Human factors

US industry codifies nuclear safety culture

15 August 2011



By the end of 2011 all US nuclear utilities will have adopted a new approach to assessing safety culture. Implementing this initiative should enhance the operational focus on safety at each plant by using a systematic approach to assessing safety culture to enable the early identification of trends.


After months of internal discussion, four year-long pilot programmes, and discussion with the US Nuclear Regulatory Commission (NRC), the US industry as a whole is adopting a new approach to safety culture to ensure the highest levels of commitment to nuclear safety. Tom Houghton, senior director of safety-focused regulation at US industry body the Nuclear Energy Institute describes the approach as ‘looking through a different lens.’ Using this scheme, utilities operating nuclear plants will standardize programmes to assess nuclear safety culture on a regular basis, to provide input for senior managers and to implement improvement programmes as necessary. The industry action supplements federal oversight conducted by the NRC, which has at least two independent inspectors at each of the U.S. nuclear energy industry’s power plant sites every day. NRC safety performance findings for each of the nation’s 104 reactors are posted online (www.neimagazine.com/nrcsapf).

The initiative was led by the NEI, and a team of utility experts, and was approved by the industry’s Nuclear Strategic Issues Advisory Committee. Houghton says: “The genesis of the efforts was about two and half years ago when the chief nuclear officers of our operating reactors decided that we needed to take a stronger look at what was going on in terms of safety culture and try and improve what we were doing”. The NEI utility team identified a number of areas where there was scope for improvement:

  • Different terminology used by NRC and INPO (Institute of Nuclear Power Operations) was creating confusion
  • Industry has the responsibility but had not taken the lead
  • NRC inspection findings are a very limited set of data for determining safety culture
  • Industry was not taking full advantage of all the possible indications of safety culture weaknesses
  • There was no industry-wide guidance for conducting safety culture assessments – self, independent, or third-party

The NEI team felt that meeting its objective of achieving a strong nuclear safety culture across industry required:

  • A common language of nuclear safety culture
  • Industry responsible and leading; NRC providing appropriate and transparent oversight
  • Use of all available assessment tools and data to assess, identify and correct weaknesses
  • A common methodology for conducting ‘snapshot’ assessments, including those requiring third-party assessors.

Both the US regulator and utilities look at safety culture, but each uses different documents and different sets of terms. The NRC uses a set of ‘components’ and ‘aspects’, while the industry uses INPO’s ‘principles’ and ‘attributes’. Although the two sets of terminology overlap, different terms are used and the parameters are not identical. “This causes a lot of confusion,” Houghton says, “because the regulator might say there is a weakness in this area and we say we think it is elsewhere. So one of the objectives was to work on a common language.”

Recently the NRC completed its policy statement on safety culture (700kb ZIP file from www.neimagazine.com/nrcsacu). This will pave the way for industry (lead by INPO) and the NRC to develop the common language so important for communicating safety culture issues between the regulator and the licensees.

Another area addressed by the NEI programme is the feeling among utilities that the regulator often does not use enough data to come to a conclusion regarding significant weaknesses in safety culture. Houghton says: “When the NRC find a violation of a regulation, they look at it and may assign a safety culture cause. If they then find four violations with that cause in a year, and believe that the licensee is not doing enough to correct the issue, they often come to a conclusion that there is a ‘substantive crosscutting issue’ regarding safety culture. Our feeling is that there is insufficient data to come to such a conclusion.”

The methodology that NEI’s team of experts has produced involves the power plants looking at all the data collected, which may include other assessments, metrics, performance indicators, issues arising out of work place, human resources issues as well as NRC violations. Houghton says that the broader base of data provides a ‘more holistic integrated picture’ of the station upon which to arrive at conclusions regarding the safety culture. “We feel this is more robust method of determining where the problem lies.” An example could be that, by looking across many areas, the station may realize that plant personnel are not taking advantage of lessons learned in the corrective action program. Site leadership could then direct actions to improve the corrective action programme and how information on declining trends is communicated.

