A framework for multi-hazards risk aggregation considering risk model maturity levels

Multi-Hazards Risk Aggregation (MHRA) aggregates risk over different risk contributors and provides a final risk index that permits the comparison with safety guidelines to support Decision Making (DM). Maturity of the risk analysis are always different for different risk contributors, resulting in different degrees of confidence on the associated results. The difference in maturity, and the sources of heterogeneity that cause such differences, should be taken into account for supporting a reliable and accurate representation of risk in respect of DM. In this paper, we first propose a hierarchical framework to evaluate the level of maturity of risk assessments in respect of DM. The framework consists of different attributes (uncertainty, conservatism, knowledge and sensitivity) and sub-attributes (availability of data, consistency of data, data reliability, experience, and value ladenness) that are believed to determine the level of maturity of a risk assessment. Evaluation guidelines are pre-defined for facilitating the attributes scoring process. Then, the maturity level is integrated in MHRA by developing a two-dimensional risk aggregation method. Finally, a numerical case study for the MHRA of a Nuclear Power Plant (NPP) is carried out to show the applicability and the plausibility of the methods.

I. Introduction

Probability Safety Assessment (PSA) has been widely applied in various industries, e.g., nuclear, aerospace, defense, etc. Usually, different hazard groups are involved in a PSA (e.g., PSA of nuclear power plants usually involves hazard groups like fire, internal flooding, etc.). To make risk-informed decisions based on the results of PSA, Multi-Hazard Risk Aggregation (MHRA) is required: all relevant information on risk from different contributors is combined, arriving at an integrated risk index [1]. Usually, risk-informed decisions are made by comparing the integrated risk index (e.g., core damage frequency, large early release frequency, risk increase, etc.) to safety goals and quantitative acceptance criteria.