The Stochastic Facility Layout Problems in Production Systems with Variable Processing Times(<Special English Issue>Production and Logistics)

Abstract

In this paper, we propose a new approach to facility layout problems. Almost all manufacturing systems involve random elements such as variable inter-arrival times of orders and variable processing times, which reduce manufacturing efficiency. For such systems, instead of minimizing the traditional materials handling cost. our aim is to find an optimal layout that absorbs the randomness as effectively as possible and results in the best manufacturing efficiency. We call this type of layout problem a "Stochastic Facility Layout Problem (SFLP)." We choose the manufacturing lead-time as the performance measure, formulate an optimal facility layout and buffer space allocation problem for job-shop type production systems in which relatively largesized parts are transferred by Automated Guided Vehicles(AGVs) and processed via variable routes in First-Come-First-Served (FCFS) manner. Our approach is based on an approximate Markov analysis using a decomposition technique for evaluating the lead-time and a genetic algorithm for searching the optimal combination of buffer space allocation and layout.