Dynamic Layout in the Case of Different Shapes and Areas of Facilities with Alternative Processing Routes

Abstract

As product lifecycle becomes shorter, facility layout should be adjusted effectively to meet dynamic change in product mix under a given system capability. This paper formulates a dynamic layout problem with different shapes and areas of facilities and alternative processing routes to minimize the total cost consisting of work flow costs and facility-rearrangement costs over a given planning horizon T. Since the total work flow depends on the selection of processing routes for all products, our dynamic layout problem requires the determination of not only layout, but also a set of processing routes in each period. This paper decomposes the original layout problem into a work-flow decision problem and a layout-position decision problem. In the former problem, an approximate Closed Queueing Network model is employed for representing the capacity constraint, and the best processing routes are determined to minimize the total work flow under a given layout in each period. The layout-position decision problem is solved using GA with a static and dynamic phase. In the static phase, a one-dimensional array chromosome is first constructed by assigning each facility number to a locus one by one. Hamamoto's method is used for decoding each chromosome to generate a two-dimensional layout. The static GA provides the best S layouts in work-flow cost for each period. In the dynamic phase, T candidate layouts over the planning horizon are first picked from among the ST layouts, and are represented as a permutation chromosome with length T. The dynamic GA provides the best layouts over the planning horizon so that the total cost including facility-rearrangement costs is minimized. Some simulation experiments are demonstrated to show that the proposed method efficiently yields a near optimal layout with high accuracy.