Flow control in kanban based multicell manufacturing: II. Design of control systems and experimental results

This paper extends our study of general blocking kanban control mechanisms in multicell manufacturing. The structural results developed in part I of this twopaper sequence are used to develop an efficient design framework for the optimal configuration of kanban control systems. The structural properties of the design problems and their relationships are established. The framework consists of optimization algorithms for the design problems over neighbourhood lattice design spaces. Extensive computational analyses show that the proposed algorithms determine the optimal configuration in a neighbourhood by exploring about 3% of the neighbourhood set. The quality of the neighbourhood solutions has been demonstrated by comparing them with benchmark strategies and an upper bound on throughput. The computational efficiency and the quality of solutions show that the proposed approach is efficient and practically viable.