In Search of the priority rule for Integrated Order Management

This paper discusses the missing link between academic studies in, and practical applications of, priority dispatching rules developed since the 1960s. From extensive review of literature it appears that no single dominant priority index rule has been suggested for dynamic job shop problems with tardiness-related criteria. Quite the contrary, opportunistic testing of a legion of scheduling rules together with confusing reporting seems to have discouraged the use of all but some simple rules in industry, despite the increasing importance of shorter lead times and timely deliveries.

We report the search for a theoretically justified yet managerially applicable priority rule  actually an open protocol for integrated order management  from two perspectives. First, by analyzing large-scale simulations we provide insights into the rationale and tardiness behavior of all reasonable priority index rules in a variety of relevant job shop environments. We demonstrate that, in fact, there is not one rule but a whole family of dominating rules. These rules strike a balance between local anticipation of job tardiness, through different types of look-ahead features, and global coordination of machine utilization through rational lead time estimates. Certain conditions of the shop and the selection of performance criteria, including order-specific penalties for tardiness, inventory holding, and expediting, may favor one of these look-ahead dispatching rules over the others, thereby suggesting a tradeoff between the informational complexity of the index and the eventual impact of the inherent coordination principle.

Second, by testing different cases of implementation practices we provide comforting results for the managers worrying over the problems caused by errors in cost data and in processing time estimates, perhaps wanting to screen the flow of orders via release mechanisms, or feeling the desire to impose centralized control or visibility upon the process  the simple protocol works, the response is robust and additional features may harm rather than help. The agenda for future research still includes some critical implementation issues of the order scheduling protocols as well as the quest for different specifications of new classes of scheduling problems emanating from services and supply chain management. Information: Prof.dr. M.B.M. de Koster, e-mail: svelde@rsm.nl, Tel. 010-4082025/81719