The coronavirus disease 2019 (COVID-19) placed significant and unprecedented challenges on the Tunisian public healthcare system. Healthcare facilities were overwhelmed by a surge in demand, resulting in crisis-level shortages of essential healthcare resources, including qualified personnel, respiratory support equipment, and Intensive Care Unit (ICU) beds. This resource scarcity created a critical imbalance between demand and healthcare system capacity, severely compromising the ability to provide adequate and timely patient care during the COVID-19 crisis. Accurate prediction of future case numbers and medical equipment needs is crucial to assist healthcare facilities in optimizing resource allocation during the COVID-19 epidemic surges. This paper presents a Long Short-Term Memory (LSTM) recurrent neural network model to forecast the required number of intensive care unit (ICU) beds. Various error quantification metrics measure the LSTM model’s performance, such as the R2, MAE, MSE, and RMSE. Despite the proposed LSTM model demonstrating good predictive performance, we observe a deviation of approximately 3 beds between the predicted values and the actual number of occupied beds, which can be significant in Tunisian public hospitals during the COVID-19 outbreak.

In this study, we propose an advanced Orthogonal Genetic Algorithm (OGA) specifically developed to tackle the Traveling Salesman Problem with Drones (TSP-D), a multifaceted optimization challenge that necessitates precise synchronization between a truck and a drone for effective delivery tasks. The OGA integrates Orthogonal Crossover and Region-Based Mutation strategies, thereby enhancing the algorithm's proficiency in optimizing drone routing in a range of TSP-D scenarios. This novel approach significantly augments the algorithm's adaptability and exploratory capabilities within the intricate search space. Our comprehensive experimental analysis rigorously evaluates the performance of the proposed OGA against established algorithms in a variety of TSP-D instances. The results from these evaluations reveal that our approach substantially surpasses conventional algorithms in terms of both convergence speed and solution quality. This enhanced performance underscores the OGA's efficacy and robustness in optimizing complex paths in TSP-D scenarios.

Accurate forecasting of hospital-specific requirements during healthcare crises enables decision-makers to tackle key strategic challenges, such as determining the optimal allocation of beds, respiratory support machines, and other critical medical equipment. Identify departments that may need to suspend operations temporarily to efficiently reallocate their resources, and assess the feasibility of resource pooling, inter-hospital sharing, and patient transfers to optimize capacity during peak demand periods. This study evaluates the forecasting accuracy of LSTM, Bi-LSTM, ARIMA, Holt-Winters, and Exponential Smoothing models in predicting demand for ventilators and oxygenation devices over two weeks during the COVID-19 pandemic. Utilizing a local dataset with detailed records of respiratory support devices provides a more precise predictive analysis of critical hospital resource demand, thereby optimizing resource allocation and improving hospital preparedness during health crises. The proposed models were evaluated using MAPE, RMSE, MSE, MAE, and R-squared. While ARIMA outperformed LSTM in terms of accuracy and model fit, it struggled to capture sudden surges in resource demand, limiting its reliability during crisis peaks. LSTM exhibited lower accuracy, likely due to the lack of long-term dependencies and limitations of the dataset. Among all models, the Bi-LSTM showed the highest predictive accuracy, aligning closely with actual observations and achieving the lowest mean absolute error (MAE), making it the most suitable approach for forecasting hospital resources during critical periods.

The 3D-bin packing problem consisted of making the suitable decision on how to arrange a set of rectangular objects in their bins. It entails optimizing the allocation of a set of items into a minimal number of bins, each with a fixed capacity. In our paper, we are interested to a three-dimension bin-packing problem, which is a combinatorial optimization problem and is known to be NP-Hard problem. We refer to the recent research work to create our hierarchical hybrid optimization named GenSA-3DBPP to solve the three-Dimension bin-packing problem. The empirical study aims to highlight the improvements made with GenSA-3DBPP on three levels. To solve this issue, we proposed The threeDorientationsII heuristic, which places the item in the orientation that maximizes load while respecting weight constraints. The Genetic Algorithm level explores the search space to tackle the main purpose of the problem, which is the minimization of both number of bins and wasted space. The Simulated Annealing level utilizes the Genetic Algorithm’s best solution as input and generates a best solution.

