The intrinsic stochasticity of manufacturing is one of the main factors that hinder system resilience. Planning and scheduling problems are typical examples plagued by uncertainties such as stochastic processing time, arrivals of new orders, and breakdowns of workstations. Frequent uncertainties disturb the workflow, and their cascading effects create chaos in the whole system. The transparency and traceability analytics in Cyber-Physical Manufacturing Systems (CPMS) bring new hope to tackle uncertainties. Inspired by the core spirit of Out-of-Order (OoO) Execution in CPU, this paper proposes a novel OoO architecture for resilient planning and scheduling in CPMS with real-time data analytics. Following OoO principles, multi-level instruction queues are constructed, under which manufacturing operations (instructions) are sequenced and performed by analyzing real-time dependencies and executability. This study contributes a new perspective to enhance decision resilience using real-time data in CPMS. Results validate the effectiveness and resilience of OoO under different levels of uncertainty, showing improvements in the on-time delivery rate and reductions in the average order flow time.

Flexible and reconfigurable manufacturing systems with independent multiskilled cells are pivotal for mass customization under volatile demand, yet internal/external uncertainties persistently disrupt planning, scheduling, and execution (PSE), causing idle time and workflow instability. This study proposes a hybrid synchronization framework with hierarchical real-time data feedback to address heterogeneous demand-capacity synchronization (HDCS), reconciling demand volatility, system reconfigurability, and dynamic human capacity. The framework first adopts the ticket-enabled queuing mechanism of the graduation intelligent manufacturing system (GiMS) to enable seamless PSE integration, ensuring resilient operations. Next, it hybridizes a global optimization model with local adjustment mechanisms, utilizing multigranularity data to balance global optimality and local practicality under uncertainty. Finally, it implements a human-cyber-physical digitalization architecture to establish smart human-machine collaborative assembly. Validated through a case study, the method enhances cost-efficiency by 1.7%, punctuality by 29.5%, and resource utilization by 3.9% compared to conventional rescheduling strategies, demonstrating superior adaptability to uncertainties. The results validate its effectiveness in advancing human-centric smart manufacturing aligned with Industry 5.0 objectives, offering an integrated solution to HDCS challenges through synergistic coordination of hierarchical control, hybrid optimization, and human-machine collaboration.

In this study, we establish a new design methodology of granular models realized by augmenting the existing numeric models through analyzing and modeling their associated prediction error. Several novel approaches to the construction of granular architectures through augmenting existing numeric models by incorporating modeling errors are proposed in order to improve and quantify the numeric models' prediction abilities. The resulting construct arises as a granular model that produces granular outcomes generated as a result of the aggregation of the outputs produced by the numeric model (or its granular counterpart) and the corresponding error terms. Three different architectural developments are formulated and analyzed. In comparison with the numeric models, which strive to achieve the highest accuracy, granular models are developed in a way such that they produce comprehensive prediction outcomes realized as information granules. In virtue of the granular nature of results, the coverage and specificity of the constructed information granules express the quality of the results of prediction in a more descriptive and comprehensive manner. The performance of the granular constructs is evaluated using the criteria of coverage and specificity, which are pertinent to granular outputs produced by the granular models.

Unmanned aerial vehicles (UAVs) will be widely deployed due to their flexibility, mobility, and miniaturization, providing the necessary support for spectrum sharing between different communication systems in the space–air–ground-integrated IoT network (SAGIN). However, there are potential security threats to spectrum sharing among different communication systems due to the openness of a wireless network, the unreliability of node behavior, and the trust barriers of the networks. In this article, a secure spectrum sharing scheme based on lightweight UAV-blockchain (LUBC) is proposed to address the above security issues. First, a spectrum sharing model based on the overlay mode is developed to improve the spectrum efficiency of SAGIN, where UAVs relay signals from the satellite to the ground users in exchange for spectrum access opportunities and serve their own users simultaneously in the nonorthogonal multiple access (NOMA) mode. Second, a secure spectrum sharing framework based on LUBC is proposed to solve the security and privacy issues of spectrum trading in SAGIN. Then, aiming at maximizing the primary user’s throughput under the premise of meeting the minimum power allocation factor of UAV network, the spectrum auction based on NOMA is formulated as a multirelay selection optimization problem, which is solved by the blockchain-based sequential Vickrey auction mechanism. Finally, the security evaluation and numerical results are conducted to verify the security and effectiveness of the proposed spectrum sharing scheme for SAGIN.

