Resource-Constrained Critical Path Scheduling by a GRASP-Based Hyperheuristic

Abstract

The greedy randomized adaptive search procedure (GRASP) is a well-known computational procedure for handling combinatorial optimization problems. Hyperheuristics, or heuristics to choose heuristics, are higher-level algorithms that operate on low-level heuristics, which in turn work in the solution domain. In this paper, a GRASP-inspired hyperheuristic is proposed for solving the resource-constrained project scheduling problem (RCPSP), one of the most important challenges facing construction project schedulers. The hyperheuristic is the main component of a multilevel procedure that runs within a commercial project management software system. On the basis of simple moves such as “swap” and “replace,” the low-level heuristics operate on the priority values that the software uses for scheduling activities. The efficiency of this prioritization-based algorithm has been addressed through a series of computational experiments on randomly generated projects. Moreover, the proposed procedure shows the advantages of utilizing the hyperheuristic approach to deal with optimization problems in the construction field.