ML20205H455
| ML20205H455 | |
| Person / Time | |
|---|---|
| Issue date: | 08/13/1986 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1543, NUDOCS 8608200079 | |
| Download: ML20205H455 (169) | |
Text
'
ORIGlNAL Y N~4 3 UNITED STATES O NUCLEAR REGULATORY COMMISSION IN THE MATTER OF: DOCKET NO:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON MAINTENANCE PRACTICES AND PROCEDURES O
LOCATION: WASHINGTON, D. C. PAGES: 1 - 140 DATE: WEDNESDAY,' AUGUST 13, 1986
.- . . n.g ACRSOFRCECDPY 1.
Jo Not lemov.e ' rom AC,lS 0 Tice ACE-FEDERAL REPORTERS, INC.
OfficialReporters 444 North CapitolStreet
('
Washington, D.C. 20001 8y';E~jy '
(202)347-3700 NATIONWIDE COVERAGE
CR27846.0 1 MM/sjg UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
SUBCOMMITTEE ON MAINTENANCE PRACTICES AND PROCEDURES 5
6 Nuclear Regulatory Commission Room 1046 7 1717 H Street, N.W.
Washington, D. C.
8 Wednesday, August 13, 1986 9
The subcommittee meeting convened at 1:00 p.m.,
10 Mr. Glenn A. Reed, chairman, presiding.
12 ACRS MEMBERS PRESENT:
i U/ 13 MR. GLENN A. REED 14 MR. JESSE C. EBERSOLE 15 MR. CARLYLE MICHELSON 16 MR. CHARLES J. WYLIE 17 18 19 20 21 22 23 24 25 ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 80 4 336-6646
() PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSIONERS' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS WEDNESDAY, AUGUST 13, 1986 The contents of this stenographic transcript of the proceedings of the United States Nuclear Regulatory ,
Commission's Advisory Committee on Reactor Safeguards (ACRS), as reported herein, is an uncorrected record of the discussions recorded at the meeting held on the above date.
No member of the ACRS Staff and no participant at
) this meeting accepts any responsibility for errors or inaccuracies of statement or data contained in this transcript.
l l
l r
I I
O V
1
8460 01 01 2
(~h s / MMbw 1 PROC.EEDINGS 2 MR. REED: This is a meeting of the Maintenance 3 Practices and Procedures Subcommittee. The meeting will now 4 come to order.
5 I am Glenn Reed, the subcommittee chairnan. The 6 other ACRS members in attendance are Carlyle Michelson, 7 Charlie Wylie, Jesse Ebersole. I&E engineer on this is 8 Herman Alderman.
9 The purpose of the meeting is to review the 10 report on Phase 1 on the Maintenance and Surveillance 11 Program Plan, MSPP.
12 The rules of participation in today's meeting O
\/ 13 have been announced as part of a notice of this meeting that 14 was published in the Federal Register on July 28, 1986.
15 It is requested that each speaker first identify 16 himself or herself and speak with sufficient clarity and i 17 volume, so that he or she can be readily heard.
18 Now I just want to whet our enthusiasm a little 19 bit for the meeting and say that I have gone over all this 20 paper in front of me having to do with the Maintenance and i
i 21 Surveillance Maintenance Program Plan, and I don't find too l
22 much in it that pertains to craftsmanship or craft.
23 I find a lot in the work today that has to do 24 with what I will call maintenance office activities. That 25 is, engineering scheduling, program data, work, logging ACE-FEDERAL REPORTERS, INC.
202 Nationside Coverage 800-336-6646 l _
, - .- . , -347-3700
8460 01 02 3 7
(/ MMbw I and so on and so forth. But I always like to worry about 2 craftsmanship.
3 So coming out of Phase 1, we might decide that we 4 need more emphasis on craft. I believe that the maintenance 5 activity is made up of damn good craftsmanship, if it is 6 there, and all these other engineering and logging and 7 planning activities. And they are a lot different, one from 8 the other. Takes different kinds of people.
9 So with that, do any other members have comments?
10 MR. EBERSOLE: Yes. I was not formally 11 recognized as a member, but I appear here today mainly on 12 the topic of maintenance, in the context of maintaining and O 13 restoring environmental characteristics of equipment. By 14 and large, when you maintain a piece of equipment, a 15 maintenance man will just glory in his ability not to read 16 paper and do it the right way with the expertise that you 17 prefer.
18 All this equipment is type tested for 19 invulnerability in environmentally hostile conditions. I 20 doubt very much that in the maintenance picture, there is 21 even much attention paid to the restoration of the 22 environmental capacity, and certainly, no tests are ever 23 made to determine the invulnerability of such equipment to 24 these hostile external influences.
O 25 I would like to hear to what extent this aspect I
l ACE-FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coserage 80 4 336 6616
8460 01 03 4
(")MMbw
\m- 1 of maintenance is taken care of. Realizing you cannot test 2 it, it is a rigidized procedure. If you make the slightest 3 deviation here or there, the stuff is going to leak, if you 4 don' t tighten th gaskents, if you don't shrink the tubes, 5 whatever the heck you do. It's not a testable function.
6 It's not easily visible. You don' t know what you've got, 7 unless you rigidize the procedures and stick to them just 8 like they did in the tight testing routines.
9 So mainly, Glenn, that's what I'd like to hear 10 about.
11 MR. REED: You are sure bringing up an 12 interesting one. I doubt that the Staff will be ready to fs O 13 ansser it, but they heard your comments. It is certainly a 14 very interesting thing.
15 MR. MICHELSON: I have one general concern that I 16 would like to get the Staff's comments on. Historically, we 17 went two years ago through a maintenance program plan and so l
18 forth. I find nothing in here that indicates any startling 19 change from what was formulated in detail two years ago, but i 20 I do find, appear to find, at least, an absence of much r
21 happening in between.
22 What is the reason for the pace that we seem to 23 be moving in this area? Am I going to anticipate two more 24 years coming back and hearing this same story again, or are i ( 25 we really moving now?
ACE-FEDERAL REPORTERS, INC.
202-347-3/00 Nationwide Coverage M346(4
8460 01 04 5
,~
k_),MMbw 1 MR. WYLIE: I don't want to belabor this too 2 much, but I think, as noted in this report that has just 3 been generated, company practices do differ. They differ in 4 other ways besides just maintenance, such as th'e quality of 5 the original design job that was done, selection and 6 application of equipment, conservatism applied in those 7 applications, which later show up in failures and forced 8 outages and things like this, and the applications. While 9 those indicators that have been identified are indicators of 10 maintenance, they are, as well, an indicator of how good you 11 did the first job, the type of equipment that was selected.
12 For example, did you select sleeve bearings or
,73 l 13 ball bearings or roller bearings or whatever you did on the 14 machine? What other things you did.
15 So company practices do differ, as far as 16 applications go in the original design job, and I am afraid 17 they also influence those indicators, as much as 18 maintenance.
19 The other thing is, I do see and applaud the 20 effort toward trying to get a uniform plan to improve 21 maintenance practices in the utilities. Also within the 22 utilities, there are different maintenance organizations 23 that compete for those activities, and sometimes this is not 24 good, and maybe this will help straighten some of that out.
25 MR. REED: All right. The first person from ACE-FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coserage 804336-6M6
8460 01 05 6
('m-w'l MMbw I the Staff to make a presentation is Mr. Gregg Cwalina.
2 (Slide.)
3 MR. CWALINA: My name is Greg Cwalina. I am the 4 Section Leader of the Maintenance and Surveillance Section.
5 Just a background, maybe I can answer some of 6 Mr. Michelson's questions.
7 The Maintenance and Surveillance Program Plan was 8 approved for implementation by the EDO in January of '85.
9 At that time, the Maintenance and Surveillance Program Plan 10 wasn't final. It was made final in April. We really didn't 11 get started in the projects then until about May-June' 12 timeframe. So when you were involved in it in 1984, it was O 13 still in its early stages and nothing had actually taken 14 place.
15 Again, the purpose of Phase 1 is to establish a 16 baseline on maintenance, determine the status of maintenance 17 that exists in the industry today. As such, we did a 18 programmatic overview of maintenance, including several 19 projects that I will talk about, but we are not going into 20 the details of how to maintain specific pieces of 21 equipment. Rather, we are looking at the old management 22 organization type aspects of maintenance.
23 So with that in mind, let me briefly explain what
, 24 we are going to do today. After I give the introduction, we 25 are going to have Mr. Bernie Grenier talk about >
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nanon axle Cowerage 800-33MM6
8460 01 06 7
('T Tommie Le is going to
(_/ MMbw 1 questionnaire survey observations.
2 talk about the site surveys. Tommie was lead engineer on 3 four of the site surveys. We are going to have Pete 4 McLaughlin talk about the maintenance measures, and that is 5 maintenance type of performance measure that we started into 6 a couple of years ago. Pete has been working on that, and 7 that is now feeding into the overall NRC program on 9
8 performance indicators.
9 Drew Persinko is going to talk about the wrong 10 unit, wrong train events. He was the team leader on those, 11 and John will talk about the overall conclusions of Phase 12 1.
O
\' 13 Just as a reminder, we started off with about 4
14 nine projects in the maintenance and surveillance program 15 plan.
- 16 (Slide.)
17 The first three projects survey performance and
! human errors constitute the bulk of our report, and that is 18
! 19 what we are going to concentrate on today.
20 If you have any questions on some of the other 21 items, we will be glad to answer them, but the bulk of our l
22 presentation is really going to be on the first three l
23 items.
24 Incidentally, we have released -- we are in the
( 25 process of releasing four NUREGs. Two have already been l
ACE-FEDERAL REPORTERS, INC.
j 202 Nationwide Coverage 800 336-6M6
-- _ -347-3700-
-. . . _ ~ ._ . - -. - . - , _ . . . , . - .
8460 01 07 8 i ('\ . .
(-) MMbw 1 published. That is the status of maintenance in the nuclear 2 power industry, which I think you have all seen, NUREG 3 12.12, and also the NUREG on wrong unit, wrong train events, 4 which is NUREG 11.92. We are going to release two more 5 NUREGS, on on the human factors in service inspection, and 6 the other is going to be a trends and patterns analysis.
7 Both of those are in the process of being printed now.
8 (Slide.)
9 Just to give you a short background of the survey 10 project. It's been broken down in two parts. The purpose 11 of this was to obtain current industry information on 12 maintenance programs and practices at nuclear power plants.
O'
\
13 We took two tacks on this. The first thing we did was to 14 send out a questionnaire to all of the resident inspectors.
15 The questionnaire, as you see, was pretty expensive, 16 consisted of about 139 questions with 500 response items.
17 Resident inspectors were directed not to answer questions 18 that they did not know the answer to offhand. In other 19 words, don't go and ask the licensees for them.
20 So we got about a 50 percent response rate on all 21 the questions, and we got a 96 percent return rate on the 22 questionnaires. So we got a lot of information from those 23 inspectors.
24 The bulk of our report is based on these 25 responses from the resident inspectors.
ACE-FEDERAL REPORTERS, INC.
M 347-3700 Nationside Coverage RA.L3%4=W l
1 8460 01 08 9
( MMbw 1 The other thing we did, we went out on site )
2e surveys, and we developed a protocol, which would cover five 3 major areas of maintenance, which we broke down into these i 4 five organizations -- administration, facilities and 5 equipment, et cetera. That was just to break it down for 6 our ease.
7 We developed a protocol, so we would gather the >
8 same amount and the same types of information from each of f
9 the eight sites, so that when we were doing the comparison, 10 we were comparing apples against apples.
11 We went on the site surveys. We.had five or six 12 members in each one of the site surveys, headed up by a O/
13 member of this team of this section. Trips reports have all 14 been published, all of which are included in Volume 2 of 15 that NUREG.
16 MR. EBERSOLE: May I ask a question? As a matter.
17 of operation, as well as maintenance, there are two ways to 18 go. The operator will religiously follow a written check 19 list, or he will use his own expertise and pride in 20 operation and run it his way. And that is the second way.
21 Equally so in maintenance and probably more so 22 with a maintenance fellow who think he is extremely smart, 23 indeed, and, in fact, as one of the ones with the best 24 competence that Glenn likes to think about, best expertise, O 25 he either has the option of doing it by the book or doing ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Co,erage 800-33MM6
1 l
t 8460 01 09 l 10 MMbw 1 it his way and doing it his way has something in it that 2] defects the original objective of both the operation as well 3 as tha maintenance.
4 How did you find the industry in regard to the 5 use of written procedures and chech lists and following the 6 book like the Navy does?
7 MR. CWALINA: Mc will get into that a little bit 8 more, but we found out exactly that. We found it ranges.
9 Practices range from one extreme to the other. And that is 10 what we found out. We observed both extremes, where you had 11 to go through that procedure word for word, do the 12 check-offs, and there were other ones where procedures were 13 minimal, where the worker himself decided he knew how to do 14 it better than the procedure and would just do it that way.
15 MR. EBERSOLEt Right. So you are g6ing to tell 16 us how you cope with that.
17 MR. CWALINA: I don' t think we are going to tell 18 you how we cope with that at this point. That is future 19 activities of the program. That is more Phase 2.
20 MR. EBERSOLE: There is a little bit of an 21 inversion problem here. In fact, the better operator, the 22 better maintenance you may be, better than his counterpart 23 in the equipment. He may be too good.
i 24 MR. CWALINA: That is right, but being where we l
(,') are, we can't distinguish who is the good operator and who l 25 i
ACE-FEDERAL REPORTERS, INC.
m .m, s.g_ < -, - - .
i
8460 01 10 11
( ) MMbw
. 1 is the bad operator.
2 MR. EBERSOLE: The ocod one is one who goes by 3 the book.
4 MR. CWALINA: You might get an argument there.
5 Before you do, we'll go ahead with our presentation.
6 MR. WYLIE: Let me ask, just as a matter of 7 info rma tion , I read the literature. Maybe I missed it, but I
8 I think you explained that you picked Public Service 9 Electric & Gas because of the salem event, and then you 10 picked Toledo Edison because of the Davis-Besse.
I 11 j What was the rationale on the rest of them? Why 12 did you pick those?
{%
13 MR. CWALINA: Those were two special cases. The 14 Salem one was because they had an NRC-mandated preventive 15 maintenance program So we wanted to go out and see the 16 effects of that. The Davis-Besse one, we were requested by 17 NRR management to go out and do the same thing we did.
18 The other ones, what we did, we tried to get a
- l 19 wide range of experience. So what we looked at, we tried to 20 cet a cross section of age, size, a cross section of the 21 types of plants. We got, I think, two boilers, two t
22 Westinghouso PWRs, one CE and one B&W We also wanted to go
,l 23 to all five regions and have all five regions participate in l.
24 it. So there were a lot cf factors put in there, including ,
<-)
'- - 25 SALF ratings. We wanted to see if we could get a wide I
ACE-FEDERAL REPORTERS, INC.
I m147 3700 xation. Lie co,e, age row?wn l.
8460 01 11 12 p(,) MMbw 1 range of experience too, and we did that by asking the 2 regional representatives, who would they recommend in their 3 region that had, in their opinion, good practices, bad 4 practices. So that type of thing.
5 MR. WYLIE: You tried to pick equal on good and 6 bad?
7 MR. CWALINA: Yes. I think we wound up pretty 8 much equal, good, bad and average. And again, a very 9 subjective opinion.
10 MR. WYLIE: Just another comment. In reading the 11 abstracts in the Executive Summaries, it was not apparent, 12 in reading them the first time -- in fact, I don't think, f
13 when I read it, that you sent the questionnaires out to all 14 U.S. site representatives, which you did. You would have to 15 read further into the document to find that out. It would 16 have been well to put that, I think, up front, very
- 17 emphatically. -
18 MR. CWALINA: One other item that I don't believe 19 is on here is the wrong unit, wrong train survey. In that 20 survey, we also went to 10 site visits, and Drew will talk 21 more about that detail when he comes up.
22 So if there are no more questions for me, I will l 23 turn it over to Bernie and let him talk about the i
- 24 questionnaire methodology and the results.
i /~T
- 25 MR. REED: This is Bernie Grenier.
l ACE-FEDERAL REPORTERS, INC.
! 202-347-3700 Nationwide Coverage En336-6M6
i l
i 8460 01 12 13 ;
(- _
( /.MMbw 1 MR. GRENIER: Good afternoon. I am Bernie 2 Grenier with the Maintenance Surveillance Section. I work 3 primarily as the program manager but also contract manager.
4 Most of this work on the survey, both 5 questionnaire and the site survey was done under contract 6 with contractor support. I think it was a joint effort, and 7 I sort of managed the contract, was involved iwht a lot of 8 the administative aspects of keeping the paper flowing.
9 I believe the questionnairo was probably one of 10 our more interesting projects of the first phase of the 11 program. It gave us an opportunity to coordinate with the 12 regional people that many times headquarters do not. So we (O
k/ 13 had a lot of interface with the -- well, the IE, the 14 inspectors out in the regions and also the resident 15 inspectors. And they were quite complimentary on the 16 questionnaire about its completeness and thoroughness. And 17 I believe, when you see or if you read the report, Appendix 18 A, Volume 2, has the results of the questionnaire. I think 19 that we have got a lot of support.
20 And we also found out, from a general standpoint, 21 that a lot of the questions were not answered, even though ,
22 we did get enough information to make it statistically 23 significant. We find that even our own staff in the area of 24 maintenance did not know as much as maybe one would expect, O 25 being out there, even right on site.
ACE-FEDERAL REPORTERS, INC.
202.MT.370) Nationmide Cmerage M633 M 46
8460 01 13 14 b
s/ MMbw 1 So maintenance, I think, shows us that there is 2 an area that does need attention from that standpoint.
3 From an overall standpoint, I think what we found 4 in the questionnaire, and it shows out as you just briefly 5 go through the questionnaire results, is that there is a lot 6 of variability out there. There is a lot of ranges. We see 7 that in the scope of the programs and the practices, in the 8 depth and in quality. We don't see things alike.
9 And let me just depart by saying a little bit, 10 that I am involved on the side in the Army maintenance 11 program, and there you see a lot of consistency in the 12 regulations that come down and are followed.
13 MR. REED: Why do you feel the military, Army, 14 Navy, have this consistency? Is there a reason for that?
15 ,
MR. GRENIER: I believe the reason is because it 16 is headed by one organization. For example, the procedures 17 and regulations come out of one -- I forget which component 18 it is, but they issue Army-wide --
19 MR. REED: Let's try another tack.
20 Supposing you have rapid turnover of your 21 people. Every three or four years, you have a whole new 22 crew, which is a lot of what the Army and the Navy is like.
23 If you got rapid turnover, and let's say not great in-depth 24 involvement in the repair anyway, if it really gets sticky, D'/ 25 you get a vendor rep, don't you think natural evolution ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 804336-6M6
8460 01 14 15 MMbw I would bring you into proceduralized consistency?
2 MR. GRENIER: I think that is maybe what we find 3 the military has been around and has learned for many years.
4 Like you said, it is an evolution. They have got turnover 5 and there is also a main corps of people who stay.
6 MR. REED: Most of the guys that are doing te 7 work are turned over every four years. And that is -- we 8 will call them the swabbies, that are down there tightening 9 and loosening and moving the equipment and greasing it, and 10 so on. They are turning over. And I think that is the key 11 to something you have to keep in mind as we go through our 12 discussion.
13 MR. MICHELSON: They don't call oa vendor reps, 14 though, in the nuclear Navy. For instance, in the 15 submarines, they have to do everything themselves.
16 MR. REED: They limp back to shore, and then they 17 call the vendor reps.
18 MR. MICHELSON: They don't like to do that 19 either unless it is absolutely necessary. They apparently 20 do everything they possibly can to stay out long periods of 21 time with the crew they have aboard.
22 i
23 24 25 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(Xh3346646
8460 02 01 16
(^
A )- MM/bc s 1 MR. GRENIER: Nevertheless, that is one point I 2 think to keep in mind this afternoon, particularly in the 3 survey, which I think is the main part of our program, that 4 there is a lot of variability and, to some degree, 5 inconsistencies among the programs -- not that they're 6 necessarily negative, because we've got to adapt to the 7 situation. .
8 (Slide.)
9 The first slide I have is taken from the volume 10 two of 12, which shows the variance that we find by 11 percentage on just the management structure. And I think 12 the slide sort of speaks for itself on the variability that 13 you will find out there.
14 MR. WYLIE: Let me ask a question about that. Is 15 that at the plant level or the corporate level?
16 MR. GRENIER: This is at the plant level.
17 MR. WYLIE: That wasn't real clear, I think, when 18 you read through this thing, at times, whether or not you 19 were talking corporate or whether you were talking plant.
20 Now, some utilities have a corporate -- well, I 21 call it Nuclear Production Department level, a separate 22 maintenance organization that does the heavy maintenance 23 work.
24 And then the plants have a light maintenance 25 work, the other maintenance work. It depends on the ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8 6 336-6646
8460 02 02 17 I MM/bc 1 utility. You must not have struck on one of those 2 utilities.
3 MR. GRENIER: I'm glad you said it depends on the 4 utility. That's what we find even in our indepth survey, 5 that, when we cover some of those points, you'll find their 6 approach does vary.
7 There's a corporate level, maintenance; a lot of 8 training occurs at that level. .
9 MR. WYLIE: It depends how heavily involved they 10 are in the nuclear prog' ram, where they've got one reactor 11 plant, or whether they've got many reactor plants.
12 MR. GRENIER: From an overall perspective, the
\- 13 questionnaires were sent to the senior resident inspector 14 and it was a three-unit or a two-unit site. Only one 15 questionnaire was completed for the program at that site, 16 but it was geared to the plant level and not to the utility 17 level.
18 MR. EBERSOLE: Of all those organizations up 19 there, do you have any observations to make about which ones 20 seem to work the best?
t 21 (Slide.)
22 MR. GRENIER: Let me reemphasize what I think l l
23 Greg said with respect to the depth of our survey. Phase 24 one was to go out and find out, without going too far and 25 makino some judgment. So we have decided we will keep 1
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 800-336-6646 l
8460 02 03 18 M
(\_/ MM/bc 1 pretty much the statements as observation, rather than not.
2 But let me just quickly add John will cover this 3 in phase two activity. What we plan on doing is going into 4 areas that we feel may give us some clues about what's 5 better and what's not.
6 But our preliminary review shows that it does 7 vary on the organization and in the type of plant and, like 8 you said, the number of -- the commitment of the utility to 9 nuclear.
10 And I think we will find, as we often hear about, 11 trying to standardize the design. And since there are so 12 many variations out there, I think we will find the same way O/ 13 that it will be very difficult to be prescriptive, even in 14 organizational structure. What works for one may not work 15 for the others.
16 MR. EBERSOLE: The degree of compartmentalization 17 of the maintenance process, I think, once you get down to 18 defining the range that you see, I think it's going to be 19 interesting to find out just to what degree that 20 compartmentalization ought to optimize the intermixing of 21 all those things; I&C, electrical and mechanical.
22 I'm inclined to think, to the extent it's 23 possible, if you didn't have any discrete departments at 24 all, that you made them all work to face to face, you'd be 25 better off.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8@33M646
i-8460 02 04 19 1
(~s . l
(_)MM/bc 1 MR. REED: I might point out you've got the '
2 different organizations. They occur probably from history 3 of how the company has been organized in the plants in the 4 past. When I use the word " organized", I mean not only what 5 the management wanted, but what the union wanted.
6 The thing you just mentioned, Jesse, I don't know 7 that the government could pass a rule that everybody be 8 classified the same and could work in such a manner. So 9 it's going to be tough to cause organizations to become 10 consistent.
11 MR. GRENIER: Let me make a side note. I think 12 part of what we found in doing the first phase, in getting 13 involved with maintenance as a regulatory agency, federal l
l 14 agency, is the attention now that we are fostering. And l
15 many times, just by bringing attention to something, you can 16 cause improvement.
l l
l 17 And we have sent this report to all of the 18 utilities and what we hope as a side benefit, if not a 19 direct benefit, is that by reading some of the observations, -
20 that there may be some different ideas about how they could 21 organize their program.
22 And I have a slide. We find out in most of the 23 organizations -- in fact, my first notation is that the 24 typical organization is to have a separate maintenance 25 department reporting to the plant manager that has three ACE-FEDERAL REPORTERS, INC.
202 Nationwide Coverage 800-336-6646
_ _ . . _ . _ , _ . _ _-347-3700 . _ _ _ . - . _ _ _ . _ _ _ _ _ _ _ . . . _ . _ . _ _ . _ . . . _ . _
8460 02 05 20 (3
x_/ MM/bc 1 subsections -- mechanical, electrical and I&C. Even though 2 other people are organized differently, it may be that other 3 people might get some ideas and say, gee, if that's the 4 preponderance, maybe we ought to give it a try, or 5 whatever.
6 So it's sort of a side benefit that we might find 7 by exposing people to this phase one findings. So that's 8 typically what we found, is that type of an organization.
9 Fifty-six percent of the managers do report directly to the 10 plant manager; yet, 44 percent report to either the 11 operations or to a secondary officer at the plabt.
12 So, again, the variability comes in.
13 Programmatically, as the first bullet shows, approximately 14 one-fifth of the plants have extensive preventive 15 maintenance programs. Half of them or so have adequate 16 programs, and 22 percent have minimal programs in preventive 17 maintenance.
