ABSTRACT. The effectiveness and stealthiness analysis of sensor attacks based on reachability graph of labeled Petri nets (LPNs) is in general subject to the state explosion problem. To address this issue, this paper provides a simplified LPN method based on simplified structure. Attack structure is derived from the concept of concurrent composition. The attack structure based on the simplified structure subnet describes all possible attacks, and by removing the state with the risk of exposure, the supremal stealthy attack substructure can be obtained. If the supremal stealthy attack substructure contains a critical marking, the attack is stealthy. By using simplified rules, the number of states in the attack structure can be effectively reduced while still guaranteeing the attack’s effectiveness and stealthiness.

ABSTRACT. Although Petri nets are a well-established formalism for modeling and analyzing DEDS, their semantics needs to be extended to meet the increasing modeling challenges posed by the dynamic and complex behaviour of many real systems. In the present paper, we introduce a rewriting logic based behaviour semantics for Petri nets, which provides a powerful and flexible formalism for modeling and capturing easily the complex dynamic associated to firing rules extensions of Petri nets. We demonstrate the effectiveness of the proposed approach by illustrating the corresponding specification of the studied arcs extensions according to that enhanced rewriting logic based semantics for Petri net.

ABSTRACT. The advent of 5G networks has revolutionized wireless communication, enabling device-to-device (D2D) interactions and promoting effective group communication among users. However, in out-of-coverage scenarios where the base station is absent, managing D2D group communications becomes a challenge. This paper proposes a dynamic leader selection protocol that efficiently coordinates D2D user equipment (UE) by electing a D2D device as the group leader. The protocol ensures that each session assigns a leader from among the group members to act as a central station. To guarantee the selection of reliable leaders, we introduce robust measurements of self-confidence, risk control, and confidence. These measurements prevent malicious devices from assuming leadership roles. Furthermore, they identify colluding devices or any suspicious behavior, enabling timely intervention and mitigation of potential attacks. Through practical analysis, we demonstrate the effectiveness of our proposed protocol.

ABSTRACT. The gradual depletion of resources has become an increasingly pressing problem in our modern world. The manufacture and disposal of glass bottles, in particular, pose significant environmental challenges. This article explores the concept of bottle deposit schemes as a sustainable solution to mitigate the negative effects of resource depletion. Encourage recycling of glass bottles through consignment programs, systems designed to reduce waste, preserve raw materials and encourage recycling practices. The main goal of this paper is to present a glass bottle reuse trackability solution in a closed loop supply chain system using new technologies as QR Code and digital marketing. For this, we synthesize existing literature, present practical implementation strategies, model the problem, and highlight the positive outcomes achieved. The findings underscore the urgency of implementing consignment initiatives for glass bottles as a crucial step towards mitigating resource depletion and advancing a more sustainable future. Finally, the problem will be solved using IBM Cplex Optimization.

ABSTRACT. This paper deals with the selective maintenance problem (SMP). SM is dedicated to systems operating an alternate sequence of missions and scheduled break. These systems are known as mission-oriented systems (MOS). To improve the performance of a MOS for the next mission, its components can be maintained during the scheduled breaks. However, given the limited maintenance resources (break duration, budget, manpower...) not all components can be selected for maintenance. To meet the required performance for the next mission, the SMP consists then to select the appropriate subset of components to maintain, the level of maintenance to perform on these components, and their assignment to repairpersons. The resulting SM plans are often unbalanced as maintenance duties (workload) are usually not equally assigned to repairpersons. As a consequence, productivity but also maintenance logistic support, and maintenance team dynamics can be deteriorated. To ensure balanced SM plans, this paper develops a novel SMP where the maintenance workload balancing constraints are accounted for in SM and repairpersons assignment decisions. The resulting joint SM and workload balancing optimization problem (JSM-WLB) is formulated as a BIP and then solved by a GA. Numerical experiments are conducted and demonstrate the accuracy and the added value of the proposed approach.

ABSTRACT. This paper considers the case of leased equipment and examines the proposed pricing and maintenance strategies from the perspective of the supplier. In fact, the latter proposes the lease of equipment, ensures the maintenance activities and offers the sale of consumables for the leased equipment. During the lease duration, the supplier carries out periodic preventive maintenance and corrective maintenance actions with minimal repair and bears the associated costs. Adding to that, a special leasing contract is offered to the customer declaring the periodic rent, the unit selling price of consumables and the proposed maintenance strategy. However, in this work, the periodic rent price of the leased equipment depends on the quantity of purchased consumables in the same period. Therefore, a mathematical formulation for the considered problem is developed in a multi-periodic environment allowing the optimization of the revenues and the costs of the lease, the sale of consumables and maintenance activities during the lease duration. A two-step optimization is implemented, numerical results are presented and the sensitivity analyses are conducted in order to examine the evolution of the proposed optimal solutions according to several key parameters.

