Authors: Hany Seidgar; Sahar Tadayoni Rad; Rasoul Shafaei
Addresses: Mazandaran University of Science and Technology, P.O. Box 734, Tabarsi Street, Babol, Mazandaran, Iran ' K.N. Toosi University of Technology, P.O. Box 470, Mirdamad Street, Tehran, Tehran, Iran ' K.N. Toosi University of Technology, P.O. Box 470, Mirdamad Street, Tehran, Tehran, Iran
Abstract: We investigate two-stage assembly flow shop problems (TAFSP) with considering machines breakdown and minimisation of the expected the weighted sum of makespan and mean of completion time is as objective value. This problem is NP-hard, hence we presented genetic algorithm (GA) and new self adapted differential evolutionary (NSDE) for solving random generated test problems. Artificial neural network (ANN) is applied to set parameters of two proposed algorithms also Taguchi method is used for analysing the effect of parameters of problem. The computational results reveal that NSDE is better than GA and achieve to good solutions in a shorter time.
Keywords: two stage assembly flow shop; machine breakdowns; simulation approach; self adapted differential evolutionary algorithm; artificial neural network; ANN; Taguchi method.
International Journal of Operational Research, 2017 Vol.29 No.2, pp.273 - 293
Received: 14 Jul 2014
Accepted: 01 Mar 2015
Published online: 18 Apr 2017 *