Search

Dynamic task scheduling problem in cloud manufacturing (CMfg) is always challenging because of changing manufacturing requirements and services. To make instant decisions for task requirements, deep reinforcement learning-based (DRL-based) methods have been broadly applied to learn the scheduling policies of service providers. However, the current DRL-based scheduling methods struggle to fine-tune a pre-trained policy effectively. The resulting training from scratch takes more time and may easily overfit the environment. Additionally, most DRL-based methods with uneven action distribution and inefficient output masks largely reduce the training efficiency, thus degrading the solution quality. To this end, this paper proposes an improved DRL-based approach for dynamic task scheduling in CMfg. First, the paper uncovers the causes behind the inadequate fine-tuning ability and low training efficiency observed in existing DRL-based scheduling methods. Subsequently, a novel approach is proposed to address these issues by updating the scheduling policy while considering the distribution distance between the pre-training dataset and the in-training policy. Uncertainty weights are introduced to the loss function, and the output mask is extended to the updating procedures. Numerical experiments on thirty actual scheduling instances validate that the solution quality and generalization of the proposed approach surpass other DRL-based methods at most by 32.8% and 28.6%, respectively. Additionally, our method can effectively fine-tune a pre-trained scheduling policy, resulting in an average reward increase of up to 23.8%. [ABSTRACT FROM AUTHOR]

Production rescheduling involves re-optimizing production schedules in response to disruptions that render the initial schedule inefficient or unfeasible. This process requires simultaneous consideration of multiple objectives to develop new schedules that are both efficient and stable. However, existing review papers have paid limited attention to the multi-objective optimization techniques employed in this context. To address this gap, this paper presents a systematic literature review on multi-objective production rescheduling, examining diverse shop-floor environments. Adhering to the PRISMA guidelines, a total of 291 papers were identified. From this pool, studies meeting the inclusion criteria were selected and analyzed to provide a comprehensive overview of the problems tackled, dynamic events managed, objectives considered, and optimization approaches discussed in the literature. This review highlights the primary multi-objective optimization methods used in relation to rescheduling strategies and the dynamic disruptive events studied. Findings reveal a growing interest in this research area, with "a priori" and "a posteriori" optimization methods being the most commonly implemented and a notable rise in the use of the latter. Hybridized algorithms have shown superior performance compared to standalone algorithms by leveraging combined strengths and mitigating individual weaknesses. Additionally, "interactive" and "Pareto pruning" methods, as well as the consideration of human factors in flexible production systems, remain under-explored. [ABSTRACT FROM AUTHOR]

In many industrial real-world environments, scheduling necessitates continual reactive adjustments due to unpredictable perturbations, leading to the dynamic transformation of predefined static schedules. In this paper, we introduce a new framework named a two-level evolutionary algorithm (2LEA) as a comprehensive approach for addressing the dynamic flexible job shop scheduling problem. The 2LEA is based on a bi-level optimization design, where the upper level is dedicated to solving the general flexible job shop scheduling problem, and the lower level is used as a new evolutionary operator guided by a probability rate in the upper level, focusing on the optimization of operation sequences. This framework is capable of handling four dynamic events job insertion, job cancellation, machine breakdown, and job replacement using a predictive-reactive rescheduling strategy. By addressing the previously unexplored dynamic event of job replacement, this paper fills a significant gap in the literature and opens avenues for further research. Extensive computational experiments conducted on well-known benchmark instances from the Brandimarte and Hurink datasets demonstrate the effectiveness and efficiency of our proposed scheduling algorithm. Our results showcase the superior performance of 2LEA over state-of-the-art approaches in terms of solution quality, affirming its potential as a leading solution for both static and dynamic scheduling challenges. [ABSTRACT FROM AUTHOR]

Aiming at the problems existing in discrete manufacturing workshops of ships, such as the lack of real-time management and control ability of on-site production status, and the long time delay of real-time interaction and fusion between physical entities and virtual entities data, this paper proposes ship production workshop data mapping and fusion method based on digital twins. Taking a ship machining workshop as an application research example, the integrated business process and data mapping and fusion method of the machining workshop based on digital twins are studied. On this basis, the dynamic scheduling optimization algorithm of ship machining workshops based on digital twins is studied. The results show that the data mapping and fusion method in this study can significantly reduce the time delay of data interaction between the virtual and real objects. It can effectively improve the real-time and production flexibility of dynamic and complex tasks such as emergency insertion and task rework. The ability of real-time data interaction and synchronous real-time mapping between physical entities and virtual entities in dynamic scenes such as production line reconstruction and customized production is improved. [ABSTRACT FROM AUTHOR]

