A Discipline for Reactive Rescheduling

Proceedings Of The Third Artificial Intelligence Planning Systems Conference

Volume

Proceedings Of The Third Artificial Intelligence Planning Systems Conference

Contents

We present here a discipline for job shop rescheduling based on partial order backtracking. We show that partial order backtracking offers the rescheduler a framework for schedule repair, based upon a set of nogoods, which impose a systematic partial order on the set of activities to be repaired but allows non systematic techniques to be used within that framework. We present rescheduling as partial order backtracking by explaining the need for rescheduling, describing our framework (using a generic example), and discussing the particular implementation problems associated with repairing job shop schedules which consist of multi domain types.