Abstract—Accurate sales forecasting is essential for any successful retail company in the competitive environment we live in today, where sales are of utmost importance to companies. By limiting overstock and preventing overproduction, it may aid in inventory management. Future sales are affected by a number of significant variables. A retail store’s overall sales trends or the sales of a particular product may be examined to determine these aspects. With the use of temporal, historical, trend and seasonal data, this study develops a deep learning-based Deep Prophet Memory Neural Network (DPMNN) forecasting approach. Using M5 Forecasting and Predict Future Sales datasets in a Python context, the built system is used and evaluated. Extensive testing and comparisons to state-of-the-art research show that the suggested demand forecasting method achieves notable outcomes by obtaining lower Root Mean Squared Error (RMSE) and Mean Squared Error (MSE) rate.
 
Keywords—demand forecasting, deep prophet memory neural network, linear clipping data normalization, bivariate wrapper forward elimination, sequential Bayesian inference optimization


Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.