Pub Date : 2025-11-30DOI: 10.1016/j.cor.2025.107346
Bayi Cheng, Junwei Gao, Lingjun Wang, Mi Zhou
We study a joint scheduling problem where a single batch-processing machine is used to process jobs from two competing agents. Once the jobs are completed, they are delivered by a single vehicle to the agents with distinct delivery times. We propose a polynomial-time approximation algorithm Coordinate Scheduling Algorithm, , to optimize the joint problem. Particularly, an optimal distribution scheme is proposed for two agents problem. Three models are considered, where production constraints are different. First, for the model where jobs have identical sizes, we prove that our algorithm is optimal when the production scheme is given. Second, for the model where jobs have identical processing times, we propose a modified algorithm , which achieves a service span asymptotically at most 1.55 times the optimal. Third, for the model where jobs have arbitrary processing times and sizes, the algorithm can produce a schedule and delivery plan with a service span asymptotically at most twice the optimal. Finally, we evaluate the performance of the proposed algorithm with a set of computational experiments and the results show the effectiveness of our algorithm.
{"title":"Two-agent single batch machine scheduling with a fixed capacity distribution","authors":"Bayi Cheng, Junwei Gao, Lingjun Wang, Mi Zhou","doi":"10.1016/j.cor.2025.107346","DOIUrl":"10.1016/j.cor.2025.107346","url":null,"abstract":"<div><div>We study a joint scheduling problem where a single batch-processing machine is used to process jobs from two competing agents. Once the jobs are completed, they are delivered by a single vehicle to the agents with distinct delivery times. We propose a polynomial-time approximation algorithm Coordinate Scheduling Algorithm, <span><math><mrow><mi>C</mi><mi>S</mi><mi>A</mi></mrow></math></span>, to optimize the joint problem. Particularly, an optimal distribution scheme is proposed for two agents problem. Three models are considered, where production constraints are different. First, for the model where jobs have identical sizes, we prove that our algorithm is optimal when the production scheme is given. Second, for the model where jobs have identical processing times, we propose a modified algorithm <span><math><mrow><mi>C</mi><mi>S</mi><msub><mrow><mi>A</mi></mrow><mrow><mi>p</mi></mrow></msub></mrow></math></span>, which achieves a service span asymptotically at most 1.55 times the optimal. Third, for the model where jobs have arbitrary processing times and sizes, the algorithm can produce a schedule and delivery plan with a service span asymptotically at most twice the optimal. Finally, we evaluate the performance of the proposed algorithm with a set of computational experiments and the results show the effectiveness of our algorithm.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"187 ","pages":"Article 107346"},"PeriodicalIF":4.3,"publicationDate":"2025-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145681549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-28DOI: 10.1016/j.cor.2025.107340
Boyu Li, Weimin Wu, Shuo Wang, Dacheng Li
The growing demand for intelligent logistics accelerates the adoption of automated guided vehicles (AGVs), making the AGV scheduling problem (AGVSP) a central research focus. The matrix manufacturing system (MMS), utilizing AGVs for Just-In-Time (JIT) pickup and delivery, attracts considerable attention as a promising manufacturing paradigm. However, existing studies on AGVSP in MMS primarily focus on task assignment, while neglecting the optimization of execution schedules from a temporal perspective, which is crucial in JIT scenarios. To bridge this research gap, this paper investigates a slack-aware AGVSP in MMS, integrating both task assignment and temporal optimization. A slack-aware execution strategy (SES) is proposed, incorporating forward time slack and backward time tightening to regulate slack time for task execution. To address task assignment, an adaptive large neighborhood search (ALNS) algorithm is developed, incorporating problem-specific initialization, neighborhood operators, and an instance-adaptive acceptance criterion. Experimental results demonstrate that SES significantly reduces earliness penalty, yielding near-optimal schedules while balancing computational complexity and optimization performance. Further comparisons reveal that ALNS outperforms three state-of-the-art metaheuristics and the Gurobi solver. A sensitivity analysis further provides managerial insights into optimizing AGV and workstation parameters.
