ML20214H922
| ML20214H922 | |
| Person / Time | |
|---|---|
| Issue date: | 11/20/1986 |
| From: | NRC COMMISSION (OCM) |
| To: | |
| References | |
| REF-10CFR9.7 NUDOCS 8612010093 | |
| Download: ML20214H922 (66) | |
Text
'
l- ORIGINAL UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION f
In the matter of:
COMMISSION MEETING ,
(
Briefing on Initiatives to Improve Maintenance Performance (Public Meeting)
Docket No. f
(
Location: Washington, D. C.
Date: Thursday, November 20, 1986 Pages: 1 - 54 8612010093 861120 '
~~ ~
ANN RILEY & ASSOCIATES q Court Reporters 1625 I St., N.W. l Suite 921 i I
Washington, D.C. 20006 (202) 293-3950 .
-s= .
4 1 D i SC LA I M ER 2
3 4
5 6 This is an unofficial transcript of a meeting of the 7 United States Nuclear Regulatory Commission held on 8 11/20/86 . In the Commission's office at 1717 H Street, 9 'N . tJ . , Washington, D.C. The meeting was open to public 10 attendance and observation. This transcript has not been 11 reviewed, corrected, or edited, and it may contain 12 inaccuracies.
13 The transcript is intended solely for general 14 informational purposes. As provided by 10 CFR 9.108, it is 15 not part of the formal or informal record of decision of the I
16 matters discussed. Expressions of opinion in this transcript 17 do not necessarily reflect final determination or beliefs. No l 18 pleading or other paper may be filed with the Commission in 19 any proceeding as the result of or addressed to any statement 20 or argument contained herein, except as the Commission may 21 authorire.
22 .
23 I 24 1
25
. 1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 ---
4 BRIEFING ON INITIATIVES TO IMPROVE MAINTENANCE 5 PERFORMANCE -
6 ---
7 Public Meeting 8
9 THURSDAY, NOVEMBER 20, 1986 10 1717 H Street, N.W.
11 Washington, D.C.
12 13 The Commission met, pursuant to notice, in open 14 session at 10:00 a.m.
15 16 COMMISSIONERS PRESENT:
17 LANDO W. ZECH, JR., Chairman of the Commission 18 THOMAS M. ROBERTS, Commissioner 19 JAMES K. ASSELSTINE, Commissioner i 20 FREDERICK M. BERNTHAL, Commissioner 21 KENNETH M. CARR, Commissioner l
22 l
23 NRC STAFF AND PRESENTERS SEATED AT CGMMISSION TABLE:
24 S. Chilk W. Parker 25 B. Russell R. Vollmer
. 2 1 AUDIENCE SPEAKERS:
2 G. Cwalina 3
4 5
6 7
8 9
10 11 12 13 14 15 -
16 17 18 19 i
, 20 21 22 l
23 i
i 24 25 l
I - . . _ _ _ _ _ _ . _ . _ . - . _ _ _ _ _ .__ __ _ _ . _ _ _ _ _ .
o 3 1 PROCEEDINGS 2 CHAIRMAN ZECH: Good morning. Commissioner Bernthal 3 will be a little late this morning.
4 Today's briefing is an information briefing that has 5 the pur'osep of providing the Commission with the status and 6 plans that the staff has with respect to the maintenance and 7 surveillance program plan. This information will serve the 8 Commission in two specific ways; first, it will add to the 9 information from last week's briefing on initial startup 10 programs in preparation for this afternoon's meeting with 11 NUMARC; and second, it will assist the Commission in its 12 consideration of the proposed staff initiative to develop a 13 Commission policy statement on maintenance.
14 I believe that the execution of an effective 15 maintenance program at all of our power plants is very 16 important to the protection of public health and safety. I 17 also believe it is one of the areas of performance where the 18 licensees have a considerable variance in performance, and 19 that the standard of acceptable performance should be 20 excellence. I believe there is room for improvement across 21 the board. Each licensee should, in my view, recognize that 22 maintenance is of great value to his performance, both of 23 safety and reliability.
- 24. In addition to the proposed staff initiative 25 regarding a maintenance policy, this is an area where I
,,.m - - - - -
,, , .- __y.-_ _ .
. 4 1 believe there is an opportunity for the industry itself to 2 demonstrate that they can raise the level of performance and 3 encourage those who particularly need improvement.
4 Do any of my fellow commissioners have any opening 5 remarks?
6 COMMISSIONER ASSELSTINE: Just a couple, Lando. I 7 agree entirely with your comments on the importance of 8 maintenance. I found the papers that the staff has put 9 together over the past several months on maintenance, both in 10 this country and overseas and most notably in Japan, to be 11 most helpful documents. Quite frankly, I share your concern 12 about the lack of uniformity and the need for overall 13 improvement in maintenance throughout the nuclear power t
i 14 program in this country.
15 I guess I'm a little concerned based upon some of 16 the information in the staff's reports about the lack of 17 progress over the past several years. It doesn't look like
(
! 18 we're much better off today than we were several years ago.
19 In fact, maintenance appears to be an increasingly dominant 20 contributor to the kinds of operating programs we're seeing at 21 the plants.
22 And one of the things the staff told us yesterday 23 was that poor operations and maintenance can increase the 24 likelihood of a serious accident in this country by as much as 25 ten times or more at individual plants. So I guess I'm i
. 5 1 particularly interested in hearing what the staff has to say, 2 and the focus of my concern is going to be can we, in fact, as 3 you suggested, simply rely on the industry to try and bring 4 about the necessary improvement or do we need to do moro 5 ourselves. And I guess I'm reaching the conclusion, based 6 upon the information that I've seen, that maybe what we need 7 is a more effective and active regulatory involvement in the 8 maintenance area, particularly since, as come of the staff 9 documents point out, there's virtually nothing in our 10 regulatory framework now on the subject of maintenance.
11 CHAIRMAN ZECH: Any other comments?
12 (No response.]
13 Mr. Vollmer, will you proceed, then, please.
14 MR. VOLLMER: Thank you, Mr. Chairman. We certainly 15 likewise would strongly second the comments you made on the 16 importance of the maintenance programs. I think as 17 Commissioner Asselstine alluded, the things that we've seen l 18 over the past -- it's been about a year now, we've talked to j 19 Davis-Besse and Rancho Seco and San Onofre events -- all 20 clearly had some strong nexus to maintenance both in terms of I
l 21 inadequate programs, inadequate implementation of the prcgrams i
22 and inadequate procedures; and even to the extent in some F
23 cases, at least one case, where probably the program of 24 procedures followed certainly regulatory and code 25 requirements, but in retrospect even that was not adequate to l
l r
. 6 1 -- I'm talking about of courso San Onofre check valves, and 2 the industry has a separate effort going on those.
3 We had sent down a paper rscommending a maintenance 4 policy statement in SECY 86-316. I think we feel it's 5 important to move forward. I hope we can demonstrate perhaps 6 a fair amount of progress in this area.
7 But I might point out in conjunction with your 8 comments, Commissioner Asselstine, that a lot of this goes to 9 not just maintenance of safety-related equipment, it goes to 10 maintenance of equipment on the secondary side, 11 balance-of-plant, which we find often challenges the primary 12 plant circuit. And in that area, we still have the important 13 safety issue to deal with, i
14 So from a regulatory perspective, if we came down 15 and said I'm going to issue a maintenance rule, I think as we 16 are currently structured it would be difficult to do it except 17 for safety-related systems, and then we would be left a little 18 bit apart. And a good Appendix B program really does cover 19 the safety-related systems. I think it's a broader issue 20 than that.
21 COMMISSIONER ASSELSTINE: I agree with that 22 entirely, and quite fra? Aly I think the two go hand in hand.
23 I agree with you that one thing the operating experience is 24 showing us is that by ignoring the balance of plant, all 25 you're doing is inviting trouble. It just doesn't work to
. 7 i
1 draw this sharp distinction between the safety-related systems 2 and the non-safety related systems and hope that the plant 3 will operate safely and reliably. It just doesn't work. That 4 distinction is outmoded.
5 MR. VOLLMER: Okay, I'd like to turn it over to Bill 6 Russell and let him carry it from here.
7 CHAIRMAN ZECH: Please proceed.
8 MR. RUSSELL: My purpose is to give you a status 9 report on our activities to date with emphasis on Phase I and 10 what the staff's views are concerning industry initiatives 11 that are going on.
12 (Slide.)
,r 13 There are principally two documents that I'm using 14 as source material. The first is the Phase I report which we 15 -
provided a copy of with the Commission paper, and the second 16 is the recent revision to the staff's maintenance and 17 surveillance program plan which addresses the Phase II 18 activities.
19 (Slide.)
20 This program was developed particularly to respond 21 to the Commission's guidance contained in the program of
~
22 policy guidance. The staff chose a phased approach with the 23 first phase being one of assessing current status; that is, 24 taking a snapshot in time of what actually exists in the 25 U.S. industry to establish a baseline for judging whether
4
/ I
. 8 1 indeed there is improvement during subsequent phases as a 2 result of the industry initiatives or not.
3 Two major activities in es'tablishing that baseline 4 were site surveys where the staff got very detailed 5 information on a few sites. The second major activity had to 6 do with the survey of the resident inspectors using an 7 information-gathering tool that tras based upon the site survey 8 protocol, so that we had essent'. ally the same types of 9 information. In one case it was for a large number of plants 10 in less depth; in the other case it was for a few plants in 11 very significant depth, including interviews of people 12 involved in performing maintenance. We found those two were 13 consistent.