ABSTRACT. With the requirements of carbon peaking and carbon neutrality, low-carbon economic development pattern has been widely implemented in countries around the world. Under carbon trading market mechanism, enterprises can reduce carbon emissions through green investment to gain more green preference based market demand, and use carbon assets for innovative financing to alleviate financing problems. This paper investigates a supply chain operational decision problem, which contains purchasing, production, financing and carbon emission reduction investment. For the new problem, a mixed-integer linear programming model is established, which incorporates the decisions of green investment, carbon emission rights pledge financing, multiple products production, resource purchasing, and green preference based market demand. Then the effectiveness of the proposed model is tested by CPLEX using a series of practice-based randomly generated instances. The computational results are compared with that of the referenced model, which does not consider green investment. The comparative results show the significance of revitalizing carbon assets, and offer valuable insights for carbon abatement and sustainable development.

ABSTRACT. Urban electricity demand forecasting in the mid- and long-term with high accuracy is vital for the power systems’ operation and planning. Many factors of uncertainties and non-linearities, in addition to economic trends and seasonal cycles, should be taken into account during forecasting. This study presents a hybrid machine learning model, i.e., ARIMA-LSTM (ARLS), for accurate mid- and long-term urban electricity demand forecasting. The model adopts the following three-fold ideas: a) incorporating various factors that affect electricity load, such as economic trends and seasonal cycles, and then selecting important features from them; b) employing an autoregressive integrated moving average (ARIMA) model to decompose the time series into level, trend, and seasonal components, and a long short-term memory (LSTM) network to account for the relevant non-linear features; and c) fusing the outputs of each learning module by using ensemble mechanism to enhance the forecast accuracy. By doing so, it combines the advantages of both traditional and ML approaches for time-series analysis. The model is tested on the monthly electricity consumption data of two Chinese cities, and the results indicate that it surpasses the existing state-of-the-art models in terms of performance.

ABSTRACT. Due to global competition and ongoing technological advancements, modern manufacturing requires highly flexible and responsive systems to adapt to market changes. Reconfigurable manufacturing systems (RMS) facilitate this responsiveness by virtue of their fundamental characteristics. This study looks into RMS configuration design from the standpoint of a sustainable and scalable system in uncertain environment. This solution makes use of Reconfigurable Machine Tools (RMT). In fact, by adding/removing modules to/from a certain modular reconfigurable machine, its manufacturing capabilities might be improved while remaining cost-effective. To address this issue, a mixed integer linear multi-objective scenario-based robust programming to determine the best set of RMTs is presented. The objective functions are minimizing overall cost and energy Consumption. By solving the illustrative numerical example, the output of the singular and multi-objective model in the robust and deterministic mode is investigated.

ABSTRACT. This paper focuses on addressing the stochastic capacitated Disassembly Lot Sizing (DLS) problem, which aims to determine both the quantity of End-of-Life (EoL) products and the timing of their disassembly in an uncertain environment. The objective is to satisfy the demand for components and modules while considering random yield factors. Notably, this work introduces a two-stage stochastic integer programming model that addresses the complexities of a multi-level disassembly structure with dynamic demands for the different components at each level and the yield uncertainties. the objective of the proposed model is the minimise the holding, backlogging expected costs along the overcapacity and setup costs. Computational experiments are performed on several instances to evaluate the performance of the suggested model, and the benefits and implications of this research are highlighted.

ABSTRACT. The growing trend of population aging has resulted in an increasing demand for long-term care and medical assistance. To address this, non-contact physiological measurements have made significant advancements by harnessing the potential of the IoB in emerging technologies. While thermal imaging techniques have been explored for respiratory rate measurement, limited research focuses on scenarios where facial obstructions, like glasses or masks, are present. In this study, we propose an IoB-enhanced method that utilizes low-resolution continuous thermal images and advanced image processing techniques. By accurately segmenting the foreground and transforming the data into frequency domain information after temporal domain denoising, we obtain valuable insights. A deep neural network, trained with the fusion of IoB insights and data from a piezoelectric respiratory belt, intelligently selects the most suitable region of interest to measure respiratory rate accurately. Our research prioritizes optimizing computational efficiency to ensure the method's feasibility within low-power embedded systems, achieving realtime and accurate measurements with a power consumption of only 10 watts. By integrating IoB principles into our method, we enhance connectivity, data exchange, and behavioral insights, resulting in a more effective respiratory rate measurement system. Experimental results demonstrate the superior accuracy of our IoB-enhanced method, even in scenarios where subjects wear glasses and masks while maintaining freedom of head movement. Although our model training requires a significant amount of data, our approach substantially reduces limitations imposed by facial obstructions in thermal imagingbased respiratory rate measurement. This breakthrough sets the stage for the future adoption and advancement of IoB-driven respiratory rate monitoring.