{"title":"Slack-aware scheduling of AGVs in just-in-time matrix manufacturing systems via adaptive large neighborhood search","authors":"Boyu Li, Weimin Wu, Shuo Wang, Dacheng Li","doi":"10.1016/j.cor.2025.107340","DOIUrl":"10.1016/j.cor.2025.107340","url":null,"abstract":"<div><div>The growing demand for intelligent logistics accelerates the adoption of automated guided vehicles (AGVs), making the AGV scheduling problem (AGVSP) a central research focus. The matrix manufacturing system (MMS), utilizing AGVs for Just-In-Time (JIT) pickup and delivery, attracts considerable attention as a promising manufacturing paradigm. However, existing studies on AGVSP in MMS primarily focus on task assignment, while neglecting the optimization of execution schedules from a temporal perspective, which is crucial in JIT scenarios. To bridge this research gap, this paper investigates a slack-aware AGVSP in MMS, integrating both task assignment and temporal optimization. A slack-aware execution strategy (SES) is proposed, incorporating forward time slack and backward time tightening to regulate slack time for task execution. To address task assignment, an adaptive large neighborhood search (ALNS) algorithm is developed, incorporating problem-specific initialization, neighborhood operators, and an instance-adaptive acceptance criterion. Experimental results demonstrate that SES significantly reduces earliness penalty, yielding near-optimal schedules while balancing computational complexity and optimization performance. Further comparisons reveal that ALNS outperforms three state-of-the-art metaheuristics and the Gurobi solver. A sensitivity analysis further provides managerial insights into optimizing AGV and workstation parameters.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"187 ","pages":"Article 107340"},"PeriodicalIF":4.3,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145681550","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Local container drayage is a critical component of port-hinterland logistics, involving short-distance transportation between container terminals, inland depots, and end customers. This paper investigates the integration of semi-autonomous truck platooning into a two-echelon drayage system and proposes a novel optimization framework that captures the operational characteristics of coordinated vehicle movements in the trunk leg. A mixed-integer programming model is developed to jointly optimize vehicle routing, platoon formation, and cargo assignment, with the objective of minimizing total transportation costs. Given the computational complexity, a tailored tabu search algorithm is designed to efficiently generate high-quality solutions for large-scale instances. Extensive computational experiments on both synthetic and real-world-inspired datasets demonstrate the effectiveness of the proposed approach in terms of solution quality and computational time. The results highlight the potential of semi-autonomous platooning to reduce operational costs, enhance system efficiency, and improve resource utilization, particularly in urban logistics settings. These findings offer valuable insights for public sector decision-making and infrastructure planning in both developed and emerging economies.
