Regular policies and stability of dynamic scheduling for manufacturing systems

In modern manufacturing systems, where parts and products are produced automatically, there are buffers on each machine in order to achieve the desired trade-off between conflicting goals such as maximizing production throughput and keeping work-in-process inventory at reasonable levels. In consequence, scheduling policies that establish buffer levels and machine operations dynamics are needed. The stability of dynamic manufacturing systems depends on the type of scheduling policy. Here the authors suggest a class of policies that stabilize, in real time, all machine buffers of a dynamic manufacturing system, that is, all buffer levels remain lower than some upper bounds. Moreover, they develop a closed formula to determine a buffer capacity for each machine and part-type. For comparison purposes, the authors take scheduling policies proposed in the literature, and consider two performance criteria: upper bound and average buffer levels. Simulation results are also included to show the effectiveness of the proposed policies.