A Collaborative Approach Based on DCA and VNS for Solving Mixed Binary Linear Programs
This paper addresses the Mixed Binary Linear Programming problems (MBLPs) by a collaborative approach using two component algorithms. The first is DCA (Difference of Convex functions Algorithm), an efficient deterministic algorithm in nonconvex programming framework, and the second is VNS (Variable Neighborhood Search), a well known metaheuristic method. The DCA and VNS are executed in parallel. At the end of each cycle, the best-found solution is exchanged between these algorithms via MPI (Message Passing Interface) library. The next cycle starts with the previous best-found solution as an initial solution. The performance of the proposed approach is tested on a set of benchmarks of the Capacitated Facility Location Problem. Numerical experiments show the efficiency of our approach.
KeywordsMixed Binary Linear Programming problems DC programming and DCA Metaheuristics Parallel and distributed programming
- 1.Centre de Calcul de l’université de Bourgogne: Bases de la parallélisation par passage de message: MPI (Message Passing Interface) (2016)Google Scholar
- 10.Polacek, M., Hartl, R.F., Doerner, K., Reimann, M.: A variable neighborhood search for the multi depot vehicle routing problem with time windows. J. Heuristics 10(6), 613–627 (2004)Google Scholar
- 12.Yagouni, M., Le Thi, H.A.: A collaborative metaheuristic optimization scheme: methodological issues. In: van Do, T., Le Thi, H.A., Nguyen, N.T. (eds.) Advanced Computational Methods for Knowledge Engineering. AISC, vol. 282, pp. 3–14. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-06569-4_1Google Scholar