Scheduling resource-constrained projects when non-conformities exist

The common assumption behind most project scheduling techniques is that the work to be done will be done as required and that rework will not be needed. Because of this assumption, scheduling techniques ignore the possibility of reworking non-conforming activities and thus may generate unrealistic schedules, where project managers have a difficult time anticipating the delays when reworks actually happen. In this paper we investigate the performance of several scheduling rules from the resource-constrained project scheduling literature. Specifically, we determined the scheduling rules that generate schedules with minimum total rework times. We analyzed a total of five rules, each of which is embedded in an early start, parallel scheduling algorithm. The total rework time of the schedule generated by each rule is computed for a set of 110 test problems from the literature.