Thesis title: SAFETY INTELLIGENCE FOR THE EVOLUTION OF INDUSTRIAL SOCIO-TECHNICAL SYSTEMS
In an era characterized by rapid technological advancements and increasing complexity, any industrial operation demands for a paramount concern on occupational and operational safety. Therefore, the landscape of industrial operations has undergone a deep transformation in recent decades. The integration of cutting-edge technologies, increased automation, and globalization has led to a surge in the complexity and dynamicity of socio-technical systems. While these developments have indeed brought increased productivity and efficiency, they have also introduced new and often unforeseen safety challenges. The goal of this research is to shed light on the efficacy of the current safety reporting mechanisms and to contribute to a deeper understanding of the challenges and opportunities that arise when navigating the multifaceted landscape of modern industrial safety management.
This thesis embarks thus on a critical exploration of current Safety Reporting Systems (SRSs) within the context of complex industrial socio-technical systems. These latter encompass not only physical equipment and production processes but also human elements, organizational structures, technical-operational aspects, regulations compliance and their mutual interactions.
To achieve its goal, this thesis investigates the databases via Business Intelligence (BI) to firstly evaluate the current state of SRSs, including processes, technologies, and human factors that affect them. Then, Machine Learning (ML) algorithms have been applied to underline hidden patterns or behaviors in the most reliable safety databases. Then, the thesis shifts towards socio-technical modelling via System Theory (ST) to explore the complexity and interdependence of the variables within available SRSs, and to draft recommendations for improving future industrial SRSs.