Aiming at the problem that the disturbance factors affect the smooth development of berth hoisting, a hoisting planning architecture based on digital twin is proposed. The berth hoisting sequence planning model and its implementation technology are studied. The model includes three parts: the digital space for formulating the hoisting scheme, the physical space for executing the block hoisting, and the information interaction between the two parts. This paper discusses the berth hoisting planning method based on digital twin, constructs the information exchange platform of digital space and physical space, develops the berth hoisting information exchange platform based on digital twin, and simulates the block hoisting planning by using DELMIA software. An example is given to simulate the operation of the hoisting planning platform. The results show that the work of this paper has a certain reference significance to solve the influence of disturbance factors on berth hoisting sequence planning. [ABSTRACT FROM AUTHOR]

The cloud-edge collaborative manufacturing system (CCMS) connects distributed factories to a cloud centre through cloud-edge collaborative communication, introducing both opportunities and challenges to conventional dynamic job scheduling. Enhancing each factory's scheduling performance by sharing general scheduling knowledge among heterogeneous factories under the consideration of data privacy protection remains challenging. To this end, this paper proposes to solve the dynamic job scheduling in the context of CCMS with a novel federated deep reinforcement learning (FDRL) approach. Within each factory, the scheduling objective involves minimising the makespan and energy consumption, accounting for machine warm-up procedures and real-time dynamics. To handle heterogeneous policy structures, we aggregate their hidden parameters through FDRL, with states, actions, and rewards designed to facilitate the aggregation. The two-phase algorithm, comprising iterative local training and global aggregation, trains the scheduling policies. Constraint items are introduced to the loss functions to smooth local training, and the global aggregation considers production scales and obtained objectives. The proposed approach enhances the solution quality and generalisation of each factory's scheduling policy without exposing original production data. Numerical experiments conducted on sixty scheduling instances validate the superiority of the proposed approach compared to twelve dynamic scheduling methods. Compared to independently trained DRL-based approaches, the proposed FDRL-based approach achieves up to an 8.9% reduction in makespan and a 22.3% decrease in energy consumption through knowledge sharing. [ABSTRACT FROM AUTHOR]

With the rapid development of economic globalization and green manufacturing, traditional flexible job shop scheduling has evolved into the low-carbon heterogeneous distributed flexible job shop scheduling problem (LHDFJSP). Additionally, modern smart manufacturing processes encounter complex and diverse contingencies, necessitating the ability to address dynamic events in real-world production activities. To date, there are limited studies that comprehensively address the intricate factors associated with the LHDFJSP, including workshop heterogeneity, job insertions and transfers, and considerations of low-carbon objectives. This paper establishes a multi-objective mathematical model with the goal of minimizing the total weighted tardiness and total energy consumption. To effectively solve this problem, diverse composite scheduling rules are formulated, alongside the application of a deep reinforcement learning (DRL) framework, i.e., Rainbow deep-Q network (Rainbow DQN), to learn the optimal scheduling strategy at each decision point in a dynamic environment. To verify the effectiveness of the proposed method, this paper extends the standard dataset to adapt to the LHDFJSP. Evaluation results confirm the generalization and robustness of the presented Rainbow DQN-based method. [ABSTRACT FROM AUTHOR]

Assembly Workshops with Island Layout (AWIL) possess flexible production capabilities that realize product diversification. To cope with the complex scheduling challenges in flexible workshops, improve resource utilization, reduce waste, and enhance production efficiency, this paper proposes a production scheduling method for flexible assembly workshops with an island layout based on digital twin technology. A digital twin model of the workshop is established according to production demands to simulate scheduling operations and deal with complex scheduling issues. A workshop monitoring system is developed to quickly identify abnormal events. By employing an event-driven rolling-window rescheduling technique, a dynamic scheduling service system is constructed. The rolling window decomposes scheduling problems into consecutive static scheduling intervals based on abnormal events, and a genetic algorithm is used to optimize each interval in real time. This approach provides accurate, real-time scheduling decisions to manage disturbances in workshop production, which can enhance flexibility in the production process, and allows rapid adjustments to production plans. Therefore, the digital twin system improves the sustainability of the production system, which will provide a theoretical research foundation for the real-time and unmanned production scheduling process, thereby achieving sustainable development of production. [ABSTRACT FROM AUTHOR]

Because disruption happens unpredictably and generates serious impact in supply chain and production environments in the real world, it is important to develop approaches to handle disruption. This paper investigates disruption handling in part input sequencing of flexible manufacturing systems (FMSs). An algorithm is proposed for FMS part input sequencing to handle machine breakage. Evaluation is performed for the proposed algorithm by simulation experiments and result analyses. Finally, conclusions are summarized with managerial implications discussed and further research works suggested. [ABSTRACT FROM AUTHOR]

