Investigating the use of metaheuristics for solving single vehicle routing problems with time-varying traversal costs

Harwood, Kieran and Mumford, Christine and Eglese, Richard (2013) Investigating the use of metaheuristics for solving single vehicle routing problems with time-varying traversal costs. Journal of the Operational Research Society, 64. pp. 34-47. ISSN 0160-5682

[img]
Preview
PDF (Investigating the use of metaheuristics for solving single vehicle routing problems with time-varying traversal costs)
10.pdf - Submitted Version

Download (328kB)

Abstract

Metaheuristic algorithms, such as simulated annealing and tabu search, are popular solution techniques for vehicle routing problems (VRPs). These approaches rely on iterative improvements to a starting solution, involving slight alterations to the routes (i.e. neighbourhood moves), moving a node to a different part of a solution, swapping nodes or inverting sections of a tour, for example. When working with standard VRPs, where the costs of the arcs do not vary with advancing time, evaluating changes to the total cost following a neighbourhood move is a simple process: simply subtract the costs of the links removed from the solution and add the costs for the new links. When a time varying aspect (e.g. congestion) is included in the costs, these calculations become estimations rather than exact values. This paper focuses on a single vehicle VRP, similar to the Travelling Salesman Problem (TSP) and investigates the potential for using estimation methods on simple models with time-variant costs, mimicking the effects of road congestion.

Item Type:
Journal Article
Journal or Publication Title:
Journal of the Operational Research Society
Uncontrolled Keywords:
/dk/atira/pure/subjectarea/asjc/1400/1406
Subjects:
ID Code:
61038
Deposited By:
Deposited On:
18 Dec 2012 14:11
Refereed?:
Yes
Published?:
Published
Last Modified:
23 Sep 2020 01:19