18 Another finding we have from the residents is 70 l
19 percent spent a quarter of their time or less on their PM I
20 activities. All of them said they are aiming for a lot i
I 21 higher during the site's -- as Tommie will discuss this.
22 But, again, we find this preventive maintenance l
! 23 is certainly an area that appears to need attention; and, 24 certainly one that we might be looking to in phase two.
\' 25 MR. MICHELSON: Part of the preventive ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 2336-6M6
8460 02 06 -21 MM/bc 1 maintenance, of course, is a vendor resporsibility to the 2 -extent that he must tell the utilities what things need to 3 be done on a routine basis.
. 4 Some of these you can speculate even without a 5 . manual, or whatever. But, some of them, the vendor has to 6 tell the utility:
7
- Have you looked into how the quality of 8 information being transmitted to the utilities in order to 9 assure that there will be appropriate preventive maintenance 10 programs-for some of these componente?
11 MR. GRENIER: We didn' t go into detail. This 3 12 gets a little bit into Tommie --
i i
13 MR. MICHELSON: That's part of the problem on the 14 scram breaker, for instance, is the vendor of the product 15 really didn't tell the utility what they had to do.
16 MR. LE: I guess, during the site visit, we did 17 look into -- we did look into the interfacing between the 18 utility and the vendor. We found out that because of the 19 Salem event and the NRC generic letter 83-23 that we've put 20 out, the utility, especially the plant personnel, is aware 21 of the vendor, later information and the updated 22 information.
23 At a couple of plants we found out that they have 24 a whole new procedure like stopping a new unit to work with
! O' 25 the contract people to require the vendor to automatically ACE-FEDERAL REPORTERS, INC.
202-347-3700 Naticnwide Coverage 80 4 336-6646
. . - . . _ - _ . ~ . - . . - . . . . . _ . . _ . _ . _ . _ . _ - , . . . _ .
4 8460 02 07 22 MM/bc 1 send in the updated technical information as part of the -
2 requisition, for instance.
3 So there is an awareness of the vendor 4 involvement. On the other hand, there's the cost involved.
5 So we found out during the survey, the utility management 6 would say:
i 7 I'd rather train my own engineers and send them 8 to school and come back and maintain the piece of equipment.
9 And that's another trend that we see, in 10 improvement for the training of-the plant engineer.
i 11 MR. MICHELSON: Maybe you missed the point of my 12 question. And, probably, I just didn't ask it very well.
13 My real concern is: Does the vendor spend an 14 adequate amount of time to appropriately understand his own 15 product and tell the utility what needs to be done in terms 16 of preventive maintenance on it?
17 MR. LE: I think that's more an attitude of the 18 utility management. How long they want the vendor to come?
19 The vendor always wants to come in because they are -- they 20 aet paid on a time basis.
21 MR. MICHELSON: Have you looked into the question i
22 of the adequacy of the vendor's instructions? And are you 23 satisfied the vendor is telling the utility, and it's just 24 that utility hasn't set up a good enough program yet? Or is 25 part of the problem with the vendor? Also, that's another i
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 804336-646
_ _ _ _ _ . . _ _ _ . ~ _ _ _ _ . . . , _ . . __ . _ . _ _ . . _ ~ , _ .__ ._ _ _ _ . . _ _ _ , _ _ , . _ _ . - - - _ _ _ .
8460 02 08 23
(")T
\_ MM/bc 1 way of asking the question.
2 MR. CWALINA: Again, what we did here was a 3 programmatic look. We didn't go into the details of the 4 adequacy of vendor information. We didn't have time to look 5 into that type of thing.
6 We looked at how the licensee handles vendor 7 information. Do they take vendor information, incorporate 8 it into their procedures? Do they attach it to their 9 procedures? How do they handle it?
10 We didn' t look at that.
11 MR. MICHELSON: Do you think there's any question 12 about the quality of information being given to the utility?
r' k-- 13 or is it fully adequate and now a question of 14 implementation?
15 MR. CWALINA: Again, I think that's just like the 16 licensee's programs in that it varies frcm vendor to 17 vendor. There's a few vendors that provide quite a bit of 18 information. There's a few that don't provide any. There's 19 other problems in licensees where the vendor information 20 will come to the corporate headquarters, but it never go?.s 21 down to the plant.
22 So there's trouble in vendor information, there's 23 no question about that. The adequacy of it, we didn't 24 really look into.
O 25 MR. GRENIER: Okay. To move on, we found that ACE-FEDERAL REPORTERS, INC.
202-347 3700 Natioawide Coserage 804336-6M6
8460 02 09 24 MM/bc 1 most of tbe plants, 75 percent of the residents have a 2 formal program in predictive maintenance. Yet, we found it 1
3 was minimally performed at most plants.
4 And 25 percent of the plants don' t have a 5 predictive maintenance program. Now, from the questions, we 6 know that they do some oil analysis, monitoring, vibration 7 monitoring. That does exist. But it's not a formalized 8 procsss. And it's certainly an area that needs attention.
9 MR. EBERSOLE: What's the difference between 10 " predictive" and " preventive"?
11 MR. JANKOVICH: I'm John Jankovich of the Staf f.
12 " Preventive maintenance" is defined as a time-based service, 13 just like in an automobile. And that's how we defined it in 14 our program, on our Glossary in volume II. And this is in 15 correspondence with the import definition of " preventive 16 maintenance program".
17 " Predictive maintenance" is based on performance
- l 18 of the equipment. Therefore, it requires measuring 19 the performance of the equipment. Measurement of 4
20 demography. Measuring the temperature of bearings.
21 Vibration analysis. Performance of motor-operated valves.
22 So these are periodically checked, records 23 checked. And if there is a deviation, then preventive 24 maintenance is performed.
25 MR. EBERSOLE: Preventive is more or less ACE-FEDERAL REPORTERS, INC.
202-347 3700 Naticnwide Cosersge N 33 N
8460 02 10 '25 t-
.; (.))'MM/bc 1 planned.
2 MR. JANKOVICH: Right.
3 MR..EBERSOLE: Depending on what you find when 4 you open it up, well, about everything can be predicted very 5 well, can't it?
6 MR. JANKOVICH: Correct.
7 MR. GRENIER: Again, let me say as I go through 8 these, some of the same questions I'm sure you have are 9 questions that we are now asking ourselves.
10 Again, phase one is just what is out there. As 11 we went'in, we asked the same questions. And that's what 12 we're studying now for phase two.
O k- 13 Again, from a programmatic standpoint and 14 organization administration, we found that the majority of 15 the plants do have a formal system for maintenance 16 performance. And that's good.
17 Yet, 40 percent of them do not have, and they 18 have trouble correlating with performance measures; finding l 19 out really what's good and what's not good. It's really not 20 done.
j 21 And if you look at it, I don't know if you're l
22 familiar with during our development of the program. We t 23 talked about a model, a maintenance model, and one of the l
l 24 important keys of maintenance model program is to have l
l 25 feedback.
l 1
l ACE-FEDERAL REPORTERS, INC.
I 202 Nationwide Coverage 8M336 6M6
-347-3700. , , _ _
8460 02 11 26 MM/bc 1 And this is where evaluation of the utilities 2 program -- is it effective? We found again 40 percent do 3 not have that.
4 Facilities and equipment. This area, we found 5 was probably the least problemmatic and one that probably we 6 will not look into in any detail. It's not to say that 7 there are no problems there.
8 Most plants have adequate facilities and 9 equipment to get the job done. But the biggest area, the 10 big problem there, is lay down space and equipment 11 accessibility.
12 But, when we talked to the people again more 13 specifically, we found it was not a major problem; they can 14 adapt to the situation.
j 15 Maintenance people were able to get the job l 16 done. A quarter of the plants have that specific problem, j 17 as identified by the residents. Twenty-five percent-of I
l 18 them, availability, accessibility and timeliness in l
l 19 obtaining spare parts has caused delays at facilities to l
l 20 varying degrees.
21 And I guess the paperwork is what we heard when l 22 we went out there was the cause of that. Sometimes, they 23 had it but getting to it, and getting to it in time, has 24 caused delays.
O 25 According to the residents, they sometimes have ACE-FEDERAL REPORTERS, INC.
l 202 Nationwide Coverage 80lk3366646
.. _ .-- ... -347-3700 --
8460 02 12 27
/3
(_/ MM/bc 1 the plant operated in an LCO in the past year because of 2 spare parts.
3 Another item in utilities that will be discussed 4 later is that while most plants do have labels on their 5 equipment, staff at 40 percent of the plants were reported 6 to add four more labels, which gives you -- raises some 7 questions about the labeling in the industry.
8 And Drew will be talking about that specific ,
9 study on labeling in a short while.
10 (Slide.)
11 Again, continuing on the five areas, the next 12 area we looked at was Procedures. Again, most plants use a O
\> 13 writer's guide primarily for input in writing their 14 procedures. And half of them use contractors.
15 We also found the maintenance engineers or 16 maintenance supervisors and managers are the ones that l
l 17 prepare the procedures, rather than the technical writers.
l
( 18 Most of the plants systematically validate their 19 procedures. Again, we find a small percentage, 15 percent, 20 do not do this routinely.
21 And, again, as you look at the questionnaire, you 22 always see the variability that comes across. There's a 23 number of plants who do not. And that's where we will be 24 concentrating in the phase two, where those problem areas
- (1) 25 are.
ACE-FEDERAL REPORTERS, INC.
1 202 347-3700 Nationwide Coserage 800-336-6M6
8460 02 13 28
,~
(,/MM/bc 1 All plants, you know, usually review and upate 2 their procedures. Only half of them have a formal process 3 to do that. I relate-that back to management and 4 organization, as well, the attitude of the management 5 towards maintenance.
6 And, again, the use of procedures. The use of 7 procedures varies. It varies by the type, you know, whether 8 it's preventive or corrective. But, most of the time, in 9 safety-related work, we find that they do use the procedures 10 as required.
11 But, in nonsafety-related work, I think, if I 12 recall, about 10 percent of the time is when procedures are
( 13 used.
14 MR. MICHELSON: They use 10 percent of the time?
15 MR. GRENIER: Ten percent of the time, as 16 reported by the residents, from their observations.
17 MR. MICHELSON: What do you mean "used"? It's 18 one thing to read the procedure, then go out and do the 19 work; it's another thing to go out and use a procedure and 20 check off item by item, and so forth.
21 So what do you mean, roughly?
22 MR. GRENIER: Even within that, it does vary from 23 the packaging, from the time they start off, do they have 24 all the right materials to perform it, to actually hold O(-) 25 points while they're performing it, or even a followup by ACE-FEDERAL REPORTERS, INC.
202 Nationwide Coserage 80k3346646
_ - . . . . _ _ . _ _ _ .._. _ -347-3700 . _ _ _ _ . _
1 l
1 l
8460 02 14 29
)MM/bc 1 the OA people.
2 MR. MICHELSON: On nonsafety-related equipment, 3 you don't run into all of that unless there is a company 4 procedure that requires OA on nonsafety equipment.
5 MR. GRENIER: That's what I'm saying. From our 6 survey, we found in only 10 percent of the cases where they 7 actually followed the procedures -- procedures are used in l
8 nonsafety-related work, maintenance work.
9 In the area of personnel, we found that the total 10 size of our one-unit facility maintenance staff is about a 11 hundred, and about 140 for a two-unit site. And each 12 increased by -- they increased by 50 percent, then 100 13 percent respectively during the outages.
14 MR. WYLIE: Let me ask you about that. I know 15 some utilities contract a lot of their people outside, a 16 lot of their maintenance outside, or from other elements of 17 their own organization.
t i
18 Does that account in the difference from 100 to
, 19 140?
?
20 MR. GRENIER: Yes. It does include people -- I 21 forget what they are called -- who move from site to site.
22 And, many times, they do have the contractors come in. So 23 that's where the increase comes in.
24 What we're saying is here is a normal staff of 25 100 people. You will find during the outages that there is, ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage sn336-6646
8460 02 15 30 r
km) MM/bc 1 you know, 50 more people at it from a combination of 2 contractors and utility --
3 MR. WYLIE: This doesn' t imply that one company 4 does it with 100 people and the other does it with 140? In 5 fact, they actually contract for their services?
6 MR. GRENIER: Tha t's correct.
7 MR. REED: In your report, I notice a couple of 8 plants on the plans. Two or three of them run up to 300 9 people.
10 I also noticed in this Battelle report that the 11 smaller -- it sort of says the smaller the maintenance 12 staff, the better the maintenance is performed. That's kind 13 of interesting.
14 MR. WYLIE: Maybe they used a lot of contract 15 people.
16 MR. GRENIER: For this particular bullet, we just 17 thought we'd give people an average of what we found.
18 We also found that about half of them have people i
19 on site af ter normal day shif t hours and half are on call.
20 This is again an observation.
I 21 Almost in all plants, the craft workers spend 22 10-20 percent of their time in some sort of training. And 23 virtually all plants have classroom training.
t 24 But, the area we found the weakest in is the l')
k' 25 requalification training, and training on plant ACE-FEDERAL REPORTERS, INC.
3 4 347 3700 Nationside Coscrage 2 336-6686 g-ge, , . . -v., + - - , + - - . ,
- -- - -----+,-w------s--
,. -*- ,,%-- y .--.-e-i--
8460 02 16 31 O
V MM/bc 1 modification. In fact, only about a third of the plants do 2 have a formal process for a requalification or training on 3 plant modifications.
4 Typical craftmen work overtime. That's always an 5 issue that sometimes comes up -- too much overcome. Does 6 overtime have an effect? And from the questionnaire, we 7 found that overtime is not a problem, that maintenance 8 people do like to work a few hours of overtime and it does 9 not impact on their work.
10 It is average.
11 12 13 14 l 15 l 16 l
l 17 18 i
l 19 l
l 20 l
l 21 l 22 L
l 23
! 24 25 ACE-FEDERAL REPORTERS, INC.
! -- . -. -_ -_ 2"'"?"'__. _ _ f *""""!"*1 :"*** _ _ ___ _"!'"". - _ _ .. _ .- -
8460 03 01 32 O
's_) MMbw 1 MR. REED: I worry about that finding, that 2 maintenance people like to work overtime. They may like to 3 sit in the cafeteria on overtime pay, but watch out for the 4 difference between work and dog robbing. That is a very 5 tricky one. If you follow up in Phase 2 of that, you better 6 watch it very carefully. In fact, many companies went a 7 lona time about whether they are going to have in refueling 8 outages, ten-hour shifts, two tens, two twelves, three 9 eights, or what not. And it all relates to the overtime 10 game, and whether they are going to work seven days a week 11 or six and just declare a holiday one day a week.
12 (Slide.)
13 MR. GRENIER: The fifth area is work control. In 14 most plants, we found they have some automation, 15 particularly in the scheduling, but again, we found a lot of 16 variability. Some plants, as reported by the residents, 17 have everything automated. They even have, when the work 18 order is submitted, it comes up on the manager's desk,
- 19 maintenance manager's desk, completely automated, with all 20 of the needed information, where to get it, what equipment 21 or tools are needed and work documentation for fcllow-up is 22 all computerized.
23 Other plants we find complete manual system, I
24 complete manual system. But you know, with today's computer 25 tools available, we see that, and even in our site surveys, 1
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 80 4 334 6646
8460 03 02 33 O
\_f MMbw 1 we found people are increasingly using automated systems.
2 MR. MICHELSON: Do you find any tendency to use 3 certain people for, shall we say, safety-related work and 4 others for nonsafety? Was there any indication of that kind 5 of discrimination?
6 MR. GRENIER: We did not have that kind of
, 7 question here, but we did get into it on the site survey.
8 If Tommie wants to respond to that.
9 MR. LE: Yes, During the site survey, we did 10 check into the qualification of the people who work on 11 safety-related by procedure. The plant had to call out 12 what kind of publication to work out --
13 MR. MICHELSON: Do they classify people as being 14 qualified to work on safety versus nonsafety?
15 MR. LE: They have a classification like Class 1, 16 Class 2 and journeyman.
17 MR. MICHELSON: That is a union classification, 18 generally. What I am asking is, within that classification, 19 do they have people who they put on safety-related and so 20 identify it, as opposed to nonsafety?
21 MR. LE: At the majority of the plants that I've 22 been to, the supervisor, when he receives the work order, he 23 looks at the job scope and finds out which of these workers 24 is qualified, and he would assign them. So if he had a
( 25 tendency to assign the best-qualified worker to the job ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-3364M6
8460 03 03 34
()MMbw 1 procedure requirement for equipment qualification -- I don't 2 mean qualification, in the sense of EO, but the work 3 qualified for diesel or for valves. At certain plants, they 4 do have dedicated crafts to work on a certain piece of 5 eculpment.
6 MR. MICHELSON: I don't know if this answers my 7 question or not, though.
8 MR. LE: I don't know how to say --
9 MR. MICHELSON: I just wondered if you were able 10 to determine. You know, you've got some people even at a 11 qiven level in your union progression of skills there are 12 some people who are better. They are not all equal, just (O./ 13 because they have reached that level.
14 Do you -- is there any attempt of the utilities?
15 Do you know, Glenn?
16 MR. REED: As he says, the first thing that makes 17 the decision for the foreman on assignment is 18 qualification. What does the job require?
19 MR. MICHELSON: By qualification, it means union 20 level -- journeymen as needed for the job, or whatever? t 21 MR. REED: First class, second class, whatever.
22 Or whether he is a welder, heliarc, whatever it is. He 23 makes qualifications. The second thing has to do with 24 radiation equalization. What's his radiation exposure on 25 this auy at this point or these guys? And that ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationside Coserage 800-336- % 86
.8460 03 04 35
(~)
(./ MMbw I censrally --
2 MR. MICHELSON: Does skill ever come into it, or 3 does it assume if you've reached the required level that you 4 are automatically skilled?
5 MR. REED: Objective evaluation on skill doesn't 6 come into it, and we will probably run into trouble.
7 MR. MICHELSON: You wouldn't dare, I guess.
8 MR. REED: What with equalization requirements?
9 MR. MICHELSON: Yes. So that the union 10 requirements, the level that you've worked up through under 11 union promotional scheme is the measure of skill.
12 MR. REED: Yes.
13 MR. MICHELSON: That varies a little bit from 14 union to union too, as to what the requirements are, and so 15 forth, to each of them. That's not on a national basis.
16 MR. REED: In training them, you've got a 17 check-off list that goes with the training, if you have 18 heltarc weldina, and so forth, that's all on the list.
19 MR. WYLIE: You have to remember, a lot of 20 utilities are not using lists.
21 MR. MICHELSON: I don't know what they do.
22 They've got their own method of grade identification and 23 skill identification.
24 MR. GRENIER: By no means scientific, but I did 25 ao through one site survey, and I did -- I believe that's ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800 336-6M6
8460 03 05 36
(,) MMbw 1 one of the areas in protocol, one of the site surveys we 2 had. We talked to the maintenance manager there. He did a 3 very sensitive and very complex job that he did assign some 4 specific people. Usually had one of his key leaders 5 participating throughout the maintenance repair.
6 I know, as a maintenance manager myself, if I 7 have a problem in the military -- I am in the Army Reserves 8 -- that I do pick out the people that I want to make sure 9 perform the job. Of course, we don't have the union to 10 contend with there.
11 Okay. To move on, backlog is an area that is 12 beina looked at as a performance measure, but let me point O
kJ 13 out that it is an area that could be deceiving. I just 14 heard a recent advertisement about Dr. Scho11's foot pads, 15 where a stewardess says that she often has to spend a lot of 16 time on her feet. In fact, some days she spends 9000 miles 17 on her feet.
18 The point here is that when we looked at the 19 backloas, we found that even though some plants were 20 characterized as having a heavy backlog, we asked the 21 auestion, how much time would it take you to work it off?
22 And I believe it was Brunswick -- correct me, if I an wrong, 23 Tommie -- they run 600 work order backlog, but we also found 24 they said it takes two weeks, if everything stopped two sb
%> 25 weeks, they could work it off. That's because they have ACE-FEDERAL REPORTERS, INC.
mm. m_ m_,. - i
8460 03 06 37 iQ iust about everything written up on a single -- not on a
(_) MMbw 1 2 single, but individual work orders, even changing a light 3 bulb would be on one and doing the valve and other kinds of 4 services are all individual backlogs.
5 So iust a cautionary note. When you hear about 6 high backlogs, that there may be a different 7 interpretation. We talked about the procedures, use of 8 procedures, and a OC is where, as I point out in the last 9 bullet here, where again, it is extensively used in 10 safety -related work, but again, to various degrees, 11 dependina on the type, a lot of variance out there on the 12 use of procedures -- I mean on the use of quality control.
O 5- 13 The last item I noted just to mention to you is 14 actual time spent does vary on that work control that 15 particular plant has.
16 ,
Some plants we found that the actual packaging is 17 done by other than the maintenance personnel, where the work 18 order is prepared, there is a different group of people who 19 actually get the tools, equipment and parts, whatever is 20 needed, such that the people are spending 75 percent at the 21 work site, but we found through the questionnaire that where 22 the maintenance personnel has to do all of the packaging, 23 that they only spend about 50 percent at the work site.
24 That is sort of a interesting finding.
I
\
25 MR. REED: I wished on that OC item, you had a ACE-FEDERAL REPORTERS, INC.
202-347 37(X) Nationwide coserage Mxk346646
l 8460 03 07 38 m
i,_) MMbw s 1 question in your questionnaire having to do with, is the OC 2 performed by line foremen and leaders, or is it performed by 3 clerical personnel from the front office.
4 I think there is a big safety and efficiency 5 aspect to who really performs QC. Also an adversarial 6 climate aspect. So you didn't answer the question in the 7 questionnaire, but it would have been a nice one to ask.
8 MR. GRENIER: I thin we did have something along 9 those lines of the site surveys, and we found -- and again, 10 correct me, Tommie, since you are newly involved in the site 11 surveys, that most of the plants have a OC department with 12 engineers that do the follow-up work.
(,
( 13 -
MR. REED: That's not answering my question.
14 You're always answering QA. I am talking about OC.
15 Was it done by actual crafts people who had been 16 craft people or are even now craft people doing other 17 people's work, or is it done by a separate group that may be 18 clerical people?
19 MR. LE: Glenn, during our site survey out of 20 curiosity, we didn't ask what background did the QC 21 inspector have. We found out, some OC people have 22 background in maintenance and operation. So some had been 23 out in the plant. They look at the equipment. The trend is 24 that because OC is more or less than many time job, that is 25 why a lot of maintenance people who get tired of overtime ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8 % 336-6646
8460 03 08 39
,o
(_) MMbw I would move to QC spots.
2 So in some plants, we do find that OC people 3 were experienced linemen and operators before.
4 MR. REED: That's good, if you find that. You 5 said some. That would be an interesting thing to follow up 6 in Phase 2.
7 MR. GRENIER: Again, as we opened up, there were 8 some 139 questions, about 100 responses. A lot of 9 information there, and our task was really to sort of 10 summarize that to see where we are and establish the 11 baseline that Greg mentioned. And I think this 12 questionnaire and the cooperation we got from the residents, O
k/ 13 you know, we did meet our objective, as we continually 14 pointed out here in the last half hour. Now we've got to 15 digest some of this material, and we do_ plan to maybe even 16 have a follow up in certain of the areas and see if we see 17 some improvement. But at least we've got a baseline 18 information established.
19 And I think it was a very successful project.
l 20 Unless there are other questions, we will move on 21 to the site surveys.
22 MR. REED: Just one point. Your Executive 23 Summary and your documents that are on NUREG 12-12 really 24 -focus on findings and conclusions, and that is what you've O
N/ 25 been talkina about. Now you are faced with, how do you ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8M336-6M6
8460 03 09 40
(')
(_/ MMbw 1 turn these statistics, findings, conclusions, data, how do 2 you turn that into judgments that are sound? That is going 3 to be tough.
4 Is that what Phase 2 is all about?
5 MR. CWALINA: That is right. We will discuss a 6 little bit of Phase 2 later.
7 MR. GRENIER: Thank you very much.
8 MR. REED: The next speaker looks to me to be 9 Mr. Le. Site survey observations.
10 (Slide.)
11 MR. LE: My name is Tommie Le. I am with the 12 Staff. I am here today to brief you on the results that we 13 found during the site survey.
14 A previous speaker has revealed to you that there 15 were eight sites. Six of them were originally from a plan
- 16 and two were the additional, due to events that happened.
17 Most of the findings that Bernie has talked to l
18 vou about are pretty well concurred in by the data from the 19 site survey that we performed. There were five areas that 1
1 20 we look at, as discussed by Bernie.
l 21 (Slide.)
22 Most of these are described in the trip report in 23 Appendix C, but I will try to walk throuah here on the
! 24 highlights of those that we think are put into the 25 findinas.
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 80fk33MM6 L - . _ . _
l 1
l I
8460 03 10 41 j MMbw 1 Under organization and administration, we have 2 subcategorized it into different areas, like we wanted to 3 find out how the organization, maintenance and 4 structurewise, had been evolved.
5 Just to give you a typical example, on page C39 6- of your NUREG, Volume 2, there, typically we show Brunswick, 7 how it has evolved from 1977 all the way up to 1985. C39, 8 if you care to look at that.
9 We found out back in 1977 and '80, the 10 organizationwide, I&C and electrical together with a very 11 small staff and group of workers and mechanical maintenance 12 supervisor, and two who report independently to the plant O
k/ 13 manager.
