ML19343D377
| ML19343D377 | |
| Person / Time | |
|---|---|
| Site: | South Texas  |
| Issue date: | 04/27/1981 |
| From: | Frazar R HOUSTON LIGHTING & POWER CO. |
| To: | |
| Shared Package | |
| ML19343D370 | List: |
| References | |
| ISSUANCES-OL, NUDOCS 8105040358 | |
| Download: ML19343D377 (40) | |
Text
-..
i l
2P 3j 4j i
5, TESTIMONY OF RICHARD A. FRAZAR 6i REGARDING CHANGES MADE TO ADMINISTRATIVE CONTROLS 7j IN RESPONSE TO THE ORDER TO SHOW CAUSE 8i 9iI Q. 1 Please state your name and current occupation.
g 11 A. 1 Richard A. Frazar.
I am the Manager, South Texas 12 !
13 Project Quality Assurance of the Houston Lighting & Power 14 19 Company (HL&P).
16 :
17 i Q. 2 Describe your professional qualifications, educational I
9gl background, and involvement in the South Texas Project
{g (STP).
1 22 A. 2 These are set forth in my testimony on the current 23 t 24 !
Quality Assurance (QA) program for STP.
29 26,
Q. 3 What is the pIrpose of your testimony?
27 l l
28 i A. 3 ihe principal purpose of my testimony is to 2 9 l>
describe the changes made in the STP administrative controls 30 39 in response to items 5, 6,
9 and 9 of the NRC's Order to 37 33 l
i Show Cause of April 30, 1980, (Order) and to explain how l
34 !
35 !
these administrative controls satisfy the requirements of 36 l Appendix B to 10 CFR Part 50.
Preliminarily, however, I 37 !
38 i 39 1 will describe a number of improvements to the STP QA program 40 i 41 l that we initiated even prior to the issuance of the Order as 42 l 43 l a follow-up to meetings with the NRC in December 1979 and 44
- ^""U 1980-45 46 47 i
l 48 i
l 49 I
50 1
51 l
l
\\.
l l
8105040358 l
1 1
2.
3l 41 5
Q. 4 Please describe the improvements to the STP QA 6;
7I program that were undertaken in December 1979 and January 8i 1980.
9
.0
.1 'l A. 4 As described in Mr. oprea's testimony, after a
- 2 meeting with Mr. Seyfrit, Director of NRC's Region IV Office
.3
.4 of Inspection and Enforcement (IE) on December 28, 1979,
.5
.6 HL&P committed to a nine-point plan for specific improvements
.7 i
These can be summarized in the
'9 O,
f llowing nine elements:
}!
n First, it was committed that B&R would hold a seminar 13 to review with both Construction and QC personnel the funda-14 ;
U5 mental philosophies and standards of STP QA program.
The 16,
l7 i seminar was to emphasize the respective roles of Construction
.8 up.
and QC in assuring quality construction and a safe facility.
10 i 1
Thesycondelementofourprogramwastochangeproce-
'2
- 3 i dures to clarify when to use a Field Request for Engineering Action (FREA) as opposed to using a Nonconformance Report 5
i
- 6 (NCR).
One source of frustration on the part of the QC i
17 l lS !
Inspectors was Construction's use of the field design change i9 !
- 0 l system in instances where it might not be fully appropriate.
..I.
,2,
It appeared to the Inspectors that in some inv.ances, Construc-
-3 i
,,g l
tion personnel would avoid correcting nonconforming conditions by obtaining Engineering's approval of the deviation in the
'7 !
form of field design changes (FREA's).
Although both systems
,8 l
d3,
resulted in review by Engineering and an appropriate disposition,
- 0 il i l
1, 2,
3l 4i 5,
when the FREA system was used the conditions were not iden-6i 7i tified as nonconformances.
This limited the" ability of the 3g{'
QA system to detect underlying problems.
We changed the h0l FREA and NCR procedures to make clear when each was appropriate.
JL
- L2 i Third, B&R issued a written policy describing the L3 L4 process for resolving any disagreements between Constructio L9 L6 and QC personnel.
The policy specifically stated that L7 i Lg l threats or intimidation would not be tolerated.
L9 '!
Fourth, because NRC questioned the qualifications of 20 2'
some B&R QA and QC personnel, we undertook a management 22 23 assessment of the overall qualifications of the B&R QA and 24 '
35 QC personnel to provide the basis for upgrading the caliber 26 37 '
of personnel.
28 gg,
Fifth, HL&P directed B&R QA and Construction management
'~Og, to implement strict procedures for concrete preplanning and 32 '
33 j placement activities.
HL&P QA and Construction personnel 34 would participate in the preplanning and placement process 3a 36 :
to make sure the work was done in a thorough fashion.
37 !
38 sixth, procedu,res were revised to provide a controlled 39 i 40 l method for judging when reinspection of a concrete placement 41 l 42 !
is necessary prior to sign-off of the pour card.
43 !
44 l Seventh, three key HL&P QA personnel, including the 5j Projects QA Manager, were assigned to the site to strengthen 47 !
HL&P's role in the short term, to be directly and visibly 48 l 49 '
involved in the work in the field and to facilitate the 50 1
51 l i
1 i
1i 2.
