首页 > 最新文献

Computers & Operations Research最新文献

英文 中文
Using machine learning to identify hidden constraints in vehicle routing problems
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-26 DOI: 10.1016/j.cor.2025.107029
Anna Konovalenko , Lars Magnus Hvattum , Mohamed Kais Msakni
Last-mile delivery involves a series of complex tasks in an unpredictable environment. Decision support tools based on optimization algorithms construct efficient routes for drivers, optimizing the cost of making deliveries. However, drivers often deviate from these routes due to factors not considered in the decision-making process. This discrepancy raises the question of how to identify routes that are useable in real-world scenarios. Our research proposes using modern machine learning techniques to classify routes based on their practical usability. In a controlled environment, we demonstrate that machine learning can learn hidden factors influencing route viability by focusing on variants of the vehicle routing problem with additional constraints like time window, capacity and precedence. For each underlying constraint, we show that a machine learning model can be trained to classify routes based on whether or not they violate the constraint. Using datasets generated from well-known benchmark instances, we present computational experiments to evaluate model performance. We discuss which types of constraints are more challenging to recognize and how large a dataset must be to allow for accurate classification. This research has the potential to improve existing decision tools, enabling them to generate routes that better account for real-world complexities.
最后一英里配送涉及在不可预测的环境中执行一系列复杂任务。基于优化算法的决策支持工具为驾驶员构建了高效路线,优化了送货成本。然而,由于决策过程中未考虑的因素,司机经常会偏离这些路线。这种差异提出了一个问题:如何确定现实世界中可用的路线。我们的研究建议使用现代机器学习技术,根据实际可用性对路线进行分类。在一个受控环境中,我们证明了机器学习可以学习影响路线可行性的隐藏因素,具体方法是将重点放在带有额外约束条件(如时间窗口、容量和优先级)的车辆路由问题变体上。对于每个基本约束条件,我们都展示了机器学习模型可以根据是否违反约束条件进行分类。我们利用从知名基准实例中生成的数据集,通过计算实验来评估模型的性能。我们讨论了哪种类型的约束更难识别,以及必须有多大的数据集才能实现准确分类。这项研究有望改进现有的决策工具,使其能够生成更能反映现实世界复杂性的路线。
{"title":"Using machine learning to identify hidden constraints in vehicle routing problems","authors":"Anna Konovalenko ,&nbsp;Lars Magnus Hvattum ,&nbsp;Mohamed Kais Msakni","doi":"10.1016/j.cor.2025.107029","DOIUrl":"10.1016/j.cor.2025.107029","url":null,"abstract":"<div><div>Last-mile delivery involves a series of complex tasks in an unpredictable environment. Decision support tools based on optimization algorithms construct efficient routes for drivers, optimizing the cost of making deliveries. However, drivers often deviate from these routes due to factors not considered in the decision-making process. This discrepancy raises the question of how to identify routes that are useable in real-world scenarios. Our research proposes using modern machine learning techniques to classify routes based on their practical usability. In a controlled environment, we demonstrate that machine learning can learn hidden factors influencing route viability by focusing on variants of the vehicle routing problem with additional constraints like time window, capacity and precedence. For each underlying constraint, we show that a machine learning model can be trained to classify routes based on whether or not they violate the constraint. Using datasets generated from well-known benchmark instances, we present computational experiments to evaluate model performance. We discuss which types of constraints are more challenging to recognize and how large a dataset must be to allow for accurate classification. This research has the potential to improve existing decision tools, enabling them to generate routes that better account for real-world complexities.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"179 ","pages":"Article 107029"},"PeriodicalIF":4.1,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143526653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Multi-objective flexible job shop scheduling based on feature information optimization algorithm
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-25 DOI: 10.1016/j.cor.2025.107027
Zeyin Guo, Lixin Wei, Jinlu Zhang, Ziyu Hu, Hao Sun, Xin Li
Multi-objective optimization methods are increasingly used in job shop scheduling optimization strategies. However, in the design process of multi-objective optimization strategies, a neighborhood search is performed on all solutions in the optimization algorithm, resulting in a time-consuming search. In the algorithm selection process, feature information carried by individuals is often ignored, leading to a lack of targeted guidance ability in the algorithm. To address the limitations of the existing methods, a multi-objective flexible job shop scheduling method based on a feature information optimization algorithm (FIOA) was proposed. First, a framework of multiple group optimization algorithms was applied to construct diverse groups. Subsequently, a representative individual selection strategy was applied to mine individual offspring information and accelerate population convergence. To balance the exploration ability and computational resources of the FIOA, multiple neighborhood search rules were used to improve the utilization rate of individual offspring. In this study, the parameter configuration of the proposed algorithm was calibrated using the Taguchi method. To evaluate the effectiveness and superiority of the FIOA, each improvement of the FIOA algorithm was evaluated. In addition, it was compared with state-of-the-art algorithms in benchmark tests, and the results showed that the FIOA outperformed the other algorithms in solving flexible job shop scheduling.
