An improved adaptive fuzzy-genetic algorithm based on local search for integrated production and mobile robot scheduling in job-shop flexible manufacturing system

Abstract

The central focus of this paper is to generate optimal schedules for job operations using mobile robots with the lowest makespan in a Flexible Manufacturing System (FMS). In Job-Shop FMS, specific machines are selected for individual jobs which requires higher levels of flexibility and complexity in scheduling. The joint scheduling problem in a Job-Shop FMS primarily involving the coordination of the production machines and mobile robots in schedules. Genetic Algorithm (GA) has emerged as the most extensively implemented evolutionary algorithm to address the production scheduling due to its capability to produce high-quality, rapid, and efficient results in exploring complex and global solution spaces. Despite its strong global search capability, the algorithm is prone to trap in its local optima when tackling the complex problem of scheduling mobile robots in a job-shop setting with precedence constraints. Therefore, this paper presents an improved structure of GA by integrating it with adaptive Fuzzy-GA operators and Tabu Search (TS) algorithm to minimize the makespan. The resulting hybrid algorithm offers a novel approach that effectively balances search performance to achieve high-quality solutions in terms of fitness minimization and convergence speed. The proposed algorithm was tested using several benchmark datasets and was subjected to comparative experimental analysis. The empirical results demonstrate the improvement of the proposed algorithm over comparative methods, with improvements of 6.5%, 6.9% and 9.8% over the GA-Tabu algorithm, Fuzzy-GA, and standard GA respectively, in solving the complex scheduling problem within FMS.