3; 4
5 e ntinuing investigation by the NRC Staff.
HL&P further 61 7j added to its staff at the site during 1980 to underscore its II active role in assuring that procedures were adequate and 9i
.0 l that implementation was thorough.
Distinctive identifi-
.1 i
.2 l cation would be provided for HL&P personnel to increase
.3,
,4 their visibility on the jobsite.
5
.6 Eighth, a refresher training course would be imple-mented for B&R Construction and QC personnel to reinforce
- 9 i their understanding of their assigned duties and the proce-
.0 '
11 ;
dures governing their work.
!2 i
!3 '
Ninth, a study was conducted in which there were inter-14 !
IS.
views of B&R personnel, starting with top level management S
5'7 and proceeding throughout the B&R organization, to determine SS{g!
the cause of the perception of harassment or undue pressurc 10; on QC personnel.
32 l As also pointed out in Mr. Oprea's testimony, after the 33 34 !
exit interview with I&E on January 24, 1980, the additional 35 !
36 !
37 !
following improvements were undertaken:
33 We ins-ituted a program to upgrade the system for 39 !
40l analyzing _ rends in honconfo*mances.
A new system for 41
- 4.,T, coding NCRs and FREAs would be adopted and all past and I3 i 44 I future FREAs and NCRs would be coded to permit analysis.
I 45 There would be quarterly Trending Reports and identified j
46 47 trends would be investigated to determine if there were 48 49 j
l 50 l
51 l --
~
L 2,
3 4'5' common causes.
When common causes were identified, appro-6'
- 7i priate corrective action would be taken.
8' 9
The assessment of qualifications of QA/QC personnel was I
continued.
Only one QC Inspector was found to have question-12 !
able credentials.
13 14 Meetings were held by P4R to reemphasize to QC Inspectors 15 16 that they must take as much time as needed to perform thorough 17 ig inspections; HL&P QA and Construction personnel were directed 9a.
j~ '
to ensure by their involvement in concrete preplacement and 21 placement activities that adequate time is allowed for 22 23 inspection prior to and during the placement of concrete.
24 ;
39 In addition, a memorandum was issued to QA/QC personnel 26 27 !
directing that all nonconforming conditions were to be
' 23,
29 documented as soon as they were identified.
O g
HL&P surveillance personnel were directed to document all nonconforming conditions, even those documented by 3
34 '
others.
39 36.
EL&P Audit schedules were revised to make sure that there 37 30 was an annual corporate audit of B&R construction, and HL&P 39 i 40 j audit procedures were revised to state that procedure imple-I 4{9.
mentation is to be verified by direct observation of work 4
43 !
44 ;
being performed in the field as well as by reviews of docu-45 mentary evidence.
We also decided to have our QA program I
47 '
audited by an outside consultant at least once a year.
48 i 49 !
90 51 l
1 i
L, 2,
3I 4'
5 B&R competely revised the Welder Training Program and 6*
7-added a ceneral Superintendent to coordinate the work of the S
9, welders on the project, to monitor their capabilities and
.0 l JL '
progress, to initiate retraining where needed, and to work JZ,
- g closely with the welding engineers and welder training
- 4 department.
In addition, a new B&R Chief Welding Engineer
..a di assumed responsibility for working closely with construction,
.7,
JI I-welder training, B&R corporate welding engineering, and
.9 i lQ QA/QC groups to institute programs to further improve welding
- 2 performance.
- 34
- 4 Radiography on the site was temporarily limited to work
'y \\
S under the direct supervision of Level III QA Engineers.
All 14 site NDE personnel were retrained and recertified.
A new l9 '
procedure was implemented for the control of film processing 10 j ll "
and another new procedure prohibited the shooter of radio-12 13 graphic film from also doing the film interpretation.
In 14
- 5,
addition, B&R personnel performing liquid penetrant examina-16 '
tions were retrained.
- 7 ;
18 '
A review of all radiographs on the Project was undertaken.
- g,
Both HL&P and B&R surveillance teams were to conduct special i2 reviews and surveillance of the NDE program.
.3 i4 l Backfill procedures were changed to specify depths for iS I ys '
conducting in-place density tests and a test program was
- 7'
- g ;
initiated on site to determine whether proper density had l0 l been obtained thus far on the Project.
The results of that il ;
1 2
3' 4
5 program are described in the testimony of Mr. Pettersson and 6-7; Mr. Hedges.
8 9,
The foregoing improvements have been completed and the
}
NRC has verified their implementation in its inspection 2'
reports.
Some of the improvements related to administrative
.3
.4 controls and were carried over into items of the Show Cause
.S
.6 order that I will discuss below.
.7,g Q. 5 The first Item of the Show cause order which
[9 relates to administrative controls is Item 5, which required 0
n
{}
that HL&P define more clearly the stop work authority, f*
temporary or otherwise, including implementation of the stop IU work authority.
Please explain the meaning of the term 16 17 "stop work authority."
13 19 A. 5 This term refers to the authority of certain 10 31 Project personnel to require that construction work be 12 33 stopped.
The construction work affected by a stop work Ii' order can be of broad scope, such as stopping all construc-la 16 tion on the Project or all welding; or it can be narrow, 17 '
18 such as stopping a,particular craftsman from working or i
i 19,
l 10 l prohibiting use of a particular piece of equipment.
