Hybridization of Evolutionary and Swarm Intelligence Techniques for Job Scheduling Problem

Abstract

Scheduling is the process of deciding how to commit resources to varieties of possible jobs. It also involves the allocation of resources over a period of time to perform a collection of jobs. Job Scheduling is the sequencing of the different operations of a set of jobs in different time intervals to a set of machines. In this chapter a hybrid technique combing an evolutionary and swarm intelligence technique to the job scheduling problem is proposed. In literature, different hybrid techniques are used for performing job scheduling process in various fields. The existing systems have utilized techniques such as Artificial Bee Colony (ABC), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Simulated Annealing (SA) etc. and hybrid techniques were derived from the combination of these algorithms. These hybrid techniques mostly concentrate on two factors such as the minimization of the makespan and completion time. The performance of these hybrid techniques lack in the scheduling process because, they have not considered the efficient factors like (i) turnaround time, (ii) throughput and (iii) response time during the job scheduling process. The main aim of this work is to provide a better hybrid job scheduling technique by overcoming the problems that exist in the literary works and to minimize the completion and makespan time. The proposed technique considered the efficient factors (i) turnaround time (ii) throughput and (iii) response time which were left out in the existing hybrid techniques for job scheduling process. The performance of the proposed hybrid job scheduling technique was analyzed with the existing hybrid techniques. The experimental results proved that the proposed job scheduling technique attained high accuracy and efficiency than the existing hybrid techniques.