{"title":"Two-echelon optimization framework for semi-autonomous truck platooning in container drayage","authors":"Zhaojie Xue , Wenxiang Peng , Jialu Zhang , Rui Chen","doi":"10.1016/j.cor.2025.107343","DOIUrl":"10.1016/j.cor.2025.107343","url":null,"abstract":"<div><div>Local container drayage is a critical component of port-hinterland logistics, involving short-distance transportation between container terminals, inland depots, and end customers. This paper investigates the integration of semi-autonomous truck platooning into a two-echelon drayage system and proposes a novel optimization framework that captures the operational characteristics of coordinated vehicle movements in the trunk leg. A mixed-integer programming model is developed to jointly optimize vehicle routing, platoon formation, and cargo assignment, with the objective of minimizing total transportation costs. Given the computational complexity, a tailored tabu search algorithm is designed to efficiently generate high-quality solutions for large-scale instances. Extensive computational experiments on both synthetic and real-world-inspired datasets demonstrate the effectiveness of the proposed approach in terms of solution quality and computational time. The results highlight the potential of semi-autonomous platooning to reduce operational costs, enhance system efficiency, and improve resource utilization, particularly in urban logistics settings. These findings offer valuable insights for public sector decision-making and infrastructure planning in both developed and emerging economies.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"188 ","pages":"Article 107343"},"PeriodicalIF":4.3,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145734762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-27DOI: 10.1016/j.cor.2025.107339
Fulin Yan , François Clautiaux , Aurélien Froger , Boris Albar
In this work, we propose a generic heuristic for the resource-constrained shortest path problem derived from dynamic programming reformulations of hard combinatorial optimization problems. The approach is a machine-learning (ML)-augmented beam search, where an ML model serves as one of the scoring functions to select candidate paths for expansion, complementing lower and upper bound estimates. Our method offers several key advantages. First, the path features used by our model are generic and only derived from the reformulation, without relying on any additional problem-specific information beyond the graph structure and resource constraints. Second, we manually designed and aggregated features to obtain vectors of fixed length, enabling us to train the model on small to medium-sized instances and apply it to much larger instances. We evaluate our algorithm on two benchmark problems: the Single Machine Total Weighted Tardiness Problem and the Temporal Knapsack Problem, each under two settings: with and without access to optimal Lagrangian multipliers. Our numerical experiments show that integration of ML into an anytime beam search enhances its solution quality in most cases, with at most minor performance degradation in the other cases. They suggest that one should systematically incorporate ML into the approach.
{"title":"Generic machine-learning-augmented beam search for resource-constrained shortest path reformulations of combinatorial optimization problems","authors":"Fulin Yan , François Clautiaux , Aurélien Froger , Boris Albar","doi":"10.1016/j.cor.2025.107339","DOIUrl":"10.1016/j.cor.2025.107339","url":null,"abstract":"<div><div>In this work, we propose a generic heuristic for the resource-constrained shortest path problem derived from dynamic programming reformulations of hard combinatorial optimization problems. The approach is a machine-learning (ML)-augmented beam search, where an ML model serves as one of the scoring functions to select candidate paths for expansion, complementing lower and upper bound estimates. Our method offers several key advantages. First, the path features used by our model are generic and only derived from the reformulation, without relying on any additional problem-specific information beyond the graph structure and resource constraints. Second, we manually designed and aggregated features to obtain vectors of fixed length, enabling us to train the model on small to medium-sized instances and apply it to much larger instances. We evaluate our algorithm on two benchmark problems: the Single Machine Total Weighted Tardiness Problem and the Temporal Knapsack Problem, each under two settings: with and without access to optimal Lagrangian multipliers. Our numerical experiments show that integration of ML into an anytime beam search enhances its solution quality in most cases, with at most minor performance degradation in the other cases. They suggest that one should systematically incorporate ML into the approach.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"187 ","pages":"Article 107339"},"PeriodicalIF":4.3,"publicationDate":"2025-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145681551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-26DOI: 10.1016/j.cor.2025.107342
Roel W.M. van Os, Twan Basten
Automated material handling and transportation of semi-finished products has great potential to enhance the efficiency of Flexible Manufacturing Systems (FMSs). The resulting optimization problems for finding the shortest-makespan manufacturing schedules are complex and often hard to solve. Automated vehicles used for transportation must avoid colliding with one another while completing all required transports in the shortest possible time. The Conflict-Free-Transportation-constrained Flexible Job-Shop Scheduling Problem (CFTFJSSP) is the combination of flexible job-shop scheduling with conflict-free vehicle routing. To solve the CFTFJSSP, we propose an exact Logic-Based Benders Decomposition (LBBD) using both Constraint Programming (CP) and Integer Linear Programming (ILP) techniques. This LBBD-based approach is proven to outperform existing approaches to solving the CFTFJSSP in terms of solution quality on benchmark instances currently available in the literature. For most benchmark instances, our LBBD-based approach finds optimal makespan values. Because of time boxing, only in a few cases, the optimality of the found solution cannot be guaranteed. The solutions found by our LBBD-based approach show a makespan improvement of at least 10% for about half of the benchmarks instances, up to a 35% improvement in the best case, when compared to the heuristic solution approaches from the literature.
