Application of a hybrid genetic algorithm to airline crew scheduling

https://doi.org/10.1016/0305-0548(95)00060-7Get rights and content

Abstract

This paper discusses the development and application of a hybrid genetic algorithm to airline crew scheduling problems. The hybrid algorithm consists of a steady-state genetic algorithm and a local search heuristic. The hybrid algorithm was tested on a set of 40 real-world problems. It found the optimal solution for half the problems, and good solutions for 9 others. The results were compared to those obtained with branch-and-cut and branch-and-bound algorithms. The branch-and-cut algorithm was significantly more successful than the hybrid algorithm, and the branch-and-bound algorithm slightly better.

References (27)

  • D. Goldberg et al.

    A comparative analysis of selection schemes used in genetic algorithms

  • K. DeJong

    Genetic algorithms are NOT function optimizers

  • J. Holland

    Adaption in Natural and Artificial Systems

  • R. Anbil et al.

    A global approach to crew pairing optimization

    IBM Syst. J.

    (1992)
  • J. Barutt et al.

    Airline crew scheduling: supercomputers and algorithms

    SIAM News

    (1990)
  • I. Gershkoff

    Optimizing flight crew schedules

    INTERFACES

    (1989)
  • K. Hoffman et al.

    Solving airline crew-scheduling problems by branch-and-cut

    Mgmt Sci.

    (1993)
  • R. Marsten et al.

    Exact solution of crew scheduling problems using the set partitioning model: recent successful applications

    Networks

    (1981)
  • M. Fischer et al.

    Optimal solution of set covering/partitioning problems using dual heuristics

    Mgmt Sci.

    (1990)
  • J. Eckstein

    Parallel branch-and-bound algorithms for general mixed integer programming on the CM-5

  • J. Desrosiers et al.

    The airline crew pairing problem

  • J. Arabeyre et al.

    The airline crew scheduling problem: a survey

    Transport. Sci.

    (1969)
  • G. Nemhauser et al.

    Integer and Combinatorial Optimization

  • Cited by (89)

    • Airline crew scheduling: Models and algorithms

      2021, Transportation Research Part E: Logistics and Transportation Review
      Citation Excerpt :

      Then, a set-covering or set-partitioning problem is applied to identify a sub-set of pairings with a minimum cost through using integer programming techniques like branch-and-bound, cutting planes, and sub-gradient optimization. However, due to the high problem complexity, many studies solve the problem heuristically (Azadeh et al., 2013; Cohn and Barnhart, 2003; Levine, 1996). For example, to deal with the airline crew scheduling, Azadeh et al. (2013) developed a hybrid particle swarm optimization (PSO) algorithm synchronized with a local search heuristic and compared the performance of the hybrid PSO algorithm with other heuristic algorithms developed based on genetic algorithm and ant colony optimization.

    • Allocating and scheduling resources for a mobile photo enforcement program

      2021, Transportation Research Part C: Emerging Technologies
      Citation Excerpt :

      These mainly include branch-and-cut (B&C) and column generation (CG) based decomposition methods (Desaulniers, 2010; Desaulniers et al., 1997; Hoffman and Padberg, 1993; Lavoie et al., 1988; Nishi et al., 2011; Vance et al., 1997). Although metaheuristics such as SA (Emden-Weinert and Proksch, 1999) and some hybrid algorithms (Azadeh et al., 2013; Ke and Feng, 2013; Levine, 1996) have been employed to solve large vehicle routing and airline crew scheduling problems, Emden-Weinert and Proksch concede that feasible solutions were difficult to find due to a number of reasons. A combined approach has primarily been used to solve the largest airline crew scheduling problems: one first decomposes a schedule using the Dantzig-Wolfe algorithm, and then solves the reformulation using a column generation algorithm (Desaulniers et al., 2002; Ernst et al., 2004).

    • A survey of the literature on airline crew scheduling

      2018, Engineering Applications of Artificial Intelligence
    View all citing articles on Scopus

    This work was supported by the Mathematical, Information, and Computational Sciences Division subprogram of the Office of Computational and Technology Research, U.S. Department of Energy, under Contract W-31-109-ENG-38.

    David Levine is a member of the research staff of the Mathematics and Computer Science Division at Argonne National Laboratory. He received an M.Sc. degree in Industrial Engineering from the University of Arizona and Ph.D. degree in Computer Science from the Illinois Institute of Technology. He is the developer of the PGAPack parallel genetic algorithm library. His current research interests are computational biophysics, genetic algorithms, parallel computing and virtual reality visualization.

    View full text