Machine layout of manufacturing cell with fix-proportion and varying product mix pattern

Abstract

This research study addresses the problem of the cellular layout of machines in static and dynamic environments. Previous researchers focused on either static or dynamic environments. As the static layout is optimized to a planning period, which has particular quantitative demand, it may not be suitable for different quantitative demands in other planning periods. The problem is how to link static and dynamic layouts, and to decide which machine arrangements are most appropriate to a set of quantitative demands in dynamic environments, Therefore, a heuristic algorithm has been developed to tackle machine allocation problems in cellular manufacturing. The proposed heuristic approach is called the MAIN (Machines Allocation INter-relationship) algorithm. This algorithm uses constructive steps to allocate machines into the constrained machine location zones. This works well in a period with fixed quantitative demand and it is referred to the Static Machine Cellular Layout (SMCL). Furthermore, it bases on a pool of SMCLs to select the best one for the Dynamic Machine Cellular Layout (DMCL). The selection of DMCL is dependent on the balance between the rearrangement cost of machines and the traveling cost of parts. Two extreme cases have been used to verify the MAIN algorithm, namely steady-DMCL and modified multi-DMCLs. The significance of these results is that the average error of SMCL is only 3%; compared with the optimal solution. Also, modified multi-DMCLs are ususally better than steady-DMCLs but this depends on the proportion of rearrangement costs.