{
14 COMMISSIONER ASSELSTINE: Bill, which are the eight 15 ones you looked at in detail?
16 MR. RUSSELL: Let's see, Millstone, Rancho Seco, 17 Davis-Besse, Turkey Point, Brunswick, Kewaunee, Arkancas, and 18 Salem.
19 COMMISSIONER ASSELSTINE: So really, the full 20 spectrum in terms of performance from best to worst.
21 MR. RUSSELL: We intentionally chose the tails of 22 the distribution as we knew it at the time in order to judge, 23 using the same information-gathering scheme, why some 24 utilities' programs were successful and why others were not.
25 And I will get into some of the findings from those in just a
+ 9 1 moment.
2 We have completed our Phase I activities.
3 (Slide.)
4 COMMISSIONER ASSELSTINE: Are you going to talk 5 about your 31 indicators or measurements?
6 MR. RUSSELL: I can. In Phase I, as Commissioner 7 Asselstine indicated, we issued a number of reports describing 8 the status. I have copies here at the desk. The two more 9 significant ones we believe are the NUREG-1212, which is the 10 Phase I report, and NUREG-1192, which was the investigation 11 into contributors of wrong unit / wrong train,'which are'indeed 12 in the operations / maintenance interface. That is the 13 situation where you intend to maintain one piece of equipment 14 and you get to the wrong piece of equipment, either the wrong 15 unit or the wrong train; or you are not adequately restoring i 16 afterward. We investigated those events in detail with staff 17 and issued a report, the principal findings of which were the 18 training procedures and labeling were the dominant . .
19 contributors to events which invo.'.ve wrong unit or wrong 20 train.
21 (Slide.)
22 COMMISSIONER ROBERTS: Say that again, the last 23 sentence.
24 MR. RUSSELL: The dominant contributors to events 25 involving maintenance or surveillance testing on the wrong
4
. 10 1 unit of a two-unit site or the wrong train of equipment on a f
2 single-unit site were trair.ing, procedures and labeling. We 3 feel that there are programs underway addressing the training 4 issue through the policy statement for training and 5 qualifications, and we are seeing those events factored back 6 in.
7 We have looked at the procedures; in particular, the 8 program for development of maintenance procedures, with INPO.
9 I have corresponded with INPO and we feel that that's quite a 10 good program. We have some concern, however, in the labeling 11 area --
12 CHAIRMAN ZECH: I would agree with those. 'I think
- l3 that those are very important. But I would also think -- and
(
14 perhaps you'll ta'lk about it -- management significance, 15 management attention, management involvement. In other words, .
16 these are specific things, and I think that's probably a good 17 analysis, but it seems to me that a dominant contributor in a 18 sense has to be management emphasis.
19 MR. RUSSELL: I would agree. Particularly when you 20 go through a facility and just in making a tour of the 21 facility you see a significant number of labels that are done 22 on tape with magic markers or things like that where you don't 23 have sufficient information on the label to identify the --
24 CHAIRMAN ZECH: Sure, but what I mean is that's 25 indicative of perhaps a larger problem.
4
. 11 1 MR. RUSSELL: That's correct.
2 CHAIRMAN ZECH: And-does management really get 3 involved or do they accept that rather not so' professional way 4 of labeling and procedures training and so forth. So my 5 experience has shown that management emphasis is very 6 significant. And I don't know if you agree with that or not 7 but I'd like to hear your comments as you proceed.
8 MR. RUSSELL: I do agree with that and I believe 9 that lack of attention to detail en the part of management in 10 assuring that there is adequate labeling of all the mechanical 11 components, that there have been problems with training and 12 procedures, that that's part of the cause. What we were 13 trying to do is point out the areas we identified that were 14 weaknesses.
15 CHAIRMAN ZECH: Okay, go ahead.
16 COMMISSIONER ASSELSTINE: On this business about the 17 wrong unit / wrong train, it's sort of hard to comprehend a 18 situation where people can't find the right reactor or can't 19 find the right piece of equipment that they're supposed to be 20 working on. Why isn't that something that should have been l 21 corrected a long time ago? Where the error rate is just
! 22 extremely low or should be extremely low?
23 MR. RUSSELL: That's precisely why we did the 24 study. We went to the 35 sites and actually interviewed the l 25 people involved to find out what were the contributors as to I
I i
. 12 1 how did the individual work on the wrong diesel generator.
/
2 An example is the situation that occurred at 3 Dresden, which is one of the events we locked into, where they 4 wanted to work on the electrohydraulic control system for one 5 of the machines. The individual had previously worked on the 6 other turbine; through error, went to the same one he had 7 worked on previously which happened to be the one on line and 8 not the one he was supposed to go to, and when he isolated 9 that the unit tripped.
10 The report is full of 35 events of that character 11 where the wrong unit or wrong train was involved.
+ 12 COMMISSIONER ASSELSTINE: You talk about procedures, 13 training and labeling and I agree that those are all important 14 steps in terms of turning that kind of a situation around, but 15 does this tell you anything about the attitudes as well? The 16 importance of maintenance, the role that maintenance has been 17 given, and the perception of management in the way that they l 18 view the importance of maintenance in terms of successful l
19 plant operation?
20 MR. RUSSELL: I believe that that is a fair 21 characterization; that's one of the reasons that we've gotten l
involved in maintenance and in the operations / maintenance 22 i 23 interface. Looking to maintenance to come in and fix 24 something that's broken so that you can continue to operate it 25 is not a healthy attitude. Maintenance should be proactive, l
l l
l
\ _ __
O o 13 1 should be in there assuring availability and reliability.
2 Preventive maintenance programs need to be effective, and you 3 shouldn't continue to operate it until it breaks and then fix 4 it.
5 MR. VOLLMER: I think the reactor trip breaker thing 6 that we went through several years ago is another example 7 where maintenance activities seem to be so separate from the 8 operations activity and almost a different organization. And 9 certainly a little teamwork appeared to be existing between 10 the two organizations at some plants.
11 CHAIRMAN ZECH: Let's proceed.
12 ,
MR. RUSSELL: I had mentioned that we did site 13 surveys at eight operating reactors. Prior to those surveys
\
14 we did develop a detailed information-gathering tool which we 15 call the protocol, and that same protocol was used to gather 16 information from the resident inspectors. This was to ensure 17 that we had comparable information for comparative analysis 18 which we did then include in our reports.
19 The wrong unit / wrong train report did involve 35 20 individual events at 10 different operating reactors where we 21 gathered information on site.
22 We did also develop a computerized maintenance i 23 database, over a quarter of a million data elements looking at 24 operations, maintenance and inspection between 1980 and '85. ;
25 We came up with about 31 measures and we developed a '
l l
f
.o 14 1 menu-driven system that can be used with a PC. This was one 2 of the early' activities in the performance indicator work. We 3 have been doing this for about two to three years, and we have 4 provided the information from our early experiences to the 5 overall program on performance indicators and have been 6 dealing with that quite closely.
7 Our trends and patterns report was one of the 8 sources of information which we extracted and put into the 9 NUREG-1212, which was our basis for concluding that while 10 overall performance did appear to be improving, we were not 11 seeing the same improvement in the maintenance area. And we 12 found also that the one indicator that was most significant 13 was the frequency of forced outages caused by equipment 14 failure; how many times did something break and force you off 15 line. That appears to account for the majority of the other 1
16 indicators that we looked at in the maintenance area, and that 17 we felt was a rather significant finding and that's one that 18 we've proposed to include in the overall indicator program 19 which is with the Commission now.
20 In addition, we've had activity going on. NUMARC ha 21 also committed to and developed a set of 10 maintenance ,
22 performance indicators. The staff worked closely with them, 23 provided comments back and forth. We had two task forces that 24 worked on that, and those ten indicators are being used and 25 that information is being made available to the staff to
1 A 15 1 review at INPO on a plant-specific basis, and on a generic 2 basis is being made available to us here.
3 We also felt that this program was one that was very 4 important to make sure that the regions and resident 5 inspectors and the other offices. knew what we were doing and 6 that it was well coordinated, so we put a lot of emphasis on 7 the coordination activities. We actually briefed at resident 8 inspector counterpart meetings what the program was, why they 9 were involved, why we needed their views, and this has been 10 one of the positive feedbacks from an administrative 11 standpoint within the NRC. That is, getting the people at the 12 grass roots level that are out there in the plant seeing it to f 13 understand what we're doing, why we're doing it and to get
\
14 their feedback.
15 We have gone back and closed the loop with them on 16 the close of the Phase I report, and we have also provided 17 them the results back from the questionnaire that we got their 18 input on, and it's been a very healthy exchange and one that l
l 19 we're planning to continue into Phase II.
l 20 We've had a number of information exchange meetings 21 with INPo and with NUMARC, technical exchange meetings on some 22 of the new diagnostic tools that are available to evaluate 23 cquipment performance, thermography, advance vibration j 24 analysin using infrared techniques. These have been very 25 useful.
i
__ - + - - -
- 16 1 And we've seen -- and this is addressed in 2 Mr. Owens' letter from NUMARC to Vic Stello, at the end it 3 identifies some of the new technologies and the extent they're I
4 being used at some plants. And it does indicate that you can 5 predict when maintenance is needed and you can move from a 6 mode of waiting until breaks to doing the proactive approach 7 to maintenance. And we feel that those have been very useful 8 exchange meetinga and we've had large participation on the 9 part of NRC and the utilities in those meetings.