Satellite imagery supports critical applications such as land cover mapping, environmental monitoring, disaster assessment, and urban planning. Despite significant advancements, challenges in analyzing satellite imagery persist, primarily due to data variability, atmospheric conditions, and complex land cover patterns. Traditional handcrafted descriptors like Scale-Invariant Feature Transform (SIFT) and encoding techniques such as Bag-of-Visual-Words (BoVW) are effective but often fall short in capturing global context and spatial relationships due to their inherent local nature. The advent of deep learning (DL), propelled by ample data and computational resources, has markedly improved satellite image analysis. However, the reliance on extensive annotated data constrains the wider applicability of DL methods. This study harnesses the strengths of both deep and hand-crafted features to enhance the classification accuracy of remote sensing images. Specifically, we synergize SIFT descriptors with pretrained MobileNetV2 and VGG16 deep features. While SIFT excels in capturing local features essential for identifying specific image characteristics, pretrained DL models provide enriched representations with global context, spatial relationships, and hierarchical features. This integration aims to overcome the individual limitations of each method, enabling the model to effectively handle perturbations, scale variations, and diverse landscapes. Extensive evaluations on the EuroSAT dataset demonstrate that our approach outperforms, not only SIFT, VGG16, and MobileNetV2 when used separately, but also surpasses state-of-the-art remote sensing image classification approaches. Another salient advantage of our approach is its robust applicability in scenarios with limited labeled data - a prevalent challenge in remote sensing image classification.

Human Activity Recognition (HAR) is essential for applications like health monitoring and fall detection using mobile sensors. However, obtaining large datasets necessary for training HAR systems is prohibitively expensive. Generative Adversarial Networks (GANs) have been proposed to generate synthetic HAR data, simplifying and enhancing the development of HAR systems. While these methods have improved accuracy, existing GAN-based approaches struggle with generating clear abnormal patterns, weakening anomaly detection capabilities. Available HAR data predominantly focuses on normal activities and does not target anomalies, making it challenging to detect anomalous situations. To address this, we propose a novel controllable GAN that generates realistic HAR data as well as distinct abnormal classes. This advancement enhances anomaly detection accuracy through more standardized activity recognition and quantification by HAR systems. We evaluate our approach on the WISDM dataset, demonstrating significant improvements in the balance and quality of synthetic data, leading to better performance in anomaly detection.

This paper aims to develop an innovative solution based on deep learning to assist doctors in detecting diabetic retinopathy. Utilizing the IDRiD and APTOS 2019 datasets, this research addresses three major challenges. The first challenge involves the precise segmentation of retinal lesions, including microaneurysms, hemorrhages, and hard and soft exudates, essential for early and accurate diagnosis. The second challenge is classifying retinal images based on the severity of diabetic retinopathy, utilizing the InceptionV3 architecture. Finally, the early detection of macular edema risk, a major symptom of diabetic retinopathy, is addressed using the MobileNetV2 architecture. This research explores various techniques of data augmentation, image enhancement, and model selection to design robust and precise solutions for healthcare professionals.

Aggregate production planning and disaggregate production planning are two key steps to improve efficiency of production planning systems. Solving them separately can decrease complexity and be adapted to the standard structure of an organization. Nevertheless, interaction between these planning levels is major to prevent the obtaining of infeasible and inconsistent plans. Additionally, the optimization by sub-problem usually incurs a global problem with suboptimal results. To cope with this, an integrated model considering both levels becomes fundamental. This model is based on using linking constraints, aggregation constraints or verification constraints. Computational tests considering data from literature and real data have been done to compare the performance of hierarchical production planning with integrated one. It was shown that the integrated approach based on verification constraints outperforms efficiently the hierarchical one.

