Adjustment of generic frequencies for major accident hazards: Case of SEVESO establishments

Abstract

The direct use of generic frequencies from databases to assess Major Accident Hazards (MAH) can introduce substantial biases in risk assessment. These generic data are inherently uncertain, as they fail to reflect the specific operational conditions of a given facility. Such biases may result in either overestimation, which can lead to excessive and costly safety investments, or underestimation, which compromises the adequacy of risk mitigation measures.

To address these limitations, this paper proposes an integrated adjustment approach specifically designed for quantitative risk analysis to manage the uncertainties associated with generic data. The proposed approach enables the adjustment of generic failure frequencies by incorporating technical factors related to safety barriers (e.g., the probability of failure on demand), and operational management factors. Additionally, it considers the time-dependent degradation of equipment, i.e., the degradation factor.

The applicability and effectiveness of the proposed approach in industrial risk management are illustrated through a case study involving a pressurized equipment system (a three-phase separator). The findings highlight its significant advantages in managing industrial risks, particularly in SEVESO Establishments, when compared to existing techniques.