Sample path techniques for the control of a real assembly system

The paper presents the application of sample path analysis techniques to the optimization of an assembly system, controlled by means of a Kanban mechanism. The approach presented in the paper is based on previous work by the authors, and uses sample path analysis techniques to estimate the behavior of the system under a number of different Kanban allocations. The proposed approach is aimed at adjusting Kanban allocation in reaction to changes in operating scenarios. In view of the number of part-types and components, the optimization problem, i.e., the problem of optimal allocation of Kanban to each component and part, is rather complex. To cope with the complexity of the system, the optimization problem is solved by a decentralized implementation of control algorithms presented by the authors in previous work. The paper reports detailed simulation results, illustrating the effect of different scheduling policies, the effects of several Kanban adaptation policies, and showing the improvement on system performance achievable with the proposed scheme.