This paper introduces Dynamic Social Particle Swarm Optimization (DS-PSO), a novel adaptation of the traditional Particle Swarm Optimization (PSO) technique specifically engineered for complex optimization challenges. DS-PSO innovatively incorporates dynamic social interactions within the swarm, enhancing adaptability and addressing the typical limitations of premature convergence and limited exploration in conventional PSO. A key feature of DS-PSO is its ability to balance exploration and exploitation efficiently, making it particularly suitable for dynamic environments. The primary application highlighted in this study is automatic clustering, a crucial task in data analysis involving unsupervised data grouping without prior knowledge of cluster numbers. DS-PSO’s flexibility and improved search capability demonstrate its potential as an effective tool for automatic clustering, promising significant advancements in data-driven optimization and analysis.

In this paper, we address the problem of two stages Hybrid Flow Shop with release dates and blocking constraints. The objective is to minimize the makespan. We consider m parallel machines at the first stage and two dedicated machines at the second stage. However, the Hybrid Flow Shop Scheduling Problem with dedicated machines, blocking and release date constraint simultaneously has not yet been well studied. Incorporating these, constraints align more closely with the real configurations found in many industries, such as the pasta industry. To address this problem, we will introduce a mathematical model and contrast the outcomes against two lower bounds discussed in the literature. In addition, we will deal with this problem through three heuristics.

This paper presents a novel enhancement to Particle Swarm Optimization (PSO) by integrating parallax learning to improve convergence speed and solution quality. Comparative simulations against state-of-the-art metaheuristics on CEC 2022 demonstrate superior performance. Additionally, we apply our method to optimize Electric Vehicle (EV) Fast Charging Station (FCS) placement, achieving efficient solutions considering travel time and power loss costs on an IEEE 33-bus test system. Our algorithm not only surpasses competitors on standard benchmarks but it also provides high-quality solutions to realistic optimization problems in energy systems planning and operation.

This paper introduces a supply chain quality sustainability decision support system (QSDSS). It uses association rule techniques to provide better logistics plans and handle risk in the supply chain. An improved Multiobjective Crystal Structure Algorithm using centroid opposition based learning and gaussian perturbation is proposed to avoid premature convergence and escape from local optimal. The experimental study is carried out in two phases, using two classic benchmark suites and the “DataCo SMART SUPPLY CHAIN” dataset for big data analysis.

with the increasing frequency and sophistication of cyber-attacks, it has become essential to develop intrusion detection systems (IDS) that can detect and prevent such attacks in real-time. Traditional IDS, which are based on rule-based and signature-based detection methods, have proven to be insufficient in detecting new and unknown attacks. In recent years, machine learning (ML) techniques have been widely used to enhance IDS’s detection capabilities. However, traditional ML-based IDS faces issues such as the need for a central authority to maintain the system and ensure trust between different parties. To address these issues, we propose a machine learning and blockchain-based intrusion detection system that uses a distributed and transparent approach to improve the detection accuracy and system security.

New healthcare units called "Covid units" dedicated to the medical care of individuals infected by the Coronavirus disease have been rapidly established under enormous pressure. This reflects the strong commitment of the Tunisian national healthcare system and public action in combating the COVID-19 pandemic. From this perspective, this study aims to evaluate the effectiveness of these units in providing effective and timely responses to the affected communities. A real Covid-19 unit at the University Hospital SAHLOUL in Sousse, Tunisia is modeled and simulated using ARENA simulation software. The simulation model analyzes the performance of the current healthcare Covid unit by providing relevant statistics on the patient flow and resource utilization rates. The simulation results identify barriers to the efficiency of the Covid care unit and question the relevance of the current distribution of hospital resources during the Covid-19 pandemic. To help healthcare managers evaluate alternative choices and identify potential solutions to optimize critical resources management, "what if" models are executed to address bottlenecks in different stages of Covid service and improve resource allocation. The simulation results of the proposed scenarios demonstrate a significant improvement, reducing the average patient waiting time by 93%, and increasing, in turn, the average daily throughput to 36,51%

This paper introduces two three-level hierarchical frameworks, one for the production campaign planning and other for the production planning and scheduling for multi product multi stage batch plants. Each level in each framework is formulated as a mixed integer linear programming model. These models are solved sequentially where the output of each campaign production planning model presents the input of the corresponding planning level model. Using data from the literature, the proposed optimization models were tested and validated.

