Did we learn about risk control since Seveso? Yes, we surely did, but is it enough? An historical brief and problem analysis

Abstract Since the Seveso disaster 40 years ago, risk management methods as well as challenges to the process industry have increased along with changes in the public's risk perception globally. The Seveso incident brought to attention some critical issues, such as lack of knowledge on runaway reaction scenarios, hazards of formation of dioxins, lack of regulatory requirements, poor communication and coordination, and little or no emergency response or evacuation plans. Regulations related to process safety and risk management have evolved with time as compared to earlier when there were no specific regulations for controls of such major hazards as became evident during the Seveso incident. Meanwhile, the third improved version of the EU Directive inspired by the incident has been implemented. Although in general over the years occupational safety has vastly improved, major losses have still occurred and in fact formed the motivation to issue the new versions of the Directive. This paper highlights two limitations of the Seveso III Directive – lack of implementation of leading indicators and limited application of hazard and risk identification (for example little or no consideration of the hazards of intermediate products). Some existing gaps like lack of learning from previous incidents, scale-up issues, limitations of experiments related to real scenarios (e.g., vapour cloud explosion), uncertainties involved in complex systems and their gradual degradation by their use are described. These gaps necessitate developing and using advanced methods and an holistic approach such as resilience and advanced mathematic-statistical methods to resolve these issues. This work presents a resilience-based analysis of the Seveso incident and lays the foundation for development of a Process Resilience Analysis Framework. Implementation of this framework would advance current risk assessment and management techniques through integration of technical and social factors. The paper concludes with some cardinal rules for a systems approach to risk management and the significance of risk governance.