Relax-and-Fix and Fix-and-Optimise algorithms to solve an integrated network design problem for closing a supply chain with hybrid retailers/collection centres
IF 4.1 2区 工程技术Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS
{"title":"Relax-and-Fix and Fix-and-Optimise algorithms to solve an integrated network design problem for closing a supply chain with hybrid retailers/collection centres","authors":"Mehdi Amiri-Aref , Mahdi Doostmohammadi","doi":"10.1016/j.cor.2025.106981","DOIUrl":null,"url":null,"abstract":"<div><div>This paper studies a multi-echelon closed-loop supply chain network design problem that is characterised by a set of hybrid retailers/collection centres in a multi-period setting. This problem is motivated by the return-to-retail approach currently prevalent in the retail industry under the deposit return scheme. This paper proposes a mathematical programming model that integrates strategic decisions regarding the number and location of hybrid retailer/collection centre facilities, with dynamic decisions pertaining to manufacturing and remanufacturing/recycling, inventory level, and fleet size across the network. This optimisation problem is formulated as a mixed integer linear programming model to fulfil customers’ demands while minimising the total network costs. To solve the problem, a matheuristic solution approach is devised, incorporating Relax-and-Fix and Fix-and-Optimise heuristics augmented by novel relaxation and fixing strategies. We introduce an integrality test which accounts for possible rounding-off errors allowing a user-defined integer feasibility tolerance. Moreover, a variable partitioning is applied to shrink the problem’s dimensions and to shorten the search space. The latter is then iteratively updated to explore neighbourhoods within a given search radius size. To evaluate the validity and efficiency of the proposed model and the solution approach, 90 instances are generated using a case study within the geographical scope limited to the network of a retail chain in France. Numerical results show that the proposed solution method provides near-optimal solutions for small- and medium-size instances in a reasonable computational time and outperforms the commercial solver for large- and extra large-size problems. Managerial insights derived from the computational experiments regarding key performance indicators of the problem are presented and discussed.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"177 ","pages":"Article 106981"},"PeriodicalIF":4.1000,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers & Operations Research","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0305054825000097","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0
Abstract
This paper studies a multi-echelon closed-loop supply chain network design problem that is characterised by a set of hybrid retailers/collection centres in a multi-period setting. This problem is motivated by the return-to-retail approach currently prevalent in the retail industry under the deposit return scheme. This paper proposes a mathematical programming model that integrates strategic decisions regarding the number and location of hybrid retailer/collection centre facilities, with dynamic decisions pertaining to manufacturing and remanufacturing/recycling, inventory level, and fleet size across the network. This optimisation problem is formulated as a mixed integer linear programming model to fulfil customers’ demands while minimising the total network costs. To solve the problem, a matheuristic solution approach is devised, incorporating Relax-and-Fix and Fix-and-Optimise heuristics augmented by novel relaxation and fixing strategies. We introduce an integrality test which accounts for possible rounding-off errors allowing a user-defined integer feasibility tolerance. Moreover, a variable partitioning is applied to shrink the problem’s dimensions and to shorten the search space. The latter is then iteratively updated to explore neighbourhoods within a given search radius size. To evaluate the validity and efficiency of the proposed model and the solution approach, 90 instances are generated using a case study within the geographical scope limited to the network of a retail chain in France. Numerical results show that the proposed solution method provides near-optimal solutions for small- and medium-size instances in a reasonable computational time and outperforms the commercial solver for large- and extra large-size problems. Managerial insights derived from the computational experiments regarding key performance indicators of the problem are presented and discussed.
期刊介绍:
Operations research and computers meet in a large number of scientific fields, many of which are of vital current concern to our troubled society. These include, among others, ecology, transportation, safety, reliability, urban planning, economics, inventory control, investment strategy and logistics (including reverse logistics). Computers & Operations Research provides an international forum for the application of computers and operations research techniques to problems in these and related fields.