Risk Metrics

Purpose

Develop quantitative and qualitative metrics that can be used in risk modeling for advanced reactors to make risk-informed decisions on plant design in a similar fashion to how core damage frequency (CDF) and large early release frequency (LERF) are currently used for water reactors.

Scope

Risk metrics should be expected to evolve over time as those used to investigate qualitative risk at an early stage of design for advanced reactors may be different from the ones used to investigate quantitative risk of a more mature design and licensing phase. Different options for risk metrics are being explored that vary from a standardized form that can apply to various designs, like CDF and LERF, to site specific risk metrics that have different bases than CDF and LERF. Stakeholder feedback has suggested that whichever option is pursued should be technology neutral.

Advanced Reactor Risk Metrics

Decades of operation of water reactor technologies have developed a consensus on the risk metrics to be used for these technologies; in particular, the use of CDF and LERF as figures of merit. However, for non-LWR advanced reactor technologies not cooled by water, the water reactor metrics may not be meaningful or useful for either regulatory or operational / management purposes because of the vastly different designs of advanced reactors. A trial Reg Guide 1.247 [7] has been developed to help fill some of the gaps identified between water reactor risk metrics and advanced reactor risk metrics. In November 2024, EPRI released a new report EPRI 3002029252 [60] that addresses the research actions in determining the proper risk metrics for advanced reactors. The report identifies Quantitative Health Objective (QHOs) as the basis for the development of all risk metrics including advanced reactors. The primary questions that EPRI 3002029252 aims to answer are the following:

• What are the strengths and weaknesses of existing risk metrics for application to advanced reactors?
• What is the background that supports the use of current risk metrics as surrogates for various high-level safety objectives?
• For advanced reactor designs that are not conducive to the use of existing risk metrics, what metrics can be developed and used?

The Key Findings in EPRI 3002029252 recommend a primary technology-neutral risk metric, offsite dose emergency action frequency (ODEAF), defined as a 50 mSv (5 rem) projected mean dose at the facility boundary for a period of four days following the start of a release. This approach is consistent with existing regulatory objectives and guidance (for example, NEI 18-04 [8], the United States Quantitative Health Objectives, and various International Atomic Energy Agency guidance documents) as well as existing regulatory targets from several national regulatory authorities.

Key topics related to risk metrics for advanced reactor technologies that are to be further investigated include:

• Importance measures to correspond to different risk metrics for ARs, including changes to use absolute measures rather than relative measures (as supported by the Non-LWR PRA Standard [6])
• Guidance for the utilization of different risk metrics for ARs related to the need to support requirements to identify significant contributors for the application of requirements in the PRA standards
• Application of risk metrics for ARs to sites with multiple units/modules
• Application of risk metrics for ARs throughout the life cycle as the plant and associated PRA evolve during design, construction, and operation
• Application of risk metrics to non-core sources of radioactivity (for example, spent fuel)
• Applicability of risk metrics to research and test reactors, which do not traditionally utilize PRA risk metrics like water reactors