10 We have also, in order to ensure that we have 11 reedback with what's going on, we have had people in the staff 12 participate in regional maintenance inspection activities.
r 13 We've had individuals participate in Trojan, WNP-2, Palisades 14 and others, and so there's a good coordination between the 15 maintenance and surveillance section and the maintenance 16 activities that are going on today as a part of our inspection 17 programs.
18 And that's the scope of what we've been doing for 19 approximately the last year, year and a half with the Phase I 20 activities.
21 I'd like to now shift to the next slide and discuss 22 some of our findings.
23 (Slide.)
24 As I indicated before, we feel that overall plant 25 performance does seem to be improving. We've seen a decline
6
& 17 1 in the average number of forced outages. We have seen 2 improvements in quality assurance and surveillance SALP scores 3 through the SALP process. We've seen a reduction in 4 challenges to safety systems; that is, a reduction in the 5 scram frequency. However, we are not seeing a decline in 6 areas that would relate directly to maintenance; that is, the 7 equipment forced outages.
8 We feel that needed maintenance is not being 9 accomplished or it not being performed effectively at some 10 plants. Sixty-four percent of the forced outage time is due 11 to component failure and for maintenance-related LER's, 48 12 percent of the total for 1985. And we found through our site
~
13 visits and through the resident inspector surveys that 14 maintenance programs are fragmented; they are not well 15 organized and they are not effective at some facilities.
16 And in about 22 percent or between a fifth and a 17 quarter of the plants, the preventive maintenance programs 18 were rated by the resident inspectors as being minimal 19 programs. They were not really effective at preventing l
20 problems.
21 We have a high percentage of failures which have 22 occurred as a result of improper maintenance. Over 30 percent 23 of the abnormal occurrences, which are the most significant 24 events reported to the Congress, have been directly related to 25 maintenance type activities. And this is all since 1975.
l
s 6
i- 18 1 We've seen the proportion of maintenance-related LER's 2 increasing with time.
-3 COMMISSIONER ROBERTS: But now to be accurate and to 4 go to your NUREG-1212, Figure 2.2, it's my understanding you 5 changed the rules on reporting in 1983.
6 MR. RUSSELL: That is correct.
7 COMMISSIONER ROBERTS: So I don't believe you can 8 take a two-year timespan and put a great deal of reliance on a 9 trend for two years. If you look at the previous years, the 10 three-year period, it looks constant.
11 MR. RUSSELL: The change in the LER reporting rule, 12 in 5072, did reduce the number of reports that were coming in
.- 13 so that the denominator of the total reports did indeed change 14 and the threshold for reporting changed. And your observation 15 that reaching a conclusion on two-year timeframe may be 16 inappropriate. It is a fact that that's what the percentage 17 is doing, and we have not relied solely on that one la observation in reaching our conclusions.
19 COMMISSIONER ASSELSTINE: But I think it's fair to l
20 say that when you look at a lot of those other indicators, a 21 lot of thoco other indicators have been constant. Abnormal 22 occurrences, for example; standards for reporting abnormal 23 occurrences have not changed, and the fact that 30 percent of 24 the most serious operating events in the plants can be traced 25 directly to maintenance is a fairly significant factor.
l . - . -_- . . - - - - - --
s s
- 19 ,
1 MR. RUSSELL: The other area that I mentioned
'/
-2 earlier is the maintenance / operations interface. Sixty-five 3 percent of the loss of safety system functional events have 4 been due to human error, and we feel that this is an indicator 5 of problems in the way operations / maintenance interface is 6 being conducted. This is principally the issue of removing a 7 compcnent from service for either a surveillance test or 8 maintenance, conducting that test or maintenance, retesting it 9 and making sure it's operable, and restoring f.t to service and 10 ensuring that the valve lineups afterwards are indeed proper.
11 That's where we feel that significant improvement can be made.
12 This is also the area that relates to the operations
.- 13 personnel being responsible for 75 percent of the human error 1
14 in the wrong unit / wrong train events that we evaluated; the 35 15 events at 10 sites.
16 We feel that the maintenance-related challenges to 17 safety systems is excessive. Using only the 1984-85 data, 75 18 percent of the engineered safeguards feature actuations 19 occurred as a result of a maintenance-related challenge.
20 COMMISSIONER CARR: On the wrong train / wrong unit, 21 what's the other 25 percent? I can't imagine it wouldn't be 22 all personnel related.
23 MR. RUSSELL: There are some that were attributed to 24 procedure problem --
I j 25 COMMISSIONER CARR: Well, that's the personnel who
o Y 20 1 wrote the procedures.
. gm.
'. 2 MR. RUSSELL: Yes, sir.
3 MR. CWALINA: My name is Greg Cwalina, I'm in the 4 Maintenance Surveillance Section. The 75 percent we're 5 talking about is operations personnel who were in the process 6 of preparing the plant for maintenance activities. The other 7 25 percent were actually maintenance people or those type 8 personnel who actually made the error.
9 COMMISSIONER CARR: The 75 percent is the paperwork 10 train, then, and the 25 percent are the actual guys who went 11 out and got the wrong unit.
12 MR. CWALINA: No, the 75 percent are the operations 13 personnel who were involved in the tdgging events, mislabeling 14 or going out and turning the wrong switches. It's operations 15 people who turned those switches to turn them over to the 16 maintenance personnel.
17 The other 25 percent was actually the maintenance
- 18 personnel themselves who went out and worked on the wrong 19 component.
l 20 MR. RUSSELL: If I could give a simple example to 21 illustrate -- if Operations properly tags out and removes 22 component A from service and the work permit is signed to work t
23 on component A but the maintenance personnel goes down and l
24 starts disassembling component B, we would characterize that l 25 as a maintenance personnel error, i
e- 21 1 COMMISSIONER CARR: And that's the 25 percent.
(
2 MR. RUSSELL: Yes, sir. The other would be the 3 maintenance for the Operations Department whether he be an 4 aux operator or a licensed operator in doing a valve lineup or 5 something fails to conduct that properly; that would be 6 characterized as an operations --
7 CHAIRMAN ZECH: It's all people; whether it's 8 operations people or maintenance people. A hundred percent of 9 it is people.
4 10 MR. RUSSELL: Yes, sir.
11 COMMISSIONER CARR: But Commissioner Asselstine's
- 12 problem of how can you get to the wrong place doesn't happen
- 7 13 as often as 75 percent of the time; just 25 percent of the 14 time.
15 COMMISSIONER ASSELSTINE: Yes, in terms of the 16 maintenance guy going to the wrong piece of equipment, yes.
17 The other 75 percent are the operators.
18 COMMISSIONER CARR: Who sent him to the wrong place.
, 19 COMMISSIONER ASSELSTINE: Yes. Doesn't give me a 20 warmer, fuzzy feeling, but it is a better understanding.
21 MR. RUSSELL: What I'd like to de is go to the next 4
22 slide and address what are the industry initiatives that are 5
23 addressing some of these areas, and the staff's views on some 24 of those initiatives.
5 25 (Slide.)
1 i
, . , . - _ . _ , - . - _ - - -- .. ,-r e v e-+ -- -- - + + - -~--~*-~~+='+'--r~ * ' " * ' - ' * - ~ ' " - ' ~ " " - ' " * - " ' ' ~ - ~ ' ' ~ ~ ~ " ' ' ' --" ' ~ ~
. . _ . . ~ - . - - . _ . - - . .-. .. ~ . .. . - . _ .
o
, 22
- 1 The one that's most significant I believe is the 1
f' -
2 first one, and that is the INPO guidelines for the conduct of 3 maintenance at nuclear power stations. This,is a preliminary f
4 document, was issued in October '85. The staff has done a 5 review of that document and compared it to the protocol that i 6 we developed ourselves to ensure that it did indeed cover the 7 important areas of maintenance and the issues that we were 8 seeing. We found that there was very good agreement between i 9 those two.
l 10 We have also started to observe INPO in their plant j 11 evaluations of maintenance, and we had one individual from the 12 staff participate in a recent evaluation at Fort Calhoun. We q 1 13 feel that that approach of how it's being used in the process 14 now as I understand it, is that where INPO identifies a j 15 deficiency in a program, they provide a reference to either a 16 section of the guideline or someone else has a good practice, I
17 and that it's part of the remedial action for problems being l
- 18 found.
19 The second bullet is the one which I think is 20 proactive and is most significant and that is that there are '
- 21 some utilities now who are taking that maintenance guideline l 22 and using it for a self-assessment to judge the quality of f
i
. 23 their own maintenance program. There's a pilot program going I.
l 24 on with a few utilities.
l
- l 25 COMMISSIONER ASSELSTINE: How many are doing that, t
i f.
--.-_m._4 --....__..m._.m._,,,,-- _,.r_.-,w,_,_,,,._,,.,y m, _ _ ,~c,._.,,,_ ,_,_,m,._, ,,,_,,_em.,-~.-.,%., .__mm.__, _,_,c ,,,y-_ _ . , -,, - _e-
. 23 1 and which ones?
b 2 MR. RUSSELL: I don't have the specifics on which 3 ones. I know that Turkey Point is one, for example, which is 4 C.O. Woody's facility and he is the Chairman of the 5 Subcommittee on Maintenance with NUMARC. There are other 6 facilities that are using portions of it. That would be a 7 good question to address this afternoon to NUMARC.
