A bi-objective maximal covering location problem: a service network design application

This paper proposes a bi-objective maximal covering location problem (MCLP) that involves customer preferences and balances between covered demand and the number of uncovered customers. The first objective maximizes the weighted sum of the covered demand, in which the weights are based on customer preferences, while the second objective is to minimize the number of uncovered customers. This newly proposed bi-objective model can be applied to the design of service networks, such as post offices, health centers, delivery services, etc. Three multi-objective evolutionary algorithms (MOEAs) are adapted to the considered bi-objective MCLP and applied on the set of modified real-life MCLP test instances that include large number of customer nodes and potential facility locations. The obtained experimental results show that all three MOEAs are suitable for solving the bi-objective MCLP, as they successfully provide solutions on the considered test instances of challenging dimensions.