A methodology to schedule and optimize job shop scheduling using computational intelligence paradigms

Evolutionary computation is emerging as a novel engineering computational paradigm, which plays a significant role in several optimization problems. Job-shop scheduling problem (JSSP) is one among the common NP-hard combinatorial optimization problems. The JSSP is defined as allocation of machines for a set of jobs over time in order to optimize the performance measure satisfying certain constraints like processing time, waiting time, completion time, etc. In this paper an eminent approach based on the paradigms of evolutionary computation for solving job shop scheduling problem is proposed. The solution to the problem is alienated into three phases; planning, scheduling and optimization. Initially, the jobs are scheduled, in which the machines and jobs with respect to levels are planned. Scheduling is optimized using evolutionary computing algorithm such as Genetic Algorithm (GA), which is a powerful search technique, built on a model of the biological evolution. Like natural evolution GA deal with a population of individuals rather than a single solution and fuzzy interface is applied for planning and scheduling of jobs. The well known Fisher and Thompson 10×10 instance (FT10) problem is selected as the experiment problem. The discussion on the proposed techniques and paths of future research are summarized.