The NEI methodology is described in NEI document 09-07 Fostering a Strong Nuclear Safety Culture, (available for download on www.neimagazine.com/nei0907) which provides a systematic approach:

  • Multiple inputs are reviewed, including external assessments
  • A monitoring review panel meets quarterly to assess the multiple sources of data and seek common issues and emerging weaknesses
  • A site leadership team meeting is convened at least semi-annually to review the panel’s report and recommendations
  • Actions are entered into the corrective action programme and communicated to all site personnel
  • External review boards, the NRC, and INPO provide external review of the process through their inspections and evaluations.

Pilot projects

Mike Gaffney, manager of regulatory assurance for PSEG Nuclear, took the lead on the pilot project at the Hope Creek plant, which is operated by Public Service Electric and Gas. Gaffney adapted NEI 09-07 for local use. In 2009 he set up a monitoring process for quarterly data and started looking at the data to ‘see what it told us about ourselves, and what we could learn from it.’

Gaffney describes the process: “The approach uses eight principles of safety culture and their lower-level attributes, which further define each principle. Various plant data such as corrective action evaluations, which provides most of the data, plus data from NRC violations, station self assessments, oversight audits and employee concerns issues are also used. We align all data reviewed to a specific nuclear safety principle attribute. We then evaluate which safety culture principle is causing the most problems and take actions to improve that performance.” The NRC took an interest in the Hope Creek pilot and attended meetings and provided feedback. Gaffney says, “We all agreed that the process needed to be transparent to the regulator. We made reports and meetings open to the regulator and came up with a standard list of actions to be taken if a significant problem occurs in any area.”

A working group carries out data analysis and presents results to senior management of line departments. In response to the question how much difference does the safety culture initiative make, Gaffney says, “We found that re-tabulating incidents against the principles gives an insight into safety culture. Looking from an aspect of nuclear safety gives a different perspective such that we might take slightly different actions with more communications.” He says that although there have been no surprises, the process has given employees a different perspective on communicating and on setting expectations. For example, the methodology exposed a potential problem area that people were not using the fundamental human performance tools of self-check and peer verification.

Robin Klearman is the station nuclear safety supervisor at another of the pilot plants, Dominion’s North Anna station in Virginia. Klearman manages site safety culture data collection and analysis. Data collected by the station include corrective actions, self-assessment databases and inspection reports. She says that applying the new safety culture methodology means looking at the data for behavioural issues, rather than just technical information. For example, she says, “Analysis of some trends we noticed showed a few low-level issues around areas where we might need a more questioning attitude.”

Other examples of changes implemented from feedback include providing station leadership with more training in QVV (question, validate, verify) and in change management. Across the four pilot plants, Braidwood (operated by Exelon), Hope Creek, North Anna and South Texas Project (operated by STP Nuclear), there was a shared feeling that improving communications is important. The safety culture methodology is now being rolled out across the US nuclear power fleet.

Richard Zuercher, Dominion’s Manager Public Affairs says, “This is a self initiative that the industry has started, to match what the NRC has done. What we are trying to do is to get ahead of the regulator, so that we can self identify issues before the regulator does, and correct them.” In future, the NEI will be hold annual meetings where utilities can share lessons learned from the methodology. NEI hopes that after NRC has observed the programme in action, it will modify its substantive crosscutting issue approach, which Houghton feels is not effective in being able to truly assess safety culture. However, he also says that he believes the NRC’s inspection insights are valuable inputs to the NEI 09-07 process, and encourages NRC to continue its look at a station’s ability to self assess safety culture.

In certain circumstances, the NRC requires that a station conduct a third-party safety culture assessment. NEI has been piloting an approach originated by the nuclear power confederation Utilities Service Alliance. This approach is a snapshot of safety culture at one point in time (as opposed to NEI 09-07, which looks continuously at safety culture.) It includes a survey and onsite interviews by a team. NEI is hoping that NRC will endorse this methodology for third-party safety culture assessment in 2011.

The USA NEI has had international interest in this self-assessment from China, Spain, Slovakia and South Korea, and has had held discussions with the IAEA.



FilesFigure 1: A simplified nuclear safety flow chart



Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.