In cluster analysis, classical approaches suffer from the problem of identifying the number of clusters, known as the automatic clustering problem. Therefore, automatic clustering has become a popular research area and offers opportunities in various data analysis applications such as bioinformatics, medicine, image processing and consumer segmentation. It is considered as NP- complete problem where it is preferable to use approximate approaches. In this study, we propose an hybrid approach between chaos game optimization and particle swarm optimization (CGOPSO). The Davies-Bouldin index (DBI) is used as a main objective of the proposed approach with the purpose to find the most accurate number of cluster centroids and their positions. To assess its performance, we compared CGOPSO with different other existing algorithms in the literature over 12 classical datasets using two different validity indexes: Davies Bouldin index (DBI) and Compact-Seperated index (CSI). The experimental results have demonstrated that CGOPSO shows better performance than other algorithms.

To obtain computer procedures that intelligently guide the search process by efficiently exploring the search space corresponds to an optimization problem that is solved using meta-heuristics. However, there are many optimization problems whose objective is to extract a single efficient solution that reflects the quality of the system performance. Hence, we also encounter multi-objective problems. This paper studies a multi-objective optimization problem that addresses the complexity of problems with synchronous indices. In this paper, association rule mining (ARM) is treated as an optimization problem. In recent literature, most of the proposed methods of ARM generate a large number of redundant and irrelevant rules. Therefore, we propose an improved multi-objective crystal structure algorithm using the TOPSIS approach. The experimental study is performed in two steps. First, we use a classical benchmark suite, namely WFG, to analyze the effectiveness of the proposed approach in generating solutions close to the Pareto fronts. In a second step, to confirm the performance of our algorithm, we apply it on the database of the Bosch Production line performance, and we conduct a comparative evaluation towards recent algorithms. The obtained results demonstrate the efficiency of the proposed algorithm in terms of number of rules, average support, average confidence, average conviction, average certain factor.

For better supply chain management, it is necessary to think about better managing its cost sources. Inventory is considered the most important element of the supply chain that generates the different logistics costs, mainly the inventory holding cost and the transportation cost. One of the most widely used models to jointly solve these two problems is the Inventory Routing Problem (IRP), which will be the focus of this study. The proposed model in this work deals with a two-tier supply network. The first level contains the supplier with a single vehicle to serve a set of customers with a deterministic and periodic demand that are located at the second level. Our work consists in studying the effect of the increase of the replenishment lead time on the different logistic costs. In addition, we introduced the Lateral Transshipment (LT) technique as an option for inventory transfer if it is economical. New mathematical models corresponding to the above-mentioned problems have been developed and solved by an exact method. The obtained results show that the variation of the replenishment lead time leads to an increase of the different logistics costs and that LT can improve the total network cost and balance the customers' inventory level.

Feature selection (FS) is considered a challenging machine learning problem that handles the large size of features. The main purpose of FS is to remove irrelevant and redundant variables to improve the performance of the learning algorithms. Consequently, the FS process is considered as an optimization problem where metaheuristics approaches prove efficiency in solving it. In this paper, we propose new binary versions of a new ancient inspired metaheuristic approach called Giza Pyramids Construction (GPC) to select the most relevant subset of features. The proposed binary versions of the algorithm called BGPC-S and BGPC-V are implemented with two transfer functions, with the main objective of maximizing classification accuracy and minimizing the number of selected features. The two versions of BGPC were compared to six well-known binary metaheuristics for feature selection problem, namely Binary Atom Search Optimisation (BASO), Binary Bat Algorithm (BBA), Binary Differential Evolution (BDE), Binary Grey Wolf Optimizer (BGWO), Binary Particle Swarm Optimization (BPSO), and Binary Harris Hawks Optimizer (BHHO), and evaluated over 20 datasets from the UCI repository. Experiments have demonstrated that the proposed approaches outperformed the other algorithms in terms of classification accuracy and the number of selected features.

