Studying the Performance of Unified Particle Swarm Optimization on the Single Machine Total Weighted Tardiness Problem

Abstract

Swarm Intelligence algorithms have proved to be very effective in solving problems on many aspects of Artificial Intelligence. This paper constitutes a first study of the recently proposed Unified Particle Swarm Optimization algorithm on scheduling problems. More specifically, the Single Machine Total Weighted Tardiness problem is considered, and tackled through a scheme that combines Unified Particle Swarm Optimization and the Smallest Position Value technique for the derivation of job sequences from real–valued particles. Experiments on well–known benchmark problems are conducted with promising results, which are reported and discussed.