NONIDENTICAL MACHINE SCHEDULING IN FLEXIBLE MANUFACTURING SYSTEM USING RECURRENT NEURAL NETWORK AND GENETIC ALGORITHM

This research paper focuses on optimizing the scheduling of jobs and tools for nonidentical machines, emphasizing minimizing makespan. The study employs metaheuristic algorithms to address this challenging combinatorial optimization problem. A comprehensive comparative analysis of different metaheuristic algorithms is conducted to evaluate their effectiveness in achieving this objective. This paper presents the above problem statement, metadata, algorithmic details, and results to provide insights into selecting the most suitable approach for scheduling in nonidentical machine environments. It is observed that the phase 3 modification of the Recurrent Neural Networks (RNN) had the lowest makespan metric while also providing a similar accuracy compared to the other algorithms. The accuracy of phase 3 of the Recurrent Neural Networks (RNN) is 99.8%, and the average processing time is approximately 4.02 minutes.