Vehicle routing problem (VRP) is a hard combinatorial problem. In practice, specificities of concepts (vehicles and networks transportation) related to VRP must be explicitly considered in modeling to obtain accurate cost and feasible solutions. Each of these concepts is represented in literature for a specific purpose. In this study, we present an ontology for modeling VRP as a unified representation based on road and vehicle classification. The approach proposed aims at providing decision-makers in transport companies a consistent understanding of the field based on ontology. Moreover, it aims at generating parameters of classification of VRPs, and at facilitating later on solving these problems, in the academic or industrial context.

Abstract Risk-aware Business Process Management (R-BPM) has been addressed in research since more than a decade. However, the integration of the two independent research streams is still ongoing with a lack of research focusing on the conceptual modeling perspective. Such an integration results in an increased meta-model complexity and a higher entry barrier for modelers in creating conceptual models and for addressees of the models in comprehending them. Multi-view modeling can reduce this complexity by providing multiple interdependent viewpoints that, all together, represent a complex system. Each viewpoint only covers those concepts that are necessary to separate the different concerns of stakeholders. However, adopting multi-view modeling discloses a number of challenges particularly related to managing consistency which is threatened by semantic and syntactic overlaps between the viewpoints. Moreover, usability and efficiency of multi-view modeling have never been systematically evaluated. This paper reports on the conceptualization, implementation, and empirical evaluation of e-BPRIM, a multi-view modeling extension of the Business Process-Risk Management-Integrated Method (BPRIM). The findings of our research contribute to theory by showing, that multi-view modeling outperforms diagram-oriented modeling by means of usability and efficiency of modeling, and quality of models. Moreover, the developed modeling tool is openly available, allowing its adoption and use in R-BPM practice. Eventually, the detailed presentation of the conceptualization serves as a blueprint for other researchers aiming to harness multi-view modeling.

Recently, hyperspectral tensor modeling has been assessed by its capacity of determining more compact as well as by its useful intrinsic data representation. In this paper, to enhance the hyperspectral tensor data representation and to eliminate non-relevant spatial datum, we integrated the intrinsic decomposition (ID) as a pre-processing step. The suggested approach acts in agreement with the joint use of spectral and spatial features provided in hyperspectral scenes, and it incorporates more usefully with the spatial data in the dimensional reduction step. The suggested framework consists of three steps: firstly, the intrinsic decomposition is employed to remove useless spatial data from hyperspectral image (HSI). Secondly, after modelling ID results as a tensor structure, the tensor principal component analysis is used to reduce tensorial data redundancy. Finally, we evaluated the proposed approach during classification tasks using real hyperspectral data sets. Compared to other methods, experiment results have proved that our approach can pave the way for the best classification accuracy.

Motivated by its capacity to process complex characteristics and deal with nonlinear problems, tensor decompositions have been also introduced, recently, to treat remote sensing data. In this article, a new tensor formulation based feature extraction framework is suggested for hyperspectral target detection. The new proposed method includes the intrinsic decomposition, as tensor structures, to improve the hyperspectral data representation and get rid of the non-significant spatial proprieties. Besides of the joint exploitation of spectral and spatial content, the new proposed approach allows to extract more effective discriminative spatial features. A series of experiments, for the purpose of hyperspectral target detection, show that the suggested scheme can be conducted on hyperspectral images with satisfactory detection accuracies.

Mobile Ad-Hoc networks (MANETs) have gained much attention thanks to their efficiency. However, they appear to be more susceptible to various attacks due to the open access medium and the dynamically changing network topology. Intrusion Detection Systems (IDS) represent an important line of defense against malicious behavior. In fact, they monitor network activities to detect any malicious attempt performed by intruders. IDS datasets show limitations in their evaluation of mobile networks since these datasets cover only wired networks. We recommend a new IDS dataset that reflects the characteristics of MANET. The main contribution turns around the integration of an IDS capable of detecting the majority of security attacks occurring in MANETs. We propose a novel approach in collecting the necessary data in order to build the behavioral Database called NetBigData, which contains normal behavior and attacks scenarios. As a matter of fact, we picked up the most common attack in mobile networks, which is Denial of Service (DoS). In this paper, we use an Anomaly-based technique to monitor traffic patterns, we have simulated four attacks out of three categories which are Packet Dropping, Routing Disruption, and Resource consumption attacks. To improve the quality of the collected data, we used data preprocessing techniques to take advantage from the best performance of our dataset. To automatically generate rules from the obtained data, we chose a Support Vector Machine (SVM) classifier. The obtained results show that the proposed anomaly-based IDS is effective in detecting the DoS type attacks with a high detection rate.

