摘要

Since the schedule is often not optimal and unable to meet the promised completion dates in the critical chain project management, in this paper, we address an optimal approach of generating the critical chain schedule. The float time of the non-critical chain is used as the primary concern of setting feeding buffers, and thus the procedure of generating critical chain project plan is remarkably simplified. A critical chain project scheduling problem considering both the resource constraints and duration uncertainties is formulated, followed by a differential evolution algorithm with a new mutation strategy that combines two single strategies to obtain the global exploration as well as the fast convergence ability. At the end of the paper, numerical experiments are performed on modified standard data-sets in PSPLIB, and the results reveal the effectiveness of the algorithm.

全文