AWIP: A simulation-based feedback control algorithm for scalable design of self-regulating production control systems

A new simulation-based feedback control algorithm, called Adaptive Work In Process (AWIP), for the design of Self-regulating Production Control Systems (SPCSs) such as Kanban, CONWIP, base stock control and their generalizations is presented. The problem of minimizing average Work In Process (WIP) subject to a required throughput is solved. The AWIP algorithm is used as a feedback controller to adjust the WIP at various stages in the production system. The algorithm is synthesized based on the structural properties of SPCSs that are established analytically in this paper. In this approach simulation is used to provide the feedback to the controllers, and leads to an iterative numerical computational algorithm. Computational experiments show that the AWIP algorithm is near-optimal, and computationally efficient, which makes it an attractive approach for designing and controlling large production systems.