{"title":"An exact decomposition-based approach to the conflict-free-transportation-constrained flexible job-shop scheduling problem","authors":"Roel W.M. van Os, Twan Basten","doi":"10.1016/j.cor.2025.107342","DOIUrl":"10.1016/j.cor.2025.107342","url":null,"abstract":"<div><div>Automated material handling and transportation of semi-finished products has great potential to enhance the efficiency of Flexible Manufacturing Systems (FMSs). The resulting optimization problems for finding the shortest-makespan manufacturing schedules are complex and often hard to solve. Automated vehicles used for transportation must avoid colliding with one another while completing all required transports in the shortest possible time. The Conflict-Free-Transportation-constrained Flexible Job-Shop Scheduling Problem (CFTFJSSP) is the combination of flexible job-shop scheduling with conflict-free vehicle routing. To solve the CFTFJSSP, we propose an exact Logic-Based Benders Decomposition (LBBD) using both Constraint Programming (CP) and Integer Linear Programming (ILP) techniques. This LBBD-based approach is proven to outperform existing approaches to solving the CFTFJSSP in terms of solution quality on benchmark instances currently available in the literature. For most benchmark instances, our LBBD-based approach finds optimal makespan values. Because of time boxing, only in a few cases, the optimality of the found solution cannot be guaranteed. The solutions found by our LBBD-based approach show a makespan improvement of at least 10% for about half of the benchmarks instances, up to a 35% improvement in the best case, when compared to the heuristic solution approaches from the literature.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"187 ","pages":"Article 107342"},"PeriodicalIF":4.3,"publicationDate":"2025-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145681978","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-24DOI: 10.1016/j.cor.2025.107341
Slim Belhaiza , Gilbert Laporte
In the Dynamic Vehicle Routing Problem with Multiple Soft Time Windows (Dynamic VRPMSTW), customer requests arrive in real time and must be scheduled within flexible service intervals. This problem is complicated by operational constraints, such as vehicle capacities, travel durations, and heterogeneous fleets, which make it difficult for classical optimization methods to adapt quickly to changing conditions. Following recent trends in contextual optimization, we propose a Data-Driven Dynamic Heuristic that integrates Artificial Neural Networks for predicting travel times and demands into a Dynamic Hybrid Adaptive Large Neighborhood Search (DD-Dynamic HALNS). Using cluster assignment and genetic crossover operators, the method generates high-quality initial solutions and continuously re-optimizes them as new requests emerge, ensuring adaptability and service reliability. The effectiveness of the proposed method is evaluated on real-world logistics data and benchmark instances. Results from real-world delivery operations demonstrate an average distance reduction of 11.6% compared with the current solution, with further improvements up to 15.5% when a 10-minute time window flexibility is introduced. These findings highlight the practical benefits of integrating predictive analytics with heuristic optimization, leading to improved cost efficiency, reduced operational constraints, and enhanced service reliability.