On STP, I
11 52 as on most other projects, the term "stop work order" is l
13 !
g4 only used for the broad scope work stoppages.
The narrow i
ff scope work stops are called " holds."
t 17 i I
is i 19 i
10.
31 i -.
L 2
3i 4
5' Q. 6 What conditions led NRC to order a clarification 6i 7
of the stop work authority?
3i 9!
A.
6 The Order does not state explicitly.
- However, Oi 1i the NRC'c April 28, 1980, Investigation Report 79-19 (NRC 2'
,3 i Investigation Report) on which the Order is based, states d
that the B&R QC Inspectors expressed uncertainty regarding
.3
.6 their authority to stop work and also that a Construction
.7 i
.3 '
Foreman and QC Inspector expressed conflicting views regard-
.9 0
ing the stop work authority of QC Inspectors.
1 3
Q. 7 Are you familiar with the HL&P-B&R Task Force 3
4 described in Mr. Briskin's testimony that was established to 5
prepare responses to the Order?
I' A.
7 Yes.
I was in charge of the Task Force until
.3 9
HL&P responded to the Notice of Violation on May 23, 1980.
0 1
Thereafter Mr. Briskin was given responsibility for the Task 2-3, Force's preparation of the response to the Show Cause Order.
4-5 I was put in charge of drafting the response to Item 1, and i
6-several members of my staff participated in the other sub-7 8
groups of the Task Force.
I participated in meetings with 1l, 0
those subgroups and have reviewed working papers of the Task 2
Force subgroups that described findings a.nd proposed changes
'3 4l to procedures.
Upon completion by the Task Force of proposed 5i 6
procedure changes, I participated in the decisions on those 7i
,g ;
changes.
>9 !
- 0 il i M
L 2
3 4
5 Q. 8 What were the findings of the Task Force concerning 6
stop work authority?
7 3
A. 8 The Task Force found that HL&P had procedures in 9
10 effect which designated certain EL&P QA personnel as having L1 12 stop work authority, and that B&R had comparable procedures.
13 14 However, it also found that neither the site procedures of HL&P nor those of B&R clearly addressed the stop work author-
'7
{g ity of personnel below the Supervisors level.
13 '
Q. 9 What was EL&P's response to Item 5 of the Order?
20 31 A. 9 HL&P stated that both EL&P and B&R had defined 22 23 the stop work authority more clearly.
It went on to identify 24 gg the positions in each organization which have stop work 26 27 authority, and explained how they can exercise such authority.
23 Q. 10 What positions in the EL&P organization have gg
}'O stop work authority?
32 A.
10 The EL&P QA Staff organIization is described in 33 34 -
the answer to Question 7 of my previous testimony concerning l
30 l
36 the current QA program for STP.
37 39 The.EL&P procedures distinguish between two types of 39 t
40,
stop work authority:
a temporary oral order, called an 4;9 Emergency Stop Work Order, and a written order of indefinite 4.
43 duration called simply a stop Work Order.
The HL&P site 44 45 QA/QC supervi.sory positions and the STP QA Manager have 46 47 authority to issue a written Stop Work Order.
All other t
48 l
49 EL&P QA/QC personnel have authority to issue Emergency Stop 90 51 l !
1 l'
1 2,
3!
4 5
W rk orders, which remain in effect until an HL&P QA or QC Supervisor issues either a written Stop Work Order or a 3I written rescission of the Emergency Stop Work Order.
The 9l 0
affected organization is then prohibitea from resuming the 1'
2 work activity until receipt of a written release from HL&P 3
4 QA.
nj Q. 11 What positions in the B&R organization have stop 7
work authority?
g 9'
A.
11 The B&R QA organization is described in the 0
1 answer to Question 8 of my previous testimony concerning the 2
4 3
B&R procedures also provide two types of stop work 5
,7 authority:
the authority to issue a hold tag, which places
[S '
a hold or work stoppage on some construction activity of 9
0 narrow scope; and a stop work order, which is used to stop a
- 2 broader range of construction activities.
i3 >
l4 i QC Inspectors apply hold tags at the time they identify l
- 5
I
- 6 on a Nonconformance Report (NCR) a condition that does not 17 i is conform to design requirements.
A hold tag is a standard 19
- o.
form that is physically attached to the nonconforming material l:,;
or work.
The hold tag prohibits continuation or commencement e.
I3 of the designated activity until there is a disposition of 14,
15 I the NCR as a result of the procedures described below.
When 16 17 l the nonconforming condition is corrected or resolved in some 18}
gg other way the hold tag is removed by a QC inspector.
Thus 30 j 51 l
~
L 2,
3' 4'
5l all QA/QC personnel have the authority to stop work immedi-6' ately through identification of a nonconformance on an NCR 7;
3I and issuance of a hold tag.
O The on-site authority to issue a stop Work order is 1!
2<
limited to the B&R Project QA Manager.
These are written 3
4 orders to organizations involved in design or construction 5
~
.6 activities, which must be acknowledged in writing by the
- 7 i affected organization.
gi 3!
