Project Resource Input Optimization Problem with Combined Time Constraints Based on Node Network Diagram and Constraint Programming

Abstract

The traditional resource-constrained project scheduling problem makes the amounts of resource input fixed and ignores the joint effect of multiple time constraints, which may lead to the failure of traditional algorithms. This paper introduces a new practical problem called the resource input optimization problem with combined time constraints (RIOP/CTC), which studies the influence of resource input schemes. The new problem combines three types of time constraints, including precedence relations, resource calendars, and interruptability for the first time, which makes it closer to the actual scheduling problem. We propose a new network diagram called node network diagram and develop an optimization model based on constraint programming (CP) and the technique for order preference by similarity to the ideal solution (TOPSIS). A three-step guideline and an actual project case are provided for schedulers to help them better use the model to solve RIOP/CTC, which also proves the validity of the model. Computational experiments are carried out to show that the CP optimizer is superior to the three common metaheuristic algorithms in solving quality and speed and can provide a near-optimum solution for large-scale scheduling problems in an acceptable time. The proposed model contributes to improving the practical decision system to support the formulation of real-life project resource input schemes, scheduling plans, and employee work plans.