature convergence problem and is easily trapped into local optima. This article presents a novel JADE variant by incorporating chaotic local search (CLS) mechanisms into JADE to alleviate this problem. Taking advantages of the ergodicity and nonrepetitious nature of chaos, it can diversify the population and thus has a chance to explore a huge search space. Because of the inherent local exploitation ability, its embedded CLS can exploit a small region to refine solutions obtained by JADE. Hence, it can well balance the exploration and exploitation in a search process and further improve its performance. Four kinds of its CLS incorporation schemes are studied. Multiple chaotic maps are individually, randomly, parallelly, and memory-selectively incorporated into CLS. Experimental and statistical analyses are performed on a set of 53 benchmark functions and four real-world optimization problems. Results show that it has a superior performance in comparison with JADE and some other state-of-the-art optimization algorithms.

A rare event, such as a natural disaster, does not occur frequently, but may cause catastrophic impacts on human beings and their living environment once it happens. Social media provides people an immediate and convenient way to share their opinions. Researchers can thus use it for investigating people's actions, feelings, and attitudes during multiple rare events. By using social media data, many studies try to build the connection between rare events in a real world and people's responses, including feelings, attitudes, and behaviors, in a virtual world. In this article, we propose a feature extraction and text classification approach to Twitter text data related to a rare event, e.g., Hurricane Sandy. First, a novel feature extraction method is proposed to mine useful features from each text message. Next, a fuzzy logic-based classification method is proposed to distinguish event-related and unrelated messages. Finally, our proposed method is compared with the existing keyword search one that is widely used in analyzing the evolution of a rare event. The result reveals that the proposed approach is suitable to identify the rare event-related and unrelated text messages.

High-dimensional and sparse (HiDS) matrices generated by recommender systems contain rich knowledge regarding various desired patterns like users' potential preferences and community tendency. Latent factor (LF) analysis proves to be highly efficient in extracting such knowledge from an HiDS matrix efficiently. Stochastic gradient descent (SGD) is a highly efficient algorithm for building an LF model. However, current LF models mostly adopt a standard SGD algorithm. Can SGD be extended from various aspects in order to improve the resultant models' convergence rate and prediction accuracy for missing data? Are such SGD extensions compatible with an LF model? To answer them, this paper carefully investigates eight extended SGD algorithms to propose eight novel LF models. Experimental results on two HiDS matrices generated by real recommender systems show that compared with an LF model with a standard SGD algorithm, an LF model with extended ones can achieve: 1) higher prediction accuracy for missing data; 2) faster convergence rate; and 3) model diversity.

Non-negative latent factor (NLF) models well represent high-dimensional and sparse (HiDS) matrices filled with non-negative data, which are frequently encountered in industrial applications like recommender systems. However, current NLF models mostly adopt Euclidean distance in their objective function, which represents a special case of a β-divergence function. Hence, it is highly desired to design a β-divergence-based NLF ( β-NLF) model that uses a β-divergence function, and investigate its performance in recommender systems as β varies. To do so, we first model β-NLF's learning objective with a β-divergence function. Subsequently, we deduce a general single latent factor-dependent, non-negative and multiplicative update scheme for β-NLF, and then design an efficient β-NLF algorithm. The experimental results on HiDS matrices from industrial applications indicate that by carefully choosing the value of β, β-NLF outperforms an NLF model with Euclidean distance in terms of accuracy for missing data prediction without increasing computational time. The research outcomes show the necessity of using an optimal β-divergence function in order to achieve the best performance of an NLF model on HiDS matrices. Hence, the proposed model has both theoretical and application significance.

A kind of facilities for wafer fabrication, cluster tools (CTs) need to close down to an idle state from time to time because of periodical maintenance and switches from one type of lots to another, which is called a normal close-down process (NCDP). It is crucial to optimize such a transient process since it tends to occur more and more frequently due to customization. Also, process modules (PMs) in CTs are known to be failure-prone. Once a PM failure occurs, a tool needs to close down to an idle state as well, which is different from NCDP and is called a failure close-down process (FCDP). With wafer residency time constraints (WRTCs) being imposed, close-down process optimization for such a tool is challenging, since one needs to not only finish this process as soon as possible but also meet WRTCs during this transient process. In order to tackle this problem, this article first introduces steady state scheduling problems. Then, with a presented backward robot task sequence, a linear programming model is first proposed to optimize NCDP. To deal with the PM failures, efficient PM failure response policies are formulated for the cases in which a PM fails. Then, four linear programs are proposed to optimize an FCDP. Finally, industrial case studies are given to show the usefulness of the proposed approaches.