ABSTRACT. El Salvador has 23 volcanoes that have experienced significant volcanic activity. Volcanoes emit various gases through their crater and floor. Therefore, monitoring the presence and intensity of these gases (mainly CO2) at the volcano's rim is crucial to assess the risk of an eruption. In El Salvador, the Ministry of Environment and Natural Resources (MARN) is responsible for monitoring the $CO_{2}$ gas on volcanoes. Specifically, this government agency only have a LI-803 sensor to carry out this task. However, this equipment is over 15 years old and lacks wireless and mobile functionalities. This work proposes the design of a portable $CO_{2}$ flow measurement and monitoring IoT system using the LI-COR 830. The Bluetooth IoT system is designed for use it by volcanologists for the profiling and mapping of CO2 in a volcano with the help of an Android application, in order to verify the flow measurement in real-time. This work was developed using the IoT architecture reference model, with an ATmega328 microcontroller and an HC06 Bluetooth transceiver as central components of the IoT Node. In addition, we develop a mobile application using Java tools and Android Studio to visualize and store data. We perform a field camping in the Santa Ana volcano located in El Salvador.

ABSTRACT. With the growing deployment of smart metering systems, ensuring secure and reliable communication between smart meters and utility providers is crucial. The DLMS/COSEM Security Suite 1 offers a standardized framework for secure data exchange and authentication in smart metering applications. This paper focuses on the implementation of an PSA-based OIML R46 communication unit, following the principles of the Platform Security Architecture framework, to achieve compliance with DLMS/COSEM Security Suite 1 specifications. The proposed communication unit integrates cryptographic algorithms, secure storage mechanisms, and key management protocols to guarantee the confidentiality, integrity, and authenticity of the transmitted data. Furthermore, the unit undergoes rigorous testing using the DLMS Conformance Test Tool V3.1 to validate its compliance and interoperability. The results demonstrate the successful implementation of a secure OIML R46 communication unit that meets DLMS/COSEM Security Suite 1 requirements and has undergone thorough CTT V3.1 testing.

ABSTRACT. In today's era of information explosion, when information and knowledge are transmitted through social platforms, people often misbelieve wrong information or false information maliciously created, causing varying degrees of impact on society. This process is called "Infodemic". The term "information epidemic" first appeared during the SARS epidemic in 2003. False information spread rapidly and massively around the world through various communication channels, causing national security, economy, and politics to be affected. Therefore, this research applies the latent dirichlet allocation (LDA) method into the topic model, combined with TF and TF-IDF for COVID-19 fake news detection comparison. As the result of five classification models comparison - SVM, random forest, XGBoost and AdaBoost, the LDA combined with TF-IDF features can improve both SVM and random forest models of F1-score, among which the SVM model has the most significant improvement effect. After 10-fold cross-validation, the average F1-score growth rate of SVM increased by 1.13%, the accuracy was 98.04%, and the F1-score reached 98.10%.

ABSTRACT. Drone Routing Problem with Time Window (DRPTW) is an important extension of the logistics service routing optimization problem and has been proved to be an NP-hard problem. In this paper, DRPTW is formulated as a mixed integer nonlinear programming model considering nonlinear energy constraints. To solve this problem efficiently, we propose a hybrid metaheuristic algorithm combined with a genetic algorithm (GA) to generate a set of feasible routes. In addition, we incorporate a tabu search algorithm (TS) that uses a taboo table to store previously encountered solutions, ensuring that the best solution is not overlooked. To evaluate its performance, we test it with Solomon's VRPTW benchmark instances. The results show that a good solution can be obtained within an acceptable computation time. Furthermore, for the solution of the same delivery sequence route, we compared the impact on delivery completion with and without energy constraints, and found that the original delivery solution would be ineffective if the nonlinear energy constraints of the drone are not considered or approximated.

ABSTRACT. This paper proposes a new mode of vessel-UAV collaborative inspection to address the challenges of high cost, low frequency, and high risk in the traditional offshore oilfield inspection. In the studied problem, a patrol vessel with a UAV departs from the port, sails to a specific location, releases the UAV for oilfield inspection, and retrieves the UAV at another location after the inspection. Considering the spatio-temporal cooperation between ships and UAVs and the variation of endurance capability of UAVs, the collaborative inspection routing problem of "single vessel and UAV take off to visit multiple target points" is studied, with a target of minimizing the total variable cost of the vessel and UAV and the sailing time of the vessel. A mixed integer second-order cone programming model based on the time period and the moment of time was established, respectively, to calculate the objective function value of the inspection routing cost. The commercial solver CPLEX is used to solve the proposed models. Numerical experiments based on actual and randomly generated cases are conducted to verify the efficiency of the model. In addition, sensitivity analysis is performed in our paper, including UAV endurance and speed, which can inspire managers for scheduling optimization in practice.