{"title":"Multi-objective flexible job shop scheduling based on feature information optimization algorithm","authors":"Zeyin Guo,&nbsp;Lixin Wei,&nbsp;Jinlu Zhang,&nbsp;Ziyu Hu,&nbsp;Hao Sun,&nbsp;Xin Li","doi":"10.1016/j.cor.2025.107027","DOIUrl":"10.1016/j.cor.2025.107027","url":null,"abstract":"<div><div>Multi-objective optimization methods are increasingly used in job shop scheduling optimization strategies. However, in the design process of multi-objective optimization strategies, a neighborhood search is performed on all solutions in the optimization algorithm, resulting in a time-consuming search. In the algorithm selection process, feature information carried by individuals is often ignored, leading to a lack of targeted guidance ability in the algorithm. To address the limitations of the existing methods, a multi-objective flexible job shop scheduling method based on a feature information optimization algorithm (FIOA) was proposed. First, a framework of multiple group optimization algorithms was applied to construct diverse groups. Subsequently, a representative individual selection strategy was applied to mine individual offspring information and accelerate population convergence. To balance the exploration ability and computational resources of the FIOA, multiple neighborhood search rules were used to improve the utilization rate of individual offspring. In this study, the parameter configuration of the proposed algorithm was calibrated using the Taguchi method. To evaluate the effectiveness and superiority of the FIOA, each improvement of the FIOA algorithm was evaluated. In addition, it was compared with state-of-the-art algorithms in benchmark tests, and the results showed that the FIOA outperformed the other algorithms in solving flexible job shop scheduling.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"179 ","pages":"Article 107027"},"PeriodicalIF":4.1,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143509484","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}
引用次数: 0
A comprehensive stochastic programming model for transfer synchronization in transit networks
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-24 DOI: 10.1016/j.cor.2025.107015
Zahra Ansarilari , Merve Bodur , Amer Shalaby
We investigate the stochastic transfer synchronization problem, which seeks to synchronize the timetables of different routes in a transit network to reduce transfer waiting times, delay times, and unnecessary in-vehicle times. We present a sophisticated two-stage stochastic mixed-integer programming model that takes into account variability in passenger walking times between bus stops, bus running times, dwell times, and demand uncertainty. Our model incorporates new features related to dwell time determination by considering passenger arrival patterns at bus stops which have been neglected in the literature on transfer synchronization and timetabling. We solve a sample average approximation of our model using a problem-based scenario reduction approach, and the progressive hedging algorithm. As a proof of concept, our computational experiments on instances using transfer nodes in the City of Toronto, with a mixture of low- and high-frequency routes, demonstrate the potential advantages of the proposed model. Our findings highlight the necessity and value of incorporating stochasticity in transfer-based timetabling models.