In semiconductor manufacturing, a process module (PM) failure in cluster tools (CTs) happens from time to time. To effectively operate a CT, such a failure should be handled in a proper and timely manner. This issue becomes much more complicated because wafer residency time constraints (WRTCs) must be met to ensure the quality for some wafer fabrication processes. With such constraints, if a tool is operated under a periodic schedule and a PM fails, it is desired that the tool can still operate under a periodic schedule if it is possible. Nevertheless, the periodic schedule after a PM failure must be different from that before its failure since in this case the tool is degraded. Thus, there must be a transient process between them. It is a great challenge to operate a tool such that it can go through such a transient process with WRTCs being always satisfied. This paper aims to solve this problem by proposing PM failure response policies which can successfully transfer a CT to the feasible schedule after failure from the one before a failure. Then, efficient algorithms are developed to improve these response policies. The proposed policies are composed of simple control laws such that they can be realized in real time and online. Illustrative examples are presented to show their applications.

Multicriteria decision making (MCDM) problems are often encountered in complex system design. Most of them need to be evaluated with a large number of interactive and qualitative indices, which are difficult to be addressed effectively through the existing methods. In this paper, a novel fuzzy Choquet integral-based grey comprehensive evaluation (GCE) method, called fuzzy grey Choquet integral (FGCI), is proposed to evaluate MCDM problems with many interactive and qualitative indices. In this method, expert evaluation of qualitative indices is represented through fuzzy linguistic values. Fuzzy values are defuzzified and standardized to obtain the original evaluation matrix. The original values are replaced by the correlation coefficients, which, to a certain extent, eliminate the influence of experts' subjective preference. An improved teaching-learning-based optimization algorithm is employed to identify λ-fuzzy-measures following the weights given by experts in order to enhance the consistency of weights. Then the correlation coefficients are aggregated through Choquet integral among λ-fuzzy-measures, which can reflect interactions among indices. In addition, according to the characteristics of λ-fuzzy-measures, the construction guidelines for a corresponding index system are given to overcome the limitations of FGCI. Finally, the performance of the proposed method is demonstrated via a practical example of green design evaluation and compared with the GCE method. The results validate its feasibility and effectiveness.

Recommender systems (RSs) commonly adopt a user-item rating matrix to describe users' preferences on items. With users and items exploding, such a matrix is usually high-dimensional and sparse (HiDS). Recently, the idea of deep learning has been applied to RSs. However, current deep-structured RSs suffer from high computational complexity. Enlightened by the idea of deep forest, this paper proposes a deep latent factor model (DLFM) for building a deep-structured RS on an HiDS matrix efficiently. Its main idea is to construct a deep-structured model by sequentially connecting multiple latent factor (LF) models instead of multilayered neural networks through a nonlinear activation function. Thus, the computational complexity grows linearly with its layer count, which is easy to resolve in practice. The experimental results on four HiDS matrices from industrial RSs demonstrate that when compared with state-of-the-art LF models and deep-structured RSs, DLFM can well balance the prediction accuracy and computational efficiency, which well fits the desire of industrial RSs for fast and right recommendations.

In this paper, we propose an enhanced branch-and-bound (B&B) framework for a class of sequencing problems, which aim to find a permutation of all involved elements to minimize a given objective function. We require that the sequencing problems satisfy three conditions: 1) incrementally computable; 2) monotonic; and 3) overlapping subproblems. Our enhanced B&B framework is built on the classical B&B process by introducing two techniques, i.e., dominance rules and caching search states. Following the enhanced B&B framework, we conduct empirical studies on three typical and challenging sequencing problems, i.e., quadratic traveling salesman problem, traveling repairman problem, and talent scheduling problem. The computational results demonstrate the effectiveness of our enhanced B&B framework when compared to classical B&B and some exact approaches, such as dynamic programming and constraint programming. Additional experiments are carried out to analyze different configurations of the algorithm.

In the last ten years, cloud services provided many applications in various areas. Most of them are hosted in a heterogeneous distributed large-scale cloud computing environment and face inherent uncertainty, unreliability, and malicious attacks that trouble both service users and providers. To solve the problems of malicious attacks (including solo and collusion deception ones) in a public cloud computing environment, we for the first time propose a double-blind anonymous evaluation-based trust model. Based on it, cloud service providers and users are anonymously matched according to user requirements. It can be used to effectively handle some malicious attacks that intend to distort trust evaluations. Providers may secretly hide gain-sharing information into service results and send the results to users to ask for higher trust evaluations than their deserved ones. This paper proposes to adopt checking nodes to help detect such behavior. It then conducts gain-loss analysis for providers who intend to perform provider-user collusion deception. The proposed trust model can be used to effectively help one recognize collusion deception behavior and allow policy-makers to set suitable loss to punish malicious providers. Consequently, provider-initiated collusion deception behavior can be greatly discouraged in public cloud computing systems. Simulation results show that the proposed method outperform two updated methods, i.e., one based on fail-stop signature and another based on fuzzy mathematics in terms of malicious node detection ratio and speed.

Many ever-increasingly complex engineering optimization problems fall into the class of High-dimensional Expensive Problems (HEPs), where fitness evaluations are very time-consuming. It is extremely challenging and difficult to produce promising solutions in high-dimensional search space. In this paper, an Autoencoder-embedded Evolutionary Optimization (AEO) framework is proposed for the first time. As an efficient dimension reduction tool, an autoencoder is used to compress high-dimensional landscape to informative low-dimensional space. The search operation in this low-dimensional space can facilitate the population converge towards the optima more efficiently. To balance the exploration and exploitation ability during optimization, two sub-populations coevolve in a distributed fashion, where one is assisted by an autoencoder and the other undergoes a regular evolutionary process. The information between these two sub-populations are dynamically exchanged. The proposed algorithm is validated by testing several 200 dimensional benchmark functions. Compared with the state-of-art algorithms for HEPs, AEO shows extraordinarily high efficiency for these challenging problems.

