Genetic algorithm for solving mobile robot scheduling problem in flexible manufacturing environment

• Main Authors: Erlianasha Samsuria, Erlianasha Samsuria, Mahmud, Mohd. Saiful Azimi, Abdul Wahab, Norhaliza, Zainal Abidin, Mohamad Shukri, Buyamin, Salinda
• Format: Article
• Language: English
• Published: Penerbit UTM Press 2023
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/108641/1/MohdSaifulAzimi2023_GeneticAlgorithmforSolvingMobileRobot.pdf; http://eprints.utm.my/108641/; http://dx.doi.org/10.11113/elektrika.v22n3.485
• Institution: Universiti Teknologi Malaysia

The scheduling genuinely a complex process, aimed at optimizing operational activities in pursuit of one or more objectives by leveraging production data which may include previous schedules. The scheduling problem in Flexible Manufacturing System (FMS) is commonly categorized as Nondeterministic Polynomial (NP)-hard combinatorial optimization problems and it remains as an endure problem to industrial practitioners and researchers. As part of real production scheduling, once one task is finished processing on a machine, transportation equipment such as mobile robot transports the completed task to the next machine. The problem of scheduling mobile robot in FMS pertains to the task allocation process for the robots, considering the transportation costs and the time spent to complete all operations. In recent years, Genetic Algorithm (GA) has been a remarkably effective search algorithm for solving a wide range of scheduling problems in a manner that achieves near-optimal solutions. This paper presents the metaheuristic techniques, specifically genetic algorithm, to address the NP-hard scheduling problem of two identical mobile robots in Job-Shop FMS environment. The algorithm is developed with the aim of finding feasible solutions to the integrated problem by minimizing the amount of time it takes to finish all tasks, commonly referred to as makespan. The performance of GA is evaluated with some numerical experiments which is executed via Matlab software. The scheduling results shows that the developed GA able to obtained the near-optimal solution of minimal makespan and converge within a short period of time.