In situations where IMJL QA/QC personnel believe that a 1
Stop Work Order should be considered, they submit a written 2
3 notice of deficiencies to the B&R Project QA Manager.
If
.4
.3 the Project QA Manager decides a Stop Work Order is required,
.6
.7 '
he orally notifies the Supervisor of the organization perform-
- S,
9 ing the work and the HL&P STP QA Manager and transmits a written Stcp Work Order to the organization performing the 2!,3 work.
The Stop Work order must be signed, dated and returned i!
to the B&R Project QA Manager by the organization performing la l 16,
the work, thereby acknowledging receipt and verifying that i
17 '
l 18,
the work has been stopped.
If the Project QA Manager deter-l 19 :
10j mines that a Stop Work order is not required, he so notifies o({i the QA/QC personnel who reported the condition and the l
13 cognizant Construction Supervisor together with his reasons g
15 I for so acting.
l 16!
17 !
Q. 12 What role does HL&P have in Stop Work Orders 18{
19 i issued by B&R?
30 l 51 !
I l.
i 1,
2.
3' 4'
5' A.
12 Authorization to resume all or part of the' work 6
7 affected by a B&R Stop Work Order may be given only by the 8l 9t B&R Project QA Manager.
He will issue such authorization 0j, only after he has determined that all responses, corrective 2'
jg actions, recurrence controls and other requirements have 4
been satisfied, and the HL&P Manager, STP QA, has determined
.6 the resolution to be acceptable.
Written approval of the
.7
.3 '
HL&P Manager, STP QA is a prerequisite for issuance of a
.9 ;
l0 total or partial release to the organization performing the 11 12 work.
13
,4 Q. 13 Do the procedures of HL&P and B&R that control
'S stop work authority comply with Appendix B to 10 CFR Part
.O 50, applicable NRC regulatory guides and other applicable
(('
19 industry standards?
10 '
31 A.
13 Yes.
Appendix B Criterion II and Regulatory 32 33 Guide 1.28, which endorses ANSI N45.2, andress the require-34 !
35,
ments for stop work authority being placed within the QA 36 37 -
- 9""l**tl "'
38 l The revised HL&P and B&R procedures provide the author-39 i
40 i ity for QA/QC personnel to stop work verbally whenever a 41 '
42 safety related noncompliance is identified.
Provisions for 43 -
44,'
formal stop work documentation follow-up is also provided.
49 !
46 These procedures address all activities of fabrication and construction as well as engineering, design, hardware and QA l
47 gg
+
49 50 51 i l
1!
2 3l 4i 5i program deficiencies and identify the internal organizational 6!
7l activities as well as EL&P/B&R interface responsibilities 8i 9l for imposing and releasing the Stop Work Order.
LO l
- 3, i Q. 14 Have the new stop work procedures been properly.
f,2 l implemented?
3 lf' A. 14 The new B&R Stop Work Procedure, ST-QAP-15.2, lo L6 :
Revision 1, was made effective on January 26, 1981.
Training L7 i LE l on the procedural requirements was completed prior to that L9 !
10 !
date.
The procedure has been fully implemented since.
l' !
gj l Since the issuance of QAP-15.2,.there have been two B&R Stop 23yj Work Orders invoked.
In one case the discrepancy was resolved in one day.
The other has undergone two partial releases,
- > i ZS.'
with the remainder currently being resolved.
In both cases 19 !
the Stop Work procedure was properly implemented.
30 i 31 4 HL&P Prcject QA procedure PSQP-A-7 was issued on July 32 l 33 ;
25, 1980 and HL&P QA Department procedure QAP-12-A was 34 !
35 l issued on September 22, 1980.
These procedures more clearly 16ll delineate which positions have stop work authority.
In I&E 1
l 37 38 39l Inspection Report 80.-27, the NRC Staff reviewed the procedures i
10l and found that stop work authority is adequately described.
l 11 !
l 12 j Q. 15 Item 6 in the Order required HL&P to develop and i
13 I l
14l implement a more effective system to provide for the identi-15 l 16 }
fication and correction of the root causes of the nonconfor-17 !
gg j mances which occur.
Please explain the meaning of the term i
!9 "nonconformance."
a0 l
il l
1j 2
3l 4
A.
15 A nonconformance is a condition that is not in 56!
accord with Project requirements such as job specificatio'ns, d
7I 3l procecures for the control of special processes, etc.
That 91
.0 l could include a concrete structure that does not have all of 1l 2l the steel reinforcement' required by the design drawings or a
.3 welder not being certified for the type of weld he is doing, lB or any other deviation from Project requirements.
6,
- 7l Q. 16 How are nonconformances identified and documented?
.O j
9I A.
16 It is the job of QC Inspectors to verify that
- 0 CLl2l construction work is being done in accordance with the C3 <
applicable requirements.
When the QC Inspector identifies a
!4 :
15 condition that does not comply with the requirements there
'S
[7 !
is a procedure he must follow that is designed to assure io :
jg*
that the nonconforming condition is properly dispositioned (i.e., either correc'ted or reviewed by Engineering and found 32 I to be acceptable).
At the STP the procedure involves the 13 l 14 initiation'of a document called a Nonconformance Report 15 i 16,
(NCR).
