Simulation-based priority rules for the stochastic resource-constrained net present value and risk problem

Abstract

In this paper, the performance of activity-based and resource-based policies with newly designed priority rules for the stochastic resource-constrained project scheduling problem is considered. The aim of the paper is to maximize the expected net present value (eNPV) and to minimize the risk of the project's net present value by semivariance risk measurement. 16 (mainly cash flow-based) heuristic rules are considered, among which 10 rules are derived from the deterministic NPV literature which are referred to as direct priority rules, and 6 rules are developed based on simulation information. The performance of these rules in the mentioned objective functions has been evaluated on the benchmark data set project scheduling problem library. The priority rules have been ranked separately in maximization of eNPV and minimization of semivariance. The computational results on various test problems with 120 activities show that the top rank priority rules in optimizing each objective function under both policies are similar. Based on the obtained results, the priority rules which produce a schedule with a high eNPV, cause a schedule with high semivariance and this indicates the trade-off between eNPV and risk. The selection of the best priority rules is dependent on the risk-aversion level of the project manager. A risk-averse project manager is recommended to use the rule discounted cash flow weight based on activity duration in direct priority rules, and the rule statistical discounted cash flow weight at earliest finish time in simulation-based priority rules under both policies. But for a risk-preferent manager it is better to use the discounted cumulative cash flow weight at latest finish time in direct priority rules, and the statistical discounted cumulative cash flow weight at latest finish time in simulation-based priority rules under both policies.