Abstract

Purpose: The purpose of this study is to compare and analyze the structural characteristics of AI-based industrial safety information systems with those of conventional industrial safety management systems, and to identify a paradigm shift in industrial safety management from an information systems perspective. Research design, data and methodology: The analytical framework is a risk-aware, information-flow–based decision-making structure, which enables a comparative analysis of conventional industrial safety management systems and AI-based systems. Results: The analysis reveals that conventional systems are characterized by reactive post-incident management that focuses on documentation and inspections. In contrast, AI-based industrial safety information systems exhibit a cyclical structure that integrates real-time data collection, AI-driven analysis, immediate alerts and responses, and continuous feedback and organizational learning. This structural distinction indicates a transformation in which industrial safety management has shifted from regulation-compliance–oriented post-incident management to a data-driven prediction and proactive management system. Conclusions: This research conceptualizes the AI-based safety information system as a fundamental paradigm shift in industrial safety management structures. The findings elucidate a transition from compliance-oriented reactive protocols to data-driven proactive systems. Consequently, this study suggests that future safety frameworks must prioritize optimized information flows and structural management to ensure sustainable organizational safety.