{"title":"A comprehensive stochastic programming model for transfer synchronization in transit networks","authors":"Zahra Ansarilari ,&nbsp;Merve Bodur ,&nbsp;Amer Shalaby","doi":"10.1016/j.cor.2025.107015","DOIUrl":"10.1016/j.cor.2025.107015","url":null,"abstract":"<div><div>We investigate the stochastic transfer synchronization problem, which seeks to synchronize the timetables of different routes in a transit network to reduce transfer waiting times, delay times, and unnecessary in-vehicle times. We present a sophisticated two-stage stochastic mixed-integer programming model that takes into account variability in passenger walking times between bus stops, bus running times, dwell times, and demand uncertainty. Our model incorporates new features related to dwell time determination by considering passenger arrival patterns at bus stops which have been neglected in the literature on transfer synchronization and timetabling. We solve a sample average approximation of our model using a problem-based scenario reduction approach, and the progressive hedging algorithm. As a proof of concept, our computational experiments on instances using transfer nodes in the City of Toronto, with a mixture of low- and high-frequency routes, demonstrate the potential advantages of the proposed model. Our findings highlight the necessity and value of incorporating stochasticity in transfer-based timetabling models.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"179 ","pages":"Article 107015"},"PeriodicalIF":4.1,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143509483","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}
引用次数: 0
The two-echelon vehicle routing problem with pickups, deliveries, and deadlines
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-22 DOI: 10.1016/j.cor.2025.107016
M. Arya Zamal , Albert H. Schrotenboer , Tom Van Woensel
This paper introduces the Two-Echelon Vehicle Routing Problem with Pickups, Deliveries, and Deadlines (2E-VRP-PDD), an emerging routing variant addressing the operations of logistics companies connecting consumers and suppliers in metropolitan areas. Logistics companies typically organize their logistics in such metropolitan areas via multiple geographically dispersed two-echelon distribution systems. The 2E-VRP-PDD is the practical problem that needs to be solved within each of such a single two-echelon distribution system, thereby merging first and last-mile logistics operations. Specifically, it integrates the distribution of last-mile parcels from the hub via satellites to the consumers with the collection of first-mile parcels from the suppliers via satellites that return to the hub. Moreover, it considers deadlines before first-mile parcels arrive at the hub, which must be transported further in the network. We solve the 2E-VRP-PDD with a newly developed Adaptive Large Neighborhood Search (ALNS) combined with a post-process integer programming model. Our ALNS provides high-quality solutions on established benchmark instances from the literature. On a new benchmark set for the 2E-VRP-PDD, we find that modifying time restrictions, such as parcel delivery deadlines at the city hub, can lead to an 8.27% cost increase, highlighting the overhead associated with same-day delivery compared to next-day delivery operations. Finally, by analyzing real-life instances containing up to 2150 customers obtained from our industry collaborator in Jakarta, Indonesia, we show that our ALNS can reduce the cost of operations by up to 17.54% compared to current practice.
{"title":"The two-echelon vehicle routing problem with pickups, deliveries, and deadlines","authors":"M. Arya Zamal ,&nbsp;Albert H. Schrotenboer ,&nbsp;Tom Van Woensel","doi":"10.1016/j.cor.2025.107016","DOIUrl":"10.1016/j.cor.2025.107016","url":null,"abstract":"<div><div>This paper introduces the Two-Echelon Vehicle Routing Problem with Pickups, Deliveries, and Deadlines (2E-VRP-PDD), an emerging routing variant addressing the operations of logistics companies connecting consumers and suppliers in metropolitan areas. Logistics companies typically organize their logistics in such metropolitan areas via multiple geographically dispersed two-echelon distribution systems. The 2E-VRP-PDD is the practical problem that needs to be solved within each of such a single two-echelon distribution system, thereby merging first and last-mile logistics operations. Specifically, it integrates the distribution of last-mile parcels from the hub via satellites to the consumers with the collection of first-mile parcels from the suppliers via satellites that return to the hub. Moreover, it considers deadlines before first-mile parcels arrive at the hub, which must be transported further in the network. We solve the 2E-VRP-PDD with a newly developed Adaptive Large Neighborhood Search (ALNS) combined with a post-process integer programming model. Our ALNS provides high-quality solutions on established benchmark instances from the literature. On a new benchmark set for the 2E-VRP-PDD, we find that modifying time restrictions, such as parcel delivery deadlines at the city hub, can lead to an 8.27% cost increase, highlighting the overhead associated with same-day delivery compared to next-day delivery operations. Finally, by analyzing real-life instances containing up to 2150 customers obtained from our industry collaborator in Jakarta, Indonesia, we show that our ALNS can reduce the cost of operations by up to 17.54% compared to current practice.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"179 ","pages":"Article 107016"},"PeriodicalIF":4.1,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143487875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Bidding in day-ahead electricity markets: A dynamic programming framework
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-22 DOI: 10.1016/j.cor.2025.107024
Jérôme De Boeck , Bernard Fortz , Martine Labbé , Étienne Marcotte , Patrice Marcotte , Gilles Savard
