Machine Learning Applications in Supply Chain Management: Predictive Power Unleashed
Saadaldeen Rashid Ahmed, Ali Jabbar Hussein, Hala Adnan. Fadel, Bourair Al-Attar, Ghassan Hamada Ali Alshmeel, Abadal-Salam T. Hussain
Abstract
Supply chain management (SCM) poses significant challenges in optimizing operations and enhancing efficiency. While previous research has explored the application of machine learning (ML) techniques in SCM, there remains a gap in understanding the impact of feature engineering on model performance. In this study, we address this gap by investigating various ML algorithms, including linear regression, decision trees, random forests, support vector machines (SVM), and deep learning neural networks, in predicting key SCM metrics. Leveraging a dataset from a Fashion and Beauty startup, our approach emphasizes feature engineering and model selection to improve prediction accuracy. Our findings indicate notable enhancements in model performance post-feature engineering, with the neural networks model exhibiting the highest accuracy, achieving a mean absolute error (MAE) of 40 and a root mean square error (RMSE) of 50. This research contributes to advancing the understanding of ML-driven SCM and underscores the importance of feature engineering in optimizing predictive models for real-world applications.