Artificial Intelligence (AI) and Machine Learning (ML) techniques are gaining increasing attention in the safety field due to their ability to discover patterns from data generated in a cognitive Internet of Things (IoT) environment. Data can be reactive (i.e., collected after the occurrence of an event) or proactive (i.e., gathered before an event). Both these data are pivotal to enhancing occupational and process safety management, but this issue is scarcely addressed in the literature. Moreover, there is limited evidence on the implementation of AI solutions for safety management purposes and, in particular, in hazardous industries. The iron and steel sector is one of the most hazardous industries in the world, whose activities may expose workers to a wide range of hazards and cause incidents, accidents, injuries, or diseases. In such a context, the iSafety project focuses on enhancing safety management in the iron and steel industry by leveraging AI techniques. Essentially, iSafety aims to increase awareness and knowledge about the hazardous conditions potentially encountered in the iron and steel industry, developing methods able to consider both reactive and proactive data with the purpose of predicting and anticipating deviations of critical parameters and scenarios and defining safety management recommendations. This will strengthen the preparedness and resilience abilities of iron and steel companies.

Keywords: Occupational safety, Process safety, Risk assessment, Risk management, Anomaly
detection, Pattern recognition

Reference Group: BEHAVIORAL MODELING AND SCALABLE ANALYTICS – BMSA

Partners: Università degli studi di Brescia (UNIBS) , Istituto di Calcolo e Reti ad Alte Prestazioni del
Consiglio Nazionale delle Ricerche (ICAR-CNR)

Founding body: MUR Ministero dell’Università e della Ricerca

Total Cost: 197.281,00

ICAR Cost: 91.911,00

Start Date: 28/09/2023

End Date: 27/09/2025