14 From -- on page C39, Appendix C again, from 1972 15 to '77, again is portrayed that because communication is 16 needed to be filtered before we go to the plant 17 superintendent, so they add a maintenance supervisor between 18 the two groups, and followed up with the next page, C40, the 19 maintenance organization in '80 to '82, found out that the 20 plant -- the work in the plant has multipled, as the age of 21 the plant and the equipment are shown. So they added more 22 supervisors and unit types into Unit 1 and Unit 2.
23 Again, all of the I&C and the electrical together 24 and mechanicals and other identities. So the evolution
)
- 25 remains like somebody had mentioned. They would report to ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8M336-6M6
8460 03 11 42 !
()MMbw 1 one particular manager, before they would go to the general 2 manager in the plant.
3 So that is how we conclude that most of the ,
i 4 plants that we have been at have undergone or recently 5 underwent major changes, particularly Brunswick, the one you 6 look at, is the one that hasn't chanced, that we found.
7 Arkansas was another one that had major changes.
8 I would also point out the separation of the 9 crafts, the administration policy and controls. We found 10 out that most of the eight plants that we've been, formal 11 preventive, corrective and predictive maintenance are 1 12 well-defined in their administrative procedures.
, ( 13 Again, like Bernie had pointed out, most of the 14 staff time is spent more on the corrective maintenance 15 rather than preventive.
16 We talked to some of the plant managers. They 17 feel that the backloa has somethina to do with the time that 18 they have to devote on the corrective rather than 19 preventive, but in the future, they would like to have a 20 percentage of 50 to 90 percent on preventive maintenance.
21 But the realisteic coal would bne 40 to 60, according to one 22 vice president.
23 MR. MICHELSON: Let me ask you a question. You
! 24 go out with a survey, and you collect all these good facts.
h 25 Then you sit down and look at them. What do you do with ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nmonwide Coverare 800-3W664
8460 03 12 43 o
I(./
) MMbw 1 them then?
2 So these are all interesting.
3 Do you attempt to correlate or find or search 4 for, shall we say, good practices and somehow, based on 5 quality of performance of the utility or something, or are 6 you just trying to collect a lot of facts?
7 MR. CWALINA: The purpose of this was to collect 8 a lot of facts.
9 What we intend to do with them and what we are 10 starting in the process, I was looking at all of these 11 things that we gathered in detail and see if we.can identify 12 particular problem areas. As Bernie said, when he was up 13 there, facilities and equipment doesn't seem to be a big 14 problem. The plants can cope with what they've got.
15 MR. MICHELSON: How do you define that you've got 16 a problem? What do you look for to say there is a problem?
17 MR. CWALINA: We're also merging into performance 18 indicative worth that Pete has done, SALP reports,
! 19 qualitative reports, inspection reports, LER summary data.
20 MR. MICHELSON: Once you've got some kind of 21 performance criteria, then you can go back and look at the 22 practices of that utility to see if they somehow relate or i 23 are causal factors in performance?
! 24 MR. CWALINA: Right. For instance, we will look l
25 at a plant that we feel -- take preventive maintenance, for l
ACE-FEDERAL REPORTERS, INC.
202 Nationwide Coscrage 8M33M684 l _ _ -M7 3700_ . _ _ _ _ _ . _ _ _
8460 03 13 44
()MMbw 1 example. Based on the SALP reports and the maintenance 2 measures, LER data, component failure data, we will go back 3 .and wo will look at plants that seem to have good 4 performance or bad performance.
5 MR. MICHELSON: I could see the usefulness from 6 what you are doing. I couldn't see it just as a matter of 7 of collecting facts, figure out quite what you were headed 8 for. Thank you.
9 MR. REED: One question. I. notice we have 10 preventive maintenance, corrective maintenance and 11 predictive maintenance. But one thing that is handled by 12 maintenance departments -- at least a larae amount of it is 13 handled by maintenance, is backfitting and modification. So 14 that doesn't show there. And I am just wonderino where it 15 is slotted. Is it under the three that you have?
16 I would like a comment. And on the Brunswick
, 17 changes in organization, I will make you a bet that most of I
! 18 those changes came about as a consequence of modification 19 backfitting.
i 20 Where do you have modification backfitting? What 21 is it under?
l 22 MR. LE: That is an interesting thing that you 23 point out, Glenn.
(
24 on the first protocol that we went out, we didn't 25 have much on modification and outaces, and the plant ACE-FEDERAL REPORTERS, INC.
202 347-37(x) Nationwide Coserage Rxk33&M 86
8460 03 14 45 MMbw 1 manager at Kiwani pointed that out to us. So we went back 2 and put more questions on it. As far as the plant 3 modification, we were talkino under like maintainability and 4 equipment improvement. That would appear under facility and 5 equipment as the next phase.
6 MR. REED: Oh, it is not under this?
7 MR. LE: Not on here.
8 MR. REED: These workloads and these percentages 9 --
10 MR. LE: Yes.
11 So to go on, another point that I believe Glenn 12 had pointed out, in the smaller plant, the communication 13 tends to improve. At least one manager attributed that his 14 plan has been a small unit and communication was excellent, 15 actually, in Kiwani, was the one that we found. Most of the 16 people at Kiwani, for instance, had been there a long time.
17 The supervisor and manager. So communication again was 18 excellent in the small group.
19
- 20 21 22 23 24 25 ACE-FEDERAL REPORTERS, INC.
m.m , s.e_-_.,. u-
8460 04 01 46
) MM/bc 1 As far as -- we found out, because communication, 2 a lot of plants now have a program, for instance, during the 3 trainina, they would train maintenance people with systems 4 and the operation people with hardware. And how it works 5 when they coordinate on equipment tagout and LCO and so on.
6 They would communicate and understand.
7 And we found that is a good point, that they are 8 promoting communication between the organization. In the O.
9 past, a lot of them did operate independently. We found ou't 10 as far as information and planning, most of the plants are 11 now formina a centralized planning group.
12 Four out of eight plants that we have been in O
k-) 13 have such a aroup. To facilitate the work, they are 14 automated computerizing and work order are extensively 15 used.
16 And it's been in the recent years, the last
- 17 couple of years or so, at Arkansas, they are undergoina an 18 implementation of a very sophisticated computerized system.
19 They even use the mainframe from some people down in New 20 , Orleans to get the data from.
21 We found out to our surprise that the NPRDS 22 information has not been very well-utilized by these plants
(
23 that we visit, for the reason that some of the equipment i
! 24 they have are not in the data base. Some of them have some
(/ 25 problems extractina the information applicable to the
}
l l
l ACE-FEDERAL REPORTERS, INC.
202 347 1700 Nationmide Cmerage M G336 % m
_ . _ _ . _ _ _ . _ . _ . _ _ _ ,_ _ _ _ _ . . _ . _ . , _ . , _ _ _ _ _ . _ _ _ . . . . _ . . _ _ . _ _ . _ . . . , _ . _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ . , ~ . _ . .
I l
l 8460 04 02 47 l
/~N
(_) MM/bc 1 equipment.
2 MR. MICHELSON: How do they extract their 3 information?
4 MR. LE: I think, if they have a motor and valve 5 problem, they would go and call for a motor and valve under 6 the same vendor name.
7 MR. MICHELSON: Do they have on line cool up 8 capability to the computer?
9 MR. LE: Those plants that we've been to, they 10 do. They talk to INPO.
11 MR. MICHELSON: That's not what I said. Do they 12 talk to the computer, or do they talk to INPO and then INPO O
-l 13 does the searches for them? How does it work? How are the 14 utilities really gettina in contact with the data base?
15 Are they doing it through a third party or are 16 they doinc it directly? Because it is, of course, capable 17 of on line searching. But it takes a little bit of training
! 18 and whatever. I was iust curious as to whether utilities 19 are training their personnel to use it, or just asking I
20 INPO.
I 21 MR. LE: I think, at one plant, I think, 22 Arkansas, I asked that question about how they get the 23 information. And they do go through the direct line, that l
i 24 they can call out.
25 MR. MICHELSON: They search the data base ACE-FEDERAL REPORTERS, INC.
nw. ~ _ + < - , , ,_
8460 04 03 48 O
l ) MM/bc s
m 1 themselves?
2 MR. LE: Yes.
3 MR. MICHELSON: Because that's the only way of 4 really aettino a good feel for what's happening, is to do it 5 yourself. But it takes some training.
6 MR. LE: Yes, that's true. It micht be the 7 people are unfamiliar with that.
8 MR. MICHELSON: Do you find other utilities are 9 using it on line, or did you ask?
10 MR. LE: Only one' plant in particular, at 11 Arkansas, I know they used it. The rest of the plants, I 12 did not check in. But the ceneral question we asked was how I
k/ 13 useful was the NPRDS, or whether you use NPRDS data in your 14 trending?
15 MR. MICHELSON: Did they use the LER data in 16 their trending also?
17 MR. LE: Later on, we found out that all plants 18 are usino the LER, the operated information, to learn from.
19 MR. MICHELSON: But what are they doing? Are 20 they using the INPO data base, or are they using the NRC I
21 data base on the LER's? How are they Qetting their LER 22 information?
23 MR. LE: We didn' t get into the particulars of i
l 24 where the LER source was. I remember one question we asked f 25 was that they allowed an operational event to happen around
' ACE-FEDERAL REPORTERS, INC.
202 347-17m Nathmwide Cmerage m))MM6
._ - - _ _ . . . _ _ _ _ _ _ _ ~ - . _ _ _ _ _ _ . _ _ . _ _ . - . ~ _ _ _ _ _ _ _ _ _ _ . . - . -.
8460 04 04 49
()MM/bc 1 the country.
2 Do they have a mechanism to screen those and 3 filter that information to the people? And they said they 4 do.
5 MR. MICHELSON: You have to ask them more of a 6 question than that. They'll say, yes, we've got a 7 mechanism. But you have to find out what is the mechanism, 8 because, see, the LER screening can be done by INPO, for 9 instance, and the utility can then depend on whatever INPO 10 feeds them back in terms of SER's. And that meets the NRC's 11 requirements and, yes, it is a mechanism but isn't much of 12 what you're looking for here.
13 MR. CWALINA: In general, what the plants do is 14 they look through the LER reports that are generated by the 15 NRC, which is essentially the same as the LER's generated by l 16 the utilities.
i 17 They'll look through it and see if it's 18 applicable to the utilities.
19 MR. MICHELSON: Do the utilities receive every 20 LER?
1 21 MR. CWALINA: No, I don't know if they receive i 22 l them and they can get them if they want. There's also a 23 NUREG that's put out periodically.
24 MR. MICHELSON: They can get them. I'm surprised i 25 if many of them receive all the LER's. That's what I'm l ACE-FEDERAL REPORTERS, INC.
l 2Ll2-347 37(U Nail 0f)*ide CO)ffagt Rilkj)6(446 i
8460 04 05 50 O)
(_ MM/bc 1 asking, how they use LER's.
2 MR. LE: INPO, besides the site variation, they 3 feed the utilities their SER reports and events, evaluation 4 reports.
5 MR. MICHELSON: They feed them the SER's, the 6 SOR's and the OP/RM's and I think that's all the utilities 7 looked at in terms of LER experience. I want to know if 8 that's all they look at? Do they actually get their own 9 material? Do they go into a computer and search?
10 Just what do they do, if anything, beyond looking 11 at the NPO, SER's, LER's --
12 MR. CWALINA: We don't know that.
( 13 MR. MICHELSON: You didn't ask that.
14 MR. LE: We didn't ask that. That's a good thing 15 to do. One thing we found out is during the automation or 16 communication of the data base to help in the maintenance 17 planning, some utilities have a problem going back to their 18 own completed maintenance work order to extract the 19 information.
20 So we asked whether the old information of the 21 equipment would be put on a new computerized system. And 22 the majority of the plants we asked say they do not because 23 of the cost involved.
24 At one plant, the answer was that it might cost 25 about a million dollars just to extract information from one ACE-FEDERAL REPORTERS, INC.
202 347 37m Natkmmkle Cmerage 8533HM6
8460 04 06 51 (3
V MM/bc 1 year data to put it on the computer.
2 If you don't have any questions in that area, 3 those were the highlights that we found under area one.
4 (Slide.)
5 Under the area, the functional area of facilities 6 and equipment, we took the plant tour on most of these plants 7 that we've been to, and I've been at some couple of plants 8 before. I worked for another utility -- and I found out 9 that the housekeeping, most of the area in the turbine 10 buildings, control buildings and auxilliary buildings are 11 very well-maintained.
12 And it seems like they pay attention to 13 cleanliness. We did find that some of these plants that 14 have the space problems, because of the modification, in one 15 particular plant or two that we talked to, they say fire 16 protection requirements on additional equipment might 17 restrict the access to the equipment for maintenance and to 18 plant machinery.
19 MR. MICHELSON: Could you give me an example of 20 how fire protection rotards maintenance? I don't understand l
21 that one. Maybe Glenn -- does that seem reasonable to you, 22 that the fire protection provisions were restricting?
23 MR. REED I'm thinking of the halon system that 24 was installed at a particular plant. You got very many of 25 theso largo bottles and lots of piping that goes around with l
l ACE-FEDERAL REPORTERS, INC.
202 347 37(n Natiimwide rmerage mn34tM6
y 8460 04 07 52 O
(_) MM/bc 1 it. You've got to put it some place and you have a finite 2 structure. So it takes up space.
3 But, to some degree, I don't understand that one 4 too well myself.
5 MR. WYLIE: If you take the oil collection system 6 which was added to the reactor coolant pumps, for example, 7 those have got to be removed for maintenance.
8 MR. MICHELSON: That would be the closest, where 9 the retrofitting --
10 MR. REED: That's a good one.
11 MR. MICHELSON: That would be a good one, a good 12 example. But it's nothing like the problem of seismic
( 13 supports, and that sort of thing?
14 MR. LE: No, it's not a major problem.
15 MR. MICHELSON: I think it would be fairly minor 16 in most cases.
17 MR. LE: Along this line, we also heard comments 18 that the utilities have been putting in additional 19 ventilation for EQ to keep the equipment cool. That also 20 takes up some space to work around the equipment.
21 So those modifications sometimes restrict a 22 facility. We found out that some of the shop and two cribs 23 and contaminated tool storage are adequater if they are 24 cramped, people would learn how to live with it.
25 There are no major problems in that area.
ace-FEDERAL REPORTERS, INC.
x m. n . , m ,o. a. u ,, m ,, mm.
8460 04 08 53
()MM/bc 1 MR. WYLIE: I'd say one of the biggest 2 contributors to lack of space, lay down space --
3 MR. LE: Some plants that we've been to, Kewaunee l 4 seemed to impress me most. And we talked to the plant 5 superintendant there. He said he was involved in the 6 original layout of the plant, like the turbine building.
7 They have a room for every piece of the turbine, that you l 8 can take out and lay on the floor.
9 So if you incorporate the maintainability of 10 equipment during the plant design, that would f acilitate 11 maintenance.
i 12 MR. WYLIE: That's a good point.
n\/ 13 MR. LE: Another thing we found out under the 14 equipment, there was a small concern of maintenance delay 15 due to lack of heavy equipment, like cranes, and so on. But 16 that only happened under normal operation; when a particular
' 17 piece of equipment is out during outage, they had had the i
18 equipment on site.
19 That is part of the outage crew coming in. We 20 found out that the MT&E system is well-maintained and 21 calibrated, either automated control or manual control, to 22 make sure that there are measurements of the tests and 23 equipment are well within the calibrated period.
24 As far as the inventory, it varies. We found out 1
j 25 either by corporate cap on the money, some spare parts are ACE-FEDERAL REPORTERS, INC.
- - " ~ " " -""">
. ._ __ a 2"="
8460 04 09 54 7-)3
(_ MM/bc 1 limited. But at some other plants, we found out that 2 they're in the range of 10 million or more.
3 MR. REED: I don't know that spare parts inventor 4 is going to be an indicator for your future judgments. t 5 There are a lot of games that go on in the spare parts 6 business. Borrowing is a standard operation.
7 In fact, the Kewaunee plant that you went to 8 typically borrows about 1 to 400 pieces of spare parts every 9 time they have a refueling outage. They borrow it from the 10 Point Beach plant.
11 So I don't know what you're going to get out of 12 the spare parts thing with respect to judgments. In fact, O
(_/ 13 some companies stock well and other companies don't. But 14 the utilities help each other.
15 And so these things are made available.
16 MR. CWALINA: That's right. As Bernie and Tommie 17 said earlier, facilities and equipment doesn't seem to be a 18 big problem. These guys seem to be able to work around it.
- 19 Resident inspectors didn't cite it often as a 20 major delay in maintenance. We brought it up mainly 21 because, on our site survey at Davis-Besse, spare parts 22 inventory was mentioned time and again as a big problem at 23 Davis-Besso and accounted for a lot of delays. ,
i 24 So wo looked at it for the rest of the plants, r r~
(%) 25 too.
, ACE-FEDERAL REPORTERS, INC.
202147-17(H Ntiona kle Cmcrage N63%fM6
~
\
l 8460 04 10 55 MM/bc 1 MR. REED: They must not have any good friends.
2 (Laughter.)
3 MR. EBERSOLE: Since the rate payer carries the 4 bill, where is the primary incentive to get the maintenance 5 done and get back on line?
, 6 MR. REED: Jessie, I don't think that's a fair 7 question.
8 MR. EBERSOLE: It's accurate though, isn't it?
1 9 MR. REED: No, it isn't. Public Servics i
10 Commissions, unlike TVA, limit the amount of money in the 11 first place that private utilities can make the amount of 12 profit.
4 13 Secondly, public service commissions are getting 14 on to the back of people for poor performance and they're 4 15 charging off the stockholders.
16 What's the famous case now where stockholders are 17 going to be in the bag for --
18 MR. EBERSOLE: The question is still a valid i 19 one. What's the incentive?
20 MR. REED: The incentive is public service l
21 commissions right now.
22 MR. EBERSOLE: You're telling me that public 23 service commissions are at long last investigating --
24 MR. REED: More than they should. In fact, O 25 they're bankrupting some companies, like Long Island ACE-FEDERAL REPORTERS, INC.
nam.m u ,- . u. a,, ~ ,, nnwa
8460 04 11 56 MM/bc 1 Lighting.
2 MR. EBERSOLE: So the incentive then is the 3 pressure from the public service commissions?
4 MR. LE: Well, I would say I'm not sure about the 5 politics of the spare parts. But I think, to keep the 6 plant -- at those sites that we've been to, we found out 7 that those that either have a financial cap are not well-8 planned ahead, like in the manual systems, they do have a 9 delay in maintenance because of untimely obtaining of parts 10 to accommodato the work order that thay may need the 11 equipment to be on line.
12 So, like Bernio says, that's some LCR they might
(~)
kJ 13 put on some of this equipment that don't have sparo parts 14 readily in the warehouse.
15 MR. EBERSOLE: Wo were told in Japan that the 16 incontive was primarily the reputation of the manager. That 17 was pretty clear.
18 MR. WYLIE: That's a good point. I don't know 19 how much it contributos. But the amount of spare parts 20 inventory that the utility commissions will allow you to 21 havo on hand, that you have in a rato case. I know some 22 ;
utility commissions I've dealt with in the past are very 1
23 ' picky about, you know, things -- does this contributo I
24 directly to the generation of electric power? If it O
(_) 25 doesn't, we're not going to allow it. And this kind of l
l ACE-FEDERAL REPORTERS, INC.
umm s o ,_ ,w .,y -
8460 04 12 57
('
(_j) MM/bc 1 thing.
2 You know, that might be something that you want 3 to take another look at in more depth later in phase two, 4 because it could impact safety if you don't keep the parts 5 on hand.
6 MR. LE: Another thing about the inventory was 7 that some plants on manual systems, based on the previous 8 outages, and see how many pieces of that and how many pieces 9 of this so they can order.
10 MR. REED: I might point out, just what you're 11 talking about, Charlie, Commissioner Zech just gave a speech 12 somewhere to the public service commission, some big meeting 13 they're having. And he raised the identical point, the role 14 of public service commissions in limiting modifications, 15 backfitting, improvements in safety-related aspects.
16 MR. JANKOVICH: I'd like to add something on 17 this. In the last year and a half or two, a group of public 18 service commissions established performance incentives for 19 the utilities, and they have a number of objectives to 20 increase productivity.
21 And that's what they are monitoring, the 22 effectiveness of the utility operations. And our Office of 23 State Programs is following up that initiative. And they 24 have produced a number of reports on the Public Service 25 Commission.
ACE-FEDERAL REPORTERS, INC.
M -147 17m Nationwide Cmerage ik14314446
8460 04 13 58
)MM/bc 1 MR. EBERSOLE: Would you answer this question.
2 Suppose that a vendor or rather an operator, like Davis-3 Besse, is going to spend X-million dollars to install a 4 primary loop depressurization system to augment the safety 5 of the plant.
6 But the NRC doesn't say they have to do it. The 7 PUC's might well say NRC says you don't have to do it, 8 you're not going to get it put in the rate base.
9 So it doesn't get added as an augmentation of the 10 safety design, because it's not a regulated requirement.
11 Where do the PUC's stand on this matter? I'll 12 ask you that also, Glenn.
4 13 MR. REED: They very well may deny it, and that's 14 unfortunate. And I think, in light of what Mr. Zech said to l
15 the public service commissions and public utilities 16 commissions, it was more or less on this target.
17 First of all, I think it behooves us in the 18 regulatory ACRS staff and Commissioners all to find the true 19 and appropriate safety measures, and then we must publicize 20 those measures.
p 21 MR. EBERSOLE: But, the less-informed, which may 22 be the PUC's, will take the view, as did TVA, by the way, 23 that the only thing that is required is to follow the 24 regulations.
( 25 MR. REED: Regrettably, there's a tendency for
't ACE-FEDERAL REPORTERS, INC.
202-34737(#1 N.iiiotteide Coverage mk33t>fMA
8460 04 14 59
(_)x MM/bc 1 that because of the public utilities commissions 2 disallowances.
3 MR. EBERSOLE: And their lack of familiarity with 4 the processes, I presume. So I say that's a large issue.
5 MR. REED: Right. You left me unanswered on 6 something. I asked you where modifications fall and where 7 are the percentages that you have given for preventive 8 maintenance and corrective maintenance and so on. But you 9 didn't really tell me about the percentages and where 10 modifications are in your summary that you had previously.
11 I think that, in the period from 1979 to, let's 12 say '86, a large percentage -- 40-50 percent of all the 13 activities going on in these plants have to do with 14 modifications. Right?
15 MR. LE: Right.
16 MR. REED: So they must show up as maintenance 17 work orders.
18 MR. LE: I agree with you that modification plays 19 a major part. During our survey, we --
20 MR. REED: Did you separate them or are they all 21 lumped into PM, CM and CDM?
22 MR. LE: The ones that we looked at are those 7
23 routine work that's performed by the plant maintenance
! 24 crew. Sometimes, modifications are done by contractor.
25 MR. REED: Sometimes, big ones are but many small 1
- ACE-FEDERAL REPORTERS, INC.
202-347-370n Nationwide Cmerage *U 34 MM
8460 04 15 60 13
(_) MM/bc 1 ones aren't.
2 MR. LE: Yes.
3 MR. REED: Depending on the company, many, many 4 modifications might be done in house with the maintenance 5 crew.
6 MR. LE: Those plants that we have been on, most j 7 are contracted out for work like NSSS related equipment and 8 modification, or special skill required. They need to go 9 out and get the contract.
l 10 MR. REED: Maybe we can't do that, but I notice i 11 one of these plants you went to had something like 6,000 12 backlogged maintenance requests.
13 What I would like to know, if you can give me a
. 14 feel for, first of all, does that 6,000 include 15 modifications and what percentage of the 6,000 would have 16 been modifications?
17 Do you have any idea on that? ,
18 19
- 20 4
- 21 22 23 24 4
25
- ACE-FEDERAL REPORTERS, INC.
202-3 47-37tW) Nat to1 wide Coverage RXk) % N M
. _ , _ _ . _ _ . . _ _ _ _ _ _ ._._,_.___.____--.__-.-_.____m. , _ _ . _ _ _ - . . _ . . _ . _ _ . . _ . _ . , _ _ _ _ . _ , _ _ _ _ _
l 8460 05 01 61 l
()MM/bc 1 MR. LE: The 6,000 we mentioned were particularly 2 at Davis-Besse that had a lot of backlog.
3 MR. CWALINA: The 6,000 was at Brunswick.
4 MR. L8: The 6,000 was Brunswick.
5 MR. PERSINKO: Davis-Besse had a system set up 6 where they described corrective maintenance orders. They ,
7 would come under MWO's, the heading of MWO's. They had 8 another system which they called -- I think what you're 9 speaking of -- the modifications they called FCO's, field 10 change orders is what it stands for.
11 So they had a dif ferent category for it. A lot 12 of times, they would count that in as work that needed to be 13 done also in their backlog.
14 MR. REED: So the best you can determine is it's 15 probably uncertain where modifications are self generated in 16 plant as well as regulatory generated, where they lie.
17 MR. PERSINKO: I'm only speaking from what I've 18 seen at Davis-Besse.
19 MR. GRENIER: At Brunswick, for instance, they 20 throw everything in in one big lump.
21 MR. WYLIE: Modif ica tions , too?
22 MR. GRENIER: Yes.
23 MR. REED: That may account for some of the 24 numbers.
25 MR. LE: At Brunswick also, overy time they 4
l ACE-FEDERAL REPORTERS, INC.