This is the same NCR I mentioned in ccnnection with 17 i 33 I my discussion of stop work authority.
The QC Inspector's 19 i 10l initiation of an NCR is the primary way that nonconformances p; i are identified during construction.
j 1.
4 13l Q. 17 So that it will be clear when the subject comes 14 15 up later, please describe the field design change procedure.
16 47,
A.
17 The field design change procedure is described 18 i 49l in detail in Mr. Briskin's testimony.
Prior to the Show l
50 i 51 t
i i
L L
3-1l-Cause Order the field change procedure involved use of a document called a Field Request for Engineering Action 7I I'
(FREA); now the appropriate form is a Field Change Request 3
3i (FCR).
Field design changes are changes to design that are 1'
2' requested by personnel at the job site.
A typical reason 3
4 for such a change may be that there is an interference 9
between piping and cable trays which could not be antici-6 I
i pated by the designer, but which was realized as construction g
9i work was planned in the field.
The NCR and the FCR are 0
1 similar in that each reports a condition that appears to 2
3 require evaluation by Design Engineers and in either case a 41 5
change in the design may result.
'6
,7 Q. 18 Please explain what the systems are that identify
's,
{g and correct the und. lying causes of nonconformances.
A.
18 The idenrification of the underlying causes of l,
12 '
nonconformances involves the analysis of NCR's to determine 13 4
14 I what condition on the Project might have led to the error in 15 !
16 question.
For example, a particular type of error may 17 i 33 result from ambiguous construction procedures, from a particu, 19 ;
lar construction worker's misunderstanding of a procedure, 10 !
1}9 or from an inadverrent error by a worker.
The Project must g
33 have a system to attempt to identify which among these or 14 i
45 other possible causes is actually involved.
One tool in 46 47 identifying the casues of nonconformances is an analysis of 48 I
49 data regarding incidents of nonconformance.
Such analyses i
50 51 j
1 2
3 4
5!
as ased to detect problems that might indicate the existence 6
\\
71 of a common, underlying cause.
This process is referred to 84 as " trend analysis" or " trending."
g
.0 once the underlying cause of a nonconformance is identi-
.1
~ ~ ~. _
.2 fied, it may be possible to prevent the same type cf condition
.3
.4 from recurring by correcting the underlying cause.
T3us if 3
.6 the cause were an ambiguous procedure, the procedure :ould
.7,
,g be clarified and the personnel using that procedure rehrained C
]i to the revised procedure.
Several such instances might i,
jg suggest a need for a review of how procedures of a certain 13 class are prepared and/or reviewed prior to publication.
If 15 the cause was misunderstanding on the part of one worker.
16,
17,
that man might be retrained or reassigned.
A large number 18 gg of such instances could reflect the need for wider re-training 30 or a refresher course for workers.
There is, of course, 3,,
3j*
also the possiblity that the nonconformance is an i-!ated 3
34 i incident and that it does not demonstrate that any recurrence 3D 36 ;
control is required.
37 :
38 Q. 19 What was the reason that NRC ordered HL&P to
~
39 i 40 ;
implement a more effective system for identification and g,4j l correction of the underlying causes of nonconformances?
43 A.
19 Here again, the Order doesn't state explicitly, 44 5!
but the NRC Investigation Report found that there was no 47 !
effective system for statistically analyzing FREAs that
'I 48 '
49 '
50 51, L
L 2,
3 4;
5, might have been used instead of an NCR to document noncon-6*
7:
forming conditions.
As discussed further in Mr. Briskin's a
g testimony, inappropriate use of the field design change f
procedures (FREA's) in lieu of an NCR could inadvertently E2 mask an underlying reason for a given problem.
NRC also
.3 L4 referred to the results of past trending reports and con-L5
.6 cluded that there were trends in the data which were not
'7
.g' being picked up in the reports.
- 9
.IO, Q. 20 What was the approach of the Task Force in
- 1
{}
responding to Item 6 of the NRC Order?
f3 A.
20 The Task Force decided to review the process by 15 which nonconformances are identified, documented and resolved.
16 UT,
It then analyzed how to improve the trending process which 13 gg was the focus of the Order.
Finally, the Task Force reviewed 10 g1 the procedures for ordering that a cause of nonconformances 12 be corrected.
With respect to each of these three phases of gy I4 the process, the Task Force - which had representatives of IS 16j Construction, QA/QC, Site Engineering, and Quality Engineering -
17 18!
reviewed the objectives of the process and proposed new 19 '
~
10 l procedures to better achieve those objectives.
I. was a il :
- 2 thorough review of the entire process.
i3 !
- ,g j Q. 21 Before the Order, what was the procedure used at f
the site to identify and document nonconformances?
![l A.
21 Under the procedures in effect at the time of j
l no'9 ;
10 :i the NRC investigation, B&R QC Inspectors planned and performed ill t
i 1
1 l
2 3'
4 5'
their inspection and then recorded the inspection results on one of a variety of different forms, depending on the type 8<
of construction work inspected.
When nonconformances were 9
LO '
found in a final inspection, the QC Inspector rece: ad that LL '
L2 fact in a draft NCR. 3hich was submitted to the Lead QC i
13 14 Inspector.