In contemporary manufacturing and supply chain environments, there are various challenges that companies have to face. Many companies are facing challenges to respond quickly to customers' requirements and to provide customised products cost effectively. Mass customisation (MC) can help companies to provide customised products and services quickly at a low price. Integrated decision-making can obtain performance improvement in many situations. This paper investigates the problem of scheduling integration of flexible manufacturing systems (FMSs) for mass customisation. A mathematical model is formulated for the problem. A set-based dynamic algorithm is developed. Numerical studies are made on the proposed algorithm by simulation and by statistical analysis. Conclusions, managerial implications, and future research suggestions are provided. [ABSTRACT FROM AUTHOR]

In the real world of manufacturing systems, production planning is crucial for organizing and optimizing various manufacturing process components. The objective of this paper is to present a methodology for both static scheduling and dynamic scheduling. In the proposed method, a hybrid algorithm is utilized to optimize the static flexible job-shop scheduling problem (FJSP) and dynamic flexible job-shop scheduling problem (DFJSP). This algorithm integrates the genetic algorithm (GA) as a global optimization technique with a simulated annealing (SA) algorithm serving as a local search optimization approach to accelerate convergence and prevent getting stuck in local minima. Additionally, variable neighborhood search (VNS) is utilized for efficient neighborhood search within this hybrid algorithm framework. For the FJSP, the proposed hybrid algorithm is simulated on a 40-benchmark dataset to evaluate its performance. Comparisons among the proposed hybrid algorithm and other algorithms are provided to show the effectiveness of the proposed algorithm, ensuring that the proposed hybrid algorithm can efficiently solve the FJSP, with 38 out of 40 instances demonstrating better results. The primary objective of this study is to perform dynamic scheduling on two datasets, including both single-purpose machine and multi-purpose machine datasets, using the proposed hybrid algorithm with a rescheduling strategy. By observing the results of the DFJSP, dynamic events such as a single machine breakdown, a single job arrival, multiple machine breakdowns, and multiple job arrivals demonstrate that the proposed hybrid algorithm with the rescheduling strategy achieves significant improvement and the proposed method obtains the best new solution, resulting in a significant decrease in makespan. [ABSTRACT FROM AUTHOR]

According to [Dolgui, Alexandre, and Jean Marie Proth. 2010. Supply Chain Engineering: Useful Methods and Techniques. Vol. 539. Springer.], advancing tactical levels in production systems has led to the disappearance of static scheduling in favour of dynamic scheduling. Additionally, the evolving challenges in the supply chain paradigm have significantly impacted the organisation of production systems. This shift has moved scheduling issues from the tactical to the strategic level, resulting in linear organisations encompassing scheduling decisions. [Proth, Jean Marie. 2007. "Scheduling: New Trends in Industrial Environment." Annual Reviews in Control 31 (1): 157–166. .] emphasised that real-time scheduling in production systems has become a pivotal area of research. He presented several open problems for researchers to address in this context, including (1) the development of real-time algorithms capable of handling multiple operations on the same product and unrelated resources, (2) adapting previous schedules with certain modifications, (3) addressing unforeseen actions that arise randomly in real-time planning, and (4) exploring cyclic scheduling problems with size limits as alternative solutions to heuristic approaches. This paper reviews the evolving trends in light of J.M. Proth's predictions and advice within the aforementioned domains. [ABSTRACT FROM AUTHOR]

As an important branch of production scheduling, the flexible job shop scheduling problem (FJSP) is a typical NP-hard problem. Researchers have adopted many intelligent algorithms to solve the FJSP problem, nonetheless, the task of dynamically adapting its essential parameters during the computational process is a significant challenge, resulting in the solution efficiency and quality failing to meet the production requirements. To this end, this paper proposes an adaptive gray wolf fast optimization algorithm (SS-GWO), which adopts the gray wolf algorithm (GWO) as the basic optimization method, and the algorithm adaptively selects the global search or local search according to the degree of agglomeration of individuals. Firstly, a non-linear convergence factor strategy is employed to control the global exploration and local exploitation capabilities of the algorithm at different stages. This enhances optimization precision and accelerates convergence speed, achieving a dynamic balance between the two. Secondly, the spiral search mechanism of Whale Optimization Algorithm is used in GWO to improve the exploration capability of Gray Wolf Optimization Algorithm. Finally, the effectiveness of SS-GWO model is verified by comparison experiments. The comparison demonstrates the superiority of SS-GWO over the other five state-of-the-art algorithms in solving the 22 classical benchmark test functions. SS-GWO is applied to solve FJSP by means of the standard test function bandimarte calculus. The optimal solution and performance of SS-GWO for solving FJSP are compared with other algorithms. The experimental results show that the SS-GWO algorithm has good optimization performance, and the maximum completion time is reduced by 19% and 37% compared with that of IGWO and GWO, respectively, and the proposed SS-GWO algorithm achieves a better solution effect on flexible job shop scheduling instances, which can satisfy the actual production scheduling needs. [ABSTRACT FROM AUTHOR]