202 347-1740 Nanonwide Cmcrage &n33MM6
8460 05 02 62
()MM/bc 1 changed a light bulb, they write a work order, too. I 2 guess, to answer your question, Glenn --
3 MR. REED: I think I got my answer. You can go 4 ahead. It sounds like it's'a little confusion in numbers.
5 MR. WYLIE: Just to round it out, at Duke, of 6 course, the construction department does the modifications.
7 But they handle it separate from the maintenance. But it's 8 managed at the plant by the operator as far as the 9 scheduling of the maintenance and contractors with the 10 construction department. But it is run by a separate 11 organization.
12 MR. REED: I think, as long as staff understands 13 those numbers, that they have like 6,000 to 400, there's a 14 lot of variables in there.
15 MR. LE: Yes. That's why Bernie mentioned that 16 we were impressed with different numbers. So we asked them 17 on the duration and the translation for these big figures 18 was within two to six weeks of time.
19 MR. EBERSOLE: What's your assessment of a system 20 that has to write a work order to put a 50 watt lamp in the 21 ceiling?
22 MR. REED: Too much damned paper, I'll tell you.
23 MR. EBERSOLE: What's to be done about it?
24 MR. REED: It's up to the staff as they make 25 their judgments --
ACE-FEDERAL REPORTERS, INC.
202-387 1700 Natkmwide Coverage 8m31MM6
8460 05 03 63 r
(3_/ MM/bc 1 MR. EBERSOLE: I could understand that in the 2 backlog list, for instance, if it goes to that degree of 3 triviality.
4 MR. LE: Most of the plants that we have been to, 5 because of maintenance and personnel error, have a lately 6 identified problem. They do have some kind of task force to 7 review the practice at their own plant. At another plant, 8 not in the plant survey, that I've been to, like Palisades, 9 where they did have a problem, they had a task force that 10 identified some of the work orders, like a screw is loose.
11 So they would have their administrative procedure 12 come out if the union allowed the operator can carry the 13 pliers and screw that up and tighten it and not have to 14 write a work order. So it's plant-specific.
15 And it depends on what the management decides to 16 do with those when it gets out of hand.
17 MR. CWALINA: Just to give you an example of the 18 extent of the problem here, and Tommie has mentioned 19 Palisades, and that's a good one, because they had so many 20 work orders outstanding > they decided they were going to go 21 Larough what they called their system of the week.
22 By that, they would walk down all the systems.
23 And anything that was outstanding that was easily corrected, 24 like tightening down a knot or a bolt, they would do right 25 then and there.
ACE-FEDERAL REPORTERS, INC.
202.147 3700 Nat nm aide Coserage 80lk 33MM
8460 05 04 64
()\
(_ MM/bc 1 In the process of doing that, they got rid of 2 about 500 or so work orders, but they generated about 3 another thousand for problems that they found wrong.
4 So there's a real problem in that area. They 5 found so many more problems; at Davis-Besse, they had 6 problems. Davis-Besse was one that now has gone to the 7 point where everything is written on an individual work 8 order and that's why they have generated a massive amount of 9 work orders.
10 But they had a problem back a while ago, where 11 the maintenance people were getting so frustrated, the stuff 12 wasn't getting done. That, even if they identified the 13 problem, they decided they weren't going to bother writing 14 up a work order, because it wasn't going to get done 15 anyway.
16 MR. EBERSOLE: Is there a general fear of not 17 doing the right thing because you don't have the piece of 18 paper in your hand?
19 MR. CWALINA: That varies from plant to plant.
20 You know, the ones that have the -- the Kiwanis and the 21 Millstones, that have the good, established maintenance 22 department, I don't think that's fear.
23 I think that fear becomes generated in a plant 24 like a Davis-Besse, where maintenance has been identified as 25 the major cause of that event.
I 1
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage *G13MM6
8460 05 05 65
()MM/bc 1 MR. EBERSOLE: I could understand if the door 2 knob fell off, you'd like to fix it but you can't do it 3 until it goes through Parkinson's Law. And so it doesn't 4 get fixed.
5 MR. CWALINA: That's right. That's what's 6 happening in a lot of plants. To generate a work order, 7 especially for something like that, so trivial, the work 8 order becomes a minor order and it sits on the bottom of the 9 pile forever, so you never get the door knob.
10 MR. REED: Go ahead.
11 (Slide.)
12 MR. LE: This is the third area, the procedures 13 that I think you heard Bernie mention earlier. So I will 14 hurry through here. We found out during a site survey that 15 five plants are undergoing major upgrading of their 16 procedure because of either human error, technically i
17 upgrading of equipment due to vendor information that they l 18 need-to update.
19 We found out some of the plants, at least two of 20 them, are using INPO guidelines on how to write procedures.
21 Another plant, for instance, Brunswick, they 22 spent about a million dollars to have some contractor come i 23 in and develop the whole procedure and training the staff on
! 24 writing procedures.
25 So there is a tendency to upgrade the procedure
! ACE-FEDERAL REPORTERS, INC.
I 202.147 37tu Nationwide Cmcrage mn JE%m
. . . . _ _ _ . _ _ ~ _ . . _ . _ _ . _ . _ _ , _ _ _ . _ . . _ _ _ . , , _ . . . _ . . _ . . . _ _ . _ . . _ _ _ _ . _ _ . , _ _ _ . . _ _ . . . . _ . . _ _ , _ _ _ _ _ _ _ . _ .
8460 05 06 66
/~T
(_) MM/bc 1 and make it more livable and more simple than before. 1 2 MR. EBERSOLE: It's here that I wanted to ask the 3 question:
4 Is a pointed effort, a particular effort, made to 5 verify and provide continuity in the environmental 6 qualification picture? Because it's not tested. It's a 7 product of type testing, where you do it in a factory or 8 someplace. And declare that a procedure, if rigidly 9 followed, will, in essence, reproduce the same device.
10 So, do the procedures incorporate a special 11 section pertinent to preserving, maintaining and even 12 upgrading the environmental qualification characteristic?
13 MR. LE: I'm glad you asked that question, 14 Mr. Ebersole. There's another counterpart tnat I know right 15 now, that an EQ inspection out there asking the precise 16 question you are asking, whether maintenance-related 17 relevant information to preserve the equipment qualification 18 after the maintenance has been worked on.
l l 19 MR. EBERSOLE: Did they even ask the first i
i 20 question: Is this piece of equipment subject to l
21- environmental quaification? You know, the entire TVA 22 shutdown apparently is due to a finding that they didn't 23 know.
24 MR. LE: Yes. In our procedure, we didn't ask r)
(_ 25 that particular question, but the staff learned a lesson.
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage &&33MM6
8460 05 07 67
()MM/bc 1 So, in this round of EQ inspection, l think they are asking 2 those questions. And I think that's the subject of another 3 meeting.
4 MR. CWALINA: Just something I might add. Last 5 year, year before last, I guess, when I was in INE, I was 6 involved in important safety inspections. One of the things 7 we did, we looked at work orders, work order systems and 8 reviewed several of them.
9 And what we found is, again, it varied to a 10 degree. But some of the plants actually had a checkoff list 11 of the type of equipment, whether it was EQ or not. They 12 would check off whether it was safety-related, nonsafety-
/~
C)' 13 related, any kind of special items.
14 One of the items we noticed on several of them 15 was EO, so they did recognize EQ requirements.
16 MR. EBERSOLE: I was once guilty of suggesting 17 they bring up a steam Jennie and a fire hose and give it the 18 treatment when they got done, but I got shut out.
l 19 MR. LE: To follow up what Greg said, we did look l
l 20 at some procedure, wouldn't they stamp EQ or non-EQ on the l
t 21 work order, also on the required procedure.
22 Generally, there are three types that most of you 23 are very familiar with -- the administrative procedure, 24 which filtered down the management policy of what they
'T
(\_/ 25 wanted on the work that they do. The maintenance procedure l
l ACE-FEDERAL REPORTERS, INC.
L 202-347-3700 Nationwide Coserage 800-336-6M6 l - _
[
8460 05 08 68
()MM/bc 1 here, if you go out and do a certain thing and the procedure 2 is written up with all the pre prescribed steps the worker 3 had to follow.
4 The third one is the kind of as you go you make 5 up type of what we call maintenance instructions, which 6 involve the case of trouble-shooting or some other cases 7 where this was not preplanned in the maintenance 8 procedures.
9 I did mention about the utility aware of the 10 quality procedure directly related to the quality of work 11 performed on the equipment placed in the procedures so they 12 are upgrading it. We found out from talking to them in the
( 13 past there was very little guidance on procedure. Anybody 14 can write procedure.
15 Most of the people we talked to, they said the 16 craft wrote the procedure before, and there was no guidance; 17 or their foreman would do that.
18 Sometimes, they used the vendor manual as part of 19 the procedure. Again, at the present time, like I mentioned 20 before, they are using consultants or form their own 21 procedures writing group.
22 They spent some money on that to make sure the 23 procedures are easy to use. Again, the effect of letter 24 83-23 did very well in updating the procedures so that the 25 technical information would be of the latest type.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Narionwide Coserage FM3346646
8460 05.09 69
(~
(. MM/bc 1 Another thing we've found interesting was the 2 verification and validation. After the procedures are 3
3 written, the person who wrote the procedure would follow the 4 craft people to see whether the procedures are easy to use 5 or was there some improvement that had to be incorporated 6 during the validation.
7 Bernie mentioned procedure use. We found most of 8 the procedure requirements, at least seven out of eight 9 plants we have been to, require their workmen to carry the 10 procedures with them to the site, at the working place, not 11 the plant site.
12 The personnel -- some plants in the past, if we 13 identified an LER that did not follow procedure, then there 14 was some discipline that would follow.
15 At one plant, we found out that because of the 16 repeated problem, the management had to write an 17 administrative directive to say that the craft had to follow 18 the procedure verbatim.
19 And we found out that management's support of i 20 plant personnel when they're doing maintenance was a plus in 21 improving morale because they know the maintenance worker 22 has been looked at by higher management.
23 MR. REED: You lost me a little bit there. You
! 1 j 24 know, the word " verbatim" has a long history, way back in l
l 25 the Nuclear Navy in the early days. And there are large ACE-FEDERAL REPORTERS, INC.
202 347-3m0 Nationwide Coserage 800-336-6M6
8460 05 10 70 (m
(_)MM/bc 1 differences of opinion on verbatim compliance to 2 procedures. I can get you a lot of anti-stories. But, in 3 the generation of procedures, one has to be very careful 4 that the procedure could not be stamped in red at the top, 5 " Call me stupid," and that is a real demotivating - you 6 mentioned the word demotivating and a morale problem for 7 facilities where you have low labor turnover and you have 8 highly skilled people.
9 If they are getting procedures really stamped in 10 red, " Call me stupid", watch out for their contributions.
11 MR. LE: Yes.
12 MR. REED: Most verbatim procedures may very well
(~)/
x- 13 be that.
14 MR. LE: Yes. We did find out that during the 15 using of procedure, if the craft had a problem with it, they 16 just stopped and contacted their foreman and improved the 17 procedure or made notes --
18 MR. REED: The point I was trying to make was if 19 it really says " Call me stupid," the craft worker will do 20 just exactly what the procedure says, and that may be the 21 wrong thing and he'll be there chuckling about it.
I 22 MR. LE: Right. Well, that leads to the control 23 of procedure. In the past, the procedure might not be l
24 carried out by the craft because it's hard to find or not 25 called for in the work order.
1 l
1 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8(XL336-6646
8460 05 11 71
()MM/bc 1 So we found out these plans that they have in the 2 reorganized maintenance group, they would dedicate a group 3 to make sure that the procedure used on that particular work 4 order are the latest approved and are valid for that 5 particular equipment.
6 On the other side, where we found if the planning 7 group is not doing it and the craft or his foreman would go 8 ahead and assemble the package themselves, and we found out 9 this is where the bottle neck would be.
10 At one plant, the craft says that if he would do 11 it, 50 percent of his time going to get the plan, the 12 drawing, or waiting at the shift engineer's office to make 13 sure that the equipment would be signed out.
I L 14 MR. EBERSOLE: Tell me, doesn't the procedure
( ,
15 have to go through a process of being written by somebody 16 who knows how to write it presumably? But, then 17 rationalized in detail by the man that has to carry out the l
l 18 physical work, who will annotate it, fix it and forward it 19 back to the fellow who wrote it, and they converge to a 20 mutually-agreeable piece of paper?
21 Isn't that an absolutely essential process?
22 MR. LE: Yes. That's what would be one of the 23 absolute processes that this planning group are trying to l
! 24 do. I'm just mentioning here some of the plans where they 25 are not having that central planning group, and there is l
l ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Cmerage 2336-6M6
8460 05 12 72 O)
(_ MM/bc 1 some time that would be wasted. And we feed back this 2 information to their management.
3 We interview the craft and the foreman. The 4 foreman would say that he spent most of his time on 5 reviewing the package and he had no time to go out and 6 supervise his workers.
7 Again, we found out that on most of the plants 8 that we have been to, they do have policies to allow for 9 temporary and permanent changes in that procedure. So that 10 they are not stuck with a verbatim requirement.
11 (Slide.)
12 In the next area we have the personnel. Again, 13 it was very well pointed out by Bernie, so I will very 14 quickly walk by here.
l 15 On the staffing, one thing we noted that the 16 shift coverage time was the same that interested you. Most 17 of the plants that we have been to have the daytime shifts.
18 And they had to call out at night on an emergency at night.
19 But, at Arkansas, for instance, they want to go l
20 into a three-shift, 24-hour. But some people object to 21 that. So they are still working on that.
l l 22 Either that or Brunswick. We found out, like 1
l l 23 Bernie mentioned before, the staff size grows from the l
! 24 outage from normal operation to outage. And the different l
25 size of one unit plants and two-unit plants, the important ACE-FEDERAL REPORTERS, INC.
202-347-37(O Nationwide Coverage 80k33MM6
_. . _ _ . . _ _ . _ . . _ _ _ _ _ . . . _ __ ~__ _ ._ _ . . _ _
8460 05 13 -73 O(_/MM/bc 1 thing is the staffs are not doubled for two units compared 2 to one unit.
3 We found out that the knowledge of people in the 4 craft are mostly technical people from technical vocational 5 school, and people come out from military service. Most of 6 the plants we have been to, they are looking for people with 7 Nuclear Navy backgrounds so that they don't have to spend 8 much time to train them.
9 We note in our survey the high turnover rate in 10 I&C craft. I guess, if you look at some organization 11 charts, I&C has been looked at as a high level work as 12 compared to electrical and mechanical. And the reason for 13 turnover is because most of these people would go on to the 14 next plant and get a better paying job, and so on.
15 And that's the reason why now the trend of most 16 of the utilities that we talked to is that they would 17 recruit local people where they have some roots there.
18 Their parents are there, or they went to school there, or l 19. relatives, or something.
( 20 21 22 23 24 25 ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6M6
f q ,360 06 01 74 1 MMbw 1 MR. REED: I notice that you show there is some 2 evidence of aptitude testing at maybe 24 percent of the 3 plants on the site survey.
4 Is that site survey -- I really think you ought 5 to watch this aptitude testing very carefully and focus on 6 it in your future work. It is a very important aspect of 7 maintenance capability.
8 MR. LE: Yes. To follow up on your comment, I 9 quess we followed the training. Most of the plants we 10 talked to, would like to have people spend 10 to 20 percent 11 of their time on training, but at one plant, we found out
( 12 that they spent only 5 percent. Because of work order 13 backlog and overtime work, they had to use the craft for 14 working and resident training.
15 We found out, INPO accreditation program is an 16 ongoing thing. Five of the eight plants we visited have 17 been accredited 1986 for all level of crafts -- mechanical, 18 electrical and I&C.
19 We found out that the training facilities have 20 been very well emphasized, and they had good size of staff, 21 like 10 to 20 trainers, for each of these plants that we 22 visited.
23 Some of the trainers actually are coming from the 24 maintenance and OC people. And they become a trainer 25 because it is on the routine normal hour rather than ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Cmerage 8 % 334646
/"N
'\_j60 06 02 75 1 MMbw 1 overtime.
2 One drawback on training is that those people 3 that are used to work in the craft on overtime, and now they 4 become a trainer, the pay is less. So at one or two plants, 5 we talked to, they intend to increase the salary of these 6 people, so they can keep their good trainers.
7 Again, we mentioned before, the cross 8 communication between operation and maintenance people needs 9 to be improved. And we found out that training also 10 provided mechanisms that, if the maintenance worker has a 11 system level knowledoe, they would talk to the operator much 12 better and vice versa.
13 (Slide.)
14 Lastly, work control. We found out that most of 15 the plants, six out of eicht plants that we looked at, are
! 16 switching, converting from manual to automated computerized l
l 17 systems, one, AMS, automated maintenance management system.
I 18 The reason for this is to keep track of all the work orders l
l 19 and also log in all of the as-found conditions and the 20 problems and how the work is complete, so that trending can l
l 21 be utilized.
l 22 We found out root causes, for instance, were not l 23 very well described in the previous old work order that we 24 looked at, but when they convert to automated system. When 25 the worker completes their work, they document their l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8( o 336-6646
~
1 (UJ60 06 03 76 1 MMbw 1 findinos and the root cause immediately is turned into a 2 computer.
3 So now they have some decent data to work with.
4 So again, the OC, we also looked at that. One 5 interesting thing about OC is that we found one plant that 6 had a lot of OC. They had four that had its own 7 procedures. Another one where they could walk in and look 0 at any work they want to.
9 At another plant we looked at, they only had one 10 OC.for every 30 craft. That is a lot of territory to cover 11 for that OC. Another plant we found out that the OC
) 12 themselves had no say-so in where they would be.
13 So the practicing of OC is varied from plant to 14 plant.
15 Again, work documentation. We found out, as I 16 mentioned before, the root cause and work performed were not 17 very well documented in the past, because the procedure was 18 not there and the awareness about trending and all of that i 19 were not emphasized, even though the ANSI Standard 18.7 has 20 been out since 1977.
(
l l 21 We found out the new computerized will provide 22 the utility with the capability for future trending, and j 23 hopefully, it would tie in with predictive maintenance and i 1 24 preventive, so that corrective maintenance would, in turn, 25 be left.
i
. ACE-FEDERAL REPORTERS, INC.
l n- ma. c- --
f%
( ,260 06 04 77 1 MMbw 1 I don't have any more.
2 MR. REED: I think this is a good time to take a 3 break.
4 (Recess.)
5 MR. REED: Okay. Back on the record.
6 We now have Mr. McLaughlin, Trends and Patterns.
7 (Slide.)
8 MR. MC LAUGHLIN: The portion of the program, 9 maintenance and surveillance program, that I had 10 responsibity for and that I will report on to you today is 11 the development of a maintenance data base.- Primarily, 12 the reason for doing that development of the data base was 13 to establish essentially a baseline to assist in describing 14 the current standards of maintenance in the U.S. nuclear 15 industry, and secondly, to assist in identifying plants with 16 apparent good and bad maintenance programs and practices, 17 based upon their performance in various areas within the 18 plant.
19 Number three, here, to begin the process of 20 selecting and validating maintenance performance indicators, 21 is now in a hold pattern. Essentially, we are melding that 22 effort into Chairman Zech's Task Force effort on developing 23 agencywide performance indicators. So we are trying to do O)
\_ 24 that in conjunction with the task force effort.
25 MR. MICHELSON: What schedule is that work on?
ACE-FEDERAL REPORTERS, INC.
202-347-3'00 Nationwide Coverage 23366M6
p V 60 06 05 78 1 MMbw 1 MR. MC LAUGHLIN: The trial program will be 2 complete by the first of September, and we want 3 recommendations to the Commission --
4 MR. MICHELSON: No, no. The performance 5 indicators. How quickly are you going to have useful 6 performance indicators? September of this year?
7 MR. MC LAUGHLIN: We will have a minimum set of 8 recommended maintenance overall performance indicators by 9 the 1st of September.
10 MR. MICHELSON: These will be validated and 11 ready to use?
O)
(. 12 MR. MC LAUGHLIN: Well, they will certainly be 13 our recommendations, based upon the information in the trial 14 program as to what we would recommend to the Commission to 15 use as performance indicators that will be validated, yes.
16 MR. MICHELSON: Are these maintenance 17 performance, or are you talking about the performance 18 indicators that Zech is talking about, which is what I asked 19 about?
20 MR. MC LAUGHLIN: Overall performance indicators, 21 one of which or two of which address maintenance. One will 22 be a maintenance backlog indicator or measurement, and then 23 we have an aggrecated sum of some of the components of the 24 maintenance measures that I am going to describe to you 25 today, which we are inteoratino into that program as a ACE-FEDERAL REPORTERS, INC.
202-347-3AC Nationwide Cos erage 80tk3366M6
V 60 06 06 79 1 MMbw I maintenance effectiveness indicator.
2 MR. MICHELSON: Okay.
3 MR. MC LAUGHLIN: Okay. Essentially, the 4 creation of the maintenance data base was based on 5 information that was reviewed from outside private sources 6 such as American Nuclear Insurers, INPO information, EPRI 7 and others. We also looked at other oc vernment agencies, 8 EPA, OSHA, DOE information, and of course, the NRC 9 information that is received.
10 That information was evaluated and a decision was 11 made on the selection of the data sources that we would use
/~
(_T) 12 to develop the maintenance data base, and primarily we wound 13 up with the five NRC sources that you see here.
'14 MR. REED: Just a question. I know way back ther 15 was some talk about having a whole bunch of contributions 16 from all the plants to provide data that was 17 maintenance-related, so that the data base could be 18 created.
19 Are you telling me the maintenance data base now 20 will be accumulated by the Staff from existing inputs that 21 are already coming from the plants?
l 22 MR. MC LAUGHLIN: That is correct.
l l 23 MR. REED: Or will there be a great addition?
24 MR. MC LAUGHLIN: There will be no addition. We 25 are utilizing currently available information that is l
ACE-FEDERAL REPORTERS, INC.
L 202-347-3700 Nanonwide Coserage N33MM6
~
/ 's
(_,260 06 07 80 1 MMbw 1 submitted currently from the plants.
2 MR. REED: Thank you.
3 MR. MICHELSON: What will you do with the LERs, 4 since most of them of real interest, are event-oriented and 5 not component failure and repair-oriented.
6 So what do you expect to find in LERs that would 7 be useful?
8 MR. MC LAUGHLIN: Two pieces of information that 9 would be useful. First of all, because of the LER rule 10 change made in '84, we get now ESF actuations. So we are 11 interested in that and how many times an ESF system is 12 actuated. Also the work that is going on at Oak Ridge to 13 classify the LERs by cause codes.
14 MR. MICHELSON: I misunderstood. You are using 15 that as a source of information for performance indicators.
16 MR. MC LAUGHLIN: That is correct.
17 MR. MICHELSON: Not for maintenance.
i i
18 l MR. CWALINA: Information.
19 MR. MICHELSON: You won't get much about 20 maintenance from LERs.
21 MR. MC LAUGHLIN: When I get to the next line, I 22 think you will see, hopefully, how we have broken that 23 down. There has been a slight normalization and adjustment 24 of the data to accommodate for acino and size and numbers of 25 reactor hours critical during a period of evaluation.
l ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nanonw kle Coserage 80tk336-6n46
l d 60 06-08 ,
81 1 MMbw 1 (Slide.)
2 MR. MICHELSON: Who is doing the actual work on 3 this? Are you doing it yourself or contracting or what?
4 MR. MC LAUGHLIN: The effort, initially, is a .
5 soin-off effort from the P&L work that was done to evaluate 6 system performance as a function of management 7 oraanization. We have essentially continued that effort 8 ' with PNL support. The contract is coming to an end now, and 9 we will now continue to update and analyze this data 10 internally ourselves.
11 MR. MICHELSON: You mean PNL isn't going to do 12 it?
13 MR. MC LAUGHLIN: PNL's contract is over the 17th 14 day of September, and that will be the end of their 15 involvement.
16 MR. MICHELSON: You are doing it in-house?
17 MR. MC LAUGHLIN: We are doing it in-house.
18 MR. MICHELSON: Thank you.
l 19 MR. MC LAUGHLIN: Okay. The information from the 20 various sources that I indentified to you previously were t
21 broken down into 31 plant and maintenance measures of 22 performance. Overall characteristics of plant performance l
l 23 and then 18 measures of thinas that we assume are of the I
l {)s s 24 notion that have a relationship to a maintenance program and l
25 its effectiveness. And you will see in there, for instance, ACE-FEDERAL REPORTERS, INC.
202-347-37W Nationwide Coverage *n346M6
8460 06 09 82 q
(_j MMbw 1 Dr. Michelson, to get back to your question about how we are 2 using LERs, we had a subset of those LERs, where we can 3 actually extract the component failure-related LERs or 4 test-related LERs from the cause coding that is being 5 done by the Oak Ridge Lab for AEOD.
6 MR. MICHELSON: To what extent are you using 7 NPRDS?
8 MR. MC LAUGHLIN: At the moment, not very much.
9 We are still considering it and looking at it.
10 MR. MICHELSON: Because it is more likely to be a 11 source of failure data than the LERs.
12 MR. MC LAUGHLIN: The failure mode, I think,
(.
