Comparative Analysis of GPGPU based ACO and PSO Algorithm for Employee Scheduling Problems

Abstract: Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO) are simple, easy to implement, their robustness to control parameters, and their computational efficiency when compared with mathematical algorithms and other heuristic optimization techniques. The calculation in PSO and ACO is very simple. Compared with the other developing calculations, it occupies the bigger optimization ability and it can be completed easily. It is used to solve many NP-Hard problems. Employee Scheduling is a real-life NP-Hard problem faced by many organizations. Self-scheduling in all situations is not always practical and possible. Nurse Rostering is related to highly constrained resource allocation problem into slots in a legal shift Earlier the problem was solved using different heuristic algorithms. In this dissertation, we have proposed, GPGPU based parallelization of PSO and ACO to solve Employee scheduling problems. To parallelize both algorithms, a master-slave approach is used. The BCV 8.13.1 data set is used for experimentation purposes. Analysis of results is done based on mean, standard deviation, standard mean error. Keywords: Employee Scheduling, Parallelization, PSO, GPGPU.