Optimal dispatch in power systems is a complex mathematical model of nonlinear programming with many physical constraints, which is difficult to solve by conventional methods. Thus, intelligent algorithms are now viable options for resolving the nonlinear scheduling issues of microgrids. In this paper, we propose a double-layer optimization strategy based on the multi-point improved gray wolf algorithm (MPIGWO). The inner layer optimizes load profiles with time-of-use tariffs. The outer one achieves a fast search for the optimal solution and prevents getting stuck in a local optimum, which improves the gray wolf algorithm significantly. First, the Bernoulli map is used to randomly generate the initial population. Second, the efficiency of optimization can be improved by modifying the attenuation factor based on the Sin function and then updating the exact weight factor of the position to reasonably select the best position. Finally, an improved dimensional learning-based hunting (IDLH) search strategy is employed to determine the optimal solution. The numerical case study shows that the proposed double-layer optimization strategy can implement dynamic scheduling for distributed power sources while lowering the costs of economic operation and environmental protection significantly. [ABSTRACT FROM AUTHOR]

Production rescheduling involves re-optimizing production schedules in response to disruptions that render the initial schedule inefficient or unfeasible. This process requires simultaneous consideration of multiple objectives to develop new schedules that are both efficient and stable. However, existing review papers have paid limited attention to the multi-objective optimization techniques employed in this context. To address this gap, this paper presents a systematic literature review on multi-objective production rescheduling, examining diverse shop-floor environments. Adhering to the PRISMA guidelines, a total of 291 papers were identified. From this pool, studies meeting the inclusion criteria were selected and analyzed to provide a comprehensive overview of the problems tackled, dynamic events managed, objectives considered, and optimization approaches discussed in the literature. This review highlights the primary multi-objective optimization methods used in relation to rescheduling strategies and the dynamic disruptive events studied. Findings reveal a growing interest in this research area, with “a priori” and “a posteriori” optimization methods being the most commonly implemented and a notable rise in the use of the latter. Hybridized algorithms have shown superior performance compared to standalone algorithms by leveraging combined strengths and mitigating individual weaknesses. Additionally, “interactive” and “Pareto pruning” methods, as well as the consideration of human factors in flexible production systems, remain under-explored.

The extremely competitive shipbuilding sector is going through a digital transformation in the context of Industry 4.0. The spatiotemporal problem has gained considerable attention during hull construction, particularly for blocks. There are fewer studies on intermediate products at lower levels, such as the dynamic spatiotemporal scheduling for hull parts under complex constraints in subassembly workshops. Therefore, this paper proposes a multi-queue two-level optimization method along with a dynamic spatiotemporal scheduling model based on the multi-directed acyclic graph (multi-DAG) and queueing theory. Firstly, a mathematical description of the complex constraints imposed by processing tasks and various resource types is provided. Then, a dynamic spatiotemporal scheduling model and an objective function are proposed to minimize the average time for parts in the system. At the sequencing level, different priority determination methods for queue events based on an improved genetic algorithm and the contribution ratio for release space strategy are proposed. At the service level, the allocation strategies for space and workers are constructed to improve utilization. Finally, several simulation experiments are conducted using the data gathered from the workshop. The proposed method is compared with the heuristic rules and combination rules using several queueing indicators and the space utilization vs. time graphs. [ABSTRACT FROM AUTHOR]

Management innovation is particularly important today when technological convergence is becoming increasingly evident. Furthermore, as employees are the most important resource in work dynamic scheduling, the mobility of employees will bring extremely discrete regression particle swarm algorithm uncertainty to work projects, making its scheduling problem more complex. This paper first proposes to apply a discrete regression particle swarm algorithm based on refinement management as an enterprise management innovation to optimize three indicators of time, cost, and stability by implementing refinement personnel management. Secondly, we set constraints for dynamic tasks to implement management rescheduling to avoid overworking employees and thus maximize profits for the enterprise. Finally, the effectiveness of the discrete regression particle swarm algorithm is verified by selecting 27 benchmark examples to compare from three indicators of contribution and time and cost. The results show that the discrete regression particle swarm algorithm (S-PSO-VNS) has an average time reduction of 22.65%, a cost reduction of 15.21%, and a stability improvement of 35.74% compared to the other two control algorithms in terms of management algorithm quality and performance analysis. The research data illustrates that the discrete regression algorithm proposed in this paper can have high efficiency and effectiveness, and the discrete regression particle swarm algorithm can provide managers of SMEs with more informed decisions on work schedules, thus maximizing profits and achieving more profitability.