A large number of services provided by cloud/edge computing systems have become the most important part of Internet services. In spite of their numerous benefits, cloud/edge providers face some challenging issues, e.g., inaccurate prediction of large-scale workload and resource usage traces. However, due to the complexity of cloud computing environments, workload and resource usage traces are highly-variable, thus making it difficult for traditional models to predict them accurately. Traditional models fail to deal with nonlinear characteristics and long-term memory dependencies. To solve this problem, this work proposes an integrated prediction method that combines Bi-directional and Grid Long Short-Term Memory network (BG-LSTM) models to predict workload and resource usage traces. In this method, workload and resource usage traces are first smoothed by a Savitzky-Golay filter to eliminate their extreme points and noise interference. Then, an integrated prediction model is established to achieve accurate prediction for highly-variable traces. Using real-world workload and resource usage traces from Google cloud data centers, we have conducted extensive experiments to show the effectiveness and adaptability of BG-LSTM for different traces. The performance results well demonstrate that BG-LSTM achieves better prediction results than some typical prediction methods for highly-variable real-world cloud systems.

Timely and proper recycling of end-of-life products promotes the sustainable development of our human society. Cost-effective and energy-efficient disassembly is a crucial step to realize such recycling. A disassembly process is usually uncertain because of the different quality of subassemblies in an end-of-life product. This work considers resource constraints and operation failure in such a process. A novel mathematical model is formulated with the objectives to maximize profit, minimize energy consumption, and minimize the total need for disassembly resources. A new solution method called a multi-objective discrete migratory bird optimizer is proposed to solve it. A real-life case is given to show its effectiveness and feasibility. Experimental results show that the superiority of the proposed algorithm over a well-known nondominated sorting genetic algorithm II and a multi-objective grey wolf optimizer.

A disassembly line balancing problem (DLBP) exists in the recycling process of end-of-life (EOL) products. It involves such factors as uncertainty of disassembly time and disassembly failure risk. Effective decisions can be made by taking them into full consideration. Under the constraints of disassembly precedence relationships and cycle time, this work establishes a stochastic multi-objective DLBP model subject to disassembly failure based on a disassembly AND/OR graph of EOL products. It considers disassembly failure risk and comprehensively evaluates the profit, energy consumption, average idle time of workstations, and hazard disassembly. Then, a new multi-objective discrete brainstorming optimizer that combines stochastic simulation is proposed for obtaining high- quality feasible solutions. Experimental results show the validity of the proposed algorithm. It outperforms both nondominated sorting genetic algorithm II and multi-objective discrete grey wolf optimizer.

Cloud computing is increasingly implemented by a growing number of organizations in recent years. Their critical business applications are deployed in distributed cloud data centers (CDCs) for fast response and low cost. The ever-increasing consumption of energy makes it highly important to schedule tasks efficiently in CDCs. In addition, many factors in CDCs, e.g., the wind and solar energy and prices of power grid have spatial differences. It becomes a challenging problem of how to achieve the energy cost minimization for CDCs in such a market. This work applies a G/G/1 queuing system to evaluate the optimization of servers in each CDC. Furthermore, a single-objective constrained optimization problem is given and addressed by a proposed Simulated-annealing-based Bees Algorithm to yield a close-to-optimal solution. Based on it, a Fine-grained Task Scheduling (FTS) algorithm is designed to minimize the energy cost of CDCs by intelligently scheduling heterogeneous tasks among distributed CDCs. In addition, it also determines running speeds of servers and the number of switched-on servers in each CDC while strictly meeting tasks' delay bounds. Realistic data-driven results demonstrate that FTS outperforms its typical benchmark scheduling peers in terms of energy cost and throughput.

As an emerging framework, edge computing achieves Internet of Things by providing computing, storage and network resources. It moves computation to edge devices located near users. Nevertheless, nodes in the edge often own limited resources and constrained energy capacities. It is impossible to entirely execute tasks in the edge due to their unsatisfied quality of service. Cloud data centers (CDCs) own almost unlimited resources yet they might cause large transmission delay and high resource cost. Consequently, it is highly needed to intelligently offload tasks between CDC and edge. This work proposes a task offloading algorithm for hybrid cloud-edge systems to achieve profit maximization of a system provider with response time bound assurance. It comprehensively investigates CPU, memory and bandwidth limits of nodes in the edge, and constraints of available energy and servers in CDC. These factors are integrated into a single-objective constrained optimization problem, which is solved by a simulated-annealing-based migrating birds optimization algorithm to yield a close-to-optimal offloading policy between CDC and the edge. Real-life data-driven experimental results show that its profit outperforms its four typical peers.