OPTIMIZING MANUFACTURING SCHEDULING WITH GENETIC ALGORITHM AND LSTM NEURAL NETWORKS

Abstract

In response to Industry 4.0 and the rise of intelligent manufacturing, this study develops a system combining Long Short-Term Memory (LSTM) Neural Networks and a Multi-Objective Genetic Algorithm to improve prediction and optimization in manufacturing scheduling. A novel model predicts work-in-process (WIP) inventory using LSTM neural networks, accommodating dynamic changes in production. A manufacturing scheduling model is also created and solved using a multi-objective genetic algorithm, simplifying the resolution process and obtaining practical solutions. These methods provide a valuable approach to optimizing production scheduling in intelligent manufacturing, enhancing efficiency and economic gains.