Random Sampling Technique For Improving The Net Present Value Of Resource-Con-Strained Projects

Shou Yong-Yi
2005-01-01
Abstract:Industrial projects require a set of resources for implementation while these resources are normally limited. The scarcity of resources significantly increases the complexity of project scheduling. Meanwhile, net present value is a more appropriate performance criterion than the commonly used project duration for projects with a long time span since the adjustment of schedules could considerably influence the cash flows within the projects. For projects with resource constraints, it is a challenge to maximize the net present value through optimized resource allocation and project scheduling. A combined regret-based biased random sampling technique is proposed in this paper to improve the net present value of resource-constrained projects. The proposed technique adopts the serialschedule generation scheme, and a variety of priority rules are combined randomly to improve the project schedule. The Patterson's test problems are adopted to compare the proposed technique with other recognized heuristics in the literature. A statistic test shows that the proposed technique can significantly improve the net present value of resource-constrained projects.
What problem does this paper attempt to address?