Strategic bidding problems have gained a lot of attention with the introduction of deregulated electricity markets where producers and retailers trade electricity in a day-ahead market run by a Market Operator (MO). All actors propose bids composed of a unit production price and a quantity of electricity to the MO. Based on these bids, the MO selects the most interesting ones and defines the spot price of electricity at which all actors are paid. As the bids of all actors determine the price of electricity, a bidding Generation Company (GC) faces a high risk regarding its profit when placing bids as the bids of competitors are not known in advance. This paper proposes a novel dynamic programming framework for a GC’s Stochastic Bidding Problem (SBP) in the day-ahead market considering uncertainty over the competitor bids. We prove this problem is NP-hard and study two variants of this problem solved with the dynamic programming framework. Firstly, a relaxation provides an upper bound solved in polynomial time (SBP-R). Secondly, we consider a bidding problem using fixed bidding quantities (SBP-Q) that has previously been solved through heuristic methods. We prove that SBP-Q is NP-hard and solve it to optimality in pseudo-polynomial time. SBP-Q is solved on much larger instances than in previous studies. We show on realistic instances that its optimal value is typically under 1% of the optimal value of SBP by using the upper bound provided by SBP-R.
{"title":"Bidding in day-ahead electricity markets: A dynamic programming framework","authors":"Jérôme De Boeck ,&nbsp;Bernard Fortz ,&nbsp;Martine Labbé ,&nbsp;Étienne Marcotte ,&nbsp;Patrice Marcotte ,&nbsp;Gilles Savard","doi":"10.1016/j.cor.2025.107024","DOIUrl":"10.1016/j.cor.2025.107024","url":null,"abstract":"<div><div>Strategic bidding problems have gained a lot of attention with the introduction of deregulated electricity markets where producers and retailers trade electricity in a day-ahead market run by a Market Operator (MO). All actors propose bids composed of a unit production price and a quantity of electricity to the MO. Based on these bids, the MO selects the most interesting ones and defines the spot price of electricity at which all actors are paid. As the bids of all actors determine the price of electricity, a bidding Generation Company (GC) faces a high risk regarding its profit when placing bids as the bids of competitors are not known in advance. This paper proposes a novel dynamic programming framework for a GC’s Stochastic Bidding Problem (SBP) in the day-ahead market considering uncertainty over the competitor bids. We prove this problem is NP-hard and study two variants of this problem solved with the dynamic programming framework. Firstly, a relaxation provides an upper bound solved in polynomial time (SBP-R). Secondly, we consider a bidding problem using fixed bidding quantities (SBP-Q) that has previously been solved through heuristic methods. We prove that SBP-Q is NP-hard and solve it to optimality in pseudo-polynomial time. SBP-Q is solved on much larger instances than in previous studies. We show on realistic instances that its optimal value is typically under 1% of the optimal value of SBP by using the upper bound provided by SBP-R.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"179 ","pages":"Article 107024"},"PeriodicalIF":4.1,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143474346","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}
引用次数: 0
A novel hyper-heuristic based on surrogate genetic programming for the three-dimensional spatial resource-constrained project scheduling problem under uncertain environments
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-20 DOI: 10.1016/j.cor.2025.107013
Lubo Li , Jingwen Zhang , Haohua Zhang , Roel Leus
For a class of large and complex engineering projects with limited construction sites, three-dimensional (3D) spatial resources usually become a bottleneck that hinders their smooth implementation. A project schedule is easily disturbed by space conflicts and uncertain environments if these factors are not considered in advance. Firstly, we extend the traditional resource-constrained project scheduling problem (RCPSP) by considering 3D spatial resource constraints under uncertain environments, and propose a new three-dimensional spatial resource-constrained project scheduling problem with stochastic activity durations (3D-sRCPSPSAD). The activity schedule and the space allocation need to be decided simultaneously, so we design the first-fit and the best-fit strategies, and integrate them into the traditional resource-based policy to schedule activities and allocate 3D space. Secondly, a novel hyper-heuristic based on surrogate genetic programming (HH-SGP) is designed to evolve rules automatically for the 3D-sRCPSPSAD. The main goal of the surrogate model in HH-SGP is to construct an approximate model of the fitness function based on the random forest technique. Therefore, it can be used as an efficient alternative to the more expensive fitness function in the evolutionary process. More importantly, the weak elitism mechanism and other modified techniques are designed to improve the performance of HH-SGP. Thirdly, we configure the parameters of 3D spatial resources and generate numerical instances. Finally, from the aspects of solution quality and stability, we verify the efficiency, quality and convergence rate of HH-SGP under different uncertain environments. The effectiveness of the surrogate model, and the performance of the first-fit and the best-fit strategies are also analyzed through extensive numerical experiments. The results indicate that our designed HH-SGP algorithm performs better than traditional heuristics for the 3D-sRCPSPSAD, and the performance of the fitness function surrogate model in HH-SGP is generally better than without it. This study can also help project practitioners schedule activities and allocate spatial resources more effectively under various uncertain scenarios.