From the Lead QC Inspector, the draft NCR was
~g S
reviewed successively by the QC Supervisor, the QA Engineer 6
g 13 After the QA NCR Supervisor prepared the final typed 20 21 NCR, it was first sent to Design Engineering for disposition 22 23 and then routed successively to a number of other organiza-24 '
29 tions for review and approval of the disposition.
The NCR, 26 -
27,
with the disposition indicated, was then routed back through 23 several offices for information purposes before Construction 29 3
re eived the form and implemented the approved disposition.
3 32 Q. 22 What were the Task Force findings about the NCR g
33,
l 34 -
system?
35 36 -
A. 22 The Task Force found that the system was unneces-l 37 38 sarily slow and cumbersome and that consequently field 39 l
personnel tried to rvoid using it.
It noted th2t the inter-40 ;
4k.
views with QC Inspectors summarized in the NRC Investigation 4
43 Report showed that their morale was adversely affected by 44 l
45 i
the lack of feedback on how the nonconforming conditions l
46 47 l
they identified were corrected or why their management 48 49 decided not to process certain draft NCR's.
00 51 l
l t
l..
L, 2,
3*
4 Q. 23 What is the new procedure for identifying and 5
6 documenting nonconformances?
7 3
A. 23 The first change is in the planning of inspec-9t 0
tions.
The expanded B&R Quality Engineering organization 1
.2 l (QE) described in Mr. Oprea's testimony participates in
.3
,4 construction planning and determines inspection " hold points" f6 f r w rk activities.
When Construction reaches such a hold
'7 point it cannot proceed until a QC Inspector performs the
.5
.3 required inspection and approves the work to that point.
!O 11 Sometimes " hold points" are not practical, such as.in contin-12 13 uous inspection of concrete placement activities.
In these 14 gg cases other inspection guidelines are specified in procedures.
'S{7 ;
For each inspection hold point QE determines the inspec-f3 i
tion characteristics required, and those inspection character-30 istics are recorded on preprinted Inspection Report forms to 31 32 be used by QC Inspectors in their planned inspections.
QC 33 34 !
Inspectors record the results of all planned inspections and 35 '
36 l,
reinspections on the Inspection Report.
The Inspection Report 37 i 38 ;
is used to track all items found in planned inspections to 39 t be unsatisfactory, from the time of initial identification 40 4f to the time the items are satisfactorily resolved.
43 j When a planned inspection for acceptance of wor 3 is 44 45 {
being conducted, all checklist items noted on the In spection 46 47 Report are verified as "S" (satisfactory),
"U" (unsatisfac-48 49 tory), or "N/A" (not applicable).
The QC Inspectors, under 90 i
51 i
1, 2,
3l 45j new procedures, promptly notify the Construction Foreman or 6
General Foreman of items recorded as "U" and initiate a NCR.
8!
The NCR is a pre-numbered and control'.ed form that is acknowl-9!
.0 i edged by the signature of the Construction Foreman or General
.1 !
.2 i Foreman.
Where appropriate, hold tags or other work con-
.3 '
,,g straints are applied by the QC Inspector at the time the NCR is issued.
5
'7 Q. 24 How is the NCR processed?
g 9I A.
24 When nonconforming items or conditions can be
!O '
11 !
reworked to the original configuration or brought into
!2 l
!3 compliance through a " standard repair procedure", then no
!4 i
!5 design evaluation is required.
'S
{7 Inspector and the Construction General Foreman agree on the SS
{g
" standard repair procedure" to be used and record that I
agreement in the disposition section of the NCR.
Once the 32 '
disposition is complete and approved, the QC Inspector 13 ;
14 '
removes the hold tags and work may proceed.
The QC Inspector 15 i 16 :
also records on the Inspection Report the fact that the NCR 37 l 38 i has been resolved.
NCR's falling into this category are 19 t 10 then forwarded to the cognizant Quality Engineer for review.
The Quality Engineer must either give written approval or I3 ;
reinitiate the NCR.
44 l 45l When an item will not or cannot be reworked to the 46,
47 i "ac-designed" configuration, the NCR is submitted to Quality 48 !
49 l Engineering for review.
If it is determined that the NCR is 50 l 51 i i
1 2
3l 4,
5 no t valid (i.e., the QC Inspector misinterpreted the require-6' 7l ment), the NCR is dispositioned by Quality Engineering, and 8'
explanatory comments noted in the justification section of 9
.0l the form.
If Quality Engineering determines that the NCR
.1 '
.2,
can be dispositioned by rework or a " standard repair pro-
.3
'4 cedure", neither of which requires further design evaluation,
.S
.6 that disposition is also noted on the form.
In either of
- 7 g
these cases the form is returned so that any needed rework
{
.or standard repairs can be completed and the QC Inspector Il can record on the Inspection Report that the condition is 12 13 satisfactory.
As discussed below, this helps to assure that 14 15 QC Inspectors are aware of the disposition of their findings.
15 17 i Most nonconforming items are resolved through this 18 gg orderly interaction of QC, Construction and Quality Engineer-10 '
gg ing.
The nonconformances that cannot be resolved by these 2
organizations are the conditions that require ucsign evalua-I4 '
tion.
