Evaluating and Managing Sustainability Performance of Supply Chain and Business Process Management: An Integrated and Applied Approach

Abstract

As global supply chains become increasingly complex and environmentally conscious, the imperative for Sustainability-Driven Decision-Making (SDDM) gains paramount importance. This paper delves into the transformative potential of machine learning in reshaping sustainability practices within supply chains. Leveraging a diverse dataset encompassing provisioning, production, sales, and commercial distribution across clothing, sports, and electronic supplies, we employ a range of machine learning algorithms, including Logistic Regression, Gaussian Naive Bayes, Support Vector Machines, k-Nearest Neighbors, Linear Discriminant Analysis, Random Forest, Extra Trees, XGBoost, and Decision Trees. Our analysis spans critical dimensions of supply chain management, from fraud detection to late delivery prediction, and illuminates the pivotal role of these algorithms in improving sustainability outcomes. Through empirical experimentation, we identify optimal models for each task, revealing their strengths and limitations. Additionally, we visualize feature importance, offering insights into the factors shaping sustainability within supply chains. Our research underscores the symbiotic relationship between data-driven decision-making and sustainable practices, paving the way for more responsible, efficient, and resilient supply chains. As businesses seek to navigate an evolving landscape, the fusion of machine learning and sustainability emerges as a compelling paradigm, fostering a future where supply chains not only optimize operations but also contribute to global sustainability goals.