8 COMMISSIONER ASSELSTINE: Okay.
9 MR. RUSSELL: We feel the second bullet is the 10 proactive approach; that is, a self-assessment against those 11 guidelines, whereas reacting to deficiencies identified and 12 found would take longer to achieve the needed improvements in 13 maintenance.
14 COMMISSIONER ASSELSTINE: Why shouldn't they all be 15 doing that?
16- MR. RUSSELL: That's a good question to direct to 17 NUMARC this afternoon.
18 COMMISSIONER ASSELSTINE: I take it your view is 19 they all should be doing it.
20 MR. RUSSELL: That's why we are pushing for a 21 unified industry commitment. You're getting to my bottom line 22 which is two slides away.
23 COMMISSIONER ASSELSTINE: All right, go ahead.
24 MR. RUSSELL: Another very significant feature which 25 we feel is very beneficial to the program is the extension of
,A
24 1 the peer evaluator process to maintenance. We've seen this 2 work in the training area with the use of peer evaluators 3 through both the accreditation review process and it's now 4 being extended to the maintenance program where you're having 5 interchange between utilities. This is the way that one 6 utility can find out how someone else solved the same 7 problem. Very positive feedback on that and one we think is 8 going in the right direction.
9 There's been an increased use of the nuclear plant 10 reliability data system, looking at component failures, 11 analyzing that. This has been discussed at previous 12 Commission meetings, and we feel that this is also significant
\
13 and it gives you information on the components.
14 COMMISSIONER ASSELSTINE: Is there still the kind of 15 lack of uniformity in the use of NPRDS that we've seen in the 16 past?
17 MR. RUSSELL: I'm not able to answer that question 18 directly. That would be one for AEOD. We've seen it from the .
19 standpoint of the products coming out of the data base and how 20 they're being used. We have not addressed through this 21 program the uniformity of reporting and inputting information 22 to the data base. It's my understanding from discussions with 23 Jack Heltemes that that is improving, but that would be a i 24 question to direct to him specifically.
l 25 COMMISSIONER ASSELSTINE: How about on the using
. 25 1 side? Are they using it for anything more than to search for 2 spare parts?
3 MR. RUSSELL: Yes. It's being used now to actually 4- evaluate component performance. It's starting, and it's taken 5 us a while to get there but we are now seeing the first signs 6 that that's.being used.
7 COMMISSIONER ASSELSTINE: Good.
8 MR. RUSSELL: The next bullet is one that the staff 9 is very encouraged about; this is the human performance 10 evaluation system. This involves an on-site coordinator at 11 the utility with a reporting system that's internal to the 12 utility, where that individual goes out in a structured 13 process to interview people involved in human performance 14 problems or events, so that you're_getting feedback back at 15 the lowest level.
16 It's a program that the staff will be having an 17 observer participate in the training and evaluation of the 18 program, and it is increasing in scope. It started out as a
- 19 pilot program with only a few utilities, and it's being
! 20 expanded now to other utilities. And it provides a very good 21 set of tools for evaluating human contribution to events.
22 One, at the lowest level of maintenance things which 23 would not be collected through our LER reporting system; an
! 24 error in maintenance that's made, for instance, by a 25 technician, would be fed back through this system and would go
. 26 1 to the other technicians in his group so that they would 2 understand that and they'd gather the information. It's being 3 done at the working levels within the utilities where the 4 information needs to be fed back, and I think that's very 5 positive and much more important than feeding a data base.
6 Maintenance workshops. This is one where the staff 7 has participated in the past, and in fact we've been requested 8 to again participate in the next one. These are very 9 beneficial. They discuss current issues that are facing a 10 number of utilities. They've had them on diesel engine 11 maintenance activities, on Limitorque valves and other issues 12 as well as maintenance program evaluation. A positive way of 13 providing feedback on common problems that a number of 14 utilities are facing.
15 Training and accreditation for the positions are 16 directly related to maintenance, and instrumentation control, 17 mechanical maintenance. We're seeing improvement in that 18 area. I've addressed that in other Commission meetings and 19 other documents and I won't go into that further.
20 And the final one I mentioned also before and-that r
21 is the NUMARC/INPO activities in developing 10 indicators to 22 judge themselves; the improvement in maintenance / performance 1 23 across their units.
24 (Slide.)
25 Our assessment of those initiatives is that we feel i
- 27 1 that if they are implemented on a consistent basis that they 2 will result in improvements in maintenance programs. We feel 3 that it should be proactive and not reactive, that nn industry ,
4 consensus to a self-evaluation process, similar to what's 5 being done in training, should be developed.
6 Our reason for that is that the variability that 7 exists in maintenance programs today and the variance in 8 performance is such that we don't feel we can wait for a 9 process that is reactive through a plant evaluation assessment 10 program; that there needs to be something done proactively 11 early on. It's for that reason that we feel a policy 12 statement is needed.
13 We think such a policy statement should do four
(
14 things. First, it should specify the Commission's long-term 15 goals for maintenance and the relationship between maintenance 16 and safety, as you, Mr. Chairman, have articulated in your 17 opening comments. That is not done currently in the body of 18 regulatory documents that exist in the Code of Federal 19 Regulations today.
20 You don't find.a programmatic description of 21 maintenance or the importance of maintenance to safety.
l 22 You can find it in Appendix B in the quality assurance l 23 programs with respect to repair of components that have f, 24 failed, to ensure that the quality level is maintained that i
25 you originally started out with and specified at the time of i
4 f
en. ,,,------am
,-_-,------w,--,.,-- ,.- _ -----,--wn_, n- .--_-m.wew - - -, , , - - - - - - v- ,-. ----ww v n -
28 1 licensing of the plant. We don't address it in the context of
~
2 aging, wear-out, using diagnostics; we essentially require 3 that the component be tested. If it passes the test through 4 the surveillance program it's deemed to be acceptable and 5 operable, and we don't address degradation with time, 6 preventive maintenance.
7 We think that such a statement needs to be made and 8 I'm giving the sales pitch for that now. We believe that's i
9 very important.
10 The second thing that a policy statement should do j 11 is it should recognize the self-improvement initiatives. We j 12 feel thet the right stuff is there; it just needs to be 4
13 implemented on a uniform basis and put in place. And this is 14 an area where we feel that the industry and the staff can be '
15 in a win / win situation. Such a program would result in 16 improvements in reliability and therefore improvement in
! 17 performance of the facility, and it would improve safety by 18 minimizing the challenges to safety-related systems and 19 challenges to the plant.
20 The third point is we think we need to articulate 21 the NRC and industry roles in implementation of the policy 22 statement. What is it that industry is assuming i
23 responsibilities for doing and carrying out, what is the 24 staff's role going to be, what are we going to be doing; that l
- 25 should be specified.
I
. 29
'l And fourthly and most importantly, we need to t'
2 specify the'NRC means of oversight. We must have th9 3 mechanisms in place to taka action where we feel maintenance 4- programs are marginal or they are adversely affecting safety, 5 and be able to take action through our enforcement process and 6 using other tools to react at that time, should the outlyers 7 not be moving along with the rest of the industry.
8 We feel that the process of developing such a policy 9 statement would indeed force industry to sign up and make a 10 decision on a uniform commitment, and that's an important 11 action. We feel that that commitment is not yet well defined.
12 (Slide.)
13 I'd like to shift now to what activities we have 14 started in Phase II t' hat are described in the Commission paper 15 that have been approved by Mr. stello.
16 You'll recall that Phase II is in fact the staff's 17 assessment of the effectiveness of the industry initiative.
18 We've given our views on those programs based upon what we've .
19 seen by way of documentation evaluation; we are now looking 20 at it from the standpoint of what's happening at facilities.
21 We are past the point of establishing the baseline.
22 We propose, as a part of the Phase II program, to do 23 additional site surveys using the same informatien-gathering 24 tool at some new sites and to also follow up at plants which 25 we visited in Phase I, particularly those where our evaluation
30 1 indicated that there were indeed major changes underway.
2 Turkey Point, for example, was in the process of reorganizing 3 their maintenance program. 'I mentioned that they are doing a 4 self-evaluation using the maintenance guidelines that have 5 been issued by INPO. We need to go back to that facility 6 later to see how effective those have been, whether it has 7 solved the problem. That's what we intend to do by way of 8 site surveys.
9 Our maintenance measures. We intend to get involved 10 in some validation through site visits such that we can 11 provide correlation between the actual performance of the 12 facility and what we observe. Our concern is that there's 13 such a time lag between maintenance being performed and 14 improving maintenance programs and when components fail or 15 don't, that if we use measures today we may be looking at a 16 situation which existed six months or a year ago. So we feel 17 that this is important, to attempt to validate our 18 observations with respect to forced outages due to equipment .
19 failures, and if those are high, is that directly a result of 20 maintenance program problems or have they modified their 21 program recently and we've just not seen the benefits of that.
22 We need to coordinate this activity with the other 23 indicators, and that is indeed identified in the Commission 24 paper on performance indicators.
25 Under the assessment of industry initiatives, we
- i
. 31 1 need to complete an in-depth review of the various documents 2 that have been provided by industry. We've done that 3 preliminary review; we have not seen it in place being 4 implemented to see how it's being followed. We need to look 5 at the operation of maintenance reminders, the INPO s
6 guidelines.