{"title":"A data-driven heuristic for the dynamic vehicle routing problem with multiple soft time windows","authors":"Slim Belhaiza , Gilbert Laporte","doi":"10.1016/j.cor.2025.107341","DOIUrl":"10.1016/j.cor.2025.107341","url":null,"abstract":"<div><div>In the Dynamic Vehicle Routing Problem with Multiple Soft Time Windows (Dynamic VRPMSTW), customer requests arrive in real time and must be scheduled within flexible service intervals. This problem is complicated by operational constraints, such as vehicle capacities, travel durations, and heterogeneous fleets, which make it difficult for classical optimization methods to adapt quickly to changing conditions. Following recent trends in contextual optimization, we propose a Data-Driven Dynamic Heuristic that integrates Artificial Neural Networks for predicting travel times and demands into a Dynamic Hybrid Adaptive Large Neighborhood Search (DD-Dynamic HALNS). Using cluster assignment and genetic crossover operators, the method generates high-quality initial solutions and continuously re-optimizes them as new requests emerge, ensuring adaptability and service reliability. The effectiveness of the proposed method is evaluated on real-world logistics data and benchmark instances. Results from real-world delivery operations demonstrate an average distance reduction of 11.6% compared with the current solution, with further improvements up to 15.5% when a 10-minute time window flexibility is introduced. These findings highlight the practical benefits of integrating predictive analytics with heuristic optimization, leading to improved cost efficiency, reduced operational constraints, and enhanced service reliability.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"187 ","pages":"Article 107341"},"PeriodicalIF":4.3,"publicationDate":"2025-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145615640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-20DOI: 10.1016/j.cor.2025.107337
Jan Pablo Burgard , João Vitor Pamplona , Maria Eduarda Pinheiro
For agent based micro simulations, as used for example for epidemiological modeling during the COVID-19 pandemic, a realistic base population is crucial. Beyond demographic variables, health-related variables should also be included. In Germany, health-related surveys are typically small in scale, which presents several challenges when generating these variables. Specifically, strongly imbalanced classes and insufficient observations within sensitive groups necessitate the use of advanced synthetic data generation methods. To address these challenges, we present a method formulated as a mixed-integer linear optimization model designed to create health variables based on class probabilities. This model incorporates fairness by considering the class distribution across sensitive groups as constraints. Furthermore, we prove that the proposed model possesses unimodularity properties and present a preprocessing technique. This allows us to generate data for large populations, such as Germany’s population of over 80 million. Our numerical tests, using one of the largest German Health Survey (GEDA), demonstrate that our approach yields better classification results than a standard random forest when considering different ages as sensitive groups.
{"title":"An optimization-based algorithm for fair and calibrated synthetic data generation","authors":"Jan Pablo Burgard , João Vitor Pamplona , Maria Eduarda Pinheiro","doi":"10.1016/j.cor.2025.107337","DOIUrl":"10.1016/j.cor.2025.107337","url":null,"abstract":"<div><div>For agent based micro simulations, as used for example for epidemiological modeling during the COVID-19 pandemic, a realistic base population is crucial. Beyond demographic variables, health-related variables should also be included. In Germany, health-related surveys are typically small in scale, which presents several challenges when generating these variables. Specifically, strongly imbalanced classes and insufficient observations within sensitive groups necessitate the use of advanced synthetic data generation methods. To address these challenges, we present a method formulated as a mixed-integer linear optimization model designed to create health variables based on class probabilities. This model incorporates fairness by considering the class distribution across sensitive groups as constraints. Furthermore, we prove that the proposed model possesses unimodularity properties and present a preprocessing technique. This allows us to generate data for large populations, such as Germany’s population of over 80 million. Our numerical tests, using one of the largest German Health Survey (GEDA), demonstrate that our approach yields better classification results than a standard random forest when considering different ages as sensitive groups.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"187 ","pages":"Article 107337"},"PeriodicalIF":4.3,"publicationDate":"2025-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145615639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.cor.2025.107329
P. Hajipour, J. Behnamian
In today’s globalized manufacturing environment, producers encounter unprecedented challenges stemming from intense competition, volatile customer demands, and the imperative to deliver high-quality, customized products. Collaborative production within multi-factory networks has emerged as a vital strategy for optimizing workflows and increasing flexibility. Complementing this approach, seru systems—which segment production lines into smaller, autonomous units—enhance responsiveness to fluctuating demand patterns. However, coordinating and scheduling operations across such networks introduces significant complexity, posing a substantial challenge for production-optimization research. Moreover, preventive maintenance is essential for sustaining system productivity by minimizing unplanned downtime and extending equipment lifespan. This study examines scheduling and order acceptance in a distributed production network that integrates seru systems, requiring customer orders to be allocated among factories—some of which operate under seru configurations. A mixed‑integer nonlinear programming (MINLP) model is proposed and implemented in GAMS. For larger-scale instances, the study introduces two memetic algorithms, each integrating a distinct local-search strategy: one utilizing simulated annealing and the other applying hill-climbing. Comparative analysis against a genetic algorithm demonstrates the superior efficiency of both memetic approaches in solving complex instances. The findings offer significant practical implications, such as reduced operational costs, increased production flexibility, and accelerated order fulfillment. Results further demonstrate that integrating seru systems with preventive maintenance strategies in multi-factory networks enhances system stability and efficiency, supports customized manufacturing, and mitigates downtime.