Fault detection and localization is a central element in the management of network failures. It allows inferring the exact failure of a network from all the observed symptoms. Because failures in network systems are inevitable, detection and diagnosis efficiency is important for the stability, coherence and performance of a communication system. In this paper, we propose a unified approach to diagnosing nodes faults in a computer network. In our approach, the problems of choosing probing stations and probes, classically separated, are solved jointly in order to obtain, with lower complexity, minimum sets of probes and probing stations capable of detecting and localizing any failed node. The optimality of our new method has been tested through the simulation of the network configuration.

The components of a computer network are vulnerable to a variety of faults such as a link break or node failure. In order to prevent faulty components from impeding the execution of network applications, it is very important to detect, locate and repair faulty components. Existing approaches to fault localization in communication systems use active or passive measures. Active measures involve additional traffic for network monitoring. On the other hand, the passive measures use the existing end-to-end data in the network in order to extract the necessary information and thus introduces no additional traffic in the network. In this paper, we propose an end-to-end approach that uses passive measures for fault isolation in communication networks and formulate the issue of fault isolation as an optimization problem. In fact, we used the Boolean Particle Swarm Optimization algorithm inferring the best node(s) to be tested, with the objective of minimizing the expected cost for all the faulty elements in the network. The performance of the proposed schemes is evaluated through an intensive simulation of different network scenarios.

The Internet is known for its ability to scale and adapt its routes to every change. These routes are made by every Autonomous System (AS) peering with each other as neighbors with the Border Gateway Protocol (BGP). Unfortunately, this amount of trust, between all these Autonomous Systems allowing protocols like the BGP to function properly, is not found in the Physical world between logistic service providers. The Physical Internet, however, with its standardized PI-Containers and Internet-derived protocols, has every promising aspect to face this challenge. We propose in this paper a new routing approach based on the PI-BGP (Physical Internet-Border Gateway Protocol) equivalent of the BGP in the Internet. We developed this new protocol to offer a new perspective to tackle the problem of routing the PI-containers in the Physical Internet.

Based on the Laplacian Eigenmaps (LE) algorithm and a potential matrix, the Spatial Spectral Schroedinger Eigenmaps (SSSE) technique has proved a great yield during the hyperspectral dimensionality reduction process. Experimentally, SSSE is in deficiency of high computing time which may hinder its contribution in the remote sensing field. In this paper, a fast variant of the SSSE approach, called Fast SSSE, was proposed. The new suggested method substitutes the quadratic constraint employed during the optimization problem, by a linear constraint. This overhaul preserves the data properties in analogous way to the SSSE technique, but with a fast implementation. Two real hyperspectral data sets were adopted during the experimental process. Experiment analysis exhibited good classification accuracy with a reduced computational effort, compared with the original SSSE approach.

This paper presents a mathematical model, three heuristics and two lower bounds in order to solve the hybrid flow shop scheduling problem with parallel machines at the first stage and two dedicated machines at the second stage. Each job is subject to a release date at stage one. The objective is to minimize the makespan.

By the recent years, the use of UAV (Unmanned Aerial Vehicle) also known as drone have witnessed a remarkable increase. Drones are no longer limited for military services as in the past. They are able to carry out civilian missions like survivor search operations after natural disasters, accidents, especially in the most difficult conditions, such as meteorological conditions and inaccessible or dangerous geographical locations. Moreover with the invasion of a new era, which is the IoT (Internet of Things), carrying a big wave of connected objects, we can talk about drone based IoT. The problem is how can we control such a connected object by preserving the privacy and security of users. The principal goal is to build a new secured architecture to provide a high security level. This architecture will control drones and promote them to an upper level connected with the IoT and the big data paradigms. We propose in this paper an architecture that relies on Id-Based Signcryption and RFID tags.

