Professor Júlio Barros, a researcher at the laboratory, has recently published an article in a highly prestigious international scientific journal.
This publication represents a significant recognition of the quality and relevance of the work developed by the researcher. Below, we provide the abstract of the article for reference.
This research paper proposed a novel Machine Learning (ML)-based approach for predicting the risk of supply delay using a scalable technological Big Data Analytics (BDA) architecture. It focuses on identifying the risk of supply delay in a proactive manner by combining several variables whose dynamics may affect positively or negatively the supplier’s performance, in detriment to the risk response manner vastly proposed in the literature. A ML pipeline stood proposed and therefore used as a basis to carry out all ML experiments in the Big Data (BD) environment. Contrarily to most research studies that have overlooked supply chain performance metrics, this one proposed was evaluated in terms of predictive power (statistical accuracy) and misclassification-related costs. Hence, the selection of the ML model was based not only on the criterion of providing better predictive power but also considering the minimum inventory-related costs caused by the mistaken classifications of the model. Overall, the proposed approach offers several benefits, mainly in operational and financial performance for the organization. It proves to be very useful for logistics planners, helping them in decision-making process and enabling proactive actions regarding possible supply delays.
The full version may be accessed through the link provided. https://link.springer.com/article/10.1007/s41060-025-00969-8
RS2Lab congratulates Professor Júlio Barros on this notable achievement and highlights his valuable contribution to strengthening the scientific research carried.