Variable Neighborhood Search to solve an airline fleet sizing and fleet assignment problem

Abstract In order to enable an airline to realize its planned schedule to and perform successfully, it is very important to determine the fleet size in an appropriate way. Fleet size represents the number of aircraft in an airline’s fleet. Larger number of aircraft than it is required means lower fleet utilization, which further affects increase of costs. Lower number of aircraft than it is required means spill of demand, i.e. reallocation of passengers to competitors’ flights and missing out on the opportunity to make profit. The problem considered in this paper refers to fleet size and fleet assignment in the case of single fleet (only one type of aircraft). In order to solve the problem we proposed a metaheuristic approach based on the Variable Neighborhood Search (VNS) methodology, which solves both fleet sizing problem and fleet assignment problem simultaneously. Preliminary numerical results based on real data are compared to optimal solution obtained by CPLEX and show the validity of the proposed modelling approach.