7 We have already started our observations of INPO 8 through the plant evaluation procesa, and we will be doing 9 some independent evaluations through our site visits.
10 We are going to be interfacing with related programs 11 going on within the NRC. For example, the aging work. We 12 have identified in our program how we will conduct those 13 interfaces to ensure that we are knowledgeable and that we 14 exchange information with the other programs. That's a 15 coordination activity.
16 And we will also coordinate with industry and with 17 standards committees and others so that we keep abreast of 18 what is happening and what is changing. .
19 The two activities that were proposed which we have 20 not started; one is pending a Commission, decision as was 21 described in the commission paper. That's the policy 22 statement. The second one is an evaluation of maintenance 23 activities of c:her industries in the United States, whether 24 that be FAA's approach to maintenance or others. That would 25 be informatien we would develop if we find a need to support a '"
- 32 1 regulatory analysis, should we choose to go to rulemaking or 2 some other activity at the end of Phase II. We have deferred 3 those activities at this point in time, putting emphasis 4 instead upon evaluating the effectiveness of the industry 5 initiatives.
6 Mr. Chairman, that covers the information that I 7 wanted to present on our status. We feel that there are a 8 number of issues that need to be addressed to NUMARC. The 9 staff has discussed the issue of the policy statement with 10 them, and I think it's a fair characterization that they don't ,
11 see the need for a policy statement at this point in time.
12 COMMISSIONER ASSELSTINE: Oh, they don't even want a 13 policy statement.
14 MR. RUSSELL: That's correct, in the letter which 15 Mr. Stello received. We feel that such a policy statement and 16 a commitment and a program which is proactive is indeed 17
- needed. That's the difference in view at this point in time 18 between the staff and NUMARC, and that's why we sent the issue 19 to the Commission for a decision.
20 CHAIRMAN ZECH: Thank you very much. Mr. Vollmor, 21 do you have anything to add before we ask if there are any 22 questions?
23 MR. VOLLMER: No, sir, I think that covers it.
24 CHAIRMAN ZECH: Questions frem my fellow 25 Commissioners?
33 1 COMMISSIONER ASSELSTINE: A couple. The one report 2
you didn't talk about that you prepared is the analysis of the 3 Japanese maintenance programs. Bill, you've been to Japan, 4
I've been to Japan and the Chairman has been to Japan. It 5
seems like the hallmark of their program is heavy emphasis on 6 preventive maintenance. They do extensive prevent maintenance 7
every 12 or 13 months on a specified number of systems and 8
components in the plant, all well defined; all defined by the 9 regulatory body.
10 s The maintenance has to be done, all the reviews of 11 the -- all the maintenance work then has to be documented and 12 that's reviewed by the regulator before the plant is allowed 13
( to go back into operation. Then the plant is only allowed to 14 operate for that fueling cycle, and they have to shut down 15 again and go through the preventive maintenance program again.
16 The operating performance in Japan seems to indicate 17_ that that system really works. You don't see any of the kinds 18 of problems that you've described at the U.S. plants. Do you 19 basically agree with that?
20 MR. RUSSELL: I do agree, and I think that there are 21 some fundamental differences in the Japanese approach versus 22 the approach that's taken in the United States. One of the 23 differences that I think is significant is the amount of work 24 that's done during the time the plant is on line in the 25 amount of removal of components from service.
~. 34 1 Essentially, their approach is to do an in-depth 2' annual-inspection to assure that the condition of the 3 equipment is such that it has a high degree of reliability for 4 operation for the next operating cycle, to ensure that the 5 potential for shutdown is small. And they've been successful -
6 at that.
7 In our case, wa a substantial amount of maintenance 8 while the plant is on linet we remove systems and components, ,
9 we do a substantial amount of testing. The difference in 10 emphasis between testing and preventive maintenance is a very 11 sharp contrast between the United States and the Japanese 12 approaches. Their outages, however, are quite long.
, 13 We have been compared their approaches down to the
\
14 component level for selected systems, there is a proprietary 15 supplement to our report which indeed goes in quite a bit of 16 depth; a comparison of the Tech Spec approach for testing 17 versus the maintenance approach that's required by their 18 ten-year maintenance plans. They are very different systems.
- 19. There are also differences in approach that I would 20 characterize as being somewhat cultural. You see this when 21 you're in Japan with the material condition of the taxicabs, 22 for example.
23 COMMISSIONER ASSELSTINE: Yes. They don't run 24 things until they break.
25 MR. RUSSELL: And that philosophy, which is embodied
. 35 1 in their law, goes to other facilities other than just nuclear 2 power plants. Their thermal plants have the same annual 3 inspection requirements. The approach of preventing a failure 4 rather than accepting a failure is rather a fundamental 5 difference, and I think it's a significant one. There are 6 things that can be learned.
7 Following the staff's evaluation there was an 8 INPO-sponsored group that did go to Japan. There are lessons 9 learned in their reports, some of which are reflected in the B
10 documents in the activities being taken by INPO. Thare are 11 others which are cultural, which we have not been able to 12 address yet.
13 The lifetime employment. If you have an individual 14 that's going to work essential for his entire working career 15 at the same utility in the same types of jobs, you can spend a
.16 lot of time and investment in his training, and not having the 17 turnover. The two-man rule where a second person observes 18 what the first person is doing to minimize the potential for 19 human error, because human error in Japan is scmething which 20 causes a loss of face. They treat that very seriousl; 21 So there are things that are put in place tha: they -
22 do that are different, some of which we can learn from.
23 COMMISSIONER ASSELSTINE: One of the arguments that 24 we hear -- I hear certainly frequently whenever I visit the 25 plants and I'm sure other Commissioners have, too -- is that
4 36 1 in this country we do too much testing with the plant in 2 operation, we do too much of taking equipment in and out of 3 service when the plant is running.
4 It seems to me from what you've just said that that 5 is largely a consequence of our different emphasis; the lack 6 of emphasis on preventive maintenance ahead of time and then 7 the consequent need to constantly verify that the equipment 8 isn't about ready to break when it's in service.
9 Given that concern on the part of the industry, 10 given the heavy contribution to forced outages due to 11 equipment failures in this country, why doesn't it make sense 12 to go to a system that basically requires a heavy emphasis on 13 preventive maintenance, much as the Japanese do, and simply
(
14 require that a specified list of systems and components be 15 properly maintained so that the plant is maintained in a very 16 high material condition every refueling outage, every 18 17 months or so?
18 MR. RUSSELL: I think that philosophically, the 19 Japanese approach and the United States' approach are at two 20 ends of the spectrum. And I believe that there is a concern 21 within Japan now that they may be doing too much during their 22 annual inspection, some of which is not needed.
23 COMMISSIONER ASSELSTINE: Yeah, it's a concern on 24 the industry on that score. The government I think is much i
l 25 more cautious about backing off --
I i
- 37 1 MR. RUSSELL: They're not ready to change a program 2 that's successful.
3 MR. VOLLMER: In fa'ct they are moving toward trying 4 to reduce the outages from 135 down to around 75 days and 5 expand the fuel cycle to 12 to 18 months.
6 COMMISSIONER ASSELSTINE: Some relaxing, that's 7 right.
8 MR. VOLLMER: And some of this type of activity 9 we're talking about would necessarily be cut out if they do 10 that.
11 MR. RUSSELL: But I think more importantly, the 12 things that we're recognizing now and that are indeed 13 discussed in the other policy paper I have now pending before 14 you all which is on technical specifications is that we may 15 indeed have gone too far; that there are things that we've 16 required for surveillance testing that are of lesser 17 importance and that we are indeed looking at reducing some of 18 those testing frequencies. I believe that ue'll be completing 19 our action shortly on the GE proposals. We've done that for 20 Westinghouse. And we need to take a better look at what we're 21 requiring. Too much testing with a base level of human error 22 rate that just increases the frequency of events.
23 COMMISSIONER ASSELSTINE: I agree with that very 24 much, but it does seem to me that the quid pro quo is some 25 degree of assurance that the equipment is going to perform
- 38 1 reliably and that it's going to be maintained in a high 2 quality condition. And the fact is all of the experience 3 seems to show that we don't have that assurance now, even with 4 the level of testing that we're currently doing. And if 5 you're going to relax the amount of testing, it seems to me 6 you have to balance that out with a heavier emphasis on 7 preventive maintenance and some degree of assurance that in 8 fact, all of the utilities are going to perform in that area 9 with the kind of enthusiasm and effectiveness that is 10 necessary to cut down on this experience.
11 MR. RUSSELL: The staff certainly agrees, and we are 12 looking at trade-offs between preventive maintenance and 13 testing. An example, which is one that we discussed at the 14 last exchange meeting -- not the last one, the one before last 15 down at crystal River -- had to do with changing oil in 16 safety-related equipment. We require that periodically the 17 oil be drained and be replaced.
18 It may be better to do a preventive maintenance type ,
19 activity of a surveillance on that oil,to determine whether 20 the oil was still capable of performing its function; take a 21 sample and not remove the conponent from service to drain it, l 22 flush it and refill it, which would be a preferred approach.
- 23 So you may be able to substitute a diagnostic, predictive type l- 24' test for the rigorous preventive maintenance type test, and 25 that may be in fact better than periodically running it
39 1 without knowing what the condition of the oil is.