In this paper, the overall architecture design of the platform is proposed, and the data architecture and business architecture of the platform are discussed. On this basis, the logistics path planning method of digital collaboration is proposed, including the aspects of path demand splitting and distribution vehicle spatio-temporal network, and the demand splittable logistics path planning model is considered. Finally, the simulation of the workshop logistics dynamic scheduling example is designed, and the dynamic scheduling optimization results are analyzed and verified. The results show that: through optimization, its time penalty cost f1 = 801.38, the total distance traveled by the delivery trolley f2 = 1709m, and the total distance of empty load is 692.15 m, and the dispatching scheme increases the distance of empty load by 59 m . The research results of this paper can provide a reference for digital management and collaboration platform for civil aircraft manufacturing enterprises and provide a reference for research in related fields.

With the rapid development of economic globalization and green manufacturing, traditional flexible job shop scheduling has evolved into the low-carbon heterogeneous distributed flexible job shop scheduling problem (LHDFJSP). Additionally, modern smart manufacturing processes encounter complex and diverse contingencies, necessitating the ability to address dynamic events in real-world production activities. To date, there are limited studies that comprehensively address the intricate factors associated with the LHDFJSP, including workshop heterogeneity, job insertions and transfers, and considerations of low-carbon objectives. This paper establishes a multi-objective mathematical model with the goal of minimizing the total weighted tardiness and total energy consumption. To effectively solve this problem, diverse composite scheduling rules are formulated, alongside the application of a deep reinforcement learning (DRL) framework, i.e., Rainbow deep-Q network (Rainbow DQN), to learn the optimal scheduling strategy at each decision point in a dynamic environment. To verify the effectiveness of the proposed method, this paper extends the standard dataset to adapt to the LHDFJSP. Evaluation results confirm the generalization and robustness of the presented Rainbow DQN-based method.

Massive computational resources are required by a booming number of artificial intelligence (AI) services in the communication network of the smart grid. To alleviate the computational pressure on data centers, edge computing first network (ECFN) can serve as an effective solution to realize distributed model training based on data parallelism for AI services in smart grid. Due to AI services with diversified types, an edge data center has a changing workload in different time periods. Selfish edge data centers from different edge suppliers are reluctant to share their computing resources without a rule for fair competition. AI services-oriented dynamic computational resource scheduling of edge data centers affects both the economic profit of AI service providers and computational resource utilization. This letter mainly discusses the partition and distribution of AI data based on distributed model training and dynamic computational resource scheduling problems among multiple edge data centers for AI services. To this end, a mixed integer linear programming (MILP) model and a Deep Reinforcement Learning (DRL)-based algorithm are proposed. Simulation results show that the proposed DRL-based algorithm outperforms the benchmark in terms of profit of AI service provider, backlog of distributed model training tasks, running time and multi-objective optimization. [ABSTRACT FROM AUTHOR]

Abstract To reduce the dynamic scheduling cost of logistics supply chain and improve customer satisfaction, this paper proposes a dynamic scheduling method for logistics supply chain based on adaptive ant colony algorithm. First, determine the goal of dynamic scheduling in the logistics supply chain. Second, considering supplier satisfaction, transportation costs, and maximum delivery distance constraints, a dynamic scheduling model for logistics supply chain is constructed. Then by smoothing the pheromones and designing a transition function, adjusting factors are introduced to update the pheromone rules. Finally, based on the adaptive ant colony algorithm, the solution of the dynamic scheduling function of the logistics supply chain is solved to achieve the dynamic scheduling of the current logistics supply chain. The experimental results show that after 19 iterations, the method can search for the optimal route A1 group with a length of 33.85 km, with fewer iterations and shorter paths. The total cost is 114,290 yuan, and the degree of cargo loss is low, with a maximum of only 0.14%. The task completion time is short, customer satisfaction is above 0.85, and the scheduling accuracy is 99.9%. It can effectively control costs, improve customer satisfaction, and accurately arrange logistics supply chains.

