Selection of setup times in a hierarchically controlled manufacturing system

A hierarchical solution to the setup scheduling problem is proposed. Real-time scheduling policies for manufacturing systems with significant setup times between operations are found. Considered are isolated cells, with unreliable machines, that are capable of producing many part types, each with specified constant demand rate, processing time and setup change time. The scheduling problem is investigated in the context of a multilevel hierarchy of distinct events with distinct frequencies. At the top level, a static optimization model is constructed for allocating capacity to the different setup changes. The decision variables are the setup change rates. The next level modifies these rates in response to changes of machine repair states. The next uses the modified setup rates to determine times a which setup changes are initiated. A feedback setup policy is used for changing setups. The parameters defining the policy are chosen such that the policy converges to a repetitive sequence of production runs. The sequence and lengths of the production runs are such that the average inventory is minimized. Part loading is controlled by the lowest level