Can I read Demand Forecasting of Fast-Moving Consumer Goods by Deep Learning-Based Time Series Analysis on EtoBox?

Demand Forecasting of Fast-Moving Consumer Goods by Deep Learning-Based Time Series Analysis by José Nicolás Valbuena Godoy; Roberto Arias; Hugo Franco is a book available to read on EtoBox.

What is Demand Forecasting of Fast-Moving Consumer Goods by Deep Learning-Based Time Series Analysis about?

Demand forecasting plays an essential role in the ability of certain companies to meet future customer requirements. Current research is focused on the construction of models intended to reduce forecasting error. However, most Colombian companies using demand forecasting tools mostly apply traditional regression models to project their demands, so they obtain and employ coarse estimations. This paper presents a performance comparison between stochastic time series models (SARIMA-MLR) and Recurrent Neural Networks (RNN) for demand forecasting of a group of Fast Moving Consumer Goods (FMCG) in the beauty and make-up sector (nail varnish). A set of demand projection models based on SARIMA-MLR and RNN (LSTM) are evaluated using forecast error measures. According to the results in this work, RNN with two LSTM layers presents the highest forecasting performance for four selected goods.

Author

José Nicolás Valbuena Godoy; Roberto Arias; Hugo Franco

Publisher

Springer International Publishing

Published

2023

Language

EN

ISBN

9783031322129

Subjects

Computer Science, Engineering, Stem