ABSTRACT. With the development of the times and the change of people's lifestyles, people are opting to order food online. For the perishable food, the delivery time and quality requirements are getting stricter. However, in practice, the production of perishable food mostly relies on experience decision-making, and the distribution mainly relies on takeaway riders, which affects the efficiency of the system. This paper aims to optimize the integrated scheduling of perishable food production and distribution with unmanned aerial vehicles (UAVs). A mixed-integer linear program (MILP) is proposed to formulate the problem. We resort to CPLEX solver with branch-and-cut algorithm to solve the problem. Experiments on randomly generated instances with up to 50 customers demonstrate the correctness and effectiveness of the proposed model.

ABSTRACT. Nowadays, automated guided vehicles (AGVs) are widely used in various fields to fulfill the transportation demands of factories or workshops due to their intelligence, flexibility, and efficiency. Researchers have widely explored AGV scheduling problems (ASP) in recent years. However, most of the contributions on the topic frequently needed to pay more attention to the issues related to battery constraints and recharge. This paper studies the AGV scheduling problem with flexible charging (ASP-FC). Each AGV can only work on one transfer task involving processing time and energy consumption. The AGVs can charge at any battery level without considering the critical threshold limit. In addition, our study also finds the situation where AGVs wait at the charging stations due to the limited number of availability. Our objective is to minimize the maximum completion time(makespan). The decisions include determining which transfer tasks are assigned to which AGVs, the sequencing of transfer tasks, which charging operations of AGVs are assigned to which charging stations, the charging sequencing, and the charging time of each AGV. We first propose a mixed-integer linear programming model to solve this problem. Many test instances are solved using Cplex, and the results prove the importance of studying this problem.

ABSTRACT. In this paper, to solve the scheduling problem of discrete event systems modeled by Place-timed Petri nets, a timed state graph with time information is designed to abstract the state space of Place-timed Petri nets. The feasible firing transitions sequence with the minimal duration is kept in the timed state graph. Furthermore, a modified timed state space is further designed with smaller size than the timed state graph, and it also keeps the feasible firing transitions sequence with the minimal duration. By searching these timed state graphs, we can find the optimal solution, i.e., the feasible firing transitions sequence with minimal duration directly.

ABSTRACT. The fuzzy controller refers to a powerful intelligent control tool, which can provide control systems with a higher efficiency than other controllers, because it has in its characteristics, components that correspond to the requirements present in control problems involving complex and nonlinear systems, with the presence of computational inaccuracies related to the input parameters in the systems. This paper describes the development of a fuzzy controller for motion control of a simplified robotic lumbar spine model. The behavior of the controller was simulated using a simplified robotic model that reproduces the movement of two vertebrae. The results of the simulations of the experimental tests, show relevant characteristics of the controller about the response speed of the system, reduction of the tracking error of the system and minimal vibration in the tracking process, ensuring robustness in the motion path of the system.

ABSTRACT. When using timed Petri nets for modeling, Petri net model often tends to be huge, and this makes it difficult to design neural networks of Petri nets. To address this issue, this paper proposes a resource scheduling system automated guided vehicle(AGV) path planning method based on colored Petri nets and graph neural networks. First, the AGV system is described as a path Petri net, and different task types are assigned to the Petri net in different colors based on different process requirements, so that the same place or transition in the Petri net can represent multiple meanings, greatly simplifying the structure of the Petri net. Secondly, using the topological structure of the Petri net, a graph neural network that can be used on colored Petri nets is designed. The topological structure of the Petri net is divided into four categories, and is applied to the graph neural network as a correlation matrix for feature aggregation, achieving higher accuracy with fewer parameters. Thirdly, deep reinforcement learning is used to train the graph neural network, and the trained model is used as a heuristic to construct the A* algorithm to solve the scheduling problem of the system. Finally, several numerical experiments are conducted, and the results show that the scheduling method based on colored Petri nets and graph neural networks can simplify the structure of the Petri net while solving the joint problem of AGV task allocation and path planning in a limited time.

ABSTRACT. Breast cancer detection is the process of identifying the presence of cancer cells or tumors in the breast tissue. It is a challenging task for doctors and researchers. Early detection of breast cancer is crucial for successful treatment and improved diagnosis and prognosis. There are several methods used for breast cancer detection. The diagnostic techniques used are time consuming and often costly. The goal of this paper is to reduce the time to diagnosis breast cancer using Artificial Intelligence (AI) techniques. To this end, we propose an IA-based approach to train, test and validate three supervised Machine Learning algorithms: Logistic Regression, Random Forest Classification and Decision Tree for the classification of breast cancer into cancerous (malignant) and non-cancerous (benign). The performance of each algorithm is evaluated by some performance indicators such as accuracy and precision. A comparative study is conducted to determine which method gives the best results and outperforms the other methods. The Random Forest Machine approach outperformed the other Machine Learning algorithms with a score of 99.5% on the training data and 96.5% on testing data. As a result, it is the most effective algorithm for diagnosing breast cancer.

