ML19098B328

RIC2019 Session T-11, U.S. NRC Operating Experience Data-Applications in Risk Modeling Presentation Title: Common Cause Failure Considerations in Context of Risk Informed Decision Making Presenter: Greg Krueger, NEI

ML19098B328
Person / Time
Site:	Nuclear Energy Institute
Issue date:	03/12/2019
From:	Nuclear Energy Institute
To:	Office of Nuclear Regulatory Research
John Lane 415-2476
References
Download: ML19098B328 (9)

Text

Common Cause Failure Considerations in the Context of Risk Informed Decision Making T11 - U.S. NRC Operating Experience Data Applications in Risk Modeling Greg Krueger Senior Technical Advisor Nuclear Energy Institute March 12, 2019 ©2019 Nuclear Energy Institute

Common Cause Failure Considerations in the Context of RIDM

• The current quantitative-focused CCF approach in evaluating performance issues emphasizes numerical results with limited consideration of recent operational processes or performance

• A quantitative focus leads to significant application of analytical resources when treatment of CCF drives a Significance Determination Process (SDP) result

• Gaining alignment on how to risk-inform CCF insights in an event/condition assessment would improve decision making and clarity of performance assessment outcomes

©2019 Nuclear Energy Institute 2

Unprecedented Industry Performance Top Best

>90% 10x Performing Ever 100x Reliability Safety Excellence Oversight Margin

>90% 10-Fold Highest Fewest Fleet Capacity Factor Reduction in Performance NRC ~100 Times Sustained Average CDF Record Ever Performance Safer Than Issues Ever NRC Goals Exceptional Safety Culture

©2019 Nuclear Energy Institute 3

Common Cause Failures - Data

• Current component failure events and common cause failures are not representative of the improvement in industry wide performance over the most recent 10 years

• Recent industry evaluation of the average number of failure events associated with Motor Driven Pumps (MDPs) and Emergency Diesel Generators (EDGs) shows a significant decrease from the period 1996 to 2005 compared to 2006 to 2015

• For example, a reduction of up to 1/3 fewer pump failure to start events in the most recent 10 years has been observed

• While MDP failure rates have generally decreased over time, the rate of decrease in CCF is more dramatic (40% to 60% decrease).

©2019 Nuclear Energy Institute 4

Industry CCF White Paper

• Expand the framework for crediting observed defense against common cause qualitatively independent of the quantitative risk analysis Recommendation 1: Perform sensitivities between baseline or nominal CCF calculation and conditional CCF calculation to illustrate the quantitative range of results Recommendation 2: Expand Ground Rule 3 in the Risk Assessment of Operational Events Handbook (RASP Handbook) to reflect use qualitative factors to reflect utility defenses employed to limit CCF

©2019 Nuclear Energy Institute 5

Common Cause Failures - Qualitative Factors

• Use of qualitative CCF factors within the existing regulatory performance assessment process has potential to:

Improve characterization of performance deficiencies Better focus on higher organizational deficiencies Optimize resource expenditures by focus on performance rather than quantitative refinements Add perspective to the evaluation of a performance deficiency Increase the recognition and use of common cause defenses by utilities, thereby promoting improved good practices

©2019 Nuclear Energy Institute 6

Common Cause Failures - Qualitative Factors

• A suggested application table reflecting the strength of defense against CCF is suggested for incorporation into the RASP handbook Strength of Defenses Qualitative Factor Complete Nominal or baseline CCF Superior Nominal CCF - 0.5 Conditional CCF Moderate 0.5 Conditional CCF - Conditional CCF Weak Conditional CCF (Current NRC quantitative process)

• Matrix could be used in conjunction with quantitative SDP results to reflect utility action taken to prevent the potential for common cause

©2019 Nuclear Energy Institute 7

Conclusions

• The combined impact of improved data and defenses employed to reduce common cause events will result in a reduction of resources used to evaluate performance issues

• Industry performance and improved processes should influence the evaluation of CCF using an integrated RIDM approach

©2019 Nuclear Energy Institute 8

Contact:

Greg Krueger gak@nei.org

©2019 Nuclear Energy Institute 9