Multi-Criteria Optimization in Fuel Distribution

This paper deals with some computational study of multi-criteria optimization in fuel distribution problem. We consider vehicle routing problem, e.g. routing of a fleet of tank trucks, for one of the famous polish petroleum companies. For the purpose of solving this problem, we developed the random search algorithm, which explores the broad range of feasible sets of routes and searches for non-dominated multi-criteria solutions. The problem is examined on real data, which contains distances between 50 petrol stations and one central warehouse (refinery). Obtained results indicate, that it is possible to obtain single solution and satisfy both optimization criteria. Based on the analysis of the collected data, we formulate a number of proposals useful in future for construction of algorithms for multi-criteria fuel distribution optimization.