Optimising the scheduling and planning of urban milk deliveries.

Urquhart, Neil B

Abstract

This paper investigates the optimisation of the delivery of dairy products to households in three urban areas. The requirement for the optimisation to be part of the existing business process has determined the approach taken. The solution is maintained in an existing customer database, with manual amendments as customers are added and deleted. The optimisation challenge is to take this solution, reduce the distance travelled, and balance the load across rounds making the minimum number of changes to the delivery network. The approach taken utilises an Evolutionary Algorithm for ordering deliveries and a multi-agent approach to reassigning deliveries between rounds. The case study suggests that distance travelled may be reduced by up to 19%, the deviation between round lengths may be considerably reduced, with only 10 % of customers being moved between rounds.

Citation

Urquhart, N. B. (2015, April). Optimising the scheduling and planning of urban milk deliveries. Presented at European Conference on the Applications of Evolutionary Computation EvoApplications 2015, Copenhagen, Denmark

• Presentation Conference Type: Conference Paper (published)
• Conference Name: European Conference on the Applications of Evolutionary Computation EvoApplications 2015
• Start Date: Apr 8, 2015
• End Date: Apr 10, 2015
• Acceptance Date: Jan 9, 2015
• Publication Date: Mar 17, 2015
• Deposit Date: Mar 23, 2015
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• Volume: 9028
• Pages: 604-615
• Series Title: Lecture Notes in Computer Science
• Series Number: 9028
• Series ISSN: 0302-9743
• Book Title: Applications of Evolutionary Computation
• ISBN: 978-3-319-16548-6
• DOI: https://doi.org/10.1007/978-3-319-16549-3_49
• Keywords: Evolutionary algorithm; route optimisation
• Public URL: http://researchrepository.napier.ac.uk/id/eprint/7683