Please use this identifier to cite or link to this item: http://bura.brunel.ac.uk/handle/2438/29350
Full metadata record
DC FieldValueLanguage
dc.contributor.authorShahmanzari, M-
dc.contributor.authorMansini, R-
dc.date.accessioned2024-07-14T15:26:58Z-
dc.date.available2024-07-14T15:26:58Z-
dc.date.issued2024-06-11-
dc.identifierORCiD: Masoud Shahmanzari https://orcid.org/0000-0003-2019-4490-
dc.identifierORCiD: Renata Mansini https://orcid.org/0000-0002-2194-0339-
dc.identifier.citationShahmanzari, M. and Mansini, R. (2024) 'A learning-based granular variable neighborhood search for a multi-period election logistics problem with time-dependent profits', European Journal of Operational Research, 0 (in press, corrected proof), pp. 1 - 18. doi: 10.1016/j.ejor.2024.06.009.en_US
dc.identifier.issn0377-2217-
dc.identifier.urihttps://bura.brunel.ac.uk/handle/2438/29350-
dc.descriptionSupplementary data are available online at: https://www.sciencedirect.com/science/article/pii/S0377221724004545#:~:text=Appendix%20A.-,Supplementary%20data,-References .en_US
dc.descriptionProduction, Manufacturing, Transportation and Logistics.-
dc.description.abstractPlanning the election campaign for leaders of a political party is a complex problem. The party representatives, running mates, and campaign managers have to design an efficient routing and scheduling plan to visit multiple locations while respecting time and budget constraints. Given the limited time of election campaigns in most countries, every minute should be used effectively, and there is very little room for error. In this paper, we formalize this problem as the multiple Roaming Salesman Problem (mRSP), a new variant of the recently introduced Roaming Salesman Problem (RSP), where a predefined number of political representatives visit a set of cities during a planning horizon to maximize collected rewards, subject to budget and time constraints. Cities can be visited more than once and associated rewards are time-dependent (increasing over time) according to the day of the visit and the recency of previous visits. We develop a compact Mixed Integer Linear Programming (MILP) formulation complemented with effective valid inequalities. Since commercial solvers can obtain optimal solutions only for small-sized instances, we develop a Learning-based Granular Variable Neighborhood Search and demonstrate its capability of providing high-quality solutions in short CPU times on real-world instances. The adaptive nature of our algorithm refers to its ability to dynamically adjust the neighborhood structure based on the progress of the search. Our algorithm generates the best-known results for many instances.en_US
dc.format.extent1 - 18-
dc.format.mediumPrint-Electronic-
dc.languageEnglish-
dc.language.isoen_USen_US
dc.publisherElsevieren_US
dc.rightsCopyright © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (https://creativecommons.org/licenses/by/4.0/).-
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/-
dc.subjectelection logisticsen_US
dc.subjectmulti-period routingen_US
dc.subjectmultiple roaming salesman problemen_US
dc.subjecttime-dependent profitsen_US
dc.subjectadaptive variable neighborhood searchen_US
dc.titleA learning-based granular variable neighborhood search for a multi-period election logistics problem with time-dependent profitsen_US
dc.typeArticleen_US
dc.date.dateAccepted2024-06-07-
dc.identifier.doihttps://doi.org/10.1016/j.ejor.2024.06.009-
dc.relation.isPartOfEuropean Journal of Operational Research-
pubs.issuein press, corrected proof-
pubs.publication-statusPublished-
pubs.volume0-
dc.identifier.eissn1872-6860-
dc.rights.licensehttps://creativecommons.org/licenses/by/4.0/legalcode.en-
dc.rights.holderThe Authors-
Appears in Collections:Brunel Business School Research Papers

Files in This Item:
File Description SizeFormat 
FullText.pdfCopyright © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (https://creativecommons.org/licenses/by/4.0/).3.19 MBAdobe PDFView/Open


This item is licensed under a Creative Commons License Creative Commons