This study aims to address a novel production and intralogistics synchronization (PiLSync) problem in multiresource-constrained assembly systems (MRCASs). PiLSync is highly complex due to the presence of multiple resources (operators, workstations, and materials) and the dynamic interactions between the resources and decisions. Moreover, various uncertainties, such as new/urgent job arrivals, uncertain operating times, equipment failures, and operator absences, further complicate this problem. Emerging Industry 4.0 technologies help capture and monitor the real-time PiL workflow, while existing methods are typically formulated to handle static or periodically updated inputs. Thus, it is difficult to explicitly incorporate real-time data into models, leading to a gap between production and intralogistic (PiL) decisions and actual executions under various uncertainties. In this article, out-of-order synchronization (OoOSync) is developed to facilitate flexible, resilient, and coordinated PiL operations in MRCASs. A sliding synchronization window is proposed as an effective decomposition tool to limit system complexity and localize demand uncertainty. Ticket validation governs interactions between resources and constraints under uncertainty. Finally, real-time data-driven synchronization mechanisms are developed for PiL decision/adjustment. The superiority of OoOSync is confirmed in computational experiments. This article is a pioneering study considering multiresource-constrained PiL operations under uncertainty. Additionally, this article offers a new perspective on using real-time data for decisions and operations subject to the dynamic interactions of resources.

Abstract In this work, we introduce our project on Dunhuang cultural heritage protection using artificial intelligence. The Dunhuang Mogao Grottoes in China, also known as the Grottoes of the Thousand Buddhas, is a religious and cultural heritage located on the Silk Road. The grottoes were built from the 4th century to the 14th century. After thousands of years, the in grottoes decaying is serious. In addition, numerous historical records were destroyed throughout the years, making it difficult for archaeologists to reconstruct history. We aim to use modern computer vision and machine learning technologies to solve such challenges. First, we propose to use deep networks to automatically perform the restoration. Through out experiments, we find the automated restoration can provide comparable quality as those manually restored from an archaeologist. This can significantly speed up the restoration given the enormous size of the historical paintings. Second, we propose to use detection and retrieval for further analyzing the tremendously large amount of objects because it is unreasonable to manually label and analyze them. Several state-of-the-art methods are rigorously tested and quantitatively compared in different criteria and categorically. In this work, we created a new dataset, namely, AI for Dunhuang, to facilitate the research. Version v1.0 of the dataset comprises of data and label for the restoration, style transfer, detection, and retrieval. Specifically, the dataset has 10,000 images for restoration, 3455 for style transfer, and 6147 for property retrieval. Lastly, we propose to use style transfer to link and analyze the styles over time, given that the grottoes were build over 1000 years by numerous artists. This enables the possibly to analyze and study the art styles over 1000 years and further enable future researches on cross-era style analysis. We benchmark representative methods and conduct a comparative study on the results for our solution. The dataset will be publicly available along with this paper.

Demanufacturing aims to recover value and conserve energy from end-of-life (EOL) products, contributing to sustainable manufacturing. To make the full use of EOL products, they are usually disassembled into components that have different values and embodied energy at different EOL options. This article studies a disassembly planning (DP) that integrates the decisions on disassembly sequence and EOL strategy to maximize the recovered value and energy conservation from EOL products. We propose a multiobjective DP based on the value recovery and energy conservation (MDPVE) model, which is different from the existing DP models by focusing on the embodied energy rather than the energy consumption during disassembly. An adapted multiobjective artificial bee colony (ABC) algorithm [multiobjective ABC (MOABC)] is developed to identify the Pareto solutions for the MDPVE and is compared with a well-known metaheuristic algorithm, Non-dominated Sorting Genetic Algorithm-II (NSGA-II). A real-world case study demonstrated the superior solution quality and computational efficiency of MOABC. Note to Practitioners-There is often more than one treatment option for EOL products or components, including reuse, remanufacturing, and recycling. However, the decision on which EOL option to select is not considered in most of the DP studies by assuming an EOL option given for each component. Hence, the disassembly plan with the EOL decision is focused in this article. As energy sustainability gains an increasing attention, it is essential to assess the profitability and energy conservation simultaneously for EOL products. Since there could be a tradeoff between recovered profit and conserved energy, a multiobjective evolutionary algorithm is developed for generating Pareto solutions which help decision-makers to find good solutions for both evaluation indicators.

This paper deals with the medium between two reconfigurable sensor nodes characterized by radio interfaces that support multiple channels for exchanging real-time messages under energy constraints. These constraints are violated if the consumed energy in transmission is higher than the remaining quantity of energy. A reconfiguration, i.e., any addition or removal of tasks in devices and consequently of messages on the medium, can cause the violation of real-time or energy constraints at run time. To achieve a feasible scheduling in time (i.e., message deadlines will be respected) and energy (i.e., there is available energy) on the medium, we propose new dynamic solutions: Balance, Dilute, and a Combination of them to manage any addition or removal of messages. The proposed approach utilizes the energy harvesting techniques and the PowerControl algorithm to reduce the nonharvested consumed energy. The proposed strategies achieve significant improvement over existing methods and provide the highest percentage of adding messages, with a lower average in response time and energy consumption. They reach a percentage of success in adding the highest priority messages while meeting deadlines up to 85%.