A new on-site committee, called the Materials Review 15 16 '
Board, has been created to coordinate the resolutisn of 17 l 18 '
NCR's requiring design evaluation.
The Materials Review 19,
p3 j Board consists of a senior representative from QA (Chairman),
p' gj, Design Engineering, and Construction.
An Authorized Nuclear 13 9;
Inspector (ASMI) and representatives from Westinghouse, I3!
Purchasing, and Materials control are available on call.
I 66 II The Materials Review Board members coordinate the 68 l 19 '
determination of a proposed disposition based on consultation 10,
il ',
i 1
1; 1
! 'i y
with their respective organizations.
When the memberu 4
07 f concur on the disposition, each member signs the NCR.
(Desim. Ch' age Notices which may be required in the case of LO l "use-a,-is" or " repair" dispositions are discussed further L1 :
12 in Mr. Briskin's testimony, in connection with Item 7 of the 13 14 Order).
When necessary, on-call members are consulted 13 16 '
before approval of a proposed disposition.
The signed NCR, 17 i with disposition noted, is then sent to EL&P QA for review ig e
'9 30 f the disposition to ensure concurrence that QA requirements 21,
have been implemented.
Thereafter it is sent to B&R Construc-22 23 '
tion for implementation and to B&R QC fcr inspection.
24 29 '
Q. 25 In what wa' s does the n. " Nr.P c acedure represent 26 ;
27 i an improvement?
28 29 ;
A 25 appeared that in the past, there were times
}' O i when an NCR was processed and the condition corrected, but g
32 the QC Inspector who had written the NCR never found out 33 ;
34 '
about the corrective action.
Because they had not heard 35 I 36 i otherwise, some of the B&R QC inspectors believed that the 37 !
l 38 i nonconformances.they had identified were not corrected.
Also 39 !
40 i the NRC Investigation Report said that because NCRs were 4j :
occasionally revised or discarded by QC management in the course 49 l
43 +
some QC Inspectors felt that they were not supported or review, 44 l 45 !
by their management.
Now NCR's are prepared by the QC 46 47 Inspectors without review and the new procedure does not 48 l
49 l
permit discarding invalid NCR's.
90 l
51 l
i t
L 2
3 4i 5;
The Inspection Report form was adopted so that there 6i will be a document which will record the NCR, its resolution 7i 89j and the final acceptance by QC.
Inspectors will now know
.0 exactly where they can look to see how a matter was resolved.
.1 i
.2 !
When an NCR is improperly written, the disposition of "use
.3 4
as is" is noted on it by the Lead QC Inspector or the Quality
.D
.6 Engineer and the justification is documented.
Thus, there
.7l
,g '
is an explanation available to the QC Inspector of how and 9l why the matter was resolved.
This is also the reason QC 0
!1 ;
will be notified of the disposition of all NCRs.
!2
!3 The NRC Investigation Report said that there was a 14
!S tendency on the Project to document nonconforming conditions 16 17,
through the field design change system instead of the NCR LS.
gg system.
The Task Force concluded that the reason for this 10 ',
was that the NCR process was cumbersome and slow, and Project 31 32 '
personnel found it faster and easier to use the field design 33 34f change system.
The new NCR system is very easy to use, and 3a i l
36 !.
with the addition of the Materials Review Board the NCR l
37 !
l 38 i process is much fas,ter.
In addition, as explained in 39 !
toj Mr. Briskin's testimony in connection with the discussion of 4'
43 the field design change system (Item 7 of the Order), the 43 procedures now clearly require that all nonconformances be 44 ;
45 documented through the NCR system.
47 !
l 48 !
49 l 50 !
51 i
L 2,
3l 4
Q. 26 Does the revised procedure comply with Appendix 5
6 B to 10 CFR Part 50, applicable NRC Regulatory Guides and 7
8, applicable industry standards?
9:
,0 l A. 26 Yes.
Our procedures in the past and the new
.1 !
,2,
procedure are consistent with Appendix B Criterion XV and
- 3 ANSI N45.2 Section 16 which deal directly with control of 4
f=,j nonconformances, as well as other associated criteria such L7,
as QA records (Criterion XVII).
La L9 !
Q. 27 How well have the new nonconformance reporting 20 '
21 ;
and disposition procedures been 3orking?
22 23 A. 27 The new nonconformance crocedure was initiated n}a
}
on October 15, 1980.
Since that time the Material Review f,
Board (MRB) has been in operation.
EL&P Discipline QA 23 !
personnel have also been directly involved in the approval 29 30 cycle of all NCRs initiated by HL&P as well as all B&R NCRs 34 32 i requiring disposition by the MRB (i.e., all NCRs other than 33 I
34
" rework" or " standard repair").
In addition, the MRB has 3S 36 been providing HL&P QA with MRB meeting minutes which list action taken on all NCRs.
EL&P extends its involvement 39 through the use of the B&R NCR system for resolution of
(
40 1
41 nonconformances identified by HL&P QA personnel; EL&P QA has 42 also performed reviews of the nonconformance control system 43 i
44 45 l
and we have found that it is functioning adequately.
46 i
47 Q. 28 Before the Order what was the procedure used on 48 i
1 the Project to analyze the trends in nonconformances?
