An Efficient Hybrid Algorithm for Dynamic Facility Layout Problem using Simulation Technique and PSO

One of the most important and effective issues of todayʹs manufacturing companies, is how to arrange their facilities in the most economic manner. Different methods and approaches have been examined by researchers to address the so-called facility layout problem (FLP). As a combinatorial problem FLP is a NP-Hard one, and so the classic and exact optimization methods just can be useful for small scale problems. Therefore, developing meta-heuristic algorithms in order to solve and analyze such problems has been focused by many researchers. In this research, a new hybrid heuristic method is developed by the combination of discrete Particle Swarm Optimization (PSO) algorithm and simulation technique to address Dynamic Facility Layout Problem (DFLP) which is the main contribution of the current study. The proposed algorithm uses simulation modelling to produce some feasible solutions for the problem and then the PSO algorithm starts with these solutions as initial particles and moves toward the near optimum solutions. To indicate the efficiency of the proposed algorithm, several test problems taken from the literature have been examined and the results were compared to other algorithms. As the computational results reveal, the quality of solutions and the algorithm speed are suitable enough to be used in the real world problems. Also, the proposed algorithm could be used in other combinatorial problems as a general heuristic algorithm.