ABSTRACT. As the key link to realize the function drive of mechanical equipment and its control system, the engine needs to have good operation and fault diagnosis and maintenance capability. Based on digital twin technology and long and short term memory recurrent neural network, the article proposes an engine fatigue monitoring method and life prediction model. The method uses digital twin technology to establish an engine control system simulation model to obtain the engine state parameters and fatigue value life related data set according to the engine composition structure and working principle; the obtained data set is trained by LSTM neural network to determine the engine life prediction model, and the real-time data of DT model is used as the test data set of LSTM network to complete the testing work, so as to achieve the prediction of the remaining engine life results. Finally, we select the Turbofan Engine Degradation dataset and compare and analyze the prediction results of BP and RNN networks through simulation results to prove the effectiveness of the proposed method.

ABSTRACT. The recent rise of urbanization and population, as well as the emergence of Just In Time (JIT) manufacturing strategy, have driven businesses such as distribution firms and logistic service providers into major problems. This results in traffic congestion and high competition between companies. This paper outlines how operations research methodologies are used to guide the decision-making process and improve operational efficiency in distribution tasks. It treats a complex routing problem faced by a Tunisian ice distribution firm called LE GLACON. This problem may be defined as a Heterogeneous Fleet Vehicle Routing Problem (HFVRP). We tackle the problem using a two-phase Tabu Search (TS) algorithm. Based on the company’s case study, the resulting distribution plan from our approach shows remarkable cost savings compared to their handcrafted approach. Moreover, the approach converges fast, allowing the logistic planner to focus on higher-value activities.

ABSTRACT. Planning of production orders in an assembly-based industry highly depends on the available components. Typically, the production plan is followed by a material requirement planning to ensure the availability of the required component and order the needed quantities from suppliers. This is why routing these elements and their availability greatly impacts the production plan and vice versa. The coordination of both plans simultaneously can improve the quality of the planning decisions and reduce operational costs. This research studies the effect of integrating routing decisions with the production plan considering environmental impact. We developed a mixed integer linear programming model; the objective function of the problems includes the operational inbound routing costs, the holding cost of components in the manufacturing site's inventory, the lateness cost of the production order, and minimizing greenhouse gas emissions. Our study is exemplified in a numerical example where we present three different scenarios, highlighting how prioritizing the manufacturing site's different objectives can directly impact the production plan of the manufacturing orders.

ABSTRACT. Drones or Unmanned Aerial Vehicles (UAVs) have the potential to modernize and transform the logistics industry. They are compact, highly mobile, economical to operate, and most importantly, have a significantly smaller carbon footprint than traditional means of transportation. As a result, they have the capability to navigate our increasingly congested cities with ease and thus propel us toward a more sustainable and environmentally friendly future. In this study, we develop an energy optimization routing model for vertical drone delivery with load dependent energy consumption in a multi depot setting. We solve a model problem and carry out numerical experiments to understand how different parameters affect the optimization model. Although energy optimization studies have been carried out for drones operating in horizontal planes, we were not able to find much literature about deliveries in vertical planes considering the gravitational penalty, which makes this study unique. Hence, this serves as a foundational investigation for further experimentation to be carried out in this domain. The paper also discusses a practical example in the form of a problem of a real-life application of this study.

ABSTRACT. This study consists in developing an analytical model in order to determine an optimal planning for manufacturing reconditioning transport supply chain MRTSC.

ABSTRACT. Producer diversification and order allocation are problems in the operational research field that require decision-making skills from the side retailers so that they can choose from who and when to place orders to meet their demand. These tasks are especially challenging in the agricultural field since orders need to be placed months in advance for some products to take into account the time required for planting and farming the produce. This work presents an exploratory model that tackles the challenge of farmer selection and order allocation. We offer a linear programming model that minimizes the demand cost while considering quantity-dependent pricing and a variety of constraints that agricultural retailers face.

ABSTRACT. Inspired by the fact that production capacity in the food industry is uncertain due to unexpected events, such as downtime, this work investigates a stochastic closed-loop supply chain for multiple perishable foods that are carried by returnable transport items, where production capacity is uncertain. For the new problem, we propose a non-linear chance-constrained model. Then it is transformed into a mixed-integer linear programming model via the sampling average approximation and the distribution-free methods such that it can be solved by commercial solvers, like CPLEX. Numerical experiments on randomly generated instances show that the distribution-free method performs better than the average approximation method. Some managerial insights are also presented and discussed.

