Filtering heuristics for rescheduling open orders in MRP systems

This paper is concerned with the problem of maintaining internal open order due dates in manufacturing firms that plan and schedule their operations using material requirements planning (MRP) systems. Earlier studies have focused on two due date maintenance procedures: fixed and dynamic. When using the fixed procedure, order due dates are never changed. Under the dynamic procedure, all shop due dates are changed to reflect the most current order due dates. In this study, a simulation model of an MRP-driven job shop is used to investigate the middle ground between these two extremes. Due date uncertainty is induced in the model by modifying the master production schedule each week. Four filtering heuristics that pass only ‘important’ due date changes to the shop are developed and then analysed using the simulation model. Average customer service level and average total inventory level are used to measure the performance of the heuristics. The results indicate that the performance of the dynamic due date procedure can be matched, and that of the fixed procedure can be exceeded, by filtering heuristics that pass only a fraction of all recommended due date adjustments to the shop. In particular, the performance of all four heuristics was comparable to that of the dynamic procedure when about half of the due date changes were implemented. For one of the filtering heuristics, a comparable performance was achieved when as few as one-fourth of recommended due date changes were acted upon.