A hierarchical multiobjective approach to production planning and scheduling problems in cellular manufacturing systems

Group technology (GT) is an innovative approach to batch-type production which seeks to rationalize small-lot production by capitalizing on the similarities that exist among component parts. Cellular manufacturing (CM), which is a subset and derivative of GT, is the physical division of the manufacturing facilities into production cells, representing the basis for both just-in-time and flexible manufacturing systems. The thrust of this research effort is the development of a hierarchical, multiobjective modeling approach and associated, efficient solution techniques to solve production planning and scheduling problems in CM systems. It is designed to minimize the disadvantages of a CM system, such as possible increased tardiness, and extended flow times.;The primary motivation is to reduce the computational burden of the resulting large-scale planning problem by utilizing a hierarchical approach which decomposes the overall problem into a hierarchy of smaller, more manageable problems. In addition, the solution procedures presented at each level utilize decomposition principles such as an aggregation/disaggregation scheme and Lagrangean relaxation to further simplify the problem. The decomposition of the manufacturing system is accomplished in three dimensions: by floor space, by product structure, and by time scale. In the time scale decomposition, the levels of the decision hierarchy differ by complexity, scope, and time horizon. The time scale decomposition employed in this research consists of the following three levels: (i) Cell loading, (ii) Joint economic lot scheduling and scheduling among groups, and (iii) Cell scheduling.;An experimental design analysis shows that the proposed hierarchical-multiobjective scheduling approach performs significantly better than previously existing techniques over a wide range of conditions and scheduling complexity due to increased flexibility. Analysis of variance (ANOVA) tables are developed to explain the relationships between the various performance criteria, such as the average tardiness and the mean flow time, and the system parameters, such as number of GT families and GT cells, and the number of items in each family. ANOVA tables indicate that there is a significant relationship between the size of lots and completion times, and the output of classification and coding systems has significant impact on the multiple performance criteria.