ABSTRACT. Evaluating supply chain (SC) disruption risks in the context of data scarcity and ripple effects is essential because uncertain disruptions can propagate throughout the SC, resulting in negative impacts on SC performance. Considering that different decision-makers have varying risk tolerances, this can affect the expected disruption risk assessment results. To tackle the problem, this paper proposes a methodology that employs probability intervals to capture uncertain parameters, utilizes dynamic Bayesian networks (DBNs) to model risk propagation, and incorporates risk deviation variables to quantify decision-makers' risk tolerance. Subsequently, a bi-objective optimization model is developed to evaluate the optimal Pareto front with respect to both SC disruption risk and deviation budget. For solving the studied problem, the linearisation and epsilon-constraint methods are developed. To demonstrate the feasibility and effectiveness of the proposed model, the numerical experiment is carried out, and the results are analysed to draw managerial insights.

ABSTRACT. The increasing demand for high-speed wireless communication has made it imperative to provide reliable Handover(HO) methods in 5G and beyond cellular networks. With the proliferation of data-hungry applications, increasing cell density (small cells) has become necessary to achieve higher data rates. However, this poses a challenge in terms of maintaining Quality of Experience (QoE) for end-users, as there is a risk of call drops during hand-offs. To address this issue, we propose an optimized algorithm that leverages ant colony optimization technique to reduce call drop rate. The effectiveness of the proposed algorithm is evaluated through simulations in MATLAB, and the results demonstrate its potential to improve the throughput for end-users. This paper presents a novel approach towards achieving reliable hand-offs in next-generation 5G and beyond heterogeneous networks.

ABSTRACT. Surrogate-Assisted Evolution Algorithms (SAEAs) use surrogate models to approximate true objective functions and they have been successfully used to solve expensive problems. However, when solving expensive problems, the complicated search space makes solvers hard to locate the promising areas. To effectively solve expensive problems, this paper attempts to improve the solving ability and find more promising solutions by learning experience from performing easier tasks. And thus, an Easy And Similar Task-aided Evolutionary Optimization (EASTEO) framework is proposed. Specifically, an easy and similar task is specially designed to obtain more useful knowledge and experience. And then, the solution experience of this created task can be transferred to the original expensive task, which can help evolve the expensive task. This work uses commonly used benchmark test problems to verify the performance of the proposed framework.

ABSTRACT. Recently, there has been a growing interest in using Machine Learning (ML) techniques to design efficient algorithms for solving combinatorial optimization problems. Integrating ML models has brought greater intelligence to the search process and enhanced solution quality. In this study, we demonstrate the effectiveness of Logistic Regression (LR) in estimating the probability of an element being a part of the optimal solution in binary combinatorial optimization problems. This probability estimation provides a valuable tool for guiding the search process of the Simulated Annealing (SA) algorithm toward promising regions of the solution space. Experiments conducted on Multidimensional Knapsack Problem (MKP) and Subset Sum Problem (SSP) instances reveal that combining the LR model with the SA algorithm enhances the search process, which leads to higher-quality solutions.

ABSTRACT. This paper addresses an enhancement of the recently proposed method to overcome the unmeasurable premise variables in TS nonlinear discrete time systems in observers design \cite{Ichalal2016}. The approach is based on immersion and state extension for tow main objectives: the first one is that the new state vector of the target system after immersion should contain all the state vector of the original system in order to avoid the inverse transformation. Secondly, express the nonlinear system in an adequate form for its transformation in TS form without any unmeasured premise variable. This state extension is performed by introducing simple functions depending on measured variables and their delayed measures. Examples are presented to illustrate the proposed approach.

ABSTRACT. This paper develops a novel high-gain observer for nonlinear systems that has a lower gain compared to the standard high-gain observer. This new observer called Artificial Delayed High Gain Observer (ADHGO) is obtained by combining the standard High-Gain methodology with two steps: the first one is defining artificial delayed outputs of the system and incorporating them in the proposed observer’s structure. The second approach is applying Taylor’s series to these artificially delayed measurements in order to annihilate the nonlinear unknown or uncertain term and then allows to reduce the gains of the high gain observer. Through analytical developments the conditions of existence of the observer and the choice of the artificial delays are determined. Consequently, the obtained lower gains allow to answer to the problems of peaking phenomenon and noise amplification. The performances of the proposed observer are illustrated through a numerical example with comparisons and discussions.

ABSTRACT. This paper explores the issue of H∞ control problem for continuous-time fractional-order descriptor systems with a fractional-order derivative α in the range of 1 ≤ α < 2. First,we propose matrix inequalities to characterize the admissibility of the closed-loop system by a state feedback control and an observer-based controller. Using these results, we show that the solvability of a set of matrix inequalities is necessary and sufficient to the existence of a dynamic output feedback controller that guaranties the admissibility and the H∞ norm condition for the closed-loop system. Furthermore, several numerical examples are presented to demonstrate the practicality and effectiveness of the theoretical findings derived in this paper.