{"title":"Seru order acceptance and scheduling in the distributed production network considering preventive maintenance","authors":"P. Hajipour, J. Behnamian","doi":"10.1016/j.cor.2025.107329","DOIUrl":"10.1016/j.cor.2025.107329","url":null,"abstract":"<div><div>In today’s globalized manufacturing environment, producers encounter unprecedented challenges stemming from intense competition, volatile customer demands, and the imperative to deliver high-quality, customized products. Collaborative production within multi-factory networks has emerged as a vital strategy for optimizing workflows and increasing flexibility. Complementing this approach, seru systems—which segment production lines into smaller, autonomous units—enhance responsiveness to fluctuating demand patterns. However, coordinating and scheduling operations across such networks introduces significant complexity, posing a substantial challenge for production-optimization research. Moreover, preventive maintenance is essential for sustaining system productivity by minimizing unplanned downtime and extending equipment lifespan. This study examines scheduling and order acceptance in a distributed production network that integrates seru systems, requiring customer orders to be allocated among factories—some of which operate under seru configurations. A mixed‑integer nonlinear programming (MINLP) model is proposed and implemented in GAMS. For larger-scale instances, the study introduces two memetic algorithms, each integrating a distinct local-search strategy: one utilizing simulated annealing and the other applying hill-climbing. Comparative analysis against a genetic algorithm demonstrates the superior efficiency of both memetic approaches in solving complex instances. The findings offer significant practical implications, such as reduced operational costs, increased production flexibility, and accelerated order fulfillment. Results further demonstrate that integrating seru systems with preventive maintenance strategies in multi-factory networks enhances system stability and efficiency, supports customized manufacturing, and mitigates downtime.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"188 ","pages":"Article 107329"},"PeriodicalIF":4.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145683735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-10DOI: 10.1016/j.cor.2025.107319
Zheng Wang , Huiran Liu , Xiaojun Fan
This paper addresses a multi-scenario multi-mode resource-constrained project scheduling problem with the goal of minimizing both the makespan and cost of the project. In order to visualize the changing process of modes and priority relationships in a project, a dynamic activity-mode network graph is introduced. Based on this network, a deep reinforcement learning model based on dynamic heterogeneous graph neural network is designed, and 12 solving models are obtained by training this model using the proximal policy optimization algorithm. The convergence of the model was verified using benchmark instances from the Project Scheduling Problem Library. Meanwhile, based on the characteristics of the solved problem, 360 instances are generated by reproducing the algorithm for generating benchmark instances. The problems are addressed using these 12 solution models and 9 additional comparison algorithms. Furthermore, a sensitivity analysis is conducted regarding the configuration parameters of the problem. The results validate the optimal effectiveness, stability, and generalization ability of the proposed learning model. It also demonstrates that this model can be a robustly better solving model and scheduling scheme according to actual demands.
