Solving an Integrated Job-Shop – Mobile Robot Scheduling Problem in Flexible Manufacturing System using Enhanced Genetic Algorithm Structure with Local Search Method

Abstract

In a highly automated Flexible Manufacturing System (FMS), optimal utilization and scheduling of resources and equipment are paramount. This necessity underpins the full utilization of automation capabilities, leading to increased productivity and minimized downtime. Efficient resource allocation and scheduling also contribute to better overall performance, allowing the FMS to meet production demands effectively while maintaining a high level of operational efficiency. In this paper, the job-shop production scheduling problem is studied which involve with the concurrent scheduling of jobs processing and mobile robot assignment for job transportation within FMS environment. The hybrid Genetic Algorithm and Tabu Search algorithm is proposed to solve the combinatorial NP-hard job-shop and mobile robot scheduling problem. The primary objective is to search for the best scheduling plan that includes job allocation and mobile robot assignment, aiming to minimize the overall time necessary to complete all tasks, also known as makespan, to the minimum as possible. The developed algorithm has been evaluated and compared with the classical (or standard) genetic algorithm and other hybrid GA (with Simulated Annealing (SA) algorithm) using two job datasets spanning from small to large-scale problems adopted from renowned benchmark job instances. The results of computer experiments substantiate the effectiveness of the proposed hybrid algorithm, showcasing superior-quality solutions with an approximate up to 2.86% and 3.55% improvements compared to the hybrid GA – SA and standard algorithm, respectively. The developed algorithm has been run and tested in the Matlab software environment.

Keywords

• flexible manufacturing system
• genetic algorithm
• hybrid optimization
• job-shop
• scheduling
• tabu search