2 And so those trade-offs need to be looked at. We 3 feel that we need to move cautiously in that area because 4 we're not convinced that we're getting the performance we need 5 today with the systems we have in place today. And those 6 trade-offs are difficult.
7 COMMISSIONER ASSELSTINE: I guess the other broad 8 question I have has to do with the policy statement approach.
9 If I understand what you're saying, you're saying there is a 10 broad range of performance within the U.S. industry, has been 11 for some time, and overall doesn't seem to be getting 12 dramatically better at least in terms of the contribution of s
13 maintenance to equipment failures, forced outages and 14 operations, personnel errors.
15 If that's the case, what makes you believe that
- 16 simply issuing a policy statement which doesn't have any 17 regulatory teeth in it, doesn't require anything, isn't 18 enforceable, is going to bring about the kind of uniform 19 improvement that you seem to feel is necessary, and that I 20 certainly agree is necessary? Why would you expect anything 21 more than the same kind of performance we've been seeing, 22 simply by issuing a policy statement?
23 Isn't it time to say enough is enought you have to 24 perform and you have to bring the performance of these plants 25 up to an acceptable level. Everybody has.to. And that means i
. 40 1 a heavy dose of preventive maintenance and some real attention 2 to the kinds of root causes that have led to the continuing 3 experience that we've seen.
4 MR. RUSSELL: I think there are two aspects that at 5 least influenced me personally in making the recommendation 6 for a policy statement instead of an alternative approach.
7 One is that I've been very closely involved with evaluating 8 the industry programs through training and accreditation and 9 the effectiveness of the policy statement approach in that 10 area. And I think it's a very healthy one. I think that that 11 approach is showing results. And the commitments that were 12 made are being met. That's why I tied the combination of 13 commitment to policy statement.
14 The second aspect is a more pragmatic one and that 15 is because of the wide variability in what exists today, I 16 would have a difficult time writing a rule that would describe 17 what is the best approach to solving it. And I'm not sure 18 putting the staff in a loop of review and approve would be 19 either the most timely or the most effective way of solving 20 the problem.
21 I personally believe that a self-assessment by a -
22 utility that's committed to learn what their own problems are 23 and doing that effectively, with an industry group that's 24 looking over their shoulder to judge whether you have dono 25 that job offectively or not, with us in a role of monitoring
O
. 41 1 and providing constructive criticism or feedback back into I
2 that process is the one that will get there the fastest.
3 I don't think we can afford -- and I characterize 4 that as it relates to the industry as a whole. I think we 5 have a very important and primary role with those that are at 6 the tail and of the spectrum, the poorer performing plants. I 7 don't think that we can wait for that process. I think those 8 who are performing poorly need to have~the regulatory 9 emphasis, that the enforcement activity needs to be conducted, 10 and that we need to direct in that case, because I don't 11 believe they've seen the light yet.
12 COMMISSIONER ASSELSTINE: What's your basis for 13 doing that? Before something has happened, what's your basis 14 for stepping in and telling them, You're not doing enough on 15 maintenance; you have to do more.
16 MR. RUSSELL: Our approach is, in fact, a reactive 17 approach from that standpoint. -
18 COMMISSIONER ASSELSTINEt That's right. .
! 19 MR. RUSSELL: You wait for an event or something to 1
20 occur.
21 COMMISSIONER ASSELSTINE: Wait for something bad to j 22 happen. ,P 23 MR. RUSSELL: Unless, through our evaluation of 24 performance indicators, we can see that there is a plant which j 25 appears to be on the bottom, through some of the other I
e a 42 1 mechanisms.that have been put in place, with the periodic 2 meetings of senior management to discuss these plants. I 3 think that there are some facilitier, that are on the current 4 list with the Commission that fall into the category'of being 5 marginal maintenance performers that have not had a 6 significant event yet t. hat clerirly we've made recommendations 7 in that area for taking action. And I believe that process 8 can work.
9 I am, I guess, hopeful that they could apply the 10 successes and the lessons learned from the accreditation 11 process into the m.aintenance process. Whether they would 12 commit to an accreditation of a maintenance facility similar 13 to what FAA does with their maintenance facilities, I'm not 14 sure what the right answer is from the industry but I believe 15 some type of a proactive, self-assessment to move forward is 16 needed and that that ought to be articulated in a policy 17 statement and that should be a primary objective for 18 accomplishment in the very near term.
', 19 COMMISSIONER ASSELSTINE: But as you point out, 20 unlike the training area, there isn't any industry commitment 21 to do that.
22' MR. RUSSELL: Not yet.
23 COMMISSIONER ASSELSTINE: In fact, there isn't even 24 a willingness to agree to a policy statement which isn't even 25 any kind of a binding requirement on them. Which I find a
r 4
43 1 rather shocking, I guess, in light of the recent experience.
2 MR. RUSSELL: That's why we highlighted that issue 3 in the Commission paper and we feel that's an important issue 4 for the Commission to discuss with NUMARC this afternoon.
i 5 COMMISSIONER ASSELSTINE: One last question. You 6 said if you had to develop a regulation, you weren't sure what 7 you would do, given the differences. If you can't figure out 8 a way to improve the individual maintenance programs given the 9 possible differences that might exist, once again, why doesn't 10 the Japanese approach provide one assured way -- it may not be 11 the optimal way from the industry's. standpoint, but one 12 assured way to bring about a rapid and dramatic improvement in 13 the maintenance areas, simply by saying what you're going to 14 have to do effective at some point in the future during your 15 annual outages is do maintenance'on the following programs, 16 the following components and systems; we will ensure that 17 you've brought those systems and components up to an 18 appropriate level of quality, and you can't restart the plant 19 after your outage until we've signed off on it. And you only 20 get the authority to run the plant until your next outage.
21 where you have to do the very same thing.
22 Why wouldn't that provide real assurance that in 23 fact, the quality of maintenance is being maintained at a very 24 high level?
25 MR. RUSSELL: I think the difference is -- and I
.. -. z _
- 44 1 believe in Japan there is an infrastructure already in place
- 2. that is used to that system that understands the importance of 3 the quality of the maintenance when it's being performed, the 4 individual responsibility and accountability. When I look at 5 some of our own experiences in construction and maintenance, 6 I'm concerned about requiring more until we improve the 7 quality of what we're already doing, because more may result 8 in more errors, more failures, and that trade-off in not 9 having that structure in place gives me some concern.
10 I'm also concerned that a mandatory program that may 11 be well intended might miss the mark as to what is needed at 12 the lower levels. And I think management focusing on their 13 own problems and deciding what's needed on a very specific 14 basis at their facility, the self-assessment process, is a 15 much more effective one. We've seen that in training. I 16 don't see why the model that's being used in training cannot 17 apply'to maintenance.
18 COMMISSIONER ASSELSTINE: Okay. Well you're right. ,
19 You certainly have to make sure that the people that are doing 20 the work do it right, and if that's your concern, that's a 21 valid concern. I guess certainly given the kind of experience 22 where people can't even find the right unit or the right 23 train, it's a legitimate concern.
4 24 But it doesn't seem to me that the difficulty in
- 25 ensuring that people do the work right is necessarily a valid
a ;
45 i
1 objection to carrying forward a program that has high 2 prospects of bringing the quality up if the people are going 3 to be allowed to run the plant.
4 MR. RUSSELL: It also has a potential effect in that 5 we don't want to co-opt or assume the responsibility ourselves 6 for the quality of individual components. I believe that that 7 is primarily a utility responsibility, to ensure the quality 8 and the safety of their components. ,
9 MR. PARLER: Well of course, putting out a 10 regulation requiring something doesn't mean that we'll assume 11 the responsibility of --
4 12 MR. RUSSELL: I understand that. But in some cases
~
13 when we've required things in the past, utilities have done 14 that and no more, and it sort of establishes the minimum 15 levels rather than a level of attempting to seek excellence in
- 16 the way that's being maintained.
l 17 COMMISSIONER ASSELSTINE: Yes. Although my standard 18 would be something akin to what the Japanese do, and I would .
19 argue that is far from a minimal level of quality. They've 20 got the same kinds of plants, they've identified the systems 21 and components, they've done the work for us, basically. And i
22 if you aimed for that level, you sure wouldn't be aiming for l
l 23 the minimum.
24 MR. RUSSELL: In fact, that ten-year maintenance 25 plan is submitted by the utility prior to getting a license,
e 46 1 and they specify what they're going to do over the ten 2 operating cycles for all the important components in the 3 plant. It is indeed a prevent maintenance program plan on a 4 component basis.
5 COMMISSIONER ASSELSTINE: That's right. That's the 6 only questions that I have. I'd just say that although I'm 7 somewhat skeptical of the policy statement approach and 8 whether we're doing as much as we should be in this area, I 9 think that the staff is to be commended for what you've been 10 doing. I found the reports informative, I think you've 11 highlighted probably one of the most significant problems in 12 plant operations in this country today, and I think that at 13 least focusing attention on that problem is a big step 14 forward, and I think you've done a good job in doing that.
15 MR. VOLLMER: Well certainly, it would help -- the 16 Commission can help to overcome the industry's reluctance to l 17 pursue this. I think the policy statement --
l 18 COMMISSIONER ASSELSTINE:- That's a good topic for 19 this afternoon.
l 20 MR. VOLLMER: The policy statement certainly appears 21 to be the best short-term goal to achieve.
l 22
~
COMMISSIONER ASSELSTINE: You and I may differ about 23 that.