\
13 information reliability, the statistics that are in there, 14 is something that we are going to be interested in 15 extracting.
16 MR. MICHELSON: You have on-line capability to 17 search the NPRDS in your office?
18 MR. MC LAUGHLIN: No, we don't. I've been going 19 over it with the AEOD folks --
20 MR. MICHELSON: You are going to get it yourself, i
l 21 aren't you?
22 MR. MC LAUGHLIN: Yes.
23 MR. MICHELSON: You almost have to hav" t, if 24 you are going to do something like this.
25 MR. MC LAUGHLIN: We have used this data in ACE-FEDERAL REPORTERS, INC.
202-347-3hD Nationwide Coverage 8f6336-6t46
A)60 06 10
(_ 83 1 MMbw I several ways to date. I will oive you an example of each.
1 2 MR. REED: I had a problem when I was reading 3 the Executive Summary and then the backup document, Volume 4 2. And it relates to the bottom line down there," man-rem 5 exposure due to maintenance."
6 For a long while I said, gee, radiation exposure 7 due to maintenance means the maintenance people have caused 8 this man-rem exposure, but that is not really it, and I 4
9 found out later, it is worded differently in Volume 2 from 10 the Executive Summary in Volume 1.
11 What you are talking about is man-rem exposure 12 caused by maintenance need.
13 MR. MC LAUGHLIN: Cor rect.
14 MR. REED: Somebody could think that is due to 15 maintenance people's activities, but it is really what is 16 caused by maintenance need, like steam generators.
17 MR. MC LAUGHLIN: Essentially, what we are trying 18 to capture there is the amount of exposure that maintenance 19 workers receive in the normal course of their duties. That 20 turns out to be 50 percent of total exposure in plants.
21 MR. REED: That is great. The way you said it 22 then is great, but I have a problem with "due to 23 maintenance."
,]
k 24 Check the wording in the Executive Summary. I 25 can give you the pages, if you want, where it varies. In
- ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 8@336-6M6
(o,360 06 11 84 1 MMbw 1 fact, I will do that later.
2 (Slide.)
3 MR. MC LAUGHLIN: We have used this data in 4 seve,ral ways. One is a soon-to-be-published NUREG CR, in 5 which we have done -- I am deviating somewhat from the 6 handout here. The slides that I am showing now are more 7 recent slides. So the data actually on the Vugraph that I 8 am showing goes through '85, where I think your handout only 9 goes through '84. So don't try and track from the handout 10 to the Vuoraph. This is one that I iust pulled out at 11 random.
) 12 It is kind of an interesting one, however, 13 because it is an error that I am particularly interested 14 in. The mean time between component failure forced 15 outages. On this one, I am comparing PWRs versus BWR 16 performance, and then the solid line is the industry mean 17 for each year as an overlay. And you will see a continual 18 gradual increase in that mean time between component l
19 failure, time, and then a steep rise in '85.
20 We are publishing this information as NUREG 21 CR 4611, and it should be out probably September, and it l
22 essentially goes through on each one of the 31 measures that 23 I showed you previously and does this kind of analysis and !
l
( 24 breaks it out, not only by NSSS vendor, reactor type, age 1
l 25 and size of plants. l l
l ACE-FEDERAL REPORTERS, INC. l 202 347-37(1) Nationwide Coverage 800-336 4 86 l
()600612 85 1 -MMbw 1 MR. MICHELSON: What is component failure forced 2 outage?
3 Let me ask that question by giving you an 4 example.
5 If you get a spurious scram, is that a component 6 failure forced outace? Nobody ever figured out why it 7 scrammed. It just scrammed. They had to turn it around, ,
8 bring it back up. It certainly forced the plant to off 9 line, but it came back up again.
10 Is that a component failure?
11 MR. MC LAUGHLIN: No. That would be classified C)
\_/ 12 under a cause code of "other," I would presume.
13 MR. MICHELSON: Component failures, when there is 14 a genuine breakdown of a component, such that it forced you 15 to go down.
16 MR. MC LAUGHLIN: Correct.
- 17 MR. MICHELSON: First of all, this puts you under 18 tech spec limit for a while, and then you eventually exceed 19 the limiting condition of operation and that forces an 20 outage.
21 Is that the kind of forced outage you are talking 22 about or are you talking about where the equipment breaks 23 down, you must come down immediately?
24 MR. MC LAUGHLIN: It also includes regulatory
- 25 imposed reductions.
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage F00-336-6M6
260 06 13 86 1 MMbw 1 MR. MICHELSON: Is a power reduction a forced 2 outage?
3 MR. MC LAUGHLIN: Not unless they come all the 4 way down to an outage.
5 MR. MICHELSON: They have to exceed the LCO and 6 actually come down to hot standby or something, before they 7 are in forced outage?
8 MR. MC LAUGHLIN: True.
9l (Slide.)
10 Another way that we have presented the data is in 11 a plant-specific format, by year, by region and by NRR
( 12 Licensing Directorate.
13 I am sorry, Glenn, because I changed slides on 14 him from the one that is in the handout, because that had 15 Point Beach in it, and I didn't want to see him looking at 16 those numbers askance, so I substituted some plans in 17 Region 2.
18 The kind of relationships that I guess we would l
19 be lookina for here that we are lookino for here, are the l
20 components. For instance, here you see number of forced 21 outages in line 2. And then down further in the list, you 22 will see number of component failure forced outages. Trying 23 to understand the relationship of those two figures, well, I k 24 you see 90 percent, for instance, of North Anna 1, number of 25 forced outaces are, in fact, component failure.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cos erage 804336-6M6
8460 07 01 87
- 0) MM/bc
(_ 1 (Slide.)
2 MR. MICHELSON: What else besides a component 3 failure would give you a forced outage?
4 MR. MCLAUGHLIN: Personnel error, test 5 maintenance.
6 MR. MICHELSON: Wait a minute now. My SCRAM 7 example, where it was spurious, you said, okay, that doesn't 8 count as a forced outage.
9 If an operator accidently pushes the SCRAM 10 button, that doesn't count as a forced outage either, does 11 it?
! 12 MR. CWALINA: It doesn't count as a component 13 failure forced outage.
14 MR. MICHELSON: These are more than componet 15 failure now. I was thinking of the previous slide when I 16 asked the question.
17 MR. MCLAUGHLIN: Forced outages, as they are 18 coded in NUREG 0020, are classified into various
! 19 categories. One is personnel error. There's another 20 category, there's maintenance and test. It depends on how l
21 that information was provided to the NRC and classified by 22 the utility itself as to whether we would include it in the 23 component failure forced outage area.
~
! 24 If there was a personal error forced outage, we I
I' 25 would not show it --
l l ACE-FEDERAL REPORTERS, INC.
1
l 8460 07 02 88 I g
(,) MM/bc 1 MR. MICHELSON: If the instrument mechanic 2 rattles the cage of the SCRAM circuitry and causes a 3 spurious SCRAM, what do you classify that as?
4 MR. CWALINA: That would probably be personal 5 error. Other examples are an instrument technician goes out l
6 and accidentally works on the wrong instrument channel B 7 instead of channel A -- and trips it.
8 MR. MCLAUGHLIN: Okay. For ease in analysis and 9 in direct support of trying to provide information into the 10 NUREG 1212 report on the maintenance and surveillance 11 program plant phase one accomplishments, we broke down the 12 measures into five categories to try and show their
(~) 13 relationship to component -- overall component reliability,
(./
14 safety system reliability, challenges to safety system, 15 radiological exposure and regulatory assessment.
16 And we've got percentage of maintenance-related 17 LER's increase in '84 and '85, even though the overall 18 number of LER's submitted, even taking into account the rule 19 change, was decreasing. The percentage of thoso LER's being 20 submitted attributable to maintenance factors were in fact 21 increasing.
22 MR. REED: I notice there in the top section, the 23 third line, older plants had fewer forced outages but longer 24 to return.
("5
(_.) 25 That goes right back to modifications, how to t
l ACE-FEDERAL REPORTERS, INC.
mmm ~.- m c_., -
j
1 V('860 07 03 89 1 MM/bc 1 upgrade. You've got to upgrade. You know, there's always a 2 load of upgrades.
3 MR. MCLAUGHLIN: I'm intrigued by that one 4 somewhat. I don't know whether-that longer to return time 5 is because they do a better job, or they have a spare parts 6 problem and can't get the parts on line to. bring it back up 7 more quickly, because of the age of the plant. I don't-know 8 what the problem there is.
9 MR. REED: I would think, once they have a forced 10 outace, they may decide or somebody may have been frichtened 11 to the extent they've got to do certain upgrades. And some 12 uparades are committed to on the basis of the next outage.
13 And I don't know whether it should be forced 14 outage, refueling outaces or scheduled outage or what.
15 MR. MCLAUGHLIN: Certainly, nome utilitian would .
16 take advantage; even if they have a scheduled outage 17 upcoming and they go into a forced outage condition, 18 they'll take advantace of it and say we won't start up.
19 We'll start our outage now.
20 It could well be that they're taking advantage of 21 that condition.
22 MR. WYLIE: I think that is a good observation.
23 I believe one of- the Duke plants just came back. One of the 28 Duke plants had a reactor coolant pump seal, failure of the 25 seal. So they elected -- they were close to a refueling ACE-FEDERAL REPORTERS, INC.
202 Nationwide Coserage 8(D 33MM6
_-347-3700 , , . , _ . . _ ___.
~
((_J')60 07 04 90 1 MM/bc 1 outage, so, basically, it's my understanding they went on 2 down to refuel.
3 MR. REED: Let me tell you, I think the way the 4 EEI system of forced outages works, if they just had a 5 forced outage and it was two weeks in advance of the 6 scheduled outage, that whole period is going to figure as 7 forced outage rate for the two weeks until they get to the 8 scheduled outage time.
9 So, you know, there are certain definitions in 10 forced outage, and I don't know whether you're using that in 11 your work or not. I guess the EEI definitions, if that b'
x/ 12 two-week period was -- it will be charged the forced outage 13 rate.
14 MR. MCLAUGHLIN: We have looked at that 15 document. It's the IEEE document on Definitions for 16 Scheduled and Forced Outage Calculations and Time. I think 17 that 762 is the document number.
18 We vary a little bit in our definition of " forced 19 outage" as we apply it in the gray book information, for 20 instance, to that being requested by INPO on this same 21 performance measure.
22 So we're sensitive to that. And we are trying to 23 adiust our definition so that we have comparability between 24 ours and the information that INPO is using.
25 The two underlying conclusions that came from the ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationside Coverage 80th3346M6
f 360 07 05 91 1 MM/bc 1 information that we looked at in this portion of the program 2 was that there seemed to be an increasing improvement in 3 reliability and availability in the industry as a whole, but ,
4 that the contribution of maintenance appears to be 5 increasing.
6 And, consequently, if we could improve in the 7 maintenance area, we would have a much greater improvement 8 in overall reliability.
9 MR. REED: That's going to take a lot of using 10 amount as you make your judgment. I think it would have a 11 lot to do with the age of plants now, the average age of (m
s 12 plants now, versus not many new plants coming on the line.
13 And then you don' t get these startup type outages versus 14 mature plant type outages.
15 A lot of analysis will have to go into that to 16 make a iudgment for application.
- l. 17 MR. MCLAUGHLIN: If there are no further I
( 18 questions, Mr. Persinko is going to present the next section i
19 on wrong unit, wrong train events.
l 20 MR. PERSINKO: One element of the maintenance 21 surveillance program plan for which I was responsible was i
j 22 the investigation of the contributors to wrono unit or wrong 23 train events.
( 24 (Slide.)
25 Wrong unit, wrong train event is a human error ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 804336 686
()600706 92 2 MM/bc 1 that was done by an individual whereby the individual 2 performed an incorrect action on an incorrect unit of a 3 multi-unit facility, or an incorrect train of either a 4 sinale or dual unit facility.
5 As we were going through this in this 6 investication, there's also another category which we did 7 include because we believe the contributors were the same, 8 and the events were very similar. We called it wrong 9 component events. It didn't have to be necessarily have to 10 be wrono unit or wrono train. But, beyond the same train, 11 just an incorrect complement.
12 MR. MICHELSON: That was included in your study 13 area, you say?
14 MR. PERSINKO: Yes. We counted them in there 15 also. We primarily were interested in wrong unit, wrong 16 train, but through our discussions with utilities, they 17 would talk to us about other events. And some of those were 18 actually wrong complement events.
19 The investication was conducted by an inter-20 office team consisting of three people. There was one 21 member from AEOD, two people from DHFT, one of the two 22 people being an engineering psychologist. We investigated 23 35 events at 10 sites.
24 The investigation consisted of extensive 25 interviewing of personnel at the sites, people knowledge ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
rx d600707 93 2 MM/bc 1 with the event. And many times, we were also able to speak 2 to the individual who actually made the error.
3 You have to keep in mind also that many of these 4 events were in the '83, '82 -- '82 '83 time frame. So they 5 were sometime ago. But the individuals remembered the 6 events quite well usually.
7 MR. MICHELSON: If there were three people, how 8 come there's only two authors?
'9 MR. PERSINKO: Because we were the primary -- we 10 were DHFT. We were doing the investication. We wrote 11 the -- the other individual participated on the site visit, ts/ 12 but did not participate in writing the NUREG.
13 MR. REED: You interviewed -- I assume you never 14 did interview iust the individual that was involved in the 15 wrong unit, wrong train error. I assume that you 16 interviewed someone, lot's say, other than the individual as 17 well? f' 18 MR. PERSINKO: Many times, we did interview the 19 individual who performed the actual error.
20 MR. REED: And no one else present?
21 MR. PERSINKO: No, we always had other people.
22 We started off with other people who knew about the event.
23 It varied in the organizations. We spoke with the 24 production manager, two people who were may iust in the 25 engineerino support oraanization, or operations, who were ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6M6
8460 07 08 94
()MM/bc 1 familiar --
2 MR. REED: Bet you would never take the 3 . subjective view of just the individual alone for your 4 conclusions?
5 MR. PERSINKO: No, it was the opposite way. In 6 some instances, we were not able -- the individual either 7 left the site, was no longer available to speak to. So --
8 in all cases, we spoke to somebody other than the individual 9 who made the error.
10 And, in most cases, we got to speak to that 11 individual also. I think they were very open with us, too, 12 once we made clear to them that we were not there to O
(.) 13 penalize them for their error. You know, you could tell 14 when they first came into the room they were a little upset 15 that, here it is, three years later, and three people from 16 te NRC are investigating an event that they did three years 17 ago.
18 But, once they realized our purpose and that we 19 were not there for any punitive type things, they became 20 very open with us, I believe.
21 (Slide.) 1 22 The contributors we found were labeling, poor 23 labeling. The labels were either nonexistent. They were 24 even difficult to read. Maybe they were not unique between
() 25 the two units. And also they may not have been QA-ed f
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
- Y,
'y.4 8460 07 09 _, 95
()MM/bc li labbls. Many ti.nes, you go through a plant, you'll see
- 1. ,
g 2 m5g'ic marker ^1abeling on equipment. And there were some
.s 5-3 i'nstances where it was incorrectly labeled and the
. i ,
4 individual thought he was doing the right thing, but he was T
- } . 5 usinggthe wrong label.
a
6 Training and experience We tied the two of these 1
7 .together because we felt that training and a person's s .8 experience level should equal the desired level of 9 campetence that you are after.
10 In many of the events, we were told that the m +
11 individual was new to his position -- not necessarily new to 12 th company, but new to that particular job that he was
~
, 14 Another contributor was procedures. The primary 15 reason. The deficiency with procedures were that they were 16 ' applicable to multiple units or multiple trains. This can l 17 be done in one of either two ways.
18 You can have a component number in parenthesis, Y 19 two component numbers in parenthesis, depending on whether 20 it was unit one or unit two. And every time you were to go l
l 21 to use that procedure, the individual, the operation 22 .
supervisor, I guess, would cross off the one that he was not 3-
- s. 23 doing that day.
i t~ i
,; , 24 or it could be done.another way whereby you go to
. \/f') 25 train A, or train B would be on different pages farther back ACE-FEDERAL REPORTERS, INC.
, 202-347 3700 N.uionwide Coserage 8m336-6M6
,, .. _ ._ _ _ _ - - . . _ _ . - . . . _ . _ . _ . . ~ . _ - _ _ _ _ . __ . _ _ _ -
1 I
i l
l 8460 07 10 96 (O_jMM/bc 1 in the procedure, and errors had occurred there where the l
2 individual, in going back to the page that he -- to the ;
3 pages in the rear of the procedure, he accidentally ended up 4 on the wrong page.
5 So there were two ways being applicable to 6 multiple units or wrong trains were actually affected.
7 Another problem we came across with procedures 8 other than that human factor deficiency, there were other 9 human factor flaws in the procedure. They were either long, 10 sometimes maybe one or a combination of the following:
11 They could be long, either contain insufficient 12 detail for the individual. They didn't contain appropriate
) 13 cautions where they would be useful.
14 Also, I guess it's not on the slide, but 15 sometimes they were designed in such a way that the 16 individual had to keep flipping between the back of the 17 procedure and the front of the procedure. And, in'so doing, 18 he made a mistake on what step of the procedure he was at.
19 Another contributor we came across was what we 20 I called mind set. We defined it as fixed state of mind 21 whereby an individual's next work assignment is seen by the 22 individual to be similar to his previous work assignment, 23 or previous to what he was doing just before he was given 24 this new assignment. In which case, he hadn't quite (3
's,_/ 25 . separated himself. He was in the habit of doing something ACE-FEDERAL REPORTERS, INC.
202-347-37m Nationwide Coserage an3b6M6
8460 07 11 97 O)
(_ MM/bc 1 and he wasn't able to get his mind on the new subject.
2 There were also layout and equipment design 3 deficiencies, communication problems whereby either verbal 4 or informal, handwritten communications were incorrect.
5 Physical stress also was a contributor.
6 Heat, noise, cramped areas, mental cramped 7 areas. Drawings were incorrect in some areas. They were 8 not updated to reflect modifications done at the plant.
9 Also, a contributor, we felt, was interruption of 10 work flow. In this instance, the individual was doing an 11 action and something came around. Something happened. He 12 was asked to do something else which interrupted his 13 assigned task.
14 And when he came down to the task, he either 15 started in the wrong place in the procedure or he did 16 something wrong that led to the event.
17 Finally, we felt that rushed is another one.
18 Rushed is a sense of urgency, either perceived or real. It 1
19 didn't necessarily have to be real.
20 MR. MICHELSON: May I ask a question?
21 I could understand how you could get messed up 22 reading through procedures and getting on to the wrong 23 pages, or whatever.
24 But, having finally gotten to the equipment to do 25 what you're going to do, having got to the wrong train and ACE-FEDERAL REPORTERS, INC.
202 Nationwide Coserage An33MM6
.___ _. . _ . . . _ __ _ -347 3700 _
8460 07 12 98
()MM/bc 1 whatever, would labeling be the full answer for those cases?
2 In other words, if the equipment had been 3 properly and clearly and fully labeled, would the mistake 4 have occurred even though he got screwed up in reading the i
5 procedure?
6 MR. PERSINKO: Yes.
7 MR. MICHELSON: That's why I'm wondering about 8 labeling alone is not the answer. Even the best of labeling 9 would not have prevented some of the examples.
10 Is that what you're saying?
11 MR. PERSINKO: Yes, that's correct.
12 MR. MICHELSON: Could you give me a kind of a b
s/ 13 feel for how you can have clearly-labeled equipment and yet
! 14 have the procedure get you into the wrong mode?
15 MR. PERSINKO: Perhaps doing a hydrogen -- I'm i
16 thinking of an incident that occurred I think at Dresden,
! 17 where the I&C technician was performing a surveillance test l
18 behind one of the back panels. Had to do with hydrogen l
1 19 injection as part of it, t
l 20 But I think he went to the wrong step in the 21 procedure and he ended up tripping two out of three 1
22 channels.
l 23 MR. MICHELSON: That's not the same as wrong l
24 train, wrong unit. .That's not even the same as wrong j 25 component. So there is yet another set of something in the i
j ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationside Coverage 804 336 6646
. .. . . _ , _ . _ _ _ _ . ~ _ . . . _ . _ . . _ . , . _ . . . . _ _ . . _
8460 07 13 99 '
'q q,) MM/bc 1 title, that's missing. Types of examples perhaps then?
2 MR. PERSINKO: I believe you're correct. I think 3 we considered that as being a wrong train type of event we 4 looked at even though it's really not.
5 MR. MICHELSON: It wasn't really a wrong train, 6 was it?
7 MR. PERSINKO: You're right.
8 MR. MICHELSON: Maybe there's something missing 9 in the title. I always thought wrong train, wrong unit, 10 well, gee, labeling is the answer to that. You can't 11 possibly -- if it were adequately labeled, how could you 12 possibly walk up and not realize when you got there that you 13 were in the wrong train or wrong unit?
14 But it wouldn't answer some of these other kinds, 15 admittedly. So I guess it's partly in the semantics of the 16 title.
17 MR. PERSINKO: I'm trying to recall an event --
18 I'm drawing a blank right now -- where procedures, even 19 though it was labeled correctly, procedures were the cause.
20 I know there are -- let me explain the next step.
21 After we talked to everybody about these events--
22 MR. MICHELSON: I can understand how procedures 23 could cause you to do the wrong thing, but not get to the 24 wrong unit, wrong train, especially if it were properly 25 labeled.
- ACE-FEDERAL REPORTERS, INC.
l 202 347-3700 Nationwide Coserage 800-336- % 46
A- -_g+
8460 07 14 100
()MM/bc 1 MR. REED: You would be surprised. This may help 2 answer your question. I notice the title of this is 3 investigation of the contributors to wrong unit, wrong train 4 events. And that's both operations and maintenance. It's 5 not just maintenance. Right?
6 MR. PERSINKO: That's correct.
, 7 MR. REED: I'm thinking of a very famous 8 operations event where a person -- and I notice that this 9 root cause is not listed -- where a person did about eight 10 things in sequence wrong, including wrong unit, wrong train, 11 wrong valve and wrong everything, wrong health physics, and 12 so on.
) 13 Yet, the person was an experienced operator. And 14 this gets back to why did he do all these things wrong? He 15 certainly had the best interests at heart. The label his 16 mind was turned off at was used rather than mind set. But 17 more than that, it turned out that the person was very 18 marginal in aptitude in his original aptitude testing. He 19 had been taken, hired, but he did not really pass.
20 He passed with a 26th percentile rather than 28, 21 which was the cutoff. And, here, again, we leave out this 22 very important thing of aptitude when we think of 23 contributors.
24 I really think you're making a mistake in your 25 work if you do not look at natural ability, inherent ACE-FEDERAL REPORTERS, INC.
202 347 37to Nationwide Cmcrage N63364M6
l I
8460 07 15 101 MM/bc 1 aptitude, and evaluate for it, because that's the way the 2 Japanese make cars, by having evaluated aptitude personnel 3 do the manufacturing.
4 5
i 6
7 i
8 9
10 2 11 1'
12 13 14 ,
15 J 16 lf 17 .
18 19
, 20 21
] 22 e
! 23 24
!O4 25
, ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(n336-6M6
8450 08 01 102 m
(_)MMbw 1 More and more utilities are going to aptitude 2 testing, and I think you have to, in your maintenance 3 evaluations, look at that.
4 MR. PERSINKO: There was one event which I 5 explained, where doing train A, you are supposed to do this 6 step from the procedure and this component -- component 7 would be in parentheses. Then there would be a slash, then 8 there would be a second component.
9 The individual, the auxiliary equipment operator 10 who got this procedure, got it from the operations person.
11 And none of them were crossed off, but the person thought 12 they knew what train they were working on. So they went 13 ahead and did it anyway. So we viewed that as a procedure
! 14 type problem, because it applied to multiple units. You 15 could say it was a communications problem too, I guess, but 16 we thought it was more of a procedures type problem.
17 MR. MICHELSON: Labeling wouldn't have done that.
18 MR. PERSINKO: Right.
19 MR. REED: Let me talk about thing a little bit 4
20 more. You know, if you are asking directions to drive your 21 car someplace and the person starts to tell you something 22 about you turn down Ridgeway Road, up this road, and you 23 make a left here and you do that. In your mind, you 24 visualize that, what they're telling you.
U 25 You think you visualize what they're telling ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(43.466:6
. - , _ , _ . . _ . _ _ _ . , _ _ _ . , _ _ _ _ . _ _ _ . . _ . , _ _ . _ . . . _ _ . . , . . _ . . . _.___. _ ._ ,, _ ~. ._ _,,
i I
i 1
8460 08 02 103
. ()MMbw 1 you, but you've really got it wrong.
2 Now you are doing the same thing in a plant. An 3 auxiliary operator gets instructions from a control 4 operator, or the shift supervisor says go out to pipe 5 so-and-so this and 1 C pump, circulating water pump here, 6 safety injector pump here. And do these kinds of things.
7 He thinks he is visualizing what the shift supervisor is 8 saying, but he isn't quite. And that is why you get into 9 these wrong unit, wrong train, wrong component situations.
10 MR. MICHELSON: It is more than labeling. The 11 answer has to be much more than labeling.
12 MR. REED: Yes. I think what you have to do is, 13 not only in these kind of operational maneuvers, does the 14 supervisor have to tell the man orally, I think he has to 15 give him a piece of paper, no matter how small the 16 instructions, saying, "Okay, here's your follow-up piece of 17 paper for this manipulation I want you to do."