Resource scheduling is an important part of military operation, especially in key-point air defense under saturation attack. Many achievements have been made in the field of radar resource scheduling and multi-aircraft scheduling. However, there is little research on the integrated scheduling of detection, tracking and attack, which will greatly increase the resource utilization to deal with the resource shortage problem caused by saturation attacks on key places. In this paper, we propose to autonomously accomplish real-time resource dispatching through end-to-end deep reinforcement learning (DRL), while allowing the commander's intervention to accomplish a variety of complex tactics. First, an integrated scheduling model of detection, tracking and interception is proposed and transformed into a sequential decision problem by introducing a disjunctive graph and a graph neural network (GNN) to extract node features. Subsequently, the Proximal Policy Optimization (PPO) algorithm is applied to learn the air defense environment (ADE), which is modeled as an Markov decision process (MDP). Benefitting from the powerful generalization capability of the policy network, our algorithm can adapt to scheduling missions of different sizes. Moreover, we propose a novel Human-Intelligence collaborative dynamic scheduling framework for emergency response. Simulation results indicate that our algorithm generates high-quality scheduling policies for defense resources, exhibiting superior performance than existing methods. In addition, the dynamic scheduling performance of the human and intelligence collaboration approach in response to multiple contingencies is proven. [ABSTRACT FROM AUTHOR]

Workflow scheduling problems have been widely studied in cloud computing and edge computing, which aim to exploit cloud-edge resources to execute workflow tasks considering several constraints and optimization goals. However, in the era of Internet of things, the load of each computing task and the amount of data transferred between computing tasks will fluctuate, which changes the original workflow and needs for a new scheduling plan correspondingly. Existing methods are difficult to quickly cope with these dynamic changes and there are few studies applying neural networks to solve problems in workflow scheduling. To bridge the gap, we propose an innovative supervised learning method which leverages function-fitting strategy of neural networks to link the workflow environment and its optimal scheduling plan. Specifically, our approach can be divided into two steps, the first one is to generate dataset and train a seq2seq-based prediction models. In this step, we develop an algorithm for generating a significant amount of workflow instances while ensuring dataset diversity based on complexity estimation. Then we apply GA, NSGA, NSGA-NN three different types GA-based optimization methods to search optimal solutions. Finally, we construct dataset which includes {workflow, environment configurations → obtained optimal solution} and train a seq2seq-based model. The other part is real-time generation of scheduling plans based on trained seq2seq-based model. Simulation experiments have confirmed that our method is both effective and efficient, demonstrating its ability to adapt to changes in the execution environment, workflow task load, and task data transmission, and effectively schedule tasks in real-time. The simulation results show that the seq2seq-based prediction method can approach 90% of the optimal scheme. [ABSTRACT FROM AUTHOR]

Because disruption happens unpredictably and generates serious impact in supply chain and production environments in the real world, it is important to develop approaches to handle disruption. This paper investigates disruption handling in part input sequencing of flexible manufacturing systems (FMSs). An algorithm is proposed for FMS part input sequencing to handle machine breakage. Evaluation is performed for the proposed algorithm by simulation experiments and result analyses. Finally, conclusions are summarized with managerial implications discussed and further research works suggested.

• The dynamic distributed reconfigurable flowshop scheduling problem is studied via DRL. • The DRL-based intelligent scheduling and reconfiguration systems are modelled. • An expected deep Q-network algorithm with precise state-action values is proposed. • The DRL-based training and execution processes for the studied problem are proposed. • Our proposed expected deep Q-network outperforms some popular DRL algorithms. Distributed manufacturing has become common in the globalized production environment. Facing the tendency of mass customization, distributed workshops are required to have the reconfiguration capability to produce multiple products. Based on the scheduling demands of distributed workshops, this paper studied the dynamic distributed reconfigurable flowshop scheduling problem (DRFSP) with new job arrivals using deep reinforcement learning (DRL) for the first time. The DRL-based intelligent scheduling and reconfiguration systems are modeled by designing efficient rewards, actions, and state features for both scheduling and reconfiguration agents. A new state feature normalization approach is proposed to address the variability in production configurations and dynamics. A novel expected deep Q-network (EDQN) algorithm is proposed to provide more precise state-action values rather than using estimations. Moreover, the DRL-based training and execution procedures for the dynamic DRFSP are provided by combining decision-making, agent learning, and reconfigurable production. Training curves validate the reasonableness of system modeling and the good learning efficiency of the EDQN algorithm. Extensive comparison experiments show that our EDQN algorithm outperforms three popular DRL algorithms and four well-known priority dispatching rules (PDRs) by around 20 % and 89 % respectively, with good generalization ability and time efficiency. The average decision time under dynamics is only 0.65 ms, which enables real-time scheduling. [ABSTRACT FROM AUTHOR]