{"title":"A novel hyper-heuristic based on surrogate genetic programming for the three-dimensional spatial resource-constrained project scheduling problem under uncertain environments","authors":"Lubo Li ,&nbsp;Jingwen Zhang ,&nbsp;Haohua Zhang ,&nbsp;Roel Leus","doi":"10.1016/j.cor.2025.107013","DOIUrl":"10.1016/j.cor.2025.107013","url":null,"abstract":"<div><div>For a class of large and complex engineering projects with limited construction sites, three-dimensional (3D) spatial resources usually become a bottleneck that hinders their smooth implementation. A project schedule is easily disturbed by space conflicts and uncertain environments if these factors are not considered in advance. Firstly, we extend the traditional resource-constrained project scheduling problem (RCPSP) by considering 3D spatial resource constraints under uncertain environments, and propose a new three-dimensional spatial resource-constrained project scheduling problem with stochastic activity durations (3D-sRCPSPSAD). The activity schedule and the space allocation need to be decided simultaneously, so we design the first-fit and the best-fit strategies, and integrate them into the traditional resource-based policy to schedule activities and allocate 3D space. Secondly, a novel hyper-heuristic based on surrogate genetic programming (HH-SGP) is designed to evolve rules automatically for the 3D-sRCPSPSAD. The main goal of the surrogate model in HH-SGP is to construct an approximate model of the fitness function based on the random forest technique. Therefore, it can be used as an efficient alternative to the more expensive fitness function in the evolutionary process. More importantly, the weak elitism mechanism and other modified techniques are designed to improve the performance of HH-SGP. Thirdly, we configure the parameters of 3D spatial resources and generate numerical instances. Finally, from the aspects of solution quality and stability, we verify the efficiency, quality and convergence rate of HH-SGP under different uncertain environments. The effectiveness of the surrogate model, and the performance of the first-fit and the best-fit strategies are also analyzed through extensive numerical experiments. The results indicate that our designed HH-SGP algorithm performs better than traditional heuristics for the 3D-sRCPSPSAD, and the performance of the fitness function surrogate model in HH-SGP is generally better than without it. This study can also help project practitioners schedule activities and allocate spatial resources more effectively under various uncertain scenarios.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"179 ","pages":"Article 107013"},"PeriodicalIF":4.1,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143474345","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}
引用次数: 0
General Polyhedral Approximation of two-stage robust linear programming for budgeted uncertainty
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-19 DOI: 10.1016/j.cor.2025.107014
Lukas Grunau , Tim Niemann , Sebastian Stiller
We consider two-stage robust linear programs with uncertain righthand side. We develop a General Polyhedral Approximation (GPA), in which the uncertainty set U is substituted by a finite set of polytopes derived from the vertex set of an arbitrary polytope that dominates U. The union of the polytopes need not contain U. We analyze and computationally test the performance of GPA for the frequently used budgeted uncertainty set U (with m rows). For budgeted uncertainty affine policies are known to be best possible approximations (if coefficients in the constraints are nonnegative for the second-stage decision). In practice calculating affine policies typically requires inhibitive running times. Therefore an approximation of U by a single simplex has been proposed in the literature. GPA maintains the low practical running times of the simplex based approach while improving the quality of approximation by a constant factor. The generality of our method allows to use any polytope dominating U (including the simplex). We provide a family of polytopes that allows for a trade-off between running time and approximation factor. The previous simplex based approach reaches a threshold at Γ>m after which it is not better than a quasi nominal solution. Before this threshold, GPA significantly improves the approximation factor. After the threshold, it is the first fast method to outperform the quasi nominal solution. We exemplify the superiority of our method by a fundamental logistics problem, namely, the Transportation Location Problem, for which we also specifically adapt the method and show stronger results.
