An Improved Genetic Algorithm for Fleet Assignment and Maintenance Routing Problems

Abstract

Airline scheduling consists of several decision-making tasks The entire problem scale is large and these decision tasks are closely related This makes optimally solving the entire problem a challenging problem Former researchers have divided the problem into five sub-problems which could be solved sequentially while sacrificing the global optimality The ordering of these five sub-problems are route development flight scheduling fleet assignment problem (FAP) aircraft maintenance routing problem (AMRP) and crew scheduling In this study we aim to deal with a problem of solving FAP and AMRP simultaneously A specially designed GA and integer linear programming are combined for such problem In the improved GA the chromosomes are arranged to represent the flight paths for each aircraft New crossover and mutation processes can then be conducted in a more effective way to finding near-optimal solutions Real data from airlines are used to demonstrate the effectiveness of the improved method

Date of Award	2018 Jul 4
Original language	English
Supervisor	Ta-Chung Wang (Supervisor)