24 MR. VOLLMER: Well, your arguments about 25 enforceability nad other things are certainly true. But l
L 47 1 defining what that adequate program would be is likewise 2 difficult to the process. -
3 COMMISSIONER ASSELSTINE: I think it really boils 4 down to whether we're going to be proactive or not, whether 5 we're going to wait for something bad to happen before we fix 6 the prcblems, or whether we're going to say enough is enough; 7 we simply can't tolerate the kind of continued performance 8 that we've seen in the past. We have to have a proactive 1
9 approach that's going to fix these things before serious 10 operating events or accidents occur.
11 MR. VOLLMER: But I think our proactive approach 12 really is in the area of performance indicators and looking 13 hard at plant performance now, which certainly envelopes this 14 activity.-
15 COMMISSIONER ASSELSTINE: It's a big step forward; I 16 wouldn't disagree with that. That's ell I have.
17 CHAIRMAN ZECH: Thank you. Commissioner Carr?
18 COMMISSIONER CARR: Yes. Do the plants with the 19 so-called standard tech specs have better maintenance?
20 (Commissioner Bernthal is entering the room.]
21 MR. RUSSELL: We have not really explicitly looked 22 at those which have standard tech specs versus those which 23 have custom tech specs. As an observation, however, the older 24 plants generally have the custom tech specs, the less 25 prescriptivo technical specification. The plants that we
6 48 1 looked at when we were looking at who were the better in 2- performance and maintenance tended to be the older facilities, i
3 In that context, one could argue that less 4 specificity and not having a stack of tech specs that require 5 100,000 or more surveillances a year, putting more 6 responsibility back on the utility for assuring that the 7 equipment operates, would tend to indicate a better 8 maintenance program.
9 However, another answer could very well be that the 10 older plants have been operating at it longer; their own 11 self-interest in keeping the plant operating has caused them 12 to have a more effective maintenance program, and that some of 13 the newer plants that are having problems are having problems
~
14 because they haven't gotten through the initial operating 15 phase of learning how the plant operates and solving those 16 early problems, so they're not facing a maintenance issue yet.
17 COMMISSIONER CARR: Well, I've got a couple of 18 comments. One is that I think this effort is about 20 years'
, 19 overdue, and it's a single area, it looks to me like, where i
20 we've got the most to gain for the least amount of money in 21 safety and performance. And I don't know why we didn't get on l
! 22 with it five or ten years ago.
I 23 The second thing I'd like to say is that obviously, l 24 the incentive for the utility to correct this is just equal to l
25 ours, because it will not only put money in their pockets, it l
. 49 1 will make them safer, too. So I am hesitant to see what kind l 2 of reluctance they would have to implementing a program.
3 I have two concerns. One is that personnel turnover 4 in the maintenance area -- if you don't have an experienced 5 maintenance organization, no matter how much you do, it may 6 not be done properly. And maintenance, if it's not done 7 properly, is Worse than not doing it at all, in many cases.
8 So I encourage you in your effort. We have to do 9 something. This is the quickest thing, it looks like, to do.
10 But if you want my encouragement to the utilities to get a 11 better maintenance program, I'll certain subscribe to that.
12 MR. VOLLMER: Commissioner Carr, I might mention on 13 the order of 10 to 12 years ago there was an effort that the 14 NRC participated in, and that was the industry standard group-15 working on quality assurance for operations. And I was part 16 of that activity, and at that time when you got into the areas j 17 of preventive maintenance there was considerable industry 18 reluctance in defining that, or accepting a definition of what l 19 they had to do in the working operation of a plant; i.e.,
20 working on particularly the secondary side.
I
! 21 COMMISSIONER CARR: Well, there are all kinds of i
l 22 thecries in maintenance and we go through them lots of ways.
l 23 The Japanese I would call on overhaul type maintenance, as 24 opposed to a continuing type maintenance. I'm a firm believer 25 personally in fix it now; when it's down, fix it and keep it
6 o- 50 1 so, as the man says, you're playing with a full deck all the 2 time. So that all your equipment is working all the time.
i 3 And you can't do that if you. wait for a year to go in and.do 4 it. You'll fix a lot of things but you'll obviously open up a 5 lot of things that didn't need work.
6 But the philosophy is not the important thing; the 7 important thing is to have a good maintenance program, 8 whichever philosophy you pick.
9 CHAIRMAN ZECH: Commissioner Bernthal?
i 10 COMMISSIONER BERNTHAL: Well, I apologize for 11 missing what is a very important subject-that probably it's 12 about time we've had a meeting on. It's probably the first 13 meeting that I can ever remember scheduled for an hour and a 14 half that seems to be done a half hour ahead of time; 15 otherwise, I would have made it here for part of it.
16 I understand that you've discussed what I find to be i
17 a most interesting chart here, comparing France and Japan and
]
18 the U.S. records. Maybe you haven't discussed Paris, but in 19 the first place it's a very good display, it shows the 20 scheduled and unscheduled outages in a two-dimensional plot 21 here. It shows that Japan has always been very high in 22 scheduled outage, ve?/ low in unscheduled, and is now dropping 23 dc.en in scheduled outages as well. France has a history of 24 fairly high unscheduled but is now rapidly dropping below, way 3
25 below the U.S. unscheduled outage time. And the U.S. is
-M yw *t m= y y-m-*g-m-y g-% w-- -pyy 9 -- w -+--w y- y
d 51 1 floundering around.
2 I just find that to be a very revealing plot.
3 Everyone else with major programs seems to be improving 4 dramatically in every respect, and we aren't. So even granted 5 the complexities and differences between our plants, we do 6 have a bigger problem in that respect. I think we need to get 7 on with it, and I apologize that I wasn't able to be here and 8 listen to your presentation but I'll look at the paper that 9 you presented. Thank you.
10 CHAIRMAN ZECH: Let me just make a couple of 11 comments. First of all, I do agree that we should look at the 12 international programs. We have mentioned Japan and f 13 Commissioner Bernthal just mentioned France and other i
14 countries. But international maintenance programs is 15 something that we should look at and analyze, evaluate and 16 gain from because there are differences.
17 I also think we should look domestically. I think 18 we've alluded very briefly to the FAA and to other industries 19 in our country that involve safety especially. So I do think 20 we should look at domestic programs, too, to see where 21 maintenance policies and programs can give us some insights.
22 As we've mentioned also briefly today, balance of 23 plant, in my view and in looking over performance in the past 24 few years, balance-of-plant maintenance as well as overall 25 performance is something that we certainly need to emphasize
52 1 and focus on. -
2 We can't divorce in my judgment balance of plant 3 from the steam supply system; it's one system, in my judgment, 4 and they inter-related, and it doesn't make any difference 5 whether you call it safety related or important to safety or 6 whatever, those are terminologies, but the plant doesn't know 7 the difference. And balance of plant can give you as much 8 problems in many areas, and has in my judgment. So we should 9 focus on balance of plant when you're talking maintenance as 10 well as the steam supply system.
11 Also we've mentioned briefly on-line testing,
- 12 testing at power, surveillance at power. I must say I think b
13 we do too much of that. I think there are tests and 14 surveillances we can go when the plant is shut down. I think 15 when we have the plant running, I agree with Commissioner Carr i 16 that we should focus on fixing it now and keep it running and 17 have a program of competence so that when the plant is running 18 we expect it to keep running and running in a safe manner.
19 So I think maintenance is something that we really 20 need to focus on -- I think certainly most of us would agree 21 with that -- and has not been focused on over the years in 22 this industry as well as it should have been.
23 I think what's going on in industry ,- and this is 24 just my impression over the past couple of years -- is that 25 there is an awareness now in industry of the importance of
.., _. .___._ _ ,,__ _ _ , _ , , , , , . . , . . . _ - , . - , _ y -_._ y.,. -
__ . . _ . , . - _ _ , -. , _ _ _ _ _ , _ . . - _ . m.
l 53 l
)
1 maintenance. I think I've seen that, but what I've seen I l 2 think more than anything is an awareness of it and even, in 3 some cases, programs and pilot programs that you've pointed 4 out here. But I haven't seen a lot of results yet, haven't 5 seen a lot of implementation, haven't seen a lot of action, 6 and that's what we're looking for.
7 I think that we will find from the utilities and 8 from the industry that there is an awareness of it, but that's 9 not enough. And I think we have to do what we can to move 10 forward to see that action and results come and come 11 reasonably soon. It is a safety item and that's our concern, 12 and it's also a reliability factor, and I think it's in the
(
- 13 industry's, the utilities' best interest to improve in this 14 regard. It just makes common sense to me.
15 So I think we're involved in something that is 16 extremely important, and I'm sure this afternoon we'll discuss 17 it more with NUMARC. I appreciate the staff's fine l
18 presentation and good, thorough work. I do think that looking .
19 internationally, domestically, perhaps somewhat carefully may 20 give us little better insights into which way to go. But 21 there's no question in my judgment there's room for 22 improvement, so I think we're about something that is very 23 important.
24 Are there any other final comments from my fellow 25 Commissioners?
l i
(
. 54 1 (No response.]
2 With that, we stand adjourned.
3 [Whereupon, at 11:15 a.m., the commission meeting 4 was adjourned.]