{"title":"General Polyhedral Approximation of two-stage robust linear programming for budgeted uncertainty","authors":"Lukas Grunau ,&nbsp;Tim Niemann ,&nbsp;Sebastian Stiller","doi":"10.1016/j.cor.2025.107014","DOIUrl":"10.1016/j.cor.2025.107014","url":null,"abstract":"<div><div>We consider two-stage robust linear programs with uncertain righthand side. We develop a General Polyhedral Approximation (GPA), in which the uncertainty set <span><math><mi>U</mi></math></span> is substituted by a finite set of polytopes derived from the vertex set of an arbitrary polytope that dominates <span><math><mi>U</mi></math></span>. The union of the polytopes need not contain <span><math><mi>U</mi></math></span>. We analyze and computationally test the performance of GPA for the frequently used budgeted uncertainty set <span><math><mi>U</mi></math></span> (with <span><math><mi>m</mi></math></span> rows). For budgeted uncertainty affine policies are known to be best possible approximations (if coefficients in the constraints are nonnegative for the second-stage decision). In practice calculating affine policies typically requires inhibitive running times. Therefore an approximation of <span><math><mi>U</mi></math></span> by a single simplex has been proposed in the literature. GPA maintains the low practical running times of the simplex based approach while improving the quality of approximation by a constant factor. The generality of our method allows to use any polytope dominating <span><math><mi>U</mi></math></span> (including the simplex). We provide a family of polytopes that allows for a trade-off between running time and approximation factor. The previous simplex based approach reaches a threshold at <span><math><mrow><mi>Γ</mi><mo>&gt;</mo><msqrt><mrow><mi>m</mi></mrow></msqrt></mrow></math></span> after which it is not better than a quasi nominal solution. Before this threshold, GPA significantly improves the approximation factor. After the threshold, it is the first fast method to outperform the quasi nominal solution. We exemplify the superiority of our method by a fundamental logistics problem, namely, the Transportation Location Problem, for which we also specifically adapt the method and show stronger results.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"179 ","pages":"Article 107014"},"PeriodicalIF":4.1,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143474281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Predictive model-based multi-objective optimization with life-long meta-learning for designing unreliable production systems
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-15 DOI: 10.1016/j.cor.2025.107011
Ehsan Mahmoodi , Masood Fathi , Amos H.C. Ng , Alexandre Dolgui
Owing to the realization of advanced manufacturing systems, manufacturers have more flexibility in improving their processes through design decisions. Design decisions in production lines primarily involve two complex problems: buffer and resource allocation (B&RA). The main aim of B&RA is to determine the best location and size of buffers in the production line and optimally allocate production resources, such as operators and machines, to workstations. Inspired by a real-world case from the marine engine production industry, this study addresses B&RA in high-mix, low-volume hybrid flow shops (HFSs) with feed-forward quality inspection. These HFSs can be characterized by uncertainties in demand, material handling, processing times, and quality control. In this study, the production environment is modeled via discrete-event simulation, which reflects the features of the actual system without requiring unreasonable or restrictive assumptions. To replace the expensive simulation runs, five widely used regressor machine learning algorithms in manufacturing are trained on data sampled from the simulation model, and the best-performing algorithm is selected as the predictive model. To obtain high-quality solutions, the predictive model is coupled with an enhanced non-dominated sorting genetic algorithm (En-NSGA-II) that incorporates lifelong meta-learning and features a customized representation and a variable neighborhood search. Additionally, a post-optimality analysis using a pattern-mining algorithm is performed to generate knowledge for allocating buffers and operators based on the optimization results, thus providing promising managerial insights.
