A multi-MLP prediction for inventory management in manufacturing execution system
Love Allen Chijioke Ahakonye, Ahmad Zainudin, Md Javed Ahmed Shanto, Jae‐Min Lee, Dong‐Seong Kim, Taesoo Jun
Abstract
Artificial intelligence (AI) positively remodels industrial processes, notably inventory management (IM), from planning, scheduling, and optimization to logistics. Intelligent technologies such as AI have enabled innovative processes in the production line of manufacturing execution systems (MES), particularly in predicting IM. This study proposes a Multi-MLP model with LightGBM feature selection technique for MES IM prediction to enable high prediction accuracy, minimal computation cost, low prediction error, and minimum time cost. The proposed model is evaluated using publicly available Product Backorder datasets to prove its reliability. Investigating varying feature selection techniques results in identifying appropriate data features relevant to building an AI-based solution for the IM prediction in MES. The experiment results demonstrate efficient decision-making of the proposed system with a low error prediction MAE of 0.2331, MSE of 0.1225, and RMSE of 0.3504.