18 MR. PERSINKO: That may be true, but I can think 19 of one instance where it was informal -- that fits into our 20 communication problem too. The equipment control room 21 operator had written down the circuit breaker for the 22 individual to go to, but he used what was a commonly used 23 term in the plant rather than a unique mark number of the 24 circuit breaker. And the commonly used term referred to t 25 two different circuit breakers. So the individual went ACE-FEDERAL REPORTERS, INC.
202 NationwiJe Coserage 80rF3 % f666
-347 3700,_
1 8460 08 03 104
(~)'\
(. MMbw 1 to the wrong circuit breaker. Since then, they use the 2 unique mark numbers whenever they give instructions now.
3 MR. REED: It is a difficult area to avoid wrong 4 unit, wrong train, wrong component. And it is going to take 5 all of these inputs and study and analysis, as well as the 6 aptitude in it.
7 MR. PERSINKO: Maybe I could go a little 8 further. Maybe you'll see some of these in our next step, 9 what we did with the contributors for each event -- each 10 event, by the way, is described in detail and the individual 11 trip reports how we viewed it, how the utility viewed it.
12 We took the contributors then, and through our (r
( 13 judgment, we applied what we thought was the primary 14 contributor to that event. One per event. We tried to say 15 what was the one main contributor to that event and we 16 assigned it as the primary.
17 For that same event, if there were other 18 contributors, secondary contributors, we would then denote 19 them as secondary contributors with no limit on how many 20 secondary contributors could be applied to the event.
21 (Slide.)
22 The results we obtained then were as such, what 23 you see, the contributors in terms of total percentage. The 24 total percentage -- it is broken down into secondary and 25 primary.
ACE-FEDERAL REPORTERS, INC.
202-347 37(R) Nationside Coverage MXK))46N6
8460 08 04 105 O
(_) MMbw 1 So as you were pointing out, there is more than 2 just labeling. I think it could be seen from this slide, 3 becaust if you just go to primary contributors, what we 4 thought was the main contributor to the event, you will see 5 that all of that, except the rushed contributor, we thought 6 was the primary contributor in at least one of the events.
7 Another way of showing it -- you can see from 8 this one -- another way of showing was to take just the 9 primary contributors and show how they relate to each 10 other.
11 (Slide.)
12 You can see that in this pie graph. An 13 interesting point here -- these are all primaries now, keep 14 in mind, not secondaries. Labeling, training and 15 interf'erence of procedures contributing to 60 percent of the 16 wrong unit, wrong train events, which we thought is a large 17 percentage, just by three of the primary contributors.
18 One other thing we did then, after we broke it 19 down in this fashion, we then said, okay, what are the 20 relationships between primary and secondary contributors?
21 Is there some kind of relationship between them?
22 (Slide.)
23 So we formed a matrix trying to find out whether I
24 it showed anything in relationships. It kind of looks like 25 a scatter at first, I guess, but some interesting things ACE-FEDERAL REPORTERS, INC.
M -347 3? m Nationwide Coserage Rh33M44
8460 08 05 106
)MMbw I come out of it.
2 First of all, you can see by the numbers in here 3 that a wrong unit, wrong train event usually consists of a 4 combination of contributors rather than just one sole 5 contributort however, by this column out here, that doesn't 6 necessarily have to be the case.
7 For instance, when mindset was a primary 8 contributor, there was an event where there was no other 9 contributor. So these contributors can also, because of 10 this last column out here, contribute to a wrong unit, wrong 11 train event without any other contributors involved.
12 So that is the two main conclusions to be drawn O 13 'from this matrix.
14 I think a few other thinga to be noted are that 15 physical stress was not a primary contributor, except for 16 this one instance; however, it was a secondary contributor 17 in many events. Labeling was both a primary contributor in 18 a number of events, as well as a secondary contributor. So 19 labeling came in both ways.
20 That concludes my talk on the subject.
1 21 We are going to be discussing the NUREG with 22 NUMARC in the future to see industry initiatives in this 23 area and there -- what their initiatives are on the 24 subject.
25 MR. REED: This NUREG is issued, is it?
ACE-FEDERAL REPORTERS, INC.
202 347-)?m Nanonaide Coserage Nn)h6M6 v - __- - -
r3 Cl60 08 06 107 1 MMbw 1 MR. PERSINKO: Yes.
2 MR. REED: It is all out, and you never got the 3 word " aptitude" in it.
4 MR. MICHELSON: You can't win them all.
5 MR. REED: I can't win it at all.
6 MR. MC LAUGHLIN: We are just learning how to 7 spell " manual dexterity."
8 MR. PERSINKO: We were asking people's opinions 9 on what do they think were the contributors in this and 10 that. I guess it never really showed up -- maybe the close 11 part of it was an experience.
^
(( )/ 12 MR. REED: Well, if one of your 35 events is that 13 famous one at Point Beach, and you went to the industrial 14 relations experts, they would pull right out the papers and 15 show you there is a contributor, aptitude, he didn't score 16 to the cut-off line, but were hard up for personnel, so we 17 took them anyway.
18 MR. PERSINKO: I can think of one event we were J
19 at where it was pointed out to us that perhaps the 20 individual was " error prone," but that is hard to say.
21 MR. REED: People are different in their 22 aptitudes, and we must face this in this nation some day, l
l 23 that not everyone should be a salesman or not everyone 24 should be a mechanic. They just don't have the manual l
25 dexterity, aptitude, nervous system, et cetera, to be l
ACE-FEDERAL REPORTERS, INC.
202d47 .I7(Il Nationwide Owerage mIHhue
l l
1
/
C 60 08 07 108 2 MMbw 1 that.
2 MR. PERSINKO: Station relationships, maybe.
3 MR. REED: Okay. The next one is projects 4 without full coverage --
5 MR. JANKOVICH: No, we deviated from that --
6 conclusions.
7 MR. REED: Oh, conclusions.
8 John Jankovich.
9 MR. MICHELSON: Herman, did we get a copy of this 10 NUREG 11.92 in the package?
11 MR. REED: It is in this handout.
b s_/ 12 MR. MICHELSON: That is just a little bit of it, 13 but I assume it is a big, thick document.
14 Did we get a copy of it? I didn't remember 15 looking at it. I looked at a lot of papers, but I didn't 16 remember that one, somehow.
17 MR. ALDERMAN: I don't believe we have that one.
18 MR. MICHELSON: If we didn't get it, could you at 4
19 least arrange to send me a copy of it? I would like a copy 20 of it.
21 I don't think we got it. I went through all the 22 papers. It sounds interesting.
23 (Slide.)
t 24 MR. JANKOVICH: Earlier, many questions were 25 addressed to what have we accomplished, actually, with this l
ACE-FEDERAL REPORTERS, INC.
202 4474 700 Nationnide Coverage mnInfM6
. _ _ _ . _ . _ _ _ _ _ _ _ _ . _ . . _ _ _ - _ _ _ . ,_4 -. _
()600808 109 1 MMbw I work in Phase 1 of the maintenance and surveillance program.
2 I would like to recap and give an answer to that 3 auestion in these concluding remarks.
4 You have heard the details of many of the 5 specific projects that we have completed and our overall 6 goal was to establish a baseline, how maintenance is being 7 accomplished presently in the United States nuclear 8 industry. .
9 We used a number of methods, means to achieve 10 that. To put it into simple terms, when we look at it, as
! 11 when we go for a health checkup, medical science has some 12 measures, body temperature and blood pressure, for example.
13 We try to establish that kind of a baseline. Of course, 14 when we started, we didn't even have methods, as far as
. 15 maintenance was concerned, and we don' t know what is a 16 normal baseline. Our objective was to establish a baseline 17 for 1985 and then develop some methods testing instruments.
18 What were those instruments? A 13estionnaire 19 that we sent out to our resident inspectors, the site survey 20 protocol, which gave us deeper insight into the maintenance 21 programs at the various licensees.
22 To put it in the picture really, imagine that now 23 we have a limited amount of information on the maintenance 24 procrams at each site. That is a broad spectrum covering 25 100 plants, but relatively narrow amount of information l
ACE-FEDERAL REPORTERS, INC.
M Nationnide roserage flott336 6M6
. __ _-347 37m_ . _ _ _ . _ _ _ . _ . _ ,
l I
1 (3
( ,j60 08 09 110 )
1 MMbw I which came from the questionnaires, from the LER data, from 1
2 the SALP ratings. 1 3 By going out to the sites during those eight 4 visits, we greatly increased our amount of knowledge 5 concerning that one site. So in the spectrum now, we have 6 peaks as far as the knowledge is concerned. These peaks 7 serve as representative samples, we hope. And from that, if 8 we carry on further, later in time, we can develop some 9 recommended practices.
10 Beyond looking at the licensees, we also looked 11 at programs that the NRC and the industry carry on, and the 12 objective of this program was not just to scrutinize the 13 plants but try to integrate all those activities which are 14 related to maintenance, both within the NRC and without.
15 For example, we coordinated goalc, objective 16 schedules with a number of other programs. They are all 17 listed in the report, specifically, but just think about,
, 18 for example, aging program, reliability program that 19 research carries on.
20 We, of course, coordinated our activities with 21 the training procram and the technical specifications
! 22 program, the technical specifications update program.
l 23 As far as the industry is concerned, we were in 24 close touch with NUMARC. For example, we had information l 25 exchange meetings at Crystal River and Charlotte, North l
ACE-FEDERAL REPORTERS, INC.
Nationwide Coverage Nn34(M6 l
l- -- _
,_ _ _._ _ _ ._ _ . ._3C
_, .__ - 347-3 7(X)__
-s b600810 111 1 MMbw 1 Carolina in the past year, 1985. We have another meeting 2 scheduled for this coming Friday on August 15, where we want 3 to discuss our plans with NUMARC for Phase 2 and also 4 provide them a short briefing in this about our 5 accomplishments.
6 We have coordinated our activities with EPRI.
7 They have a lot of programs going on in maintenance, both in 8 the human factors area and in the mechanical subjects area.
I 9 We are familiar with their projects, and we try to 10 complement our programs with theirs to avoid any duplication 11 of effort.
12 We have established effective interoffice 13 communication within the NRC. You probably are familiar 14 with how closely we work with the Office of I&E. One I&E 15 staff member is always on our team.
16 We also established closer cooperation with the 17 regional offices, as far as maintenance is concerned. They 18 also participated as team members in our site visits, with 19 one of their staff members. We have one designated i
i 20 representative at each regional office, who is following our 21 activities and is responsible for disseminating the 22 information originating from this program. And in my view,
- 23 with these activities during this past year, we established 24 a leading role for NRC in the maintenance area. And our 25 objective is to keep the momentum going.
ACE-FEDERAL REPORTERS, INC.
202 4 47 3700 Nationwide Coserage 80tk346646
_- . _ _ _ _ _ _ , _ . _ _ . _ _ - ~ ~ . _ . _ , , . . _ _ . _ _ _ _ , . _ _ _ _ . . _ _ _ _ . . _ . . _ _ . _ . _ _ , _ _ _ _ . - _ _ _ _ _ _ , . . _ _ - _ - _ .
(Q,60 08 11 112 1 MMbw 1 (Slide.)
2 If you want to see really what are the findings 3 that we have observed in Phase 1, I can list it. This is 4 coming from the final report. These are the major items 5 which are there to be remembered. Maintenance is not 6 accomplished in a number of cases, or it is not performed 7 effectively. And this has been documented and pointed out, 8 for example, 64 percent of the forced outages are due to 9 component failures. In the LERs, 39 percent, the last year, 10 was due to maintenance-related activities. In '85, it was 11 48 percent. And we also documented that approach to 12 effective maintenance is fragmented, largely differing from 13 site to site. Sometimes at certain licensees, a narrow 14 point of view is taken at maintenance, but only the daily 15 corrective activities are classified as maintenance. We 16 think that major management commitment to good maintenance 17 procram is necessary for maintenance to be effective.
18 High percentage of failures are do to improper 19 performance of maintenance. For example, even if we go to 20 large -- like abnormal occurrences reported to Congress, 21 one-third of the cases are due to maintenance. We observed 22 that the maintenance operations interfaces are inaccurate in i
23 many cases. Loss of safety system function was due to 24 maintenance in 87 cases out of 133. 75 percent of the wrong I
l 25 units, wrong train events, were due to human error. And ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage mXh33MM6
.~ ---
4 60 08 12 113 1 MMbw 1 these are daily events.
2 MR. REED: I have got a problem sometimes with 3 words. I try to avoid the use of the word "none" or "all" 4 or these extreme modifiers, and sometimes I think that you 5 need to have modifiers. I just looked quickly at page 16 in l
6 your Executive Summary, Volume 1, and the same words appear 7 there as appear in the top line here, but you said them 8 -differently a few minutes ago.
9 ' You said words to the effect, much needed 10 maintenance is not being accomplished or is not being 11 performed effectively.
O's_/ 12 When you say "needed maintenance is not being J
13 accomplished," you know, you could just as well be saying 14 "all maintenance is not being accomplished," but you know 15 that is wrong.
16 so I am quibbling with words. You know, we are 17 oftentimes quoted out of context, and sometimes we are 18 quoted in context, and sometimes we put out a word that we l
19 ought not to put down that way, but certainly you are right, 20 much needed maintenance is not being accomplished or is 21 being delayed and maybe it is not being performed l 22 effectively, but certainly that could imply "all needed i 23 maintenance."
! 24 Do you see what I am saying?
l 25 MR. JANKOVICH: Yes, I see.
- ACE-FEDERAL REPORTERS, INC.
l 202 347 37(O Nationwide Coserage 80tk336-6M6
j60 08 13 114 1 MMbw 1 MR. REED: And you said it differently.
2 MR. JANKOVICH: Certainly, that is what we meant 3 in the report. We don't say that it is not being done.
4 MR. REED: Well, it is a matter of record now.
5 6
7 8
9 10 11 12
, 13 14
- j. 15 l 16 i
17 18
! 19 l
20 21 22 I 23 l 24 i'
25 ACE-FEDERAL REPORTERS, INC.
202-347 3700 Narumwide Coverage &#L33M44
('600901
) 115 1 MM/bc 1 MR. JANKOVICH: We also found 75 percent of the 2 engineering safety feature actuation are due to maintenance 3 activities, and so on and so on. Occupational exposure, 4 personnel, amounts to about half, 46 percent maintenance 5 persons.
6 Out of these findings, of course, we come up with 7 some sort of recommendations.
8 (Slide.)
4 9 Recommendations mostly for us, what to do. Or 10 you can look at them this way. These are the lessons we 11 learned during this past year. Really, the words are here
() 12 all completely from the final report, so I won' t read them.
13 I would rather like to highlight some of these 14 recommendations, or what we want to accomplish. Or we can 15 look at it, these are the issues that we identified which 16 need attention. We establish that baseline.
17 And our function is and was to monitor how 18 maintenance is being done. And from these activities we are 1
19 able to focus our attention on these major issues.
20 First of all, we need somewhat better 21 maintenance performance measures. These are the tools.
22 This is one of the methodologies that we have been using.
23 They are available now. Anybody can follow it up.
! 24 But we are not completely satisfied. We probably 25 never will be because it's based on a data base, sources of ace-FEDERAL REPORTERS, INC.
202 347 37(I) Natanta kte Coserage R4uk3)MA46
. . _ . . . - -- . . - , , - , . ~ ~ _ - . - - - . . - - . - _ _ _ _ - - - - - - - --
._- -.. -_-. _. _ . ~ - .
n 116
(_j60 09 02 1 MM/bc 1 the information have limitations.
2 But we want to work and refine them. We also 3 want to see really what is the effect and the impact of 4 human errors in maintenance and the overall performance and 5 safety?
6 Some people say that human error cannot be 7 completely eliminated; it will always be there. This is the 8 second item here on the list.
9 I don't agree with that. Human error, the rate 10 of human errors can be reduced and it has been shown. Just 11 look at air line pilot performance going back to the early 12 ages of aviation. And if you follow that development up to 13 the present time, we see a tremendous amount of 14 improvement.
15 If we analyze these situations, we can improve 16 the human error either by feeding the task better to the 17 human operators or changing the conditions for performing 18 that task, or making the individual perform his task 1
19 differently, either by train;ng him, giving instructions, 20 improving the procedures, and so on.
21 So this situation can be improved.
22 Item three, here we are primarily concerned with 23 preventive maintenance. As we see, still 25 percent of the
, 24 plans don't have a formal preventive maintenance program.
25 That's a shortcoming, obviously.
ACE-FEDERAL REPORTERS, INC.
3,_ ,, ~_ u. o_,, ..o-
em
( ,,560 09 03 117 1 MM/bc 1 Just yesterday, I got a phonecall from one of our 2 senior resident inspectors. And he said he is in a 3 desperate situation. The technical specifications at that 4 site requires for safety systems a leak rate testing and 5 walk-down and visual inspection.
6 And that is defined in writing as preventive 7 maintenance program. And he asked me is this an acceptable 8 preventive maintenance program?
9 Unfortunately, I had to tell him, yes, it is 10 because it is in technical specifications, and they got a 11 license based on that. Certainly, this is an area which
() 12 would need refinement and further explanation.
13 Item four, we want management commitment to 14 maintenance. We see there's an important area where there 15 is a general dedicated leadership which pays enough 16 attention to maintenance we can see significant 17 improvements.
18 Item 5 down here, we need some sort of criteteria 19 or standards for maintenance. Here we can talk a lot not 20 just about strict measures, not just about IEEE or ASME 21 standards. But, I think guidelines which were generated by 22 INPO during this past year, or for overall maintenance 23 procrams, or procedure writing guidelines serve also as some
( 24 measure for some sort of criteria and standards by which a 25 licensee can improve maintenance performance.
ACE-FEDERAL REPORTERS, INC.
20244137m Nanon* kle Ometage kn34 AM6
O C 60 09 04 118 2 MM/bc 1 Item 1 before the last discusses the problem 2 which we found between -- and the interface between 3 maintenance and operations. Our previous presentations 4 amply demonstrated that this is an area which needs further 5 attention.
6 And then we want to continue our coordination 7 with the industry. With these slides, we can conclude our 8 activities concerning phase one.
9 Of course you are aware that we are proposing and 10 we have proposed to the EDO a phase two.
11 (Slide.)
12 And this is how we view it at the present. We 13 would want to evaluate industry progress. That is, looking 14 at the baseline that we established in '85 and see what 15 further developments we can have.
16 We have now some measures and we want to carry on 17 those measures as test instruments and see what we can find 18 in the near future; use that information and then determine 19 what other actions should an RC take.
20 To carry out our work, we have to modify our 21 operations because we will not have a contract for next 22 year. We just have a limited number of resources, five of 23 this in this section, who work on maintenance as dedicated 24 staff. And we want to keep the momentum going on what the 25 industry started, what the NRC accomplished during this past ACE-FEDERAL REPORTERS, INC.
202 347 )?00 Nationside Ometage kn 316-fM6
Iv 160 09 05 119 1 MM/bc 1 year.
2 So we proposed a number of specific projects to 3 date, which I show on the next slide.
4 -(Slide.)
5 We proposed to continue our site surveys. We 6 want to visit some other plants that we haven't seen.
7 Recall that picture I painted at the beginning of my 8 presentation, the spectrum of the knowledge of information i 9 peaks. We want to have more peaks, more indepth analysis.
10 We want to use our protocol, that set of 11 questions, the basic principle we want to explore at each
() 12 site. However, we want to modify that somewhat for phase 13 two because we know now some certain areas which need closer 14 attention.
15 Those were the issues which I had on the previous 16 slide. We want to concentrate on those areas. For example, 17 we see maintenance operations interfaces a problem area. We 18 see that as work control, another problem area.
19 So we want to concentrate somewhat in these 20 critical subjects. We also want to visit some of the plants 21 that we have seen previously. Our record clearly shows that 22 maior new programs and programs were initiated by the 23 licensees id 1985. Reorganizations are going on, so we want
() 24 to give some time for these programs to come to fruition, a 25 year and a half or so.
ACE-FEDERAL REPORTERS, INC.
202 347 37(X) Nationu6de Cmerage Nuk33MM6
10
(_j60 09 06 120 1 MM/bc 1 Then we want to go back to those sites and see 2 how those programs are implemented, what improvements they 3 resulted in. We think about visiting six new plants and 4 revisiting maybe four. The time frame for these visits will 5 be for the new plants to be done in the next 12 months, 10 6 months. And then, in another couple of months, we will do 7 the revisits.
8 We certainly want to improve our maintenance 9 measures. We want to integrate, of course, completely with 10 the NRC performance indicators. We have a huge data base.
11 We have to update that. We have to establish relationships i 12 between those single individual measures.
13 We want to see some correlations, hopefully, 14 enhanced measures by combining some of them. We want to 15 validate them in the future. That is project two.
16 We want to continue our assessment of industry 17 initiatives. There is another approach that we baven' t done 18 in the past, but we would like to explore in the future, to 19 participate in the INPO plant evaluations early, two years 20 before we went to two INPO plant evaluations and how 21 maintenance is being reviewed; what recommendations come up i
22 from INPO team members.
! 23 Since then, INPO has published new evaluation 24 criteria, new plant visit criteria. And we don't see how 25 they are being implemented and what are the reasons.
l
- ACE-FEDERAL REPORTERS, INC.
M 347 37m Nationwide Coverage RG3)M686
I\_-
D60 09 07 121 1 MM/bc 1 Hopefully, we can complete this project, too. We 2 also want to look at the Aviation Industry, Project number 3 four, FAA study.-
4 Now, the. relationship between the regulated 5 industry, the airlines, and their regulating body, the 6 Federal Aviation Administration, is very similar to our 7 relationship to the nuclear industry.
8 However, FAA has been involved in maintenance 9 from its very beginning. They have a long history of how 10 they monitor. And they have criteria and standards and 11 bulletins to regulate airline maintenance programs.
() 12 We want to study and see what benefit NRC could 13 cet from that relationship and from that experience that FAA 14 has developed over the years.
15 MR. REED: You have to be very careful again in 16 applying the airlines' experience to the nuclear utility 17 situation because, in the airlines business, you've got 18 millions of standardized parts and replacement kits. And 19 vou have thousands of airplanes that are the same, and 20 hundreds of thousands of engines that are the same.
21 i And in the nuclear electric utility plant, we i 22 never did get to that because we are cut off at 100 plants 23 and almost everything is different. Not everything, but
() 24 almost.
25 MR. JANKOVICH: Yes, I understand that aspect of ACE-FEDERAL REPORTERS, INC.
g 202 347-3NX) Nationwide Cmerage *Xb31M686
( )60 09 08 122 1 MM/bc 1 it.
2 MR. REED: So we've got a great big data bank and 3 a whole bunch of PDM, did you call it, to factor in. And' 4 here in the nuclear business, we don' t have that. We've got 5 all these differences in the parts and pieces and procedures 6 can' t be as standardized. All kinds of problems.
7 MR. JANKOVICH: But there is a lesson to be #
8 learned certainly in the approach to regulation because each 9 airline is responsible for their own maintenance work. And 10 they submit their maintenance plan to FAA. FAA reviews it, 11 approves it or asks for modifications before it goes into s 12 effect.
13 Right now, NRC has really no regulation 14 concerning maintenance. If we were to have any, that 15 approach would be one of the possible options.
16 MR. MICHELSON: One of the things that perhaps 17 you discussed in the phase one, but I didn' t get a clear 18 indication of it, that is, it's important that a utility ,
19 looks very carefully at its failures when they occur.
20 Component failures. -
21 Part of that looking, of course, means to lean 3
22 over the shoulder of the maintenance man to see how things 23 really looked when he took them apart and fixed them and put 24 them all back together.
25 There are many messages, many things to be ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 804336-6M6 ,
. - ~ , - - . . . . - . . , _ - , - _ - . .. --. , .
l l
,, \
'(_)60 09 09 123 1 MM/bc 1 ' learned from such observations of your failures. l
$: 2 L To what extent 90 you determine that the utility 3 'is organized to look at such information and try to glean 4 the most it can from it?
I 5 MR. JANKOVICH: Could I ask for a clarification 6 concerding your question?
~
71 MR. MICHELSON: Sure.
i 8I l MR. JANKOVICH: Do you mean that all NRC is 9 looking at the FAA practice? Or do we intend to look at all 10 our licensees?
11 MR. MICHELSON: It has nothing to do with FAA.
(~T
(,) 12 I
In phase one of your examination of NRC licensees, that's 13 .what I'm interested in. hnd in the process of doing phase 14 .one, did you look at what kind of organizations the 15 individual licensee has to really look at its own
! 16 experience?
17 We discussed earlier and I questioned a little i 18 bit how it looked at other people's experiences. But' it's
., '19 very important that you take a very careful lo'ok at your own 20 experience.
21 , Did you verify that utilities are really doing l
22 this?
!. i I j jj 23 MR. JANKOVICH: We looked at the management of 24 organization and at certain places, maintenance management j
25 has in their performance evaluation some criteria which i
l 1 J' . ACE-FEDERAL REPORTERS, INC.
, t 1R-347-3700 ' Nationwide Coverage 800 336-6M6
- 7. .. C . _ _ _ . _. . _ _ . _ . . . _ . _ _ . _ ., . _ _ _ _ _ _ _ . _ _ _ _ . .
