Cat Swarm Optimization-Based Schemes for Resource-Constrained Project Scheduling

Abstract

This paper presents a cat swarm optimization (CSO)-based method for resource constrained project scheduling problem (RCPSP). The CSO simulates the behavior of cats in two sub-models and potential solution to the RCPSP is presented by the multidimensional positions of cats. CSO-based scheme for the RCPSP has three main stages: first randomly initialize the parameters of cats, then update the position in iteration and calculate the fitness through serial SGS method, finally terminate the process if the condition is satisfied. Compared to the other widely used heuristic methods, CSO is easy to understand and to implement. The adoption of CSO in solving RCPSP indicates the universality of CSO in solving operational problems. When solving RCPSP, some refinement of original CSO are made. The performance of the proposed algorithm is compared against a set of heuristic and meta-heuristic methods, and it is tested on standard problem sets called PSPLIB which is freely available on the Internet. The empirical results show that CSO has an average good performance among the other compared methods.