Scheduling with co-evolving resource availability profiles

Abstract

Existing resource allocation models schedule activities by balancing resource requirement with resource availability using the resource available profile as a constraint to guide the scheduling process. The given resource available profile, however, may not be the best configuration for the project. Consequently, the resource available profile may have to be modified to produce improved resource utilization. With each modification the activities have to be rescheduled to meet resource availability constraints. A model is presented herein which is able to generate activity schedules that is optimal with respect to the resource available profile that is configured interactively with the schedule for minimum resource idling. The entire project is subdivided into separate resource contracts. The model then employs genetic algorithm (GA) to determine the start times of the activities and simultaneously self-adjust the resource hiring levels and end-dates for each contract in order to minimize the objective function. Results show that the model can provide schedules and corresponding resource available profiles that yield considerable savings in idle resources even in multi-resource problems. © 1996 OPA (Overseas Publishers Association) Amsterdam B.V. Published in The Netherlands under license by Gordon and Breach Science Publishers SA.