Abstract
Decades ago, German chemical companies, including Bayer AG, adopted a semi-deterministic approach to risk assessment through the application of a risk matrix in Hazard and Operability Studies (HAZOPs). This matrix classifies the likelihood of various initiating events and their severity, thereby defining the level of risk. In this way, the reliability of safety measures is determined. The likelihood of initiating events is primarily derived from statistical data, while severity levels are based on guiding principles enhanced by individual judgments from process and plant safety professionals. In complex scenarios, severity assessments are supplemented by modeling the physical consequences. The severity of toxic material releases or thermal radiation from fires is typically estimated using simplified methods or consequence modeling. However, modeling the consequences of physical explosions is complex, resource-intensive, and necessitates multiple boundary conditions and assumptions. To enhance this process, we developed a method that effectively correlates equipment design data with the resulting severity classifications. These classifications are based on calculations of gas-generated pressure waves for diverse equipment configurations and conservative assumptions, specifically for ductile equipment. Consequently, we developed a user-friendly table that enables process and plant safety professionals to easily determine the appropriate severity level for different design scenarios, ultimately improving safety practices in risk assessments and aiming for better harmonization of severity level determination and safety concepts throughout Bayer.
