The Inventory of Spare Parts Prediction based on LSTM for Large Data and a Short Replacement Cycle

Abstract

Large variations in the replacement cycle and data size of different auto parts present a challenge to auto part companies. To predict the inventory of spare parts, an prediction method based on long short-term memory (LSTM) is proposed. The LSTM-based predictive model is mainly adopted for auto parts with large data and a short replacement cycle. We verified by experimentation that our prediction method based on LSTM can accurately forecast the inventory of spare parts with large data and a short replacement period. The method provides an efficient solution for improving the inventory performance of auto parts companies.