• Solve a multi-equipment integrated scheduling problem in a U-shaped port. • Consider double-cycling mode and uncertain operation time with a cascade effect. • Incorporate anti-cascade effect into a new robust-proactive scheduling framework. • Integrate a mixed-integer programming model with a bi-objective solution algorithm. • Examine real case study and explore implications in port or other industries. The integrated scheduling of quay cranes, internal vehicles, and yard cranes in container terminals aims to improve port operations and often requires scheduling robustness under uncertainty with cascade effects. In container terminal operations, uncertainty in equipment operating time poses challenges to effective scheduling, as even small fluctuations can create cascade effects throughout the operations, rendering the original schedule ineffective. This research aims to develop a new method that enables a balance between optimization and robustness in container terminal scheduling. Additionally, the double-cycling operation mode and U-shaped port layout, known for their improved efficiency in container terminals, are gaining increasing attention and hence are incorporated into this study. It creates a three-stage hybrid flow shop scheduling problem with bi-directional flows, waiting time, and uncertain operation time. To address the complex problem, a mixed integer programming model is proposed to characterize the integrated scheduling problem, and an index based on complex network structure entropy is designed to evaluate the anti-cascade effect as well as the robustness of the schedule. The index and makespan serve as the bi-objectives, transforming the original problem into a bi-objective optimization one. The non-dominated sorting genetic algorithm-Ⅱ with appropriate coding and decoding rules is utilized to solve the model and obtain a set of Pareto frontier solutions. The feasibility of the new method is verified through a real case analysis. Specifically, comparative analysis with basic stochastic programming, basic robust optimization, triangular fuzzy programming, and maximum gap method are used to demonstrate the effectiveness of the new method. The paper also provides insightful practical implications for port managers, and the genericity of the method could also contribute to its practical values spreading to a wider scope of beneficiaries, such as manufacturing warehousing and distribution management. [ABSTRACT FROM AUTHOR]

System disturbances, such as the change of required service durations, the failure of resources, and temporary tasks during the scheduling process of data relay satellite network (DRSN), are difficult to be predicted, which may lead to unsuccessful scheduling of tasks. A high-efficiency and robust DRSN calls for smarter and more flexible disturbances elimination strategies. Here, we unify the above three system disturbances as temporary task arrival and extend the static scheduling model of DRSN. Specifically, we derive and define a scheduling model that unifies the static scheduling and dynamic scheduling processes. Meanwhile, we propose a k-step dynamic scheduling algorithm considering breakpoint transmission (k-steps-BT) to solve the above model. Based on the principle of backtracking algorithm and search tree, k-steps-BT can eliminate disturbances quickly by rescheduling tasks and can determine the rescheduling scheme when temporary tasks arrive. Finally, extensive experiments are carried out to verify the proposed model and algorithm. The results show that the proposed model and algorithm can significantly improve the task completion rate of dynamic scheduling without drastic adjustments to the static scheduling scheme. [ABSTRACT FROM AUTHOR]

Traditional scheduling techniques are designed to reduce processing times while disregarding energy costs. One way of lowering energy usage is to implement scheduling strategies that distribute tasks to specified resources, which influence the processing time and power usage. Among the primary objectives to be achieved in cloud computing, power and energy consumption of the cloud environment have become issues due to ecological and economic reasons. Despite the existence of research efforts from the past pertaining to the same topic, an ideal solution to this problem has not yet been found. One of the main drawbacks of utilizing cloud computing is that the cloud data centers hosting cloud computing applications use higher volumes of energy, adding to the increased functioning cost and carbon footprint in the environment, which in turn increases the need for energy-efficient systems. In this research work, a hybrid optimization algorithm is presented, which is intended to minimize energy in the cloud computing environment. Thus, the crow search algorithm (CSA) and the sparrow search algorithm (SSA) are combined to obtain the proposed hybrid model. The hybrid approach achieves the optimal position in the shortest period of time and with the least amount of energy, load, and makespan, hence improving system performance. The experiments were conducted using three setups with different task sizes. The analysis while using the task size = 300 shows that the proposed method improved QoS, Resource Utilization (RU) and decreased makespan, energy usage, and load at a rate of 0.5073, 4.4035, 0.0331, and 0.0014, respectively. [ABSTRACT FROM AUTHOR]