A maximally permissive (or optimal) supervisory control of an automated manufacturing system (AMS) modeled by Petri nets (PNs) can be usually implemented by imposing constraints in the form of a set of linear inequalities. To find such a set of linear constraints, in the existing work, an integer linear programming (ILP) problem is generally formulated and solved for some dominant markings obtained by reachability analysis and vector covering theory, which is computationally inefficient due to the combinatorial nature of solving ILPs to decide the coefficients of optimal linear constraints. This paper addresses the deadlock prevention problem for AMSs by developing efficient methods to reduce the computational overhead through the establishment of conditions on deciding the maximal permissiveness of linear constraints imposed on a system. By taking the advantage of structural properties of a PN model under consideration, we identify the most part of minimal covered illegal markings that can be optimally controlled via policies obtained by specific linear inequality constraints. Also, algorithms are developed to implement the proposed policies. The proposed approach can verify the optimality of a linear constraint efficiently without solving ILP problems. A linear programming method is developed to deal with the markings that cannot be processed by the proposed structural analysis. It is shown that for the considered class of PNs, called system of simple sequential processes with resources (S3PR), no mixed ILP problem needs to solve and the computational burden is dramatically reduced. Two examples are employed to demonstrate the efficiency of the developed method.

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.

This paper deals with the formal verification of reconfigurable discrete event control systems (RDECSs) using reconfigurable timed net condition/event systems (R-TNCESs) formalism. A reconfigurable system switches from a mode to another during its working process to adapt its behavior to the related environment. By including such a feature, RDECSs become complex and their verification is often expensive in terms of computation time and memory. In this paper, a new methodology for formal verification of RDECSs is proposed in order to ensure the correctness of these systems with a reduced cost (decreasing the verification time and memory occupation). The proposed contribution includes an improved modeling and verification of RDECSs. The modeling with R-TNCESs is enriched with all reconfiguration forms, and the verification involves an improvement method that avoids any redundancy and cancels unnecessary calculations. In addition, a visual tool called Rec-AG based on the proposed methodology is developed. The performance evaluation of this paper is achieved by measuring computation time and memory for several systems and different sizes of the problem. This paper's contribution is applied to the benchmark production system FESTO.

Scheduling a cluster tool with wafer residency time constraints is challenging and important in wafer manufacturing. With a backward strategy, the scheduling problem of such singlerobot-arm cluster tools is well-studied in the literature. It is much more challenging to schedule a more general case whose optimal scheduling strategy is not limited to the backward one. This work uses a timed Petri net (PN) to model the dynamic behavior of the system and presents a method to determine the optimal scheduling strategy for the system. Based on its PN model and the obtained strategy, it reveals that the key issue to schedule such a tool is to determine when and how long the robot should wait for. Based on this finding, this work establishes for the first time the necessary and sufficient conditions regarding the existence of an optimal and feasible one-wafer cyclic schedule for singlerobot-arm cluster tools. It then formulates a computationally efficient linear program to find it if existing, and finally gives industrial examples to show the application and power of the proposed method.

Wafer revisiting and residency time constraints complicate the scheduling problem of cluster tools in semiconductor manufacturing. Random disturbance to the activity time in operating a tool further complicates such a scheduling problem. To solve this challenging problem, this paper proposes a robust real-time schedule which consists of a real-time controller (RTC) and an off-line schedule. The former is developed to offset the activity time disturbance such that the wafer sojourn time fluctuation in a process module is minimized. With the RTC, to find the off-line schedule, necessary and sufficient schedulability conditions under which a feasible schedule exists are derived and these conditions can be easily checked. Then, the off-line schedule can be efficiently found by the proposed algorithms based on nondisturbed activity time if a feasible schedule exists. With the obtained real-time schedule, it is shown that the productivity of the system is maximized. Finally, examples are used to illustrate the proposed approach.

Based on the process of augmented Lagrangian coordination (ALC) solving, a solution framework of ALC for the complex system problem is proposed. Oriented to order fulfillment modes and the process of production outsourcing modes of process manufacturing enterprises in an industrial cluster and according to the hypothesis condition, an optimal configuration model of process cluster supply chain (PCSC) considering order horizontal subcontracting and cross-chain process production outsourcing is set up. By dividing the independent decision-making rights of the participants in single supply chains, a distributed optimal configuration model is established. Meanwhile, ALC is employed to solve the distributed model and solving results are analyzed and compared with the traditional method. Results comparison confirms the accuracy and efficiency of ALC for the PCSC configuration problem.