Robust scheduling in an advanced planning and scheduling environment

摘要

A new scheduling problem that appears in an advanced planning and scheduling (APS) environment is discussed. Under the precondition that all materials are available when needed, the problem is formulated as follows: min Σ from i=1 to n of (D_i - E_i)/N subject to D_i - C_i≥0, for all i, where N is the number of customer orders arriving randomly at the shop during a certain period, D_i is the estimated due-date for customer order i, E_i is the due-date estimation time for customer order i and C_i is the completion time for customer order i. Then, D_i and C_i are endogenous variables and E_i is an exogenous variable. The ability to construct a flexible scheduling process that avoids the need to fix ongoing schedules is essential to all APS systems. Therefore, the concept of a due-date buffer is introduced which is expected to enable the production schedules for each customer order to be flexible at the beginning and gradually become fixed as the processing of the order progresses. A simulation-based scheduling algorithm using the concept of a due-date buffer is developed here and subsequently examined through a series of numerical experiments. The obtained computation results show that the performance of due-date buffers is outstanding with respect to complicated production processes having higher utilizations.