Mobile edge computing as an emerging technique can provide services for mobile terminals, and meanwhile the mobility of users brings new challenges. When a user moves across different areas, the system needs to determine whether to migrate service so as to guarantee quality of experience for the user. However, it is difficult to obtain the optimal migration policy in real time due to the huge state space. Considering delay‐sensitive data‐intensive applications run by mobile terminals with limited battery power, an efficient service migration policy should be able to make a good tradeoff among service cost, service delay and terminal energy consumption. Here, an online Monte Carlo‐based service migration (MCSM) policy is proposed to minimize service cost under constraints of deadline and terminal energy consumption. A penalty mechanism is designed to update reward when partial or all constraints are not meet. State‐action value estimation and policy improvement are triggered only on the completion of each episode. Each episode is traversed reversely to calculate the average cumulative reward so as to improve policy. Experimental results show that the proposed approach can improve service success ratio and reduce average service cost compared to the existing service migration policies. [ABSTRACT FROM AUTHOR]

This thesis combines the ABC classification method and hierarchical analysis method to propose the evaluation method of logistics spatial layout and applies it to actual automobile manufacturing enterprises. The spatial layout scheme of logistics nodes in the park has been adjusted and optimized to achieve dynamic inventory classification and improve inventory management. When the demand of the decision-making body decreases, the optimal production and transportation scheduling plan for the automobile industry supply chain is formulated using the dynamic scheduling decision-making model. The results show that the spatial layout optimization scheme of supply chain logistics nodes in the automobile industrial park is RDC-B3, VDC-B5, and PDC-B1. The total cost of supply chain logistics after optimization is reduced by 3.8 × 108 yuan. The inventory items in the storage center are divided into 3 categories: A category for inventory items numbered 1-6, B category for 7-9, and C category for 10-11, to improve the turnover rate of the whole inventory and reduce the impact of inventory risk. When the demand side of the supply chain is reduced, the total cost of scheduling the optimal solution is 138.01 million. The research results of this paper have some reference significance.

In the real world of manufacturing systems, production planning is crucial for organizing and optimizing various manufacturing process components. The objective of this paper is to present a methodology for both static scheduling and dynamic scheduling. In the proposed method, a hybrid algorithm is utilized to optimize the static flexible job-shop scheduling problem (FJSP) and dynamic flexible job-shop scheduling problem (DFJSP). This algorithm integrates the genetic algorithm (GA) as a global optimization technique with a simulated annealing (SA) algorithm serving as a local search optimization approach to accelerate convergence and prevent getting stuck in local minima. Additionally, variable neighborhood search (VNS) is utilized for efficient neighborhood search within this hybrid algorithm framework. For the FJSP, the proposed hybrid algorithm is simulated on a 40-benchmark dataset to evaluate its performance. Comparisons among the proposed hybrid algorithm and other algorithms are provided to show the effectiveness of the proposed algorithm, ensuring that the proposed hybrid algorithm can efficiently solve the FJSP, with 38 out of 40 instances demonstrating better results. The primary objective of this study is to perform dynamic scheduling on two datasets, including both single-purpose machine and multi-purpose machine datasets, using the proposed hybrid algorithm with a rescheduling strategy. By observing the results of the DFJSP, dynamic events such as a single machine breakdown, a single job arrival, multiple machine breakdowns, and multiple job arrivals demonstrate that the proposed hybrid algorithm with the rescheduling strategy achieves significant improvement and the proposed method obtains the best new solution, resulting in a significant decrease in makespan.

In mass manufacturing of OLED displays, photolithography process is a critical bottleneck in production capacity. As a result, effective scheduling of the photolithography process is crucial to the overall throughput, productivity, and efficiency of the production. The processing of OLED panels depends upon heterogeneous resources. Aside from that, the complexity, stochastic nature, and highly dynamic characteristic of the processpresent considerable challenges. To address these, we propose a new framework employing a heterogeneous graph neural networkbased Rainbow algorithm. This framework optimizes the schedule used in photolithography process. Considering the constraints on machines, masks, and the stochastic nature of arrival processes, we optimize for maximizing productivity while minimizing the associated costs. We model the interactions among product lot steps, machines, and masks as a heterogeneous graph. This graph encodes the information of lot steps, machines, and masks into distinctive nodes. We then implement a Graph Attention Networkbased architecture for deep representation learning. This transforms state information into node embeddings for each step of each product lot, facilitating decision‐making. Reinforcement learning agents leverage these embeddings to prioritize products via a parameterized Q‐function. Our extensive experiments demonstrate the superior performance of our approach across multiple evaluation metrics in both static and dynamic environment benchmarks, with a higher winning rate and scheduling reward return when compared to existing reinforcement learning methods. [ABSTRACT FROM AUTHOR]