ABSTRACT. Unknown disturbances are one of the most critical factors impacting control performances, and the design of globally convergent disturbance observers for estimation of disturbances in real-time plays an important role in improving the performance of control systems. This work presents the design of a novel disturbance observer based on model reference adaptive control to relax the restrictive gain conditions in conventional nonlinear disturbance observer design. The proposed Model Reference Adaptive Control-based Nonlinear Disturbance Observer (MRAC-NDO) features a simple structure with theoretically guaranteed global convergence for estimating unknown disturbances appearing in affine systems. The proposed strategy is applied to unmanned aerial vehicles (UAVs) based on cascaded control incorporated by MRAC-NDO disturbance compensation. Numerical results validate the effectiveness of the proposed solution.

ABSTRACT. The fifth generation of wireless network (5G) is a recent networking technology able to provide numerous advantages compared to the previous generations of wireless networks. 5G is a packet switched wireless system with extended coverage and high speed. It is designed to reduce latency and improve service flexibility. However, the number of mobile users and services has substantially increased over the last years, hence the need in improving the network capacity and coverage. The goal is to ensure a good quality of service and satisfy the increased user’s request. In this work, we are interested in coverage planning for a 5G network. The aim is to determine the optimal number of base stations (gNodeB) to be used in the network with minimum total power consumption. These stations must be well positioned so as to have maximum coverage of the area to be covered. We try to solve this problem by using a meta-heuristic approach. The proposed approach is based on the Grey wolf optimizer. We consider as a geographical area the City of Oran situated in the north of Algeria to evaluate our approach. The numerical results are interesting and show the applicability of the proposed approach for wireless networks optimization.

ABSTRACT. Conventional wireless networks are half-duplex and most of them use contention-based protocols. These protocols usually adopt a principle of contention with collision avoidance and infer a collision occurrence very late from the absence of an acknowledgement after data transmission, thus causing low network performance. Wireless full-duplex enables simultaneous transmission (TX) and reception (RX) on the same channel. By exploiting this functionality, this work proposes a novel design that enables contention with collision detection (CCD) to improve the network performance. With this design, in contention, a node exploits its TX antenna to transmit a signal for channel contention, while exploiting its RX antenna to sense if other nodes are transmitting too. By checking the status of TX and RX antennas, the node can detect the contention collision before data transmission and hence obtains an opportunity to avoid data collision effectively. This work then develops a theoretical model to analyze the collision probability of this design. Extensive simulation results verify the effectiveness of our proposed design in improving wireless network performance.

ABSTRACT. In the context of HWN (Heterogeneous Wireless Networks), the claim for the best connectivity is a mandatory challenge for the users. As all of us know, there is a panoply of RATs (Radio Access Technologies) that users can connect to due to the multi-interface property of the actual user-devices. Indeed, these multi-interfaces users are always seeking for reaching the ABC concept (Always Best Connected). This challenge is also in the heart of the NGN Next-Generation Networks, where scholars are proposing models and algorithms to make in practical the ABC paradigm. So, the idea is to make user-devices capable to select the best RAT instantaneously each time the users need to use the network. This process is known as the network selection problem. Simply, the network selection consists on the automatic and transparent selection of the best RAT at a given time; many papers exist in this field using mainly MADM (Multi- Attribute Decision-Making) methods, different implementation of the MADM methods are proposed with various objectives, increasing the QoS, reducing the energy consumption, reducing the cost. The objective of this paper is to address this challenge by proposing a modified machine learning algorithm (an adapted K-means) to make the network selection, the results show that the proposal is more performer compared to the legacy MADM methods.

ABSTRACT. Wireless technology and sensing methodologies have made it possible for wireless signals to detect human behavior and activity, as demonstrated in numerous studies. The recognition of human activity using channel state information (CSI) in commercial WiFi devices is particularly important in applications such as health monitoring, smart home automation, and security monitoring. In this paper, we propose a device-free activity recognition system that utilizes only commercial off-the-shelf WiFi infrastructure and devices, with a focus on practicality, efficiency, and real-world deployment. The system operates in the 5 GHz band. Our experiments, conducted in an indoor environment, demonstrate the effectiveness and robustness of the proposed system in recognizing human activity.

ABSTRACT. In simulation optimization, a ranking and selection (R\&S) problem aims to select the best from candidate solutions, subject to a limited budget of simulation runs. Existing R&S literature focuses on selecting the best solution, based on a ranking criterion defined by the mean performance. Ignoring performance variance in the ranking criterion definition, however, may lead to selecting a very risky solution, with low average performance but high variation. In this paper, we address a new risk-averse R&S problem, which is a generalization of the classic (risk-neutral) R&S problem, by ranking the solutions via the weighted sum of the mean and variance of the performance. For this novel problem, a new approach is developed based on Karush-Kuhn-Tucker conditions, which is a generalization of optimal computing budget allocation (OCBA). Numerical experiments are conducted to show its efficiency.