The connectivity of mobile networks is increasing heavily and the evolution of the risks is highly dynamic. In Mobile Ad Hoc Network (MANET), attacks and digital attacks are becoming increasingly complex. Due to their nature, these networks make some information unavailable and/or incomplete needed for attacks detection process. Several solutions have been made to ensure the security of mobile networks specially intrusion detection systems (IDS). This solution allows enhancing IDS detection efficiency even with incomplete information about occurred attacks. In this paper, we propose an IDS based on three algorithms NCF, FNF and DPA allowing special traffic abstraction and data collection. We used these algorithms to generate a "behavioral database" for supervised nodes in the network. We study and implement four types of Denial of Service (DoS) attacks, which could disturb the routing process in MANET. These attacks are Blackhole, Grayhole, Wormhole, and Flooding attack. We generate these types of attacks by modifying the normal AODV routing algorithm behavior. We have implemented these attacks using Opnet Modeler 14.5. We proposed a set of IDS nodes to supervise the network behavior using Fuzzy Inference System (FIS). These nodes identify a pattern for each attack behavior to be stored in the "behavioral database". The performance of a network under attack is investigated.

Schroedinger Eigenmap Projections (SEP) is a well-established dimensionality reduction technique for hyperspectral images; it is the linear approximation of Schroedinger Eigenmap (SE) method indeed. The SEP approach is based on Locality Preserving Projections (LPP) algorithm in which the adjacency graph is created in advance without taking in consideration the number of data points in ground objects. This is can negatively affect on hyperspectral reduction and classification process. In this paper, to resolve the problem, we adopted a variant of LPP termed modified LPP (MLPP) instead of original LPP. MLPP adopts an adaptive strategy to create the adjacency graph in which the number of neighbors for each data point can be chosen adaptively. The proposed feature extraction technique uses the Schroedinger operator in the MLPP framework. Indian Pines scene was used for this study. The classification results show effective classification accuracies according to the SEP and other dimensionality reduction methods.

Efficient and reliable fingerprint matching is crucial for many civilian and forensic applications. Fingerprints are characterized by large intra-class variations (i.e. variability in different impressions from the same finger). This variability manifests itself by the missing or the displacement of genuine minutiae and the detection of spurious minutiae. Displaced, missing and spurious minutiae make fingerprint matching a very challenging pattern recognition problem. Although significant improvements have been made in minutiae-based matching, most of minutiae matching algorithms lack of robustness with respect to displaced, missing and spurious minutiae [4]. This paper introduces EDT-C, a new fingerprint matcher based on minutiae triplets. The proposed matcher uses an extended form of Delaunay Triangulation that allows to take into account, not only genuine minutiae, but also spurious, displaced and missing ones. EDT-C characterizes minutiae triplets by a set of innovative geometric features that help in tolerating linear and non-linear distortions. Finally, EDT-C includes some optimizations that allows to quickly consolidate local matchings and filter non-matching minutia triplets. Experiments show that EDT-C has a reasonable computational cost and is far more accurate than its main competitors.

This study considers the application of a hybrid genetic algorithm (HGA) to the Inventory Routing Problem (IRP), that aims to minimize the cost of the total distance traveled over a time horizon discretized in periods, while guaranteeing that the customers do not incur a stock-out event. The proposed algorithm is tested using the instances proposed by Archetti et al. [1] and [2] where we consider only one vehicle available at the supplier. This instance is the biggest one solved so far. In this paper, computational results are given by an HGA, showing that this approach is competitive with other search and simulated annealing in terms of solution time and quality.

