Joint Design of Quality and Production Control in Multi-Stage Asynchronous Manufacturing Systems

This paper proposes an analytical method for the joint design of quality and production control parameters in unreliable asynchronous multi-stage lines. Classical statistical quality control charts are typically designed without taking into consideration the logistic performance of multi-stage systems. Similarly, production control parameters, such as the number of kanban cards at any stage of a kanban controlled system, are typically designed without taking into consideration the impact of this decision on the product quality. By doing so, it is implicitly assumed that the two controllers act as independent. However, the dynamics of real systems show that this is not true. Specifically, the proposed method optimally sets the sample size, the sampling frequency and the position of the control limits of the quality control charts as well as the number of kanban cards at any stage of production, by jointly considering the mutual relations between the controllers. Numerical results compare the solution of this integrated design with the configurations obtained by solving the two problems in isolation with existing techniques. They show that great benefits can be achieved by the proposed integrated design of the quality and production control parameters, since it fully captures the interaction between the dynamics of the two controllers.