Critical Chain Project Scheduling Problem In DTRTP Environment

Existing research is mostly carried out in the single-mode background. Only a few scholars have conducted their researches in the multi-mode context, especially in the aspect of DTRTP. In construction and software development projects, it often occurs that the activities could be executed in different modes, corresponding to different amounts of resources and thus different activity durations. This kind of problem is known as DTRTP which attracting increasing attention. But existing research mostly focuses on developing algorithms, including deterministic algorithms and heuristic algorithms. However, there are various optimal execution modes with the same minimum project duration in the DTRTP. The projects will produce different performance in the execution phase under uncertainty.DTRTP is the discrete time/resource trade-off problem. The objective of the DTRTP is to find the execution mode with an as short project duration as possible under the premise of meeting the time and resource constraints. Aim at two different-scaled problems in this thesis, the branch and bound algorithm and tabu search algorithm were used to get all the optimal execution modes. In order to generate the corresponding critical chain project schedule, the CC/BM was applied to insert feeding buffers in baseline schedule. And the impact of different optimal execution modes on the critical chain scheduling was analyzed. Two filtering indexes including total duration and resource tightness were proposed, the corresponding screening strategy to select the optimal execution modes with better project performance was also summarized. Finally, the effectiveness of the indexes and strategy were verified by large amount of simulated experiments.