Explainable artificial intelligence for supply chain forecasting : identifying the drivers of sales predictions
Loading...
Date
2026-09-28
Authors
Δουκλιάς, Χριστόφορος
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis examines the use of Explainable Artificial Intelligence for sales forecasting in the context of supply chain management. The analysis is based on a publicly available dataset containing information related to products, orders, prices, and sales. The main objective is not only to achieve high forecasting accuracy, but also to improve the understanding of the factors that influence the predictions produced by the models.
The results show that very high predictive performance can be maintained even when a smaller number of variables is used. Product price, order quantity, and product category were found to play an important role in the forecasting process. In addition, the use of explanation methods makes the predictions more understandable and transparent. Overall, the study demonstrates that accurate forecasting can be combined with greater interpretability, supporting more informed and trustworthy decision-making in supply chain management.
Description
Keywords
Artificial intelligence, Forecasting, Sales, Explainability