Conference Proceeding

A Bi-Level Approach to Vehicle Fleet Reduction: Successful Case Study in Community Healthcare



Brownlee AEI, Thomson SL & Oladapo R (2024) A Bi-Level Approach to Vehicle Fleet Reduction: Successful Case Study in Community Healthcare. In: Genetic and Evolutionary Computation Conference 2024, Melbourne, Australia, 14.07.2024-18.07.2024. GECCO '24: ACM.

We report on a case study application of metaheuristics with Argyll and Bute Health and Social Care Partnership in the West of Scotland. The Partnership maintains a fleet of pool vehicles that are available to service visits of staff to locations across a largely rural area. Maintaining such a fleet is important but costly: we show how the allocation of fleet vehicles can be formulated as a bilevel optimisation problem. At the upper level, vehicles are allocated to 'base' locations such as hospitals. At the lower level, vehicles are allocated to specific jobs. We explore local-search approaches to solving this problem. We show that some blurring of the distinction between upper and lower levels can be helpful for this problem. We also demonstrate, for our case study, a 7.1% reduction in the vehicle fleet while still being able to meet all demand.

Optimal job scheduling; evolutionary computation; bilevel optimization

FundersNHS Highland
Place of publicationGECCO '24
ConferenceGenetic and Evolutionary Computation Conference 2024
Conference locationMelbourne, Australia

People (1)


Dr Sandy Brownlee

Dr Sandy Brownlee

Senior Lecturer in Computing Science, Computing Science and Mathematics - Division

Research centres/groups