Partially feasible solution space for integrated SATS operations

Abstract

Purpose The purpose of this paper is to investigate the feasibility of solving an integrated flight scheduling, fleet assignment and crew pairing problem for an on-demand service using a small, up to 19-seater, aircraft. Design/methodology/approach Evolutionary algorithm is developed to solve the problem. Algorithm design assumes indirect solution representation that allows to evaluate partially feasible solutions only and speed up calculations. Tested algorithm implementation takes advantage of the graphic processing unit. Findings Performed tests confirm that the algorithm can successfully solve the defined integrated scheduling problem. Practical implications The presented algorithm allows to optimise on-demand transport service operation within minutes. Social implications Optimisation of operation cost contributes to better accessibility of transport. Originality/value The presented integrated formulation allows to avoid sub optimal solutions that are results of solving sequential sub problems. Indirect representation and evaluation strategy can be applied to speed up calculations in other problems as well.