Investigation of surrogate measures of robustness for project scheduling problems

Abstract

Uncertainties in systems and data necessitate plans that are flexible enough to be applicable in a real-time production environment. Such plans find a compromise between optimality and flexibility, frequently termed as robustness. Time is a well-known criterion for evaluating optimality of a schedule, but in the case of robustness, additional surrogate measures are required. These surrogate measures of robustness aid in identifying proactive schedules, which can provide flexible yet efficient plans with minimum redundancy. This study suggests new robustness measures and proposes a test bench for the systematic analysis of such criteria. Regression analysis was used for comparison of the suggested measures with traditional slack-based measures. The results show that float index (a novel measure) has a strong correlation with tardiness and is an effective measure to determine the robust nature of any schedule.