This paper considers the two-parallel-machine scheduling problem with precedence constraints. One of the machines may not always be available due to machine breakdowns or preventive maintenance during the scheduling period. All execution and communication times between tasks are considered unitary and all unavailability dates are known in advance. The considered task graph (since tasks are related by precedence constraints) is an intree, where each task can have many predecessors but only one successor. The considered objective function in this paper is the makespan denoted Cmax. To solve the problem, a new optimal algorithm entitled Scheduling Intrees with Unavailability constraints (SIwUC) is proposed. The used strategy is to find the best trade-off between the minimization of the communications and the minimization of the difference in load between the processors in order to minimize the makespan. Algorithm details and key ideas proving the optimality of the algorithm are described.

To be competitive, transportation systems must be able to analyze and to evaluate, in real-time, critical differences between the short-term planned actions and the actual performed actions generating states of undesirable or unacceptable risk. We propose a method for monitoring the dynamic evolution of risk in the operational flow of a transportation system. It consists in an approach assessing risk associated to delays affecting the transportation operations. We use the FMEA (Failure Modes and Effects Analysis) around failure scenarios rather than failure modes, and we evaluate risk using probability and cost. A scenario probability is estimated dynamically in discrete points based on events occurrences during the process execution. The proposed approach useful to transportation managers is based on consistent and meaningful risk evaluation criteria to facilitate cost-based decisions during execution. The implementation of this method is performed by monitoring a container delivery process facing delays risks.

The problem studied in this paper is to allocate and to sequence the elective operation on operating rooms (ORs). We develop a mixed integer linear programming (MILP) model to solve this problem. Decisions in this model include the allocation of operations to material resources and human resources, the starting time of them and the starting time for each surgeon. To show the efficiency of this model, we decide to compare it with a constraints programming (CP) approach. The performance of these models is tested using a benchmark of the literature. The results indicate the efficiency of the MILP model compared with the CP model in terms of computational time.

In this article two mathematical models and lower bounds are presented for the hybrid flow shop scheduling problem noted HFS. We consider meparallel machines at the first stage and two dedicated machines at the second stage. We showed that the performance depends on the choice of the decision variable types.

Computer security is far from being guaranteed due to the scalability of computer networks, the constant evolution of risks and the presence of noisy information. Several solutions were proposed to ensure the integrity, confidentiality and availability of resources, including intrusion detection systems (IDS). The main objective of the current work is first to take advantage of data mining techniques such as normalization, feature selection and eliminating redundancies in order to analyze the huge data like the KDDCUP'99. The second objective is the learning ability of neural networks and the third one is the fuzzy logic reasoning that realizes knowledge wave characters. The proposed model is the Neuro-Fuzzy model precisely the NEFCLASS (Neuro Fuzzy Classification) model of a generic fuzzy perceptron which is in the form of a combination of neuron and fuzzy system networks. This model is characterized by its powerfulness thanks to its large database and rapidity due to its parallel architecture. Moreover, it can be easily updated through the re-learning process following the scalability potential inherited in its architecture. As well as it is distinguished by an intuitive model presented by linguistic rules that are easily understood by the security operator. To test out the adaptability of our approach in detecting unknown attacks we use a test database , namely corrected test that contains new attacks that are not present in the training set and compare the results obtained by the model ANFIS. We prove that our approach based on NEFCLASS model is more powerful in classifying intrusions than the one based on ANFIS model. To highlight the motivation for using the Neuro-Fuzzy Classifier, a comparative study was conducted by using the full 10% KDD among the NEFCLASS model and other supervised classifiers which are ANN and C5.0.

This correspondence is related to the determination of both control and scheduling of flexible manufacturing systems under cyclic command. Different approaches can be found in the literature, but we focus on those which respect the optimal throughput while minimizing the work in process. So, we recall methods of performance evaluation developed during the last 20 years. The last part is devoted to a new approach of cyclic scheduling using a Petri net. This method uses algebraic tools (dioids) developed for the study of marked graphs. In this way, the problem of the scheduling is progressively transformed into a problem of the search of solution(s) on a system of equations

We focus on the analysis of the cycle scheduling problem in FMS using unfolding time Petri nets after slicing off some sub-nets using the transitive matrix. We can change an iterated cycle module into an acyclic module without changing any other behavior property in Petri nets. We first show that properties can be studied through the unfolding nets concepts. A method to analyze and optimize the control of such systems is presented and explained on an example.