Authors: Nikos C. Tsourveloudis
Addresses: Department of Production Engineering and Management, Technical University of Crete, 73100 Chania, Crete, Greece
Abstract: The production scheduling literature identifies the need for optimised fuzzy controllers for the Work-In-Process (WIP) reduction problem. One way towards fuzzy controller optimisation is the Evolutionary Algorithms (EAs). The question this paper faces is: How robust and generic are the controllers that come out of an evolutionary process? An answer is attempted here, by testing the evolutionary-tuned fuzzy controllers in demand conditions different than the ones used during controllers| optimisation. Extensive simulation runs of production lines and networks turn out that the evolutionary-fuzzy strategy achieved a significant reduction of the WIP compared to a successful heuristic approach.
Keywords: WIP inventory; work-in-process; scheduling; fuzzy control; evolutionary algorithms; controller design; fuzzy logic; simulation.
International Journal of Advanced Intelligence Paradigms, 2010 Vol.2 No.4, pp.306 - 315
Published online: 02 Nov 2010 *Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article