Bottleneck identification and transfer prediction for automated production lines based on FNN

Abstract

Abstract In a re-entrant production system, the throughput of the whole system depends on the capacity of the bottleneck machine. In this study, a new definition of bottleneck is proposed for a precision forging blade shop. The reinforcement learning algorithm is used to optimize the production scheduling to determine the most suitable scheduling scheme, which lays the foundation for bottleneck identification. Subsequently, the bottleneck identification index system was established according to the optimization objective, and the bottleneck identification problem was transformed into a multi-attribute decision-making problem. Finally, a fuzzy neural network is used for training, and the basic scheduling examples of each flow shop are utilized for bottleneck identification and prediction to verify their effectiveness.