5 6
7 8
9 10 11 12 f~ 13 14 15 16 17 18 19 1
20 21 22 23 24 25
1 1 2 REPORTER'S CERTIFICATE 3
4 This is to certify that the attached events of a 5 meeting of the U.S. Nuclear. Regulatory Commission entitled:
6 7 TITLE OF MEETING:
Briefing on Initiatives to. Improve Maintenance Performance (Public Meeting)
> 8 PLACE OF MEETING: Washington, D.C.
9 DATE OF MEETING: Thursday, November 20, 1986 10 11 were held as herein appears, and that this is the original 12 transcript thereof for the file of the Commission taken l{'
13 stenographically by me, thereafter reduced to typewriting by 14 me or under the direction of the court reporting company, and 15 that the transcript is a true and accurate record of the 16 foregoing events.
17 /
18 ----- M - ---- ------
[ Suzanne B. ung h 19 20 21 22 Ann Riley & Associates, Ltd.
23 24 25
STATUS OF ACTIVITIES TO IMPROVE NUCLEAR POWER PLANT MAINTENANCE NOVEMBER 20, 1986 l
l l
l l
l l
l
l O
BACKGROUND JANUARY 1985 -
PHASE I 0F MAINTENANCE AND SURVEILLANCE PROGRAM PLAN (MSPP) APPROVED FOR IMPLEMENTATION APRIL 1985 -
MSPP ISSUED (SECY-85-129)
JUNE 1985 -
MSPP PHASE I CONTRACTS INITIATED JULY 1985 -
SITE SURVEYS BEGUN ,
NOVEMBER 1985 - RESIDENT INSPECTOR QUESTIONNAIRE ISSUED JUNE 1986 -
PHASE I ACTIVITIES COMPLETE m m- v -e-. ,- --, - ,---,.,-- _ +
PHASE I ACTIVITIES OBJECTIVE: DETERMINE STATUS OF MAINTENANCE IN U.S. NUCLEAR POWER INDUSTRY
- SITE SURVEYS AT EIGHT OPERATING REACTORS
- DEVELOPMENT OF SURVEY PROTOCOL
- QUESTIONNAIRE RESPONSES FROM 68 RESIDENT INSPECTORS
- ON-SITE INVESTIGATION OF 35 WRONG UNIT / WRONG TRAIN EVENTS AT 10 OPERATING REACTORS
- DEVELOPMENT OF COMPUTERIZED MAINTENANCE DATA BASE
- 1/4 MILLION DATA ELEMENTS ON OPERATIONS, MAINTENANCE, INSPECTION
- 1980-1985
- MENU-DRIVEN SYSTEM FOR NRC-WIDE DISTRIBUTION
- DEVELOPMENT OF 31 MEASURES OF PLANT MAINTENANCE EFFECTIVENESS
- REVIEW / COMMENT ON NUMARC MAINTENANCE PERFORMANCE INDICATORS
- INFORMATION EXCHANGE MEETINGS WITH INP0/NUMARC
- PARTICIPATION IN REGIONAL MAINTENANCE INSPECTIONS l
e PHASE I ACCOMPLISHMENTS
- NUREG-1212, STATUS OF MAINTENANCE IN THE V.S. NUCLEAR POWER INDUSTRY 1985 (JUNE 1986)
- NUREG-1192, AN INVESTIGATION OF THE CONTRIBUTORS TO WRONG UNIT OR WRONG TRAIN EVENTS (APRIL 1986)
- NUREG/CR 4600, HUMAN FACTORS STUDY CONDUCTED IN CONJUNCTION WITH A MINI-ROUND ROBIN ASSESSMENT OF ULTRASONIC TECHNICIAN PERFORMANCE (AUGUST 1986)
- NUREG/CR-3883, ANALYSIS OF JAPANESE-U.S. NUCLEAR POWER PLANT MAINTENANCE (JUNE 1985)
- NUREG/CR-4611, TRENDS AND PATTERNS IN MAINTENANCE PERFORMANCE IN THE U.S. NUCLEAR POWER INDUSTRY: 1980-1985 (OCTOBER 1986)
% --m
a FINDINGS NEEDED MAINTENANCE NOT BEING ACCOMPLISHED OR NOT PERFORMED EFFECTIVELY AT SOME PLANTS
- 64% OF TOTAL FORCED OUTAGE TIME DUE TO COMPONENT FAILURE
- MAINTENANCE-RELATED LERS 48% OF 1985 TOTAL
- SOME MAINTENANCE PROGRAMS FRAGMENTED AND LESS THAN EFFECTIVE
- 22% OF PREVENTIVE MAINTENANCE PROGRAMS RATED MINIMAL HIGH PERCENTAGE OF FAILURES FROM IMPROPER PERFORMANCE OF MAINTENANCE
- OVER 30% OF ABNORMAL OCCURRENCES SINCE 1975
- PROPORTION OF MAINTENANCE-RELATED LERS INCREASED MAINTENANCE /0PERATIONS INTERFACE INADEQUATE
- 65% OF LOSS OF SAFETY SYSTEM FUNCTIONAL EVENTS DUE TO HUMAN ERROR
- OPERATIONS PERSONNEL RESPONSIBLE FOR 75% OF HUMAN ERROR IN WRONG UNIT / WRONG TRAIN EVENTS MAINTENANCE-RELATED CHALLENGES TO SAFETY SYSTEMS IS EXCESSIVE
- 75% OF ESF ACTUATIONS IN 1984 AND 1985
- LACK OF IMPROVEMENT IN 1981-1984 TRENDS INDUSTRY INITIATIVES ARE BEING IMPLEMENTED
.~
INDUSTRY INITIATIVES GUIDELINES FOR THE CONDUCT OF MAINTENANCE AT NUCLEAR POWER '
STATIONS (PRELIMINARY) - INP0 85-038 (OcT. 1985) i
- PILOT SELF-ASSESSMENT PROGRAM USING INPO GUIDELINES EXTENSION OF PEER EVALUATOR PROGRAM TO MAINTENANCE INCREASED USE OF NPRDS EXPANSION OF HUMAN PERFORMANCE EVALUATION SYSTEM MAINTENANCE WORKSHOPS TRAINING ACCREDITATION (MAINTENANCE PERSONNEL)
DEVELOPMENT OF PERFORMANCE INDICATORS O
P
e*
NRC ASSESSMENT INDUSTRY INITIATIVES HAVE POTENTIAL TO IMPROVE PLANT MAINTENANCE, If IMPLEMENTED ON A CONSISTENT BASIS.
- NO INDUSTRY CONSENSUS (AS IN TRAINING)
- PHASE I FOUND SUBSTANTIAL VARIABILITY IN MAINTENANCE PROGRAMS AND THEIR EFFECTIVENESS
- NEED POLICY STATEMENT
- SPECIFY LONG TERM GOALS
- RECOGNIZE SELF-IMPROVEMENT INITIATIVES
- ARTICULATE NRC/ INDUSTRY ROLES IN IMPLEMENTATION
- SPECIFY MEANS OF NRC OVERSIGHT l
l
- o PHASE II ACTIVITIES OBJECTIVES - EVALUATE EFFECTIVENESS OF INDUSTRY INITIATIVES
- DEFINE ROLE OF MAINTENANCE IN SAFETY
- DEVELOP RECOMMENDATIONS FOR GOOD PRACTICES
- SITE SURVEYS
- NEW SITES
, - FOLLOW-UP VISITS OF PHASE I PLANTS
- ANALYSIS OF MAINTENANCE MEASURES
- VALIDATION
- COORDINATED WITH NRC INDICATORS
- ASSESSMENT OF INDUSTRY INITIATIVES
- DOCUMENTS (GUIDELINES, 08MRS, ETC.)
- OBSERVATIONS
- INTERFACE WITH RELATED PROGRAMS
- NRC ,
- INDUSTRY (STANDARDS, ETC.)
1
NMWWWWGV0tWGknGWVIVfVtVGtWGVpt%;VGVit cV(tW;t gy;(:gygygygggggggig, ,
i Document Control Desk, 016 Phillips TRANSMITTAL T0: --
. 1 The Public Document Room
$! ADVANCED COPY TO:
! DATE: //!M!Eb -
3:
3 .
SECY Correspondence & Records Branch ,
3 FROM:
3 ::
l l
3: i 3 Attached are copies of a Comission meeting transcript and related meeting 3 :: !
3 !. document (s). They are being forwarded for entry on the Daily Accession List and placement in the Public Document Room. No other distribution is requested or
@j required, sl a !'
b CS ko b e.0c
]! Meeting
Title:
3(K8h e n 1
.Nah ::e, Twkwer= '+
i .
! Meeting Date: \\liolfb , Open d Closed 3 f.
3 '
3
- 3 Copies 3 :l Item Description *: DCS 3 Advanced '3 to POR g Si 3 .
3 l 1 1 S i: 1. TRANSCRIPT 3
3i, ua/O a e o c. 4.hs 3 .
I 3:
3 :.
3:
3 !' 2' -
3:
3 .
m:
3
- 3
- : 3.
'3g m:"
'3 "
3 4.
3:1 3:,
3B 3 .
3:. 5.
3
'3
, b 3' 6.
f
!
- POR is advanced one copy of each document, two of each SECY paper.
C&R Branch files the original transcript, with attachments, without SECY papers.
AJM lhl Y h hl hh
- ._ _ _ ___ _ _ _ _ _ _ __ _. . _ _ . _. .