{"title":"A novel learning model with dynamic heterogeneous graph network for uncertain multimode resource-constrained project scheduling problem","authors":"Zheng Wang , Huiran Liu , Xiaojun Fan","doi":"10.1016/j.cor.2025.107319","DOIUrl":"10.1016/j.cor.2025.107319","url":null,"abstract":"<div><div>This paper addresses a multi-scenario multi-mode resource-constrained project scheduling problem with the goal of minimizing both the makespan and cost of the project. In order to visualize the changing process of modes and priority relationships in a project, a dynamic activity-mode network graph is introduced. Based on this network, a deep reinforcement learning model based on dynamic heterogeneous graph neural network is designed, and 12 solving models are obtained by training this model using the proximal policy optimization algorithm. The convergence of the model was verified using benchmark instances from the Project Scheduling Problem Library. Meanwhile, based on the characteristics of the solved problem, 360 instances are generated by reproducing the algorithm for generating benchmark instances. The problems are addressed using these 12 solution models and 9 additional comparison algorithms. Furthermore, a sensitivity analysis is conducted regarding the configuration parameters of the problem. The results validate the optimal effectiveness, stability, and generalization ability of the proposed learning model. It also demonstrates that this model can be a robustly better solving model and scheduling scheme according to actual demands.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"186 ","pages":"Article 107319"},"PeriodicalIF":4.3,"publicationDate":"2025-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145526122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Accurate travel time forecasting for China–Europe Express (CRE) trains in the international section has become a significant challenge for railway practitioners and academics, with even mainstream deep forest (DF) models and variants encountering unresolved technical issues. This paper introduces a novel dual mechanism DF regression model (DMDFR) that enhances both the predictive performance and interpretability of the DF model to predict travel times of CRE trains in the international section. The proposed DMDFR model incorporates a dual mechanism consisting of an internal and an external mechanism. The internal mechanism addresses the problem of uneven dataset partitioning by adjusting the importance of each sub-forest during cross-validation. Meanwhile, a more interpretable and transparent external mechanism is embedded within the DF framework to tackle technical issues related to error transfer. In addition, the information transfer process utilizes an incremental information transfer approach to minimize the loss of internally represented information and improve the interpretability of the model. The DMDFR model deconstructs the DF gray-box arithmetic principle and develops model arithmetic algorithms using a straightforward, explanatory computational process. Through example analysis, we demonstrate the superiority of the DMDFR model across various statistical metrics. Given the rapid advancement of deep learning, the significant improvements achieved by the DMDFR underscore the importance of researching interpretable deep learning algorithms.
{"title":"International travel time prediction for China–Europe Express trains via interpretable deep learning models","authors":"Jingwei Guo , Xiang Guo , Zhen-Song Chen , Witold Pedrycz","doi":"10.1016/j.cor.2025.107330","DOIUrl":"10.1016/j.cor.2025.107330","url":null,"abstract":"<div><div>Accurate travel time forecasting for China–Europe Express (CRE) trains in the international section has become a significant challenge for railway practitioners and academics, with even mainstream deep forest (DF) models and variants encountering unresolved technical issues. This paper introduces a novel dual mechanism DF regression model (DMDFR) that enhances both the predictive performance and interpretability of the DF model to predict travel times of CRE trains in the international section. The proposed DMDFR model incorporates a dual mechanism consisting of an internal and an external mechanism. The internal mechanism addresses the problem of uneven dataset partitioning by adjusting the importance of each sub-forest during cross-validation. Meanwhile, a more interpretable and transparent external mechanism is embedded within the DF framework to tackle technical issues related to error transfer. In addition, the information transfer process utilizes an incremental information transfer approach to minimize the loss of internally represented information and improve the interpretability of the model. The DMDFR model deconstructs the DF gray-box arithmetic principle and develops model arithmetic algorithms using a straightforward, explanatory computational process. Through example analysis, we demonstrate the superiority of the DMDFR model across various statistical metrics. Given the rapid advancement of deep learning, the significant improvements achieved by the DMDFR underscore the importance of researching interpretable deep learning algorithms.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"187 ","pages":"Article 107330"},"PeriodicalIF":4.3,"publicationDate":"2025-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145518932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号