Scheduling of Jobs on Dissimilar Parallel Machine Using Computational Intelligence Algorithms

Abstract

The success of Computational Intelligence (CI) techniques to solve combinatorial scheduling problem critically depends on efficient modelling of the problem. In this chapter, we study the scheduling of a set of jobs with different release and due dates on a set of dissimilar parallel machine problems to minimize the processing cost. The performance of five recent CI techniques viz., Artificial Bee Colony, Dynamic Neighborhood Learning based Particle Swarm Optimizer, Genetic Algorithm, Multi-Population Ensemble Differential Evolution (MPEDE) and Sanitized Teaching-Learning based Optimization is evaluated on problems reported in the literature. It was observed from 750 unique trials (5 problems × 2 datasets × 5 algorithms × 15 runs) that MPEDE showed superior performance to the other four algorithms for larger problems.