(j60 09 10 124 1 MM/bc 1 require that their department performs well. But we didn' t 2 carry it any further.
3 We realize now we don't have any measurements, 4 standards or guidelines by which we can establish the status 5 to any licensee in this area.
6 MR. MICHELSON: Is this an area where perhaps 7 we're thinking about more carefully? And is that going to 8 be part of your phase two projects?
9 MR. JANKOVICH: I think that's a logical 10 extension --
11 MR. MICHELSON: It isn't on your list.
12 MR. JANKOVICH : Performance measures, to utilize 13 that.
14 MR. MICHELSON: It's not on your list of things 15 that you have in mind.
16 MR. JANKOVICH: Not at the moment, yet. I am l 17 coming to tell you what we anticipate to be the results of l
l 18 phase two.
l
( 19 MR. CWALINA: Some of the things you're talking 20 about are planned in phase two. We haven' t gone into the 21 details of each one of these projects yet.
22 MR. MICHELSON: I was looking at this handout.
23 MR. CWALINA: Some of those are included in 24 that. Part of our plant visits, we want to look at root 25 cause analysis, trending, that type of thing. One of the ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80lk33MM6
i l
l l
( )60 09 11 125 1 MM/bc 1 things we did look at our add-on site visits was the use of )
2 INPO maintenance indicators that were developed by NUMARC. l 3 Most of the plants, because 95 percent of the plants across 4 the country input and keep track of the NUMARCK maintenance 5 indicators, but they don' t know what to do with it.
6 MR. MICHELSON: I don' t think this has anything 7 to do with maintenance indicators. That's a different 8 subject.
9 I made a rather pointed request, or a rather 10 pointed question as to how the utility looks at its own 11 experience, not in terms of if it is as good as anybody
() 12 else's. But, rather I was thinking of San Onofre. They 13 have repeatedly maintained their check valves and have 14 repeatedly seen the damage and repeatedly fixed it, put them 15 back together and didn' t think about them.
16 Now, how are you going to avoid this San Onofre 17 type situation? And what are you doing? What can we do to 18 assure that the utilities will look a little more carefully 19 at what they' re seeing, instead of just fixing it?
20 MR. CWALINA: Part of our protocol has been and 21 will continue to be to look at trending root cause analysis 22 and what plants are doino about it. Are they doing some 23 kind of self-evaluations?
() 24 MR. MICHELSON: I'm not sure even that is what 25 I 'm af ter.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage MG336-6M6
(- 126 V 60 09 12 1 MM/bc 1 MR. CWALINA: But what we can do to make the 2 utilities do it better. We haven' t reached that yet.
3 MR. MICHELSON: What can we do to make them think 4 about what they see when they tear a piece of equipment 5 apart and fix it? What can we do to make them think about 6 it in terms of what are the implications?
7 What caused this kind of damage? Question what 8 they see, instead of just fixing it. Because it's one thing 9 to fix it but it's another thing to have it catch up with 10 you again.
11 MR. CWALINA: Right. Root cause analysis is a 12 definite problem.
13 MR. MICHELSON: The other question I have is --
14 and it has to do now a little bit with FAA because they sell 15 the same thing -- what are we doing in terms of thinking
! 16 about the common mode maintenance wherein a mechanic goes in 17 and fixes both trains incorrectly? Or he goes in and leaves i
18 the O-rinos off of three engines?
19 MR. REED: Or he puts in 60 level switches that 20 are all wrong design?
21 MR. MICHELSON: That's a little beyond the 22 mechanic's problem. That's deeper and dif ferent. I'm 23 thinking more of mechanics only.
) 24 Do we think that's a problem from what we have 25 seen so far? Common mode type of damage?
i ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6M6
O y l60 09 13 127 1 MM/bc 1 MR. CWALINA: From our experience, we didn' t go 2 into that kind of detail. I know, with the I&E special 3 inspection safety system functions, they have found a 4 potential for common mode failure.
5 In the use of incorrect parts, for instance, they 6 found equipment where two trains of equipment were 7 lubricated with the wrong oil. So they have found the 8 potential for those.
9 And plants are being caught on those to an extent 10 where the SSFI's are working.
11 MR. MICHELSON: I'm wondering if they're bigger
( 12 risk contributors than a lot of the other stuff we are 13 looking at. Even though they don' t happen very happen, when 14 they do happen, they can be potentially quite --
15 MR. CWALINA: We don' t know how often they happen l 16 either.
17 MR. MICHELSON: That's right.
18 MR. CWALINA: We need the kind of inspection that 19 I&E is doing to find these type of events.
I 20 MR. MICHELSON: Yes. But, you're not really
( 21 pointing anything specifically at common mode maintenance?
22 MR. CWALINA: No.
l 23 24 25 i
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
(h
' 460 10 01 128 2 MM/bc 1 MR. JANKOVICH: There is one more point here, the 2 last item, which I would like to discuss. At the present, 3 we are thinking that maybe a policy statement could be 4 developed stating what is the NRC's position concerning 5 maintenance and how we would go about it, a policy statement 6 like what NRC has published concerning training, or another 7 one concerning fitness for duty.
8 This item here has not been firmed up yet. When 9 we meet with NUMARC-on August 15th, we will bring this 10 subject up and see how industry sees it. Do they see a need 11 for it. And if there is a.need, and if our NRC management O 12 supports this idea, then we plan to prepare a draft policy
, 13 statement based on Phase I and part of Phase II
( 14 activities.
15 And our time frame for that policy statement i
! 16 would -- draft statement -- would be next June. At the same 17 time, next June, we plan to publish an interim report on the 18 findings of Phase II, all these projects one through 19 five. And then, in about another 10 months, we plan to 20 publish phase two final reports.
l 21 That would include revisits to the previously 22 visited sites.
l 23 MR. REED: Thinking of policy statements, I'm O 24 thinking of some that get generated and how long it takes to 25 generate them, get them all through and out by the
! ACE-FEDERAL REPORTERS, INC.
! Nationwide Coverage 8 % 336-6646 202-347-3700
f~l
\.d60 10 02 129 1 MM/bc 1 Commissioners. I wonder, Carl, as I listen to Phase I and 2 the findings and conclusions of Phase I and realizing that 3 Phase II is not going to finish until '88, I just wonder if 4 you are in the frame of work, or in the status of work that 5 can lead you to a policy statement in June of next year.
6 MR. JANKOVICH: That would be a draft policy 7 statement.
8 MR. REED: Draft.
9 MR. MICHELSON: That can come five years before 10 the next policy statement.
11 MR. REED: So we don't worry about it; it's a 12 draft next year.
13 MR. MICHELSON: I'm a little concerned about 14 asking NUMARC is we need a policy statement. I would be 15 utterly surprised if any utility representation ever said we 16 needed more regulation of any sort, even in terms of 17 policy.
18 I just think you are totally wasting your time to 19 ask them, because I think I can predetermine what the answer 20 has to be.
21 MR. CWALINA: The purpose of the policy statement 22 is to get their -- I think the utilities may be more 23 interested in the policy statement because it gives them a 24 finite time to better themselves.
25 It would be similar to the training policy ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 804336-6M6
( !
'wd60 10 03 130 1 MM/bc 1 statement, which means you go through your improvement. We 2 continue to monitor you for two years. And if you guys .
3 don't get better in two years, then we're going to go 4 through with our rulemaking.
5 MR. MICHELSON: Do you think the utilities will 6 actually ask for a policy statement? They will want one?
7 MR. CWALINA: I don't know if they will ask for 8 it or not.
9 MR. MICHELSON: I would be terribly surprised if 10 they did. I might be pleasantly surprised but rarely does 11 industry ever ask for more regulation of any sort, policy O 12 even as sort of a regulation, not a real one.
13 MR. CWALINA: It won't be their idea. We're 14 going to bring it up to them Friday.
15 MR. MICHELSON: Therefore, I think what you want 16 to do is ask them what ideas they might have to input to 17 it. But don't leave them the option of whether you should 18 have one or not.
19 I think that you decide, not them, because you 20 almost have to -- you know the answer they'll give you. The 21 answer is no, you don't need it.
22 MR. JANKOVICH: We will find it out in two days.
t . 23 MR. CWALINA: The advantage you give them is it l 24 gives them a finite period of time to go ahead and get l
I 25 better.
l ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 804336-6M6 i
l l
l
[';
'vd60 10 04 131 !
1 MM/bc 1 MR. REED: On the other hand, Carl, though, we 2 had a famous statement recently from Toledo-Edison Company 3 that said they wanted even something more than a policy 4 statement on aptitude; they wanted a regulation.
5 MR. MICHELSON: That was one utility.
6 MR. REED: That was one company, okay. I just 7 want to point it out.
8 MR. MICHELSON: That's a little different axe 9 that they're grinding.
10 MR. REED: I'm sure that if a utility felt that 11 they needed a policy statement to assist them in doing what O 12 they would like to do, and perhaps what the NRC also thought 13 they should do, they would encourage it if they got 14 roadblocks in their way.
15 MR. MICHELSON: I didn't sense any situation like 16 that.
17 MR. REED: I don't think there would be any for 18 maintenance.
19 MR. MICHELSON: It's rare for them to ask for 20 more reculation, or even more guidance.
21 MR. REED: Notice how I stuck in that aptitude 22 test thing there?
23 MR. MICHELSON: Every chance you get.
O 24 (Laughter.)
25 MR. REED: Because, here, again, in the ACE-FEDERAL REPORTERS, INC.
M-347-3700 Nationwide Coserage 804336-6M6
(~'t
s d60 10 05 132 1 MM/bc 1 maintenance work, I am concerned that the issue of aptitude 2 testing has not surfaced in the work.
3 I'll tell you what I feel. I feel that there are 4 not 10 percent of the people involved in maintenance work in 5 nuclear facilities who have been evaluated for their natural 6 abilities, skills, aptitude.
7 There are not 10 percent, and I'll take bets on 8 that -- if it's allowed in a government office.
9 MR. MICHELSON: If you want to play the odds and 10 you want to have a reasonable chance of winning, this is the 11 one to work on for aptitude testing, is the maintenance 12 area.
13 Don't work on top management of TVA, for 14 instance, because that just doesn't fit too well. But, 15 clearly, it fits here and I think it has merit here. I 16 wouldn't disagree with you.
17 MR. REED: You take over in operations. I would 18 suspect aptitude evaluations probably have been conducted 19 on, realistically, maybe 30-40 percent of the people in the 20 operations group.
21 What I think I am seeing since Three-Mile Island, 22 with all the training emphasis and license emphasis, I think 23 I am seeing that operations personnel are doing much better O 24 than the designers are doing.
25 For instance, they're overcoming design ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-33MM6
('M
\wd60 10 06 133 1 MM/bc 1 vulnerabilities at Toledo-Edison, at San Onofre, and so on.
2 So I see this moving up in the operating performance. In 3 fact, I am embarrassed a little bit, but designers are 4 civing them such difficult challenges sometimes that I don't 5 think we've got it moving in maintenance.
6 And I hope that, in your work activity, perhaps 7 y.ou would get involved in that. Maybe, in Phase II, you 8 should consider that, moving into that operation and 9 checking it out. And don't take a lot of utility -- Yeah, 10 we just started doing that type of testing or evaluating.
11 You've got to look when did they start really O 12 doing it, and are they doing it for new hires only, which is 13 probably what they're doing, off of transfers. How does it 14 really stack up in numbers in the maintenance work.
15 One last comment on an incident in my life in 16 nuclear. You spoke of the I&C group at Turkey Point with 17 this 96 percent -- I remember an I&C group, one of the very 18 earliest ones. And I'm told this story many times.
19 Twenty-three people in the group, craft people in 20 the group, one person -- they are unevaluated type people.
21 One person in that group broke more instruments every day 22 than the 24 others could repair. It's a fact.
23 MR. WYLIE: In the report, it indicates that 62 0 24 percent of the utilities are using aptitude tests. Well, it 25 doesn't say that. It says, written knowledge or aptitude ACE-FEDERAL REPORTERS, INC.
202 347 3RK) Nationwide Coverage 80tk346646
(~)
k 460 10 07 134 1 MM/bc 1. test.
2 MR. REED: Watch out for those kind of things.
3 MR. WYLIE: But, you know, in talking to INPO, 4 INPO indicated to me that a high percentage of the utilities 5 were going to aptitude testing.
6 Is that not the case?
7 MR. REED: Well, going to is one thing. And you 8 will find out, without the assistance of regulatory putting 9 out a regulation, as the Toledo-Edison people asked for, 10 without the assistance of the regulatory, they would be 11 blocked from backfitting aptitude testing for transferres or O 12 for existing personnel. And they can only do new hires.
13 And you're not going to get the impact of new 14 hires' work activity for 10 years.
15 MR. MICHELSON: You never suggested backfitting, a
16 Glenn. I never heard that one.
17 MR. REED: We certainly could do with the 18 tranferrees.
19 MR. MICHELSON: But not for backfitting to 20 present people, that's a different kind of thir.g. Now 21 you've got a lot of other arguments you have to present as 22 to why this is that important.
23 I don't want to get into it now. That is a new O 24 wrinkle on it.
25 MR. REED: Okay. Are you through?
ACE-FEDERAL REPORTERS, INC.
202-347-37tt) Nationwide Coserage Mk336-6646
\.d60 10 08 135 1 MM/bc 1 MR. MICHELSON: Could I ask a question?
2 MR. JANKOVICH: I'm ready. I can answer any 3 question, or I can try to answer any question.
4 (Laughter.)
5 MR. MICHELSON: In the question of aptitude 6 testing, and so forth, and now we are also dealing with 7 performance indicators, you're getting into the performance
! 8 indicating business. It would be nice if somewhere along 9 the way before long you could somehow take the performance 10 indicators and then ao back and check that same set of 11 utilities to see which ones have aptitude testing, and what b,
7 12 kind, and so forth, to see if there is any correlation 13 between good performance, at least in the maintenance area, 14 which would be the most useful one in this case.
15 Is there any correlation between apparently good 16 performance and the fact that they have certain types of 17 tests that they require of all employees.
18 If there is, that would be very, very useful 19 information in making a decision as to whether it's 20 important enough even to put into a policy statement or a 21 regulation.
22 I haven' t seen that done. I don't know if that's 23 what you intend to do. But it would be useful if it were O 24 done.
25 MR. JANKOVICH: We can attempt to do that because ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 80lk336-6M6
1
)
I f')
(_J60 10 09 136 1 MM/bc 1 we have resident inspectors --
2 MR. MICHELSON: Well, you claim there is this 3 relationship and maybe here's a good chance, without too 4 much effort to verify it.
5 MR. WYLIE: I think you're right. I think it 6 would be a big help.
7 MR. REED: I think you can verify that Kiwani 8 uses, has used aptitude testing from day one.
9 MR. MICHELSON: I might even believe it if they 10 can do it. But, keep it in mind.
11 MR. WYLIE: Of course you're coupling it with
\'
12 performance indicators and what it indicates.
13 MR. MICHELSON: Yes. Somehow, I'm sure they're 14 walking on thin ice yet on performance indicators as to what 15 is meaningful, or whatever. But, whatever one comes up 16 with, can it be related now to such simple things as does 17 the company have certain battery of tests that every 18 maintenance man has to pass.
19 MR. WYLIE: I agree.
20 MR. MICHELSON: It would be nice if all the good 21 performers did have it, and all the bad performers didn't.
22 And the sample were big enough, that would be convincing 23 information.
24 MR. JANKOVICH: We had the sponsors from 62 25 resident inspectors. Some of them indicated that there was ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwkie Cmcrage 800-336 4 46
/"')
kwd60 10 10 137
- 1. MM/bc 1 some sort of dexterity test. And we can look at those and 2 compare to their performance measures.
. 3 I would like-to emphasize that we had to limit 4 the number of questions we could ask from our resident 5 inspectors. We are limited to 39. Even that took them over 6 half a day to respond to.
7 So, in exploring detail, it really went with the-8 extent of those tests.
9 .MR. MICHELSON: When you ask them, can you get 10 the name of the tests that were being used so we have 11 everything on a consistent basis, instead of saying, yes, h 12 there's some kind of manual dexterity test? I'd like to 13 know which test they use.
14 MR. REED: Ask them also the test, how long? For 15 all people or just new hires, and so on and so forth? So 16 you can sort out the real impact that is happening in the 17 work place.
18 MR. MICHELSON: Yes. Maybe they just started 19 doing it.
20 MR. WYLIE: When you say "the test", you're 21 talking about the generic class of tests?
22 MR. MICHELSON: There aren' t that many in use, 23 are there?
l
, 24 MR. WYLIE: Well, a lot of utilities developed l
l l 25 years ago when the EI tests were not available, they t
i ACE-FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coserage 800-3%e86
. , - - . ~ . .__ _ .,_ _ _ . ,
p k.360 10 11 138 1 MM/bc 1 developed their own mechanical comprehensive tests, 2 dexterity tests, and other tests. And so they're not 3 generally known by a name other than certain inhouse names.
4 MR. MICHELSON: I didn't realize that they did 5 customize them.
6 MR. REED: My experience is with marketed trade 7 name tests. On mechanical comprehension, I can almost spit 8 out the particular name that goes with it.
9 MR. MICHELSON: Anyway, these gentlemen are going 10 to think about it and maybe even submit some information 11 that might give us a meaningful trend.
12 MR. REED: Is that the program for today?
13 MR. JANKOVICH: That's it.
14 MR. REED: I would like to thank you people. I 15 think one point we made to come out of this, we think --
16 Carlyle made it -- maybe aptitude should be in there as 17 performance indicator, you ought to follow up on it i 18 certainly in Phase II.
f 19 It may be very significant. I want to say that I l
l 20 think you have got a very large challenge in human factors 21 to como out with a policy statement, or to come out with l 22 Phase II, where you are going to really make judgments, and i
,_ 23 perhaps render advice or cause regulatory action.
' k_) 24 Human factors is very difficult. It takes very j
25 wise, experienced workplace people to make these kind of l
l ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336 6M6 j - - - - . - - .-___. - , . . , _ . - . - _ , , ,
[~)
\wd60 10 12 139 1 MM/bc 1 evaluations. And I have a problem that I feel maintenance 2 success comes from two aspects of the maintenance 3 organization.
4 It comes from the craft people, their evaluated 5 aptitudes, training, so on and so forth, and that maybe 6 makes up 50 percent of the success.
7 On the other hand, it comes from the leadership, 8 the planning, the scheduling, the engineering, the 9 evaluations, the analysis. And if you have demotivated 10 craft interfacing with an arrogant -- we'll call it an 7_
11 arrogant maintenance office group, you're not going to get 12 anywhere in that plant with successful maintenance.
13 So that you can't contribute in whatever the 14 regulatory does, you should not contribute to causing strife 15 and dissention between these two different aroups of .
16 people.
17 So you have a very difficult task to make your l 18 judgments, and I hope to heck you can do it.
i l
19 You made one comment about your budget is cut.
20 That means you're not going to have contractors. Do you l 21 have consultants available to you in your work activities on
! 22 maintenance?
l l 23 You mentioned you are down to five people who are
( 24 poing to try to do this challenging job.
l l 25 MR. CWALINA: No.
I ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
('wd60 10 13 140 1 MM/bc 1 MR. REED: You don't have consultants. Do you 2 ever use consultants? Is that a mode of operation?
3 MR. MICHELSON: They can if they've got the 4 money.
5 MR. CWALINA: Except we don' t have the money.
6 MR. MICHELSON: You have to have special money 7 for that. Program support, I guess is what you call it, 8 don't you?
9 MR. REED: I was just going to recommend that 10 maybe you should get yourself, if you could, and this -
11 wouldn't cost a lot of money, a real craft type person who O 12 represents the skill in the craft and is an outstanding 13 performer. And I know such a person.
, 14 Maybe you should then get a person who is a 15 maintenance office type as consultant. And listen to them.
16 Perhaps you can't.
17 MR. MICHELSON: You don't want a psychologist at 18 all?
19 MR. REED: I would like to see a real workplace, 20 a real workplace, long experienced, industrial utility 21 psychologist, sure. Three of them would be a wealth of 22 information.
- 23 That's that.
24 (Whereupon, the Open Session of the ACRS.
25 Committee Meeting was adjourned at 4:45 p.m.)
ACE-FEDERAL REPORTERS, INC.
a 202-347-3700 Nationwide Coserage 800 336-6M6
. _ _ ~ . -. - _._
CERTIFICATE OF OFFICIAL REPORTER U,73 This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of:
NAME OF PROCEEDING: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON MAINTENANCE PRACTICES AND PROCEDURES DOCKET NO.:
PLACE: WASHINGTON, D. C.
(V-) DATE: WEDNESDAY, AUGSUT 13, 1986 were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission.
11 (sigt M.
(TYPED) d MIMIE MELTZER Official Reporter ACE-FEDERAL REPORTEES, INC.
Reporter's Affiliation r
MAINTENANCE PRESENTATION TO THE ACRS SUBCOMMITTEE
({}
ON MAINTENANCE PRACTICES AND PROCEDURES 1
l l
l ACTIVITIES AND FINDINGS OF
> PHASE I, MAINTENANCE AND SURVEILLANCE PROGRAM l
TOPIC PRESENTER INTRODUCTION ;
MAINTENANCE SURVEY METHODOLOGY G. CWALINA B. GRENIER
(]) QUESTIONNAIRE SURVEY OBSERVATIONS SITE SURVEY OBSERVATIONS T. LE MAINTENANCE MEASURES: " TRENDS AND PATTERNS" P. MCLAUGHLIN STUDY OF WRONG UNIT / WRONG TRAIN EVENTS D. PERSINK0 CONCLUSIONS J. JANKOVICH
\
5
l I
l MAINTENANCE AND SURVEILLANCE PROGRAM PLAN PHASE I i
l PROJECTS I
- 1. SURVEY OF CURRENT MAINTENANCE PRACTICES
- QUESTIONNAIRE TO SRIS
- SITE SURVEYS
- 2. MAINTENANCE PERFORMANCE MEASURES
- 4. MONITOR INDUSTRY ACTIVITIES
- 5. PARTICIPATE IN STANDARDS GROUPS
(])
- 6. INTEGRATION WITH OTHER NRC PROGRAMS
- 7. ANALYSIS OF JAPANESE /U.S. NUCLEAR POWER PLANT MAINTENANCE
- 8. HUMAN FACTORS IN INSERVICE INSPECTION
- 9. DEVELOPMENT OF A MAINTENANCE MODEL O
(]) SURVEY OF CURRENT MAINTENANCE PRACTICES OBJECTIVE TO OBTAIN CURRENT INDUSTRY-WIDE INFORMATION ON MAINTENANCE PROGRAMS AND PRACTICES AT U.S. NUCLEAR POWER PLANTS i
METHODOLOGY AND SCOPE l DEVELOPMENT OF MAINTENANCE SURVEY PROTOCOL DEVELOPMENT OF QUESTIONNAIRE FIVE MAJOR AREAS OF MAINTENANCE COVERED:
- ORGANIZATION AND ADMINISTRATION
- FACILITIES AND EQUIPMENT
- PROCEDURES
- PERSONNEL
()
- WORK CONTROL QUESTIONNAIRE
- 139 QUESTIONS CONSISTING OF SOME 500 RESPON3E ITEMS
- RELIABILITY CHECK = 89%
l - ONE PER SITE; 96% RETURN RATE; 50% RESPONSE RATE ON O ITEMS SITE SURVEYS
- EIGHT SITES SELECTED: SIZE, TYPE, AND REGION
(
- COMPOSITION OF SURVEY TEAM
- CONDUCT OF SURVEY:
PREPARATION INTERVIEW AND OBSERVATION EXIT BRIEFING
- TRIP REPORTS O
() PROCEDURES
- MOST PLANTS USE A WRITER'S GUIDE OR STANDARD FOR DEVELOPING PROCEDURES AND CONTRACTORS ARE COMMONLY USED (50%).
- MOST OF THE PLANTS (77%) SYSTEMATICALLY VALIDATE (USABILITY)
THEIR PROCEDURES, ALTHOUGH ABOUT 15% DO NOT ROUTINELY PERFORM THIS FUNCTION.
- ALTHOUGH VIRTUALLY ALL PLANTS UPDATE THEIR PROCEDURES, ONLY HALF 0F THEM HAVE A FORMAL PROCESS TO REVIEW ALL PROCEDURES AT REGULAR INTERVALS.
O'
'~
PERSONNEL
- TOTAL SIZE OF A ONE UNIT FACILITY MAINTENANCE STAFF IS ABOUT 100 AND 140 FOR A TWO UNIT SITE. EACH INCREASES BY 50% AND 100%
RESPECTIVELY DURING OUTAGES.
- AT ALMOST ALL PLANTS (90%), MAINTENANCE CRAFT WORKERS SPEND 10-20% OF THEIR IN TRAINING.
- REQUALIFICATION TRAINING AND TRAINING ON PLANT MODIFICATIONS i ARE MINIMAL.
l
+- - - - - - , . - . . - - , - , . . , . . . , - . - , , , . - - - , - - - - - - - - , - - - - - - - - - . - , - . - .
l 1
I ORGANIZATION AND ADMINISTRATION
- APPROXIMATELY ONE-FIFTH (22%) 0F THE PLANTS HAVE EXTENSIVE PREVENTIVE MAINTENANCE PROGRAMS, SLIGHTLY OVER HALF (56%) HAVE ADEQUATE PROGRAMS, AND APPROXIMATELY 0NE FIFTH (22%) HAVE MINIMAL PROGRAMS. A MAJORITY (70%) 0F THE PLANTS SPEND A OR LESS OF THEIR TIME ON PM ACTIVITIES.
