A Repair-based Approach for Reactive Scheduling in Job Shop Environment

An important practical problem in real world schedule execution is the occurrence of unforeseen events. When any changes in a production environment occur, the schedule has to be revised. Considering time consuming and shop floor nervousness, reactive repair of a disrupted schedule is a better alternative to total rescheduling. Most of the researchers have focused on heuristic based schedule repair techniques, as these techniques are easy to implement. Among these, right shift rescheduling is well known. This paper discusses how GENET network may be used to model and deal with the local disruptions that are frequently encountered in job shop environment, such as machine breakdown or urgent job (i.e. due date change). A repair-based local search approach for repairing a schedule with the objective of minimizing the difference between the new and the original schedule is proposed. The approach contains progressive stochastic search scheme and several heuristics. Then the resulting repaired schedule is evaluated through a simulation study with the performance measures of efficiency and stability.