Artificial Intelligence for Forecasting in Supply Chain Management - Université Polytechnique des Hauts-de-France Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Artificial Intelligence for Forecasting in Supply Chain Management

Résumé

This paper proposes an appropriate model to forecast the trend of white sugar consumption rate in Thailand due to the fluctuation of consumption rate nowadays. This paper will focus on two main forecasting model types, which are the regression models and neural network models. Moreover, the performance is evaluated by using Root Mean Square Error (RMSE) and Theil’U statistic value. After processing the experiments, the results demonstrate that Long Short-Term Memory (LSTM) recurrent neural network provides the best performance for the forecasting, with the condition of combination between the existing consumption rate and other relevant factors like production supply, import rate, export rate, and inventory stock. Also tuning the model’s parameters is an important issue.
Fichier non déposé

Dates et versions

hal-03457244 , version 1 (30-11-2021)

Identifiants

  • HAL Id : hal-03457244 , version 1

Citer

Anirut Kantasa-Ard, Abdelghani Bekrar, Abdessamad Ait El Cadi, Yves Sallez. Artificial Intelligence for Forecasting in Supply Chain Management: A Case Study of White Sugar Consumption Rate in Thailand. 9th IFAC Conference on Manufacturing Modelling, Management and Control MIM 2019, 2019, Berlin, Germany. pp. 725-730. ⟨hal-03457244⟩
35 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More