- PREDICTIVE MAINTENANCE IS MINIMALLY PERFORMED AT MOST PLANTS AND NEEDS ATTENTION.
- A MAJORITY OF THE PLANTS (87%) DO NOT HAVE A MAINTAINABILITY IMPROVEMENT PROGRAM.
- WHILE A MAJORITY OF PLANTS HAVE A FORMAL SYSTEM FOR EVALUATING MAINTENANCE PERFORMANCE, 40% DO NOT.
{}
4 FACILITIES AND EQUIPMENT
! - MOST PLANTS HAVE ADEQUATE FACILITIES (WORK) AND EQUIPMENT (TOOLS) FOR PERFORMING MAINTENANCE WORK.
- THE AVAILABILITY, ACCESSIBILITY AND TIMELINESS IN OBTAINING SPARE PARTS HAS CAUSE DELAYS AT ALL FACILITIE3 TO VARYING DEGREES. LOCATING THE SPARE PART(S) AND THE PAPERWORK REQUIRED APPEAR TO BE THE MORE FREQUENT CAUSES.
- OVER i 0F PLANTS SOMETIMES OPERATED IN A LCO IN PAST YEAR DUE TO LACK OF SPARE PARTS.
O
O WORK CONTROL
- MOST PLANTS (76%) HAVE SOME AUTOMATION AS PART OF SCHEDULING, ;
BUT MOST FULLY AUTOMATED INFORMATION SYSTEMS ARE USED FOR PARTS AND CALIBRATED TOOLS.
- MOST PLANTS HAVE A LOW (27%) TO MODERATE (46%) BACKLOG OF MAINTENANCE WORK; APPROXIMATELY d HAD A HIGH BACKLOG.
- QC IS EXTENSIVELY USED FOR MAINTENANCE ON SAFETY-RELATED WORK BUT THE EXTENT OF USE IN THE MAINTENANCE AREAS (CM, PM, AND SURV. TEST) VARIES. A QC CHECK IS PERFORMED ON 10% OR LESS OF
{) NONSAFETY-RELATED WORK.
t l
l I
($) l i
O I, ORGANIZATION AND ADMINISTRATION STRUCTURE UNDERGOING OR RECENTLY UNDERWENT MAJOR CHANGES
- SEPARATION OF CRAFTS (MECHANICAL, ELECTRICAL AND I&C)
ADMINISTRATIVE POLICIES AND CONTROLS
- FORMAL PM, CM, PDM DEFINED (4/8 PLANTS)
STAFF TIME ON CM MORE THAN ON PM UTILITY G0AL IS 50%-90% STAFF TIME ON PM COMMUNICATION AND C0 ORDINATION MANAGEMENT COMMITMENT PROMOTE COMMUNICATION BETWEEN MAINTENANCE AND Q -
OPERATION OUTAGE MANAGEMENT GROUP INFORMATION AND PLANNING CENTRALIZED PLANNING GROUP COMPUTERIZED WORK ORDER SYSTEMS DATA BASE FOR TRENDING
O-II. FACILITIES AND EQUIPMENT ,
FACILITIES HOUSEKEEPING APPEARS ADEQUATE AND WELL MAINTAINED SPACE PROBLEMS FIRE PROTECTION RETROFITS RESTRICT ACCESS TO PLANT MACHINERY
- SIZE OF SHOPS, TOOL CRIBS, AND CONTAMINATED TOOL STORAGE DO NOT CAUSE MAINTENANCE DELAYS EQUIPMENT SOME MAINTENANCE DELAYS DUE TO LACK 0F MATERIAL
(]) -
HANDLING EQUIPMENT CALIBRATED MT8E ALMOST ALWAYS CONTROLLED USING AUTOMATED OR MANUAL SYSTEMS ,-
SPARE PARTS INVENTORY
- MANAGEMENT EXPERIENCE AND JUDGMENT, VENDOR RECOMMENDATIONS, AND COMPUTERIZED PART TRACKING SYSTEM '
MANUAL SYSTEMS ON SPARE PARTS REP 0RTED TO HAVE CAUSED MAINTENANCE DELAYS WAREHOUSE FACILITIES VARY WIDELY FROM LESS THAN 50%
NEEDED TO COMPLETELY ADEQUATE O
e
('#)
II
I. PROCEDURE
S TYPES: - ADMINISTRATIVE PROCEDURES (AP)
- MAINTENANCE PROCEDURES (MP) l
- MAINTENANCE INSTRUCTIONS (MI)
PROCEDURE DEVELOPMENT:
- UNDERGOING MAJOR UPGRADING (5/8 PLANTS) USING INP0 GUIDELINES AND UTILITY-DEVELOPED STANDARDS PROCEDURE WRITING:
- PAST: LITTLE GUIDANCE AND VENDOR MANUALS
- PRESENT: USE DEDICATED PERSONS OR GROUP TO WRITE-(5/8 PLANTS). POLICY DEFINED IN AP.
- OTHERS: - UPDATING 0F VENDOR TECHNICAL
(]) INFORMATION
- FORMALIZED VERIFICATION / VALIDATION 4
PROCEDURE USE:
- ESTABLISHED BY AP (7/8 PLANTS)
- CRAFT PERSONNEL WERE NOT FOLLOWING PROCEDURES STEP-BY-STEP
- MANAGEMENT EMPHASIS ON FOLLOWING PROCEDURE VERBATIM (2/8 PLANTS)
- PROCEDURES FOR PM, CM, SURVEILLANCE TESTING PROCEDURE CONTROL:
- PLANNING GROUP ATTACHED MOST RECENT PROCEDURE TO W0s
- CRAFT / FOREMEN ASSEMBLED PACKAGE AND PROCEDURES
- POLICIES ON TEMPORARY AND PERMANENT CHANGES T0 PROCEDURES
(])
f G,
aO" IV. PERSONNEL STAFFING:
NO EFFECTIVE METHOD TO DETERMINE PROPER SIZE OF U
MAINTENANCE STAFF (4/8 PLANTS)
NUMBER OF CRAFT VARIES:
ONE UNIT PLANT (107-164 STAFF)
TWO UNIT PLANT (141-260 STAFF)
RATIO 0F SUPERVISOR TO CRAFT VARIES 1:5 TO 1:10 SHIFT' COVERAGE AND OVERTIME
- CONTRACT, SUPPORT DURING OUTAGE
- ' QUALIFICATION AND SELECTION:
e- 'HAVE PLANT-DEVELOPED STANDARDS AND INDUSTRY STANDARDS OR UNION CONTRACTS O
- RECRUITMENT AND RETENTION:
- TECH /VOC SCHOOL, LOCAL, MILITARY NOTED HIGH TURNOVER RATE IN "IEC" CRAFT TYPICAL LENGTH OF TIME ON JOBS:
,, SUPERVISORS 5 YEARS (84%)
JOURNEYMEN 5 YEARS (45%)
. CRAFT HELPERS 2 YEARS (75%)
TRAINING:
MOST CRAFTS (40-50%). SPENT FROM 10% TO 20% OF TIME ON 4 TRAINING INP0 ACCREDITATION PROGRAM
- IMPROVEMENTS IN TRAINING FACILITIES AND INCREASE IN jg '
TRAINING TIME 4 -
SYSTEM LEVEL TRAINING ENHANCE COMMUNICATION ,
l t
'+
()
V. WORK CONTROL PLANNING AND SCHEDULING:
SWITCHING FROM MANUAL TO AUTOMATED SYSTEMS MOVING TOWARD LARGER AND MORE CENTRALIZED PLANNING AND SCHEDULING W0s BACKLOGS (SEVERAL HUNDRED TO 6000)
WORK EXECUTION AND QUALITY CONTROL (OC):
- CRAFT (50% TIME LOST), FOREMEN (30%-80% TIME LOST)
PLANNING / SCHEDULING GROUP ASSEMBLES MATERIAL AND WO PACKAGES VERIFICATION OF WORK COMPLETION:
ENGINEERING REVIEW (2/8 PLANTS)
(])
- POST MAINTENANCE TESTING (2/8 PLANTS)
SIGN-0FF BY SUPERVISORS OR OPERATIONS PERSONNEL QC INVOLVEMENT VARIES FROM PLANT TO PLANT WORK DOCUMENTATION:
USE AUTOMATED SYSTEMS (2/8)
- INFORMATION ON "AS FOUND," ROOT CAUSES, WORK PERFORMED WAS NOT WELL DOCUMENTED NPRDS AND PLANT EXPERIENCES WERE NOT FULLY UTILIZED TRENDING TO BE PERFORMED IN NEW COMPUTERIZED SYSTEMS. (0LD DATA ARE NOT BEING ENTERED IN THE NEW SYSTEMS.)
O
MAINTENANCE MEASURES
(]) TRENDS AND PATTERNS IN PLANT AND MAINTENANCE PERFORMANCE 1980 - 1985 I
PURPOSE
- TO ASSIST IN DESCRIBING THE CURRENT STATUS OF MAINTENANCE PERFORMANCE IN THE U.S. NUCLEAR POWER INDUSTRY.
- TO ASSIST IN IDENTIFYING PLANTS WITH GOOD AND BAD MAINTENANCE PROGRAMS AND PRACTICES.
4
- TO BEGIN THE PROCESS OF SELECTING AND VALIDATING MAINTENANCE PERFORMANCE INDICATORS.
, ([) METHODOLOGY AND SCOPE
- CREATION OF THE MAINTENANCE DATA BASE
- PUBLICLY AVAILABLE SOURCES:
- LICENSED OPERATING REACTORS: STATUS
SUMMARY
REPORT (NUREG-0020)
- SEQUENCE CODING AND SEARCH SYSTEM FOR LERS (NUREG/CR-3905)
- OCCUPATIONAL EXPOSURE AT COMMERCIAL NUCLEAR POWER REACTORS (NUREG-0713)
- NRC'S COMPUTERIZED
SUMMARY
OF INSPECTION AND ENFORCEMENT (766 FILE)
- 75 DATA ELEMENTS X 70 PLANTS X 5 YEARS = 26,250 DATA ITEMS
- PERIOD COVERED: 1980 - 1985
- DATA COLLECTION AND VALIDATION
- NORMALIZATION AND ADJUSTMENT OF THE DATA O .
- - - - --r, - , - - , ---, , - -- . . - - - , e-,-- - , - - - - - , - - - ,,,- -a, -- - - - - - . --- -----,-wn,r-- -- - -
+r--
() 31 PLANT AND MAINTENANCE MEASURES DEVELOPED 13 PLANT MEASURES:
- PERCENT OF UNIT AVAILABILITY )
- TOTAL NUMBER OF FORCED OUTAGES
- ADJUSTED PERCENT OF FORCED OUTAGE TIME PER YEAR l
- NUMBER OF SCRAMS PER YEAR
- NUMBER OF SCRAMS PER 1000 HOURS CRITICAL
- NUMBER OF ESF ACTUATIONS
- NUMBER OF MAJOR VIOLATIONS
- NUMBER OF MINOR VIOLATIONS
- GROSS HEAT RATE
- TOTAL NUMBER OF LERS
- TOTAL MAN-REM OF EXPOSURE
- QUALITY PROGRAMS SALP RATING
- OPERATIONS SALP RATING 18 MAINTENANCE MEASURES
(]) - NUMBER OF COMPONENT FAILURE FORCED OUTAGES
- PERCENT OF TOTAL FORCED OUTAGE TIME DUE TO COMPONENT FAILURE
- NUMBER OF SCRAMS DUE TO COMPONENT FAILURES
- COMPONENT FAILURE RELATED SCRAMS PER 1K HRS. REACTOR CRITICAL
- NUMBER OF SCRAMS DUE TO MAINTENANCE AND TESTING l
- MAINT. AND TEST RELATED SCRAMS PER IK HRS REACTOR CRITICAL
- ESF ACTUATIONS DUE TO MAINT. AND SURV. OR COMPONENT FAILURE
- PERCENT OF COMPONENT FAILURF. RELATED LERS
- NUMBER OF COMPONENT FAILURE LERS l
- PERCENT OF MAINTENANCE RELATED LERS OF TOTAL LERS l
- NUMBER OF MAINTENANCE RELATED LERS
- MAINTENANCE SALP RATING
- SURVEILLANCE SALP RATING
- PERCENT OF MAINTENANCE RELATED VIOLATIONS OF TOTAL VIOLATIONS
- NUMBER OF MAINTENANCE RELATED VIOLATIONS
- MEAN TIME BETWEEN EQUIPMENT FAILURE FORCED OUTAGES
(} - MEAN TIME TO RETURN UNIT TO SERVICE
- MAN-REM EXPOSURE DUE TO MAINTENANCE
I O.
l NUMBER OF COMPONENT FAILURE FORCED OUTAGES BY PLANT ELECTRICAL CAPACITY
- ANNUAL MEAN, 1980 - 1984 I "
Plant Capacny (Not MWe):
10-E 599 or Less E 600 799 I"7 _
,", c, 6 g : c sp, ,
l o . ,99
- L 1980 1981 1982 1983 1984 Year
'Plart Electncal Capacey - Maximum Dependable Capacity (Not NMe) i Maint. Measure 1. Number of Component Failure Forced Outages Per Year by Piart Electrical Cane *y (Net MWe) 1980 1981 1982 1983 1984 MWe 3.9 3.8 2.5 1.9 3.5 599 or Less 6.8 7.6 7.3 5.0 4.7 600 - 799 7.6 8.4 7.1 6.6 5.0 800 - 999 8.6 7.8 8.2 6.2 7.4 1000 or +
Industry Mean 6.7 7.1 6.4 5.1 5.3 '
4.9 5'.1 4.2 4.2 4.6 IM:Std. Dev.
O
O PLANT AND MAINTENANCE PERFORMANCE IN THE U.S. NUCLEAR POWER INDUSTRY DIRECTORATE: PWR-1 YEAR: 1984 g;,,,,,,,,, CINNA RETAUNEE POINT POINT PRAIRIE PRAIRIE M**** 4 e,,,e BEACH-1 DEACH-2 ISLAND-1 !$ LAND-2 4 .,ege
% Awdeb.sity
- 83.2 73.0 77.2 85.7 72.7 84.5 94.3 89.2 e Forced outages
- 7.5 2.8 2 3 0 1 8 0 f 12.3 f.8 4.2 0.2 0 0.1 4.9 0 e SCRAMa* 4.4 1.8 1 3 0 0 5 0
- 8 ",*"'e # 0.9 0.2 0.1 0.4 0 0 0.8 0 g,, c ,, ,
- 8.2 14 2 5 4 1 e ESF Actuet ne 12.9 4 e Me,ar weiseens 1.1 0.1 0' 1 0 0 0 0 e Menor wensimene'" 18.1 9.1 12 14 10 12 8 10
- ** 0 oviagn
= 5.4 1.8 2 1 0 0 5 0
7 ""*
I 85.4 40.0 100.0 47.1 0 0 94.0 0 3.0 1.0 1 1 0 0 4 0 e Comp. Fed SCRAMo*
e Como. Fed SCRA,,M, s 0.8 0.1 0.1 0.1 0 0 0.5 0 l eer 1000 krs. Crut.
0.2 0.1 0 1 0 0 0 0 e Maint. 6 Test SCRAMS
' "*'nt. a Tem SCR, Aus g,g g,g g g,g g g g g per 1000 Mrs Crit.
4.4 2 4 5 1
- ESF"' A* -tueriene
C'*# ' *d' 9.7 3 11 47.7 48.2 40.2 47.8 20.0 50.0 27.3 100.0
% Cosmo. Feel. LORE
- 30.4 35.9 38.5 33.3 20.0 50.0 45.5 33.3
% Me n Lams
- 1.8 1.4 1 1 2 2 1 1
, us nt. sAtp net.ng*.
1.4 1 1 2 2 5,n,. gatp n ,,,,*. 1.7 1 1 l
- 7.3 8.2 0 8.7 10.0 8.3 0 0 g u ,,n, y,,,,,,,n.
Seurse del feUREG4120. Denneed Operet*g A*ectare Status Summery Report 191 Seevence coding and Search System ser beenese Ennt Reports (c) SALP Management Summery System (d) 766 System Computer Fde f
- The SALP procese se n.ot neced on the cowr voor. Therefore. entries may mclude seeero*e retings
.., 69,orer,t , r,.or,s e . se-r ,.
e This plant usee greater then one stenderd devietsen stove the industry everege for outage hours en refuehng O and estected moper meeneenence.
f Empty estes enecoted no este vues owedeWe f
(])
SUMMARY
OF MAJOR FINDINGS MEASURES ANALYZE INTO FIVE CATEGORIES:
- OVERALL SYSTEM / COMPONENT RELIABILITY
- DECLINE IN NO. FORCED OUTAGES & COMPONENT FORCED OUTAGES INCREASE IN THERMAL EFFICIENCY
- OLDER PLANTS HAD FEWER FORCED OUTAGES BUT LONGER TO RETURN
- SMALLER PLANTS HAD SIGNIFICANTLY LONGER TIME BETWEEN COMPONENT FAILURE FORCED OUTAGES
- OVERALL SAFETY SYSTEM RELIABILITY,
- OLDER PLANT (PRIOR TO 75) AND SMALLER PLANTS (600 MWE OR LESS) HAD FEWEST LERS AND MAINTENANCE LERS
- PERCENTAGE OF MAINT. RELATED LERS INCREASED: 39% IN 84 AND 48% IN 85
([) - CHALLENGES TO SAFETY SYSTEMS
- SIGNIFICANT DECLINE SINCE 1980
- EVEN LESS FOR OLDER AND SMALLER PLANTS
- 75% OF ESF ACTUATIONS IN 84 8 85 WERE MAINTENANCE RELATED
- RADIOLOGICAL EXPOSURE
- EXPOSURE LEVELS STABLE, 46% MAINTENANCE RELATED
- BWRS HAVE DOUBLE MEAN EXPdSURE LEVELS OVER PWRS
- REGULATORY ASSESSMENT
- SIGNIFICANT DECREASING TREND CONCLUSIONS
- OVERALL INDUSTRY RELIABILITY APPEARS TO BE IMPROVING (DECLINING FORCED OUTAGES AND SCRAM RATES)
() - MAINTENANCE CONTINUES TO BE A SIGNIFICANT CONTRIBUTOR TO INDUSTRY RELIABILITY PROBLEMS (BY 75% ON SOME MEASURES)
NUREG-1192 oo An Investigation of the Contributors to Wrong Unit or Wrong Train Events e
U.S. Nuclear Regulatory Commission D. Persinko, A. Ramey-Smith O
f
_ _ _ . _ - .._-~--.. _ _ - . . _ -
O CONTRIBUTORS TO WRONG UNIT / WRONG TRAIN EVENTS
- LABELING NON-EXISTENT DIFFICULT TO READ NOT UNIQUE TRAINING / INEXPERIENCE PROCEDURES APPLICABLE TO MULTIPLE UNITS OR TRAINS HUMAN FACTOR FLAWS (E.G., LONG, INSUFFICIENT DETAIL, O NO CAUTIONS)
MIND SET LAYOUT / EQUIPMENT DESIGN
~
i COMMUNICATIONS PHYSICAL STRESS DRAWINGS INTERRUPTION OF WORK FLOW RUSHED O
4 Contriautors to WU/W- Events l
Primary and Secondary 50 -
ss, 40 -
Ie -
fJ % % m .
=
-p hh sg l Th!!hhyhyhy $vv l Lab T/l Pro Phy Com MS Eq Rush Dwg Int Q Primary Secondary
~
I
-__r
& 6 e
O m
N ^
@ o M o m
,_ K x F -
v "
- U h 2 N
> ' ' E W
k* o n e
C ,.
?- s v
Ds 7 3S b '
O Oo m
@ .EC y D.
o -
u b m O' x O
- Q O -
V "
3 m
_O a er o W
,- m h
s V o u
[ '
C5 .6 rn O v 0 %
a O
o.
3 1 1 1 1 1 sR 1 1 1 1
}&
eR 1 _
_ . 83 gum 3 r _
o .
&3=b 1 su rb t
oi tr _
1 1 1 ut 1
.E5" %33s 3 1 - bn io rC t
ny 1
or 8 t3*[a 2 .
Ca
' yi m .
rr aP
. d 5u8 5 1 1 nh oc ca s*f.3
~o e S eE ft h
oi y W i* iE 2 .
cd
, ne et 2 2 1 ua k1gs 2 1 qi ec ro Fs s
- A 2= F a2# 2 2 3 2 1 .
2 e
l
. g b 3 1
.S :.3 1 1 1 T
a s
,r p n
s e
. u o r n d e t o n n s n i
t S i o
. g I
/ e g a t n g r t i c l s p s
. n u e / s i a g uk o i l i n
d e
S te uD n
u i c i n ~rr vo l
eW m s w h b e
a i
a r
c o
r d
i n o.
y aq m o h y e r
t nf W L T P M LE C P B Io o
- l;j j
O PHASE I FINDINGS NEEDED MAINTENANCE IS NOT BEING ACCOMPLISHED OR IS NOT PERFORMED EFFECTIVELY.
A HIGH PERCENTAGE OF FAILURES RESULT FROM IMPROPER PERFORMANCE OF MAINTENANCE.
THE MAINTENANCE AND OPERATIONS INTERFACE IS INADEQUATE.
O
- THE NUMBER OF MAINTENANCE-RELATED CHALLENGES TO SAFETY SYSTEMS IS EXCESSIVE.
THE MAJOR PORTION OF OCCUPATIONAL RADIATION EXPOSURE AND MANY RADIOLOGICAL HAZARDS OCCUR TO PERSONNEL PERFORMING MAINTENANCE ACTIVITIES.
O
a-w PHASE I RECOMMENDATIONS (NRC AND INDUSTRY) l IMPROVE THE DEVELOPMENT AND USE OF MEASURES OF PLANT MAINTENANCE PERFORMANCE IN ORDER TO REVIEW INDUSTRY PERFORMANCE IMPROVEMENT AGAINST A BASELINE AND TO FOCUS RESOURCES ON APPROPRIATE PLANTS AND ISSUES.
DETERMINE THE CONTRIBUTION OF HUMAN PERFORMANCE IN MAINTENANCE TO PLANT RELIABILITY AND IDENTIFY THE SAFETY IMPACT OF ALTERNATIVE STRATEGIES FOR IMPROVING HUMAN RELIABILITY.
DEFINE MAINTENANCE NEEDS, INCLUDING THE ROLE OF PREVENTIVE MAINTENANCE FOR ACHIEVING ACCEPTABLE LEVELS OF RELIABILITY AND SAFETY.
1 DEFINE GOALS FOR NUCLEAR POWER PLANT RELIABILITY THAT FOSTER MANAGEMENT INVOLVEMENT IN ENSURING EFFECTIVE MAINTENANCE, INCLUDING THE USE OF PERFORMANCE MEASURES AND INCENTIVES.
DEVELOP AND IMPLEMENT PERFORMANCE-ORIENTED MAINTENANCE CRITERIA AND STANDARDS AS THE BASIS FOR EFFECTIVE MAINTENANCE PROGRAMS. FURTHER EVALUATION OF CURRENT MAINTENANCE PROGRAM PRACTICES IS NECESSARY TO DETERMINE FACTORS INFLUENCING MAINTENANCE EFFECTIVENESS.
IDENTIFY APPROACHES TO MAINTENANCE WHICH ENHANCE THE INTERFACE BETWEEN MAINTENANCE AND OPERATIONS AND ENSURE RELIABILITY.
1 i
C0 ORDINATE NRC AND INDUSTRY INITIATIVES RELATED TO
] MAINTENANCE.
_. - - - -- .- - _ - -- _ 1
O PHASE II OBJECTIVES: EVALUATE INDUSTRY PROGRESS IN IMPLEMENTATION OF SELF-IMP _ROVEMENT INITIATIVES AND THEIR EFFECTIVENESS TIMEFRAME: FY87-88 Q PROPOSED PROJECTS:
SITE SURVEYS MAINTENANCE MEASURES ASSESSMENT OF INDUSTRY INITIATIVES INTEGRATION ANALYSIS AND DOCUMENTATION O
3 l
e
( )-
l 1
MSPP PHASE II PROJECTS
- 1. SITE SURVEYS NEW PLANTS ,
REVISITS i
i
- 2. ANALYSIS OF MAINTENANCE MEASURES 3.
ASSESSMENT OF INDUSTRY INITIATIVES
- 4. FAA STUDY
]
- 5. INTEGRATION WITH RELATED PROGRAMS i
- 6. POLICY STATEMENT
t-l 1
1 MAINTENANCE AND SURVEILLANCE PROGRAM O SITE SURvevS CONoVCreo l
MAINTENANCE SURVEY SITES (8)
- KEWAUNEE
- MILLSTONE
- BRUNSWICK
- ARKANSAS NUCLEAR ONE
- TURKEY POINT
- RANCHO SECO
- DAVIS-BESSE
- SALEM O
WRONG UNIT / WRONG TRAIN SITES (10)
- DRESDEN
- SURRY
- NORTH ANNA
- PEACH BOTTOM
- SALEM
- LASALLE
- D C. COOK
- MCGUIRE
- TURKEY POINT
- OCONEE O