ABSTRACT. Accurate mid-and long-term forecast of urban electricity demand is crucial to the operation and planning of power systems. Besides economic trends and seasonal cycles, there are also many factors of uncertainties and non-linearities that should be considered during forecasting. To address these problems, a fuzzy time series forecasting model named FTS is proposed in this paper. The model first sorts the monthly electricity consumption of residents in order from smallest to largest, and then segments the monthly electricity consumption by 0.5 times the standard deviation of monthly electricity consumption. Then define fuzzy sets based on the domain interval division of each data factor, that is, define a fuzzy set for each interval. Finally, according to the fuzzy set where the data of the first two moments are located, the next monthly electricity consumption is predicted to rise or fall. The monthly electricity consumption of two Chinese cities is adopted as the benchmark dataset, and the experimental results show that the proposed model outperforms the existing state-of-the-art models in terms of monthly electricity load forecast accuracy.

ABSTRACT. This paper proposes an edge detection algorithm based on statistics, which is very suitable for workpiece positioning in industrial automated production systems; as this algorithm has no threshold, it is not sensitive to changes of light intensity, and can be well adapted to production environment of various light intensities without adjusting any parameters. This algorithm is also very robust to pollutions, water stains, shadows on the surface of the workpiece and various noise interference in the imaging process, which can significantly improve the vision system's ability to adapt to the production environment. This algorithm has high positioning accuracy, and the average deviation of angle detection of this algorithm is within 0.1 degrees under normal conditions, indicating our algorithm has good adaptability to actual working conditions.

ABSTRACT. This study addresses Unmanned Surface Vessels (USVs) task assignment and path planning problems with minimizing the maximum completion time of USVs. First, a mathematical model is developed for the concerned problems. Second, an unsupervised learning algorithm, K-Means++, is employed to assign multi-tasks to USVs. According to the assignment results, five meta-heuristics are used to solve path planning problems for USVs. Finally, experiments are executed to solve 10 cases with different scales. The effectiveness of K-Means++ for task assignment is verified. The results of five meta-heuristics for path planning are reported and analyzed. The harmony search algorithm has the strongest competitiveness among all compared algorithms for solving the concerned problems.

ABSTRACT. The efficient allocation of resources is critical for improving the performance of datacenters in cloud environments. It is considered as NP-hard problem. Traditional optimization methods for resource allocation are often inadequate due to their complexity, lack of performance guarantees and lengthy training times. The Whale Optimization Algorithm (WOA) has recently emerged as a well-accepted alternative for solving constraint optimization problems. This article further into the potential of WOA to address resource allocation problem in cloud datacenters. To overcome the limitations of WOA and constraints optimization, an adaptive penalty function-based method namely APFWOA is proposed to minimize the makespan and execution cost. The APFWOA is evaluated on real and synthetic datasets in a simulated cloud infrastructure environment and compared to other contemporary techniques. The findings of this study indicate that the proposed algorithm surpasses alternative methods in minimizing makespan and execution costs.

ABSTRACT. —Disassembly is of significant importance in maintenance, remanufacturing, and recycling, as it promotes sustainable development and resource utilization. However, due to the rapid advancement of production automation, designers often overlook the fact that manual operations offer greater flexibility and employee satisfaction during task execution, thereby impacting production efficiency and effectiveness. Therefore, considering human factors in disassembly line balancing problems holds significant research significance. In this work, we investigate the multi-objective parallel disassembly line balancing problem, taking into account factors such as worker fatigue. We establish a mathematical model that incorporates worker fatigue and employ the Pareto Envelope based Selection Algorithm II (PESA-II) for solving the proposed problem. Comparative analysis is conducted, comparing the results with other algorithms, including Pareto Archived Evolution Strategy (PESA), Carnivorous plant algorithm (CPA), Fruit Fly Optimization Algorithm (FOA), and Multivariate Singular Spectrum Analysis (MSSA), based on multiple sets of experimental cases to demonstrate the outstanding performance of PESA-II across various metrics.

ABSTRACT. With the rapid development of modern information technology, various advanced logistics distribution models and strategies have been proposed. All related stake holders have paid much attention to city express delivery. Achieving the ideal supply and demand matching of logistics resources is one of the important goals in smart cities. A dynamic logistics distribution network model based on Artificial Hummingbird Algorithm (AHA) is proposed in this paper. Dynamic Partition Network (DPN) can effectively divide a distribution area such that the total distribution consumption is minimized. It is a novel logistics region partition framework that can develop the optimal strategy for each distribution task. AHA is improved with a new coding method and the idea of the Grey Wolf Optimizer, which improves the performance of the model solver and speeds up the output of a better distribution scheme. Extensive experiments on real-world logistics and distribution tasks show that the dynamic division model is able to make substantial improvement in dividing distribution areas over existing technologies.