{"title":"Predictive model-based multi-objective optimization with life-long meta-learning for designing unreliable production systems","authors":"Ehsan Mahmoodi ,&nbsp;Masood Fathi ,&nbsp;Amos H.C. Ng ,&nbsp;Alexandre Dolgui","doi":"10.1016/j.cor.2025.107011","DOIUrl":"10.1016/j.cor.2025.107011","url":null,"abstract":"<div><div>Owing to the realization of advanced manufacturing systems, manufacturers have more flexibility in improving their processes through design decisions. Design decisions in production lines primarily involve two complex problems: buffer and resource allocation (B&amp;RA). The main aim of B&amp;RA is to determine the best location and size of buffers in the production line and optimally allocate production resources, such as operators and machines, to workstations. Inspired by a real-world case from the marine engine production industry, this study addresses B&amp;RA in high-mix, low-volume hybrid flow shops (HFSs) with feed-forward quality inspection. These HFSs can be characterized by uncertainties in demand, material handling, processing times, and quality control. In this study, the production environment is modeled via discrete-event simulation, which reflects the features of the actual system without requiring unreasonable or restrictive assumptions. To replace the expensive simulation runs, five widely used regressor machine learning algorithms in manufacturing are trained on data sampled from the simulation model, and the best-performing algorithm is selected as the predictive model. To obtain high-quality solutions, the predictive model is coupled with an enhanced non-dominated sorting genetic algorithm (En-NSGA-II) that incorporates lifelong meta-learning and features a customized representation and a variable neighborhood search. Additionally, a post-optimality analysis using a pattern-mining algorithm is performed to generate knowledge for allocating buffers and operators based on the optimization results, thus providing promising managerial insights.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"178 ","pages":"Article 107011"},"PeriodicalIF":4.1,"publicationDate":"2025-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143437255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Model-based algorithms for the 0-1 Time-Bomb Knapsack Problem
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-13 DOI: 10.1016/j.cor.2025.107010
Roberto Montemanni , Derek H. Smith
A stochastic version of the 0–1 Knapsack Problem recently introduced in the literature and named the 0–1 Time-Bomb Knapsack Problem is the topic of the present work. In this problem, in addition to profit and weight, each item is characterized by a probability of exploding, and therefore destroying all the contents of the knapsack, in case it is loaded. The optimization aims at maximizing the expected profit of the selected items, which takes into account also the probabilities of explosion, while fulfilling the capacity constraint. The problem has real-world applications in logistics and cloud computing.
In this work, two model-based algorithms are introduced. They are based on partial linearizations of a non-linear model describing the problem. Extensive computational results on the instances available in the literature are presented to position the new methods as the best-performing ones, while comparing against those previously proposed.
{"title":"Model-based algorithms for the 0-1 Time-Bomb Knapsack Problem","authors":"Roberto Montemanni ,&nbsp;Derek H. Smith","doi":"10.1016/j.cor.2025.107010","DOIUrl":"10.1016/j.cor.2025.107010","url":null,"abstract":"<div><div>A stochastic version of the 0–1 Knapsack Problem recently introduced in the literature and named the 0–1 Time-Bomb Knapsack Problem is the topic of the present work. In this problem, in addition to profit and weight, each item is characterized by a probability of exploding, and therefore destroying all the contents of the knapsack, in case it is loaded. The optimization aims at maximizing the expected profit of the selected items, which takes into account also the probabilities of explosion, while fulfilling the capacity constraint. The problem has real-world applications in logistics and cloud computing.</div><div>In this work, two model-based algorithms are introduced. They are based on partial linearizations of a non-linear model describing the problem. Extensive computational results on the instances available in the literature are presented to position the new methods as the best-performing ones, while comparing against those previously proposed.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"178 ","pages":"Article 107010"},"PeriodicalIF":4.1,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143420803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
An exact method and a heuristic for last-mile delivery drones routing with centralized graph-based airspace control
IF 4.1 2区 工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2025-02-12 DOI: 10.1016/j.cor.2025.107006
Jorge Luiz Franco , Vitor Venceslau Curtis , Edson Luiz França Senne , Filipe Alves Neto Verri
The increasing demand for efficient last-mile delivery, driven by the rise of e-commerce, has intensified the need for innovative solutions to manage the complexities of urban logistics. Among the most pressing challenges are the Multi-Agent Pathfinding (MAPF) problem and collision avoidance, both of which are NP-hard and critical for the safe and efficient operation of drones. Collision avoidance is particularly challenging due to the expected high density of drones in future urban environments, making it a problem that remains largely unsolved. Traditional approaches often rely on heuristic and metaheuristic methods to manage these complexities, as large instances are beyond the reach of exact methods. Additionally, distributed relaxations to these problems can lead to suboptimal outcomes and highlights the need for a more centralized and controlled solution. This research adopts a graph-based representation of the delivery area, transforming the centralized Last-Mile Delivery Drones (LMDD) problem into a network flow optimization problem. We propose two graph-based novelty methods in LMDD, a purely exact, NP-hard Mixed Integer Linear Programming (MILP) solution that is evaluated against a heuristic. The complexity of the heuristic is bounded by O(P1.5K), where P represents the number of permits and K is the number of drones. In contrast, the complexity of the MILP model is approximated by O(K7P5.252K2PP), making it intractable for larger instances. The findings from simulations indicate that the graph-based heuristic effectively balances computational efficiency and operational reliability, making it a viable solution for real-world LMDD applications, where large instances and practical execution times are required. This research significantly contributes to the fields of drone logistics and transportation by providing a scalable method for optimizing LMDD paths.
