Tuesday, October 27, 2015

Why Reforming the Reactor Oversight Process is the Right Thing to Do

Jim Slider
Fifteen years ago, the NRC and industry cooperated on reforming the way in which NRC decided where to focus its attention across the U.S. fleet of power reactors.  Among the guiding principles of the reform was to make NRC decisions on operating reactor oversight more transparent and predictable, and ensure that additional NRC resources were applied where they would have the greatest benefit to safety.  Combining performance indicators and inspection results, the Reactor Oversight Process (ROP) is widely regarded as far superior to the largely subjective and non-public method it replaced. 

Over its 15 year life, the ROP has evolved.  As NRC and industry learned from experience, adjustments were made in various features of the ROP to ensure the program continued to meet its objectives and adhere to its guiding principles.  Two years ago, at the Commissioners' direction, the staff undertook an independent review of the ROP that solicited feedback from stakeholders within and outside the NRC.  For the past year, the NRC has been considering the recommendations from that review plus others from a separate internal review.  One aspect of the ROP that internal and external stakeholders both mentioned was a concern about the Action Matrix.

The Action Matrix is a key feature of the ROP. The title "Action Matrix" refers to a decision-making guide published in the NRC's ROP instruction document.  The "matrix" is a table of criteria that determine where a plant falls in the scale of NRC responses.  The "action" refers to sets of planned responses NRC will take unless unique circumstances suggest a different response is more appropriate.

At the highest level of performance (Column 1 of the Action Matrix), the NRC grades all inputs to the Action Matrix (performance indicators and inspection findings) as Green.  At this level, a nuclear plant is assigned the NRC's lowest level of inspection, the so-called Baseline Inspection Program.  All operating nuclear plants are subject to the Baseline Inspection Program.

Just below that level of performance (Column 2 of the Action Matrix), a lapse in a single performance indicator or one inspection finding is judged to be of slightly greater significance (labeled "White").  This White input to the Action Matrix triggers an additional inspection focused just on that particular change in performance.

To be put into the third tier (Column 3 of the Action Matrix), a plant must experience a lapse in two performance indicators or two inspection findings that are judged to be of White significance. A plant could also be placed in Column 3 because of a single lapse judged to be of even greater ("Yellow") risk significance.  When a plant is put into Column 3, the NRC will mount a large inspection effort to understand how well the owner has investigated the problems evident in the White or Yellow inputs, corrected them, and shown positive results from the corrective actions taken.

Finally, below that level of performance (Column 4), if the lapse in performance that put the plant into Column 3 persists more than about a year or spreads to other areas, the plant is considered to have multiple or repetitive "degraded cornerstones".  A major recovery program will be launched by the owner and an intensive independent inspection by the NRC will probe the design and results of the recovery program.

In late August, the NRC staff proposed to raise the threshold for entry into Column 3 from a lapse in two indicators or inspection findings to three.  The NRC's proposal includes a thorough analysis of past performance of plants put into Column 3 that shows the use of three lapses rather than two is a better indicator of what was originally intended to put plants into this level of added oversight.  In addition, the proposal airs several arguments against the proposal raised by members of the staff who oppose the change.  As presented in the staff proposal document, the arguments against it are qualitative, and offered without substantiation or objective evidence.  This is not to say the arguments lack merit; merely that we do not know what the empirical basis for the objections might be.

The NRC's distinguished Advisory Committee on Reactor Safeguards (ACRS) recently offered their opinion on the staff's proposal.  The role of the ACRS is to provide independent, expert advice to the Commission on a wide range of technical issues.  Their October 16 letter to NRC Chairman Stephen G. Burns summarizes information on the proposal they gleaned from staff briefings held in September and early October.  Their letter further explains the technical basis for the change in Column 3 proposed by the staff and offers their support for the change.

The industry supports the staff proposal.  This recalibration of the threshold for entry into Column 3 is in keeping with the sense of the ROP founders 15 years ago about the level of performance issues that should trigger the additional inspection effort identified with that column.  Our own analysis suggests that resetting the threshold for Column 3 will have no adverse effect on safety or on the timeliness of NRC response to changes in plant performance.  In light of the supportive ACRS letter and other considerations, we remain optimistic that the Commission will soon endorse the staff's proposal.

No comments: