---

Technidigm-2000
Level One Opinions

Comments on the
Revised Reactor Oversight Process
Pilot Plant Workshop
January 2000

A lot of work is still required to get the new oversight program started up at all the nuclear plants in April 2000, and the NRC and industry appear to be determined to make that happen.  Below are some highlights of the workshop and comments on a few areas that will likely need more attention, at least from my perspective.

The purpose of the workshop was to review the results of the 13 plant pilot program and determine whether program revisions are needed prior to full implementation at all 103 plants.   During the workshop, the various issues were collected and discussed largely to determine whether they needed to be accomplished prior to the first week in April 2000, the hoped-for start time for full implementation.  There was no shortage of issues, and many of them were ear marked for completion before April.

- Charles R. Jones
  Germantown, MD
  Nuclear Safety Engineer
  cjones@Technidigm.org

The new regulatory oversight framework is dependent largely on the good will and cooperation of the various stakeholders, primarily the licensees and the NRC.  They must work closely together more than ever in dealing with a myriad of complex and transitory conditions and issues important to nuclear plant safety and, thus, important to public safety.   The licensees are, of course, the stakeholders with the most direct access to the inside knowledge needed to make the new regulatory oversight process a success.

This licensee advantage is really no different than under the previous regulatory approach, but it is now increasingly significant under this new performance indicator and threshold issue approach.   The nuclear safety culture within the NRC and within the industry has become critical, but the "safety conscious work place" was sidelined during this workshop as a cross-cutting issue to be dealt with later, like all the other cross-cutting issues.  This is somewhat alarming because, while most people understand that the so-called cross-cutting issues are able to provide early indicators of plant problems, the new oversight process has been unable to come to grips with them.  There are no thresholds identified for cross-cutting program failures, and there is no proven method that can painlessly allow cross-cutting areas to be included.

This is why I recommended a new, more creative approach in late 1998, an approach that I named "Delta Scores."  At the time I thought that it would likely be viewed as being too simple rather than elegant and creative.  Since it was NIH (not-invented-here), I am sure that the NRC simply ignored it, which is the usual reaction of most government agencies to outside help.  Keep in mind that anyone at the NRC who might have the time to study and understand such outside help would most likely not have sufficient power to accept or implement it.  With no insider ready to champion it, no such outside suggestions have any hope of being dealt with in an adequate manner.

Even with such a champion within middle management and even when paid for with many thousands of tax dollars, creative methods are inhibited by a lack of understanding or competence.  An example of this is the $350k MPPA inspection program that SCIENTECH, Inc. produced for the Department of Energy several years ago.  Everyone who studied it agreed that it was a great approach, but no one was prepared to even do a trial run to test the methods.  Perhaps it was too difficult to explain its attributes to management, and perhaps it did not take long enough to develop.  Since I was the primary creative force behind that effort, too, I made that project's documentation available to everybody by putting it on the internet.  It addresses five key nuclear facility  functional areas, plus identifies the possible management root causes for problems found in those functional areas.  It is a highly integrated team inspection methodology that maximizes the efficiency of an inspection team.

The Delta Scores and MPPA oversight and inspection processes are actually complementary, one providing the inspection method and the other providing the reporting or assessment method.  Importantly, they focus on identifying early indicators of problems while also accommodating emergent issues and their root causes.   While the new regulatory oversight process is well conceived in terms of safety issue coverage and focus on levels of importance, it is also turning out to be cumbersome, largely time-late, and diverts human resources from dealing with problems to dealing with the process.  Its primary attribute is that it is consensus based, which means it is a series of compromises agreed to among the various stakeholders, often based on raise-your-hands votes by the people who showed up at a workshop.

It is yet to be seen whether a poorly performing "red plant" will be able to slip through this largely reactive regulatory oversight process as a "green plant."   I would not be very surprised if this were to happen, but it could be too late to prevent an accident, almost by definition.   The ability of performance indicators to intercept a plant afflicted by a poor safety culture that is heading toward meltdown seems to be wishful thinking rather than logical thinking.   This is why commissioners voiced concerns two years ago regarding the performance indicator approach being a time-late approach.

It only takes one nuclear plant to have the wrong culture to cause the new regulatory oversight process to fail.  More importantly, process failure can not be tolerated since it means that public safety would be jeopardized.  Perhaps the weakest part of the new regulatory process is its time-late basis that results from the reliance on performance indicators and avoids "looking under the hood" until after problems are evident.  In the deregulated energy market, nuclear plants are perceived as being less profitable than fossil plants.  It does not take a very smart mechanic to tell you that you have a problem after your automobile crashes.  The real question is, "Who is checking the gas gage, the tire wear, the brakes, and the exhaust system as these 103 licensees speed along their life cycles, not unlike limousine drivers trying to make a buck competing with taxi drivers?"  At this point, no one is prepared to address cross-cutting programs and their issues.

The overall process is a "time-late" process, one where we start to find problems and then start looking for their causes.  We also assume that the spectrum of performance indicators and thresholds of regulatory interest are adequate such that a single event at a "green" plant would not go directly to the "red zone" or result in a significant radiological release.  The NRC is largely trusting the licensees to have good training programs, not to suppress workers' concerns, and to provide complete and objective performance indicator information.  This trust represents a vulnerability, but it is also misplaced to the extent that neither the NRC nor the licensees have come to grips with fundamentals such as nuclear safety culture.  They can articulate the problem but seem unable to deal with it, so it is relegated to the list of things to be worked on later, an obscure list without the advantage of having the kind of real corrective action program expected of the licensees.  Every NRC-responsible issue should be the subject of a Condition Report, just like the now accepted and expected process found at most nuclear plants.

As an independent consultant on nuclear safety culture issues and many years of nuclear inspection experience, I am of course far more aware of the problem and how to correct it than are most people.  Recognizing this opportunity to be part of the solution, I am pursuing the development of a simple but effective training program on nuclear safety culture, much of which I have already placed on the internet for general use for all nuclear facility workers.   <http://Technidigm.org/Technuke/nuclear.htm>

While there is significant evidence that plant performance has improved since the mid-1980s, the progress on industry and NRC management's ability to deal successfully and appropriately with the public and with internal disagreements remains questionable, as was seen during this workshop.  I made this perspective well known during the workshop, and the responsible managers seem to be interested in working more diligently on the problem.   It is just very difficult for government and industry managers to overcome the many years of inertia that form the basis for this weak area.  They are at least listening, if not hearing and understanding.  They may even be hearing and understanding, but the evidence of any real change in the attitude presented to the public must be created over time.

Nevertheless, since this was only a workshop, the NRC management representatives made it clear that they retained the final say on what needed to be done and when.   The workshop recommendations would be given consideration, although there was no promise that deviations would be documented or justified.   This had the unfortunate effect of making NRC managers appear to be a little out of touch with the stakeholder concept.  Moreover, the NRC management culture continues to avoid the fact that NRC managers are public servants and need to be much more vigorous in dealing with public concerns and recommendations.

During one of the breakout sessions, all cross-cutting issues and related concerns were earmarked for consideration by a followup panel, a panel that did not yet exist.  This deferral of key issues reflected the inability of the workshop to deal with any significant issues, but it also set up conditions under which those key issues could be downplayed even further or just ignored.  Interestingly, it appears that NRC managers believe that they do not need a followup panel, and they will likely deal with the deferred issues without any such panel.

Logic would indicate that, if NRC managers do not need any help with the most difficult issues, they really should not need help with all of the easy ones that were actually dealt with at the workshop.  The workshop was little more than an extended brainstorming session, one where the NRC could cherry pick the ideas that it liked and ignore the rest.  There is a continuing need for the NRC to come to grips with how to identify poorly performing plants early, which was the overall purpose of the proposed panel.  The new oversight process is very good at identifying plants that have problems (after the fact), but not very good at going after root causes and not very good at avoiding those causes in the first place.  The only path forward to deal with these problems does not appear on the NRC road map.

Perhaps the most disheartening treatment of public recommendations involved those made by the primary "public" representatives, those from the State of New Jersey.  The many recommendations (more than 20) made by the State of New Jersey were treated as trivial and were scheduled to be dealt with "in a couple of weeks" by NRC managers.  Also, the results of "roundtable" discussions that the NRC has held in the public sector to date as well as other public recommendations requested and received by the NRC had not been collected and dealt with in any recognizable form during the workshop.

To make sure that I was not missing something, I made several pointed comments on these topics during the breakout sessions.   While my more vigorous comments were at least received with good humor, it was apparent that the NRC culture still needs a lot of work in the area of public stakeholder input and handling.  One response that I received was, essentially, that responding to all the publicly raised concerns, recommendations, and issues would be "too hard."  Nothing else was too hard, but that was.

I would submit that responding to the public is expected to be difficult, but it is central to the NRC's stated objectives of making the new process more understandable by the public and communicating effectively with the public.  When we dealt with similar issues for the Department of Energy, we learned that vigorous involvement with the public and treating their representatives as equal players was the key to credibility.  The NRC recognizes the public as stakeholders and players, but still has great difficulty accepting that public stakeholders should be treated as equals.  I could put on my Constitution hat and point out that the NRC works for the public, making the public representatives more than equals, but most readers are not willing to accept such a concept, again out of habit.

The State of New Jersey presented more than 20 recommendations at the beginning of the workshop, most of which were not added to the agenda for specific discussion.  Indeed, even though the NRC managers conferred upon the New Jersey representatives the distinction of representing the general public, they did not appear to know what to do about it.  If the culture had been right, the NRC would have not only identified the specific recommendations being made and asked for clarification, they would have made them priority business for the workshop.

Ideally, such recommendations should have been identified and dealt with long before the workshop even began, but failing that, the workshop should have dealt with them.  While this could easily have derailed the workshop agenda, such is the price for failing to deal with such issues in a timely and orderly manner.  It does become too hard, so it does lead to inadequate regulatory oversight from the public's perspective.  From the NRC's perspective, of course, they are bending over backwards relative to their accustomed approaches, so NRC managers are not even aware when they have insulted their public stakeholders.  Fortunately, the public has become accustomed to being politely mistreated, so they do not really insist on action regarding their concerns.  They just insist on being heard, then leave with the feeling that they may have wasted their time.

Overall, the NRC has made significant improvements in dealing with public concerns, but there still seems to be a disconnect between hearing them and properly pursuing them.

Perhaps one of the most obvious regulatory oversight gaps that was apparent during the workshop was the avoidance of root causes.   The NRC continued to put off or evade dealing with root cause issues since they ultimately reach into plant management and programmatic issues.  While some attention is being given to licensee corrective action programs, the emphasis seems to be limited to the condition report generation and the issue management process.  The more complex and useful task of identifying and dealing with root causes still eludes both the NRC and the industry.

One of the fundamental root causes that is most often avoided is the resource issue.  Nuclear related facilities are expensive to design, license, build, test, operate, maintain, decontaminate, and decommission.  The most pervasive and repetitive failure in the nuclear industry is the failure to provide the needed resources to do the job right the first time.  There is usually plenty of money available once a plant is shutdown and the operational mission or profit goals are not met.  Absent such an emergency, there seems to be a continuing and overwhelming need on the part of some licensees to gradually reduce program resources until forced to do otherwise.  The new regulatory oversight process needs to be adjusted to meet this challenge if it hopes to be better than the previous approach.

Several years ago I was investigating why a plant's safety systems and auxiliary systems were no longer reliable.  Going to the heart of the problem, I discussed the maintenance of nuclear safety systems and auxiliary systems with the maintenance manager.  He told me that plant senior managers had reduced the maintenance budget ten percent each year for seven years.  He knew what should be done to guarantee future operational reliability, but senior managers only saw current operational reliability.  The cost cutting senior managers were praised for saving money and promoted to even more senior positions.  By the time I was called in to look for the root causes for plant shutdown, the damage had been done.  It was probably the shortest root cause investigation in history, taking only ten minutes to figure out.

There are some related issues that need to be stated:

(1)  The resource root cause is often cyclic, cutting back on maintenance budgets until good operations become bad, then pouring many extra millions of dollars into the plant to catch up, while losing additional millions due to being off line.  This phenomenon has repeated itself many times in the nuclear industry.

(2)  The reason that resource root causes can exist in the first place is due to the absence of an adequate nuclear safety culture that ensures managers meet the needs of the workers (vice workers meeting the needs of managers).  Managers too often assume that, based mostly on their superior position in the organization, that they are able to manage the technology as well as the people.  Frequently, managers do not understand the technology at the level needed to make good decisions about it.  Yet, they have the power to cripple the supervisors and workers who do understand the technology.  If the safety culture is not adequate, management causes plant shutdowns by inappropriate decisions years before, and the workers get blamed for it when the effects of mismanagement become evident.

(3)  Many industry and NRC managers confuse nuclear safety culture with a "safety conscious work environment," making it less likely that the more global issue of nuclear safety culture will be addressed.  There has been so much focus on the symptom of whistle blowing and retaliation against whistle blowers that some industry and NRC managers have a very narrow view and understanding of nuclear safety culture.

(4)  The NRC, as the regulator, needs to develop its ability to identify and deal with nuclear safety culture related root causes, including a range of management failures that continue to haunt the industry.  Perhaps some of this needs to be done internally, a task that exceeds the level of effort applied previously in the NRC culture area, which were mostly organizational happiness surveys.  While such limited efforts can yield some indicators of the culture problem, they do not include identifying and implementing corrective actions.  Positive culture changes depend largely on leadership, attitudes, principles, and consistency across the organization.

(5)  The nuclear industry needs to define the essential elements of its own nuclear safety culture and find a way to create and maintain those essential elements in each and every nuclear organization.  These essentials must include not only the unique organizational and management elements needed at a nuclear facility, but also the unique technical elements that enable safe operations but are seldom created and maintained as essential.  In my experience, only about ten percent of the US commercial nuclear plants are favorably comparable to the Navy program on the technical elements even though there is a lot of interest in hiring former Navy nuclear people.

Under the previous oversight process, complexity resulted from having many requirements and many inspections, with many opinions regarding what was or was not a problem.  Everything was fair game since there were no thresholds for NRC involvement.  The industry as well as the NRC tolerated the previous oversight process since there was no available alternative.  Indeed, the process was more haphazard than organized in many ways.

With the development of the new regulatory oversight process, complexity has taken on a new meaning.  Each performance indicator must be defined for each plant, and then each occurrence or issue has to be assessed in terms of the associated threshold.  The discussions of these basics in the workshop were complicated by further issues such as extended shutdown considerations, and whether inspectors should avoid documenting positive findings that might mitigate the negative findings.

During the workshop, several of the performance indicators were discussed in some detail and in very fundamental terms.  The performance indicator breakout session addressed at least 9 general issues and several specific issues under most of the general issues.  General issue 1 was "Inadequate guidance for some PI definitions," which included at least six specific performance indicators that fell into this category.  General issue 2 was "Some PI thresholds may not be set properly," which included at least five specific performance indicators, only two of which were the same as discussed under general issue 1.  General issue 3 was "PI Program Guidance lacking in some areas," covering specific concerns such as (1) when a PI is invalid, (2) extended shutdown considerations, and (3) formal processes to add or modify a performance indicator.  Other general issues addressed concerns such as meaningfulness, definitions, how to treat historical data, the short reporting period, resolving interpretation issues, and how to deal with single versus multi-unit sites.

There was also some discussion of the minimum thresholds for documenting inspection findings.  A continuing inspection program is required since not every important aspect of nuclear safety can be captured in performance indicator numbers.  The process of dealing with inspection findings was provided in a logic chart with about half a dozen decision points.  Completing the logic chart simply determines whether a finding needs to be documented at all, not the actual severity level or impact on the color coding process.  The complexity of this process is signaled to some degree by the chart "explanation" which starts with "Documenting violations that are greater than minor:  SDP type issues would be evaluated with the SDP.  Non-SDP issues would be assigned a severity level."  Included in the process is "Minor violation decision block:  If the issue is a violation, it should be compared to the guidance for minor violations."  At least seven questions are suggested for use in determining an issue's impact on the safety cornerstones, similar to a 10CFR50.59 screening list.  Also considered as part of the inspection finding minimum threshold for documenting are extenuating circumstances, the possibility that the finding is an event precursor, and whether the problem might get worse over time.

The problem with such screening logic diagrams is that they are changing the inspection process into an assessment process.  An inspector must now have the ability to find the problem (inspect) and be able to evaluate a range of potential safety impacts in areas where the inspector may have no expertise.  Rather than admit not knowing how to answers to questions such as "Can this issue reduce the plant's protection to the design basis threat of radiological sabotage?", the inspector might simply back off, the inspector might guess, or the inspector might have to spend extra time doing research.  It is likely that each inspection team will need to have an expert on dealing with findings.  Fortunately, it is apparent that the first line of issue expertise under the new regulatory oversight program may be the resident inspectors.  Perhaps team inspections will have to work a lot closer with the resident inspectors if this is to be an efficient process.

The workshop participants diligently pursued each and every concern, much to their credit.  Nevertheless, it was clear that the new regulatory oversight process involves some very complex considerations at this point, many of which will continue to be very plant specific and somewhat susceptible to debate at each step of the performance indicator process.  With additional experience in the process, most of this complexity should settle out and become more manageable.  Nevertheless, the NRC and the industry face some fairly daunting, complex work as they proceed toward the April 2000 planned implementation for all licensees.

I should point out that I recommended a simpler regulatory oversight system to the NRC more than a year ago.  My approach would eliminate much of the complexity by focusing on the health of necessary programs as evidenced by emerging issues or findings.  It would also minimize argument and would allow the licensees to take credit for their good points as well as be criticized for their bad points or issues.  The new regulatory oversight process is much more complex and requires much more intelligence, experience, knowledge, and effort to establish and maintain than either the previous system or the simpler system that I recommended to replace the previous system.

The benefit of all this from the licensee's perspective is that the plants may be able to avoid getting embroiled in oversight complexities, including additional inspections, if they hire additional people who are able to sort out the complexities and deal with the NRC's conclusions before they are reached.   This is also the downside of the process.  Rather than working on actual nuclear safety issues and correcting them, the plants will need to divert at least some highly skilled resources to dealing with the oversight process.

The same could be said of the NRC oversight personnel since they still have to deal with a myriad of plant designs, organizations, issues, and refueling cycle status.  The NRC efforts will be more focused, but they may also require some adjustments in staffing and level of effort to be successful.  As a nuclear safety consultant, I was pleased to hear during the workshop that it is no longer a given that the new regulatory oversight process will allow the NRC to reduce the involvement of contractors in carrying out its inspections.  I would not be surprised to hear the same from some of the licensees.

One of the most interesting aspects of the January 2000 Pilot Plant Workshop was the interaction of the NRC with the State of New Jersey "public" contingent, lead by Dr. Jill Lipoti.   She provided a very interesting and challenging presentation on the first day of the workshop, using 24 slides to raise a comparable number of concerns and issues.  Although representing a state government and having been involved with the development of the new regulatory oversight process, the concerns and issues raised by this stakeholder were relegated to the back burner, at least from my perspective.  The NRC workshop managers expected the State of New Jersey to "defend" its concerns and issues rather than taking them aboard as central issues to be addressed even without the continued presence of this presenter.

I counted 29 potential topics raised by Dr. Lipoti and attempted to followup on them in the later breakout sessions.  She also listed a number of additional weaknesses that she perceives as needing to be addressed.  Each of the points that she made needs to be formally addressed.  Someone in authority at the NRC needs to take personal responsibility for the disposition of each of her points, much like what the NRC expects the licensees to do in response to Condition Reports generated by plant workers.

The primary feedback from the NRC during the work shop was that these issues would be considered separately in a couple of weeks.   The list of concerns raised by the State of New Jersey included things such a the need to have a second pilot plant program to further define and refine the process.  This key suggestion was subsequently inadvertently validated by the workshop results, by the way, but as far as I can tell it was never directly discussed further during the workshop and may never be taken seriously.

Overall, Dr. Lipoti seemed to understand the essential vulnerabilities of the new process better than most of the other workshop participants.  Within the context of my own technical decision making framework that I have named Technidigm-2000, Dr. Lipoti would be a Level Four participant in the generation of a good solution for the new regulatory oversight process.  That is, given that the NRC, NEI, and industry participants are inherently polarized and somewhat political in some areas, it would be useful to have key decisions made by such an outsider.  In the case of the new regulatory oversight program, this would not only go a long way toward improving the process, it would also solidify the NRC's public credibility objective by using the good offices of a third party participant.  Again, I do not concur with the NRC view of the State of New Jersey as representing the public, but state governments at least have the advantage of being as close to a third party as we are likely to get.

For the record, below are the 29 questions and points that I found in Dr. Lipoti's presentation, based on my notes and my understanding of the substance of the points.  I have taken the liberty of adding my own comments to the first seven of her points, hopefully such that the reader will be motivated to consider the other 22 questions and points to be equally interesting and relevant to the discussion, a discussion that is definitely ongoing and not quite as complete as some stakeholders might hope.
 

1.  Is objectivity going to be lost in negotiations between the NRC and the licensees?  The point here seems to be that Dr. Lipoti is concerned that negotiating definitions and threshold crossing points between NEI and the NRC in the field will result in differing threshold criteria for each plant.  While the intent is to avoid this to the extent feasible, the responsibility for avoiding it does not appear to be clear.   Moreover, even less constrained by the process are the negotiations that are likely to take place regarding inspection findings (discussed above).   Perhaps the response to this concern is that the process is as objective as it is likely to be.   The process might be more objective in such areas with a third party mediator or judge, perhaps using a knowledgeable state official or other independent entity.

2.  Is the significance determination process (SDP) too complex?  The SDP is obviously very complex, but it is difficult to determine whether it is "too" complex without further evidence or consequences.  I have already noted that the new regulatory process will require the involvement of some very knowledgeable and experienced people to deal with these complexities.  Thus, I would maintain that the answer to this question has to be created within the context of the available talent and the continuing development and maintenance of that talent throughout the NRC and other the stakeholders who might be involved in understanding the SDP.  Understanding the SDP is potentially central to public confidence in the new process, so it needs to be as scrutable as possible.  It is apparent that this is a continuing issue that does not lend itself to an easy solution.

3.  What is the validity of the underlying risk assumptions and conclusions?  This has always been a concern for me also.  As shown during the post-work-shop discussion on the finer points of the risk determination process, this is not a very scrutable area of the new regulatory oversight process.  I tested this perspective at the end of that discussion by commenting, "Now that we have addressed the scrutable part of the process...".  I paused to see the response from the audience, which was a general laughter, indicating that they agreed that the discussion had been quite complex and not particularly scrutable.   Nevertheless, the question remains as to the validity of this not-so-scrutable process.  The underlying data used for risk determinations has always been pretty limited, and the translation of that data to plant-specific applications by risk mathematicians is not very comforting to me at least.  Even when these insights are translated and applied through the considerable plant-specific knowledge of a qualified plant SRO, perfection is not assured.  I have experience applying such SRO-informed risk considerations for 150 physical plant issues at a Watch List plant, and found that even that input was periodically questionable.  Thus, some risk conclusions need to be reached by experienced and knowledgeable people, which means all such conclusions need such attention.  The new regulatory oversight process needs to include people who are certified, trusted agents in risk applications as well as risk calculations.  I do not know at this point whether this has been addressed.

4.  Do colors add burden due to undesired debates on colors?  This question reaches to the heart of several other points already made.  Dr. Lipoti was apparently pointing out the possible need to just use the actual numerical risk values rather than combining counts of risk colors to get away from the intricacies and potential inconsistencies of converting color blocks across thresholds.  The answer to the question is probably that the color process adds burden and, if done properly, removes some of the mathematical complexity so that plant status becomes more apparent to the general public.   The NRC could respond that this is an acceptable trade off needed to meet somewhat conflicting goals of being objective and scrutable.  Again, the added complexity seems to be acceptable to the NRC and to the industry.

5.  Will utilities take corrective actions if no thresholds are threatened or will they just add to the backlog list?  It is apparent from the work shop that the NRC is struggling with how to assess corrective action programs, and it is likely that this concern will be addressed in that context.  My experience is that the NRC is not very good at inspecting and assessing corrective action programs, and their inspectors have been reluctant to take such programmatic (cross-cutting) issues on due to the thankless complexity involved.  Just like it is difficult to inspect and assess a plant's training program, the NRC is apparently currently planning to deal with corrective action programs reactively.   It is possible to look at the results of such programs without getting involved with their underlying substance.  This imposes a time-late consequence that the NRC seems to be ready to embrace as the path  of least resistance.  Again, programmatic or functional area issues would more easily be dealt with using the mutual-disagreement strategy that I described in Delta Scores.  As I described in that document, the licensees should be tasked to state on a quantitative scale how good they are in their various functional areas, and the NRC should look for evidence to the contrary.  This would result in the licensees taking more responsibility for such programs while allowing the NRC to evaluate both the programs and the apparent ability of the licensees to self evaluate and present accurate programmatic assessments.   The answer to the question is that some plants would be good at dealing with issues on their backlogs and others will just allow the backlog to expand until forced to deal with it.

6.  Are all of the applicable strategic performance areas addressed?  Given the extraordinary level of effort that has gone into developing the new process, I think we would all be surprised if any additional strategic performance areas could be identified.  Some of them might have been discarded along the way, and some are cross-cutting areas currently on the back burner, but the identification process has been rather extensive.  If there are other such areas, they should be presented and argued.

7.  Should economic performance be included in the oversight process?  This question gets to some of the points I raised above about cost cutting actions that have historically frustrated continued good plant performance, especially for safety and auxiliary systems.   Dr. Lipoti suggested that it might be appropriate to have criteria for the minimum number of employees needed for safe plant operation and maintenance and any similar resource criteria.  It would seem to be a relatively trivial task to identify the necessary personnel resources (numbers and credentials) needed at each plant.   Such criteria would eliminate the temptation to cut resources during periods of good performance, often using arbitrary budget cut percentages, which has historical initiated a performance nose dive that only became apparent years later.  Thus, the likely very unpopular answer to the question is yes, plant budget allocations and changes need to be watched and justified systematically.  This seems to be an optional consideration under the new oversight process even though it is central to the root causes of inconsistent performance at more than a few nuclear plants.
 

8.  The plant performance information for each pilot plant should be stated and compared with the results of the pilot.

9.  Why should the public be comfortable with fewer inspectors?

10.  How will the public adopt or figure out risk?  Is there a strategy or mechanism?

11.  Are performance indicators set in stone already?

12.  Absolute risk values should be reported and trended.

13.  What do the performance results look like when presented in more familiar graphical terms such as using a bell curve?

14.  Is the plant shutdown decision matrix adequate?

15.  The NRC should select about 12 more plants and do another pilot program.  Full implementation in April 2000 seems premature.

16.  The NRC should collect and assess more data before committing to irrevocable decisions.

17.  The oversight process needs to be improved such that the plant and NRC culture allows concerns to be raised.

18.  How many non-utility / non-NRC people attended the workshop?

19.  Where are the inputs from commenters and the NRC roundtables?

20.  Some indicators are grouped together since some sites have multiple units.  What are the impacts of this on the oversight process, such as in the area of emergency planning?

21.  The new regulatory process seems to require the use of more experienced inspectors, although the NRC appears to be determined to use less experienced people than in the past.

22.  Were the performance indicator results from the pilot program related to plant performance or not?  Was the pilot plant program only used to prove that the process can be executed while ignoring the safety validity of the approach?

23.  If all the plants in the country are in the green zone, does this mean that the new process is a success?

24.  Will green plants tend to lose attention to detail due to being in the green zone?

25.  Will utilities learn to "manage" the indicators and the culture such that problems are obscured while not violating the letter of the process?

26.  What about the plants that do not volunteer to subject themselves to the new regulatory oversight process?

27.  From a statistical perspective, how does the NRC justify the pilot plant sample size and the performance indicator sample sizes?

28.  Is 100 hours adequate for the baseline inspection process?

29.  What are the root causes of the the pilot plant findings?  Was the SDP successful in the pilot program, and did the SDP address and incorporate the generic importance of identified root causes?

Go to the Technidigm.org home page.