{"title":"An exact method and a heuristic for last-mile delivery drones routing with centralized graph-based airspace control","authors":"Jorge Luiz Franco ,&nbsp;Vitor Venceslau Curtis ,&nbsp;Edson Luiz França Senne ,&nbsp;Filipe Alves Neto Verri","doi":"10.1016/j.cor.2025.107006","DOIUrl":"10.1016/j.cor.2025.107006","url":null,"abstract":"<div><div>The increasing demand for efficient last-mile delivery, driven by the rise of e-commerce, has intensified the need for innovative solutions to manage the complexities of urban logistics. Among the most pressing challenges are the Multi-Agent Pathfinding (MAPF) problem and collision avoidance, both of which are NP-hard and critical for the safe and efficient operation of drones. Collision avoidance is particularly challenging due to the expected high density of drones in future urban environments, making it a problem that remains largely unsolved. Traditional approaches often rely on heuristic and metaheuristic methods to manage these complexities, as large instances are beyond the reach of exact methods. Additionally, distributed relaxations to these problems can lead to suboptimal outcomes and highlights the need for a more centralized and controlled solution. This research adopts a graph-based representation of the delivery area, transforming the centralized Last-Mile Delivery Drones (LMDD) problem into a network flow optimization problem. We propose two graph-based novelty methods in LMDD, a purely exact, NP-hard Mixed Integer Linear Programming (MILP) solution that is evaluated against a heuristic. The complexity of the heuristic is bounded by <span><math><mrow><mi>O</mi><mrow><mo>(</mo><msup><mrow><mi>P</mi></mrow><mrow><mn>1</mn><mo>.</mo><mn>5</mn></mrow></msup><mi>K</mi><mo>)</mo></mrow></mrow></math></span>, where <span><math><mi>P</mi></math></span> represents the number of permits and <span><math><mi>K</mi></math></span> is the number of drones. In contrast, the complexity of the MILP model is approximated by <span><math><mrow><mi>O</mi><mrow><mo>(</mo><msup><mrow><mi>K</mi></mrow><mrow><mn>7</mn></mrow></msup><msup><mrow><mi>P</mi></mrow><mrow><mn>5</mn><mo>.</mo><mn>25</mn></mrow></msup><msup><mrow><mn>2</mn></mrow><mrow><msup><mrow><mi>K</mi></mrow><mrow><mn>2</mn></mrow></msup><mi>P</mi><msqrt><mrow><mi>P</mi></mrow></msqrt></mrow></msup><mo>)</mo></mrow></mrow></math></span>, making it intractable for larger instances. The findings from simulations indicate that the graph-based heuristic effectively balances computational efficiency and operational reliability, making it a viable solution for real-world LMDD applications, where large instances and practical execution times are required. This research significantly contributes to the fields of drone logistics and transportation by providing a scalable method for optimizing LMDD paths.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"178 ","pages":"Article 107006"},"PeriodicalIF":4.1,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143403707","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}
引用次数: 0
期刊
Computers & Operations Research
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1