49 SO l
51
-100-
l L
2, 3;
I I
4l 5
A. 38 Prior to February 1980, the trend analysis 6
7; function at STP was an informal process undertaken by various 89l members of B&R Project management.
Results were reflected primarily in the corrective action taken.
The formal trend
.2 analysis that was performed analyzed a portion of the STP
.3 4
inspection reports, NCR's, and selected Field Requests for
?
6 Engineering Action (FREAs).
These data were not normalized.
.7,g l The identification of a trend was based solely on the number Q
{~ '
of incidents reported in a given category of activity.
HL&P n
{-
reviewed B&R trending reports and did no independent trend analysis.
In February 1980, we began to develop a formal 15,
trending program.
This was committed to in our early response 16 17,
to information from the January 24, 1980 exit interview with 13 gg the NRC.
l 30 Whst additional changes were made in trending 31 m.
32 !
33l j b performance in response to the Order?
34 i A.
29 A new Data Analysis Group, comprised of Quality 39 36 j Systems Engineers within B&R QA, was established in July, 37 1
38 :
1980, and is now responsible for identifying the methods to
~
l 39 !
40 j be used to collect data, the ways to categorize and monitor 41 '
42 deficient conditions by the use of quality indicators, data 43 normalization and graphic representation, and the methods of 44 4f !
reporting this information to manageatent.
Formal procedures 47 require the collection of data from all QA records that j
48 I
49 record nonconforming conditions.
50 i
51 l
i
-101-1
L 2,
3; 4i 5;
A new unif rm coding system has been developed and all those charged with collecting data for trending have been 8
trained in the use of the system to assure uniform applica-9;
.0 '
tion.
Standardized codes have been developed for a number
.1 !
.2l of categories, including:
building or area, activity,
.3 '
i,4 failure type and time interval.
t-
'l6 Procedures specifically identify the types of documents L7 l to be trended and the B&R organizations responsible for
,g l9k review of each type of document, collection of data and 20 :
11 '
submission to the Data Analysis Group.
Examples of documents 22 23 included in the list are:
NCRs, Corrective Action Requests 24 23 (CARS), Vendor Surveillance Reports and Audit Deficiency 26 27 f Reports.
The Data Analysis Group is responsible for reporting 23 ;l the results of its analyses to B&R Quality Engineering.
29 3
l Quality Engineering reviews suspect areas to determine 32 !
whether the incidents that make up a trend have a common 33,
34 !
cause.
35 !
36 l The trend analysis performed by the Data Analysis Group 37 3g is made available to HL&P QA at the same time that it is 39 I distributed to B&R management.
The HL&P Supervisor, Quality 40 ifl Systems reviews B&R data and other data collected by HL&P 43 '
implementation reviews, and performs an independent trend 44 45 review.
I include this information in the monthly report 46 47 i that is sent to Mr. Oprea and use it in assessing the per-1 48 49 formance of the B&R QA program.
50 51
-102-l
[
1 2,
3 4
5 Q. 30- Why does this new procedure represent an improve-6 7
ment over the previous system?
o 89' A. 30 The Task Force found that the trending efforts prior to the NRC investigation were not effective because 2
there was no centralized responsibility for collecting data
.3
.4 and perform.ng trend analysis; and trending was not controlled
.D
.6 by formal procedures.
Now those formal procedures have been
.7
.g promulgated and a competent engineering staff has been
'9 O
established with primary responsibility for collecting and
}
analyzing the data.
3 In addition, the NRC Investigation Report found that 4
5 the effectiveness of the trending efford was reduced by the 6
7 use of the field design change procedure in situations in
.8 9
which an NCR would have been more appropriate.
That affected 0
1 trending because not all field design changes were included 2
in the trending.
By procedurally assuring that nonconforming 3
conditions are not handled as field design changes, we have 6
eliminated this potential problem, i.e.,
we have taken steps s
S to eliminate the use of the field design change process for 9
0.
the resolution of what are really nonconformances.
1' 2
Q. 31 Does the trending program now in use on the 3
.,g Project comply with 10 CFR Part 50, Appendix 3, and appli-5i cable NRC Regulatory Guides and industry standards?
6
.7 i 8l 29
- 0,
il ;
-103-
L 2
3, 4l 5
A. 31 Yes.
Appendix B and ANSI N45.2 do not specifi-6' cally require a trend analysis program or establish any 7i 8
criteria relative to the mandatory content of a trending 9 i 10 l program.
However, Criterion XVI of Appendix B requires that L1 12 '
corrective action be taken to preclude repetition of signifi-13,
14 cant conditions adverse to quality.
The new trend analysis 15 16 system provides a method for detecting adverse trends and 17 l for initiating investigations to determine whether or not gg corrective action to preclude repetition is required.
B&R 21 performs trending rf the B&R indentified deficiencies and 22 23 HL&P trends EL&P identified deficiencies, evaluates the 24 35 B&R trend analysis and compares the results of both.
26,
27 ;
Q. 32 How well have the new trending procedures been 38 i 29 working?
30 i l
g Thetrendingprgcedureswereimplementedin A.
32 32 f July, 1980 and the B&R Data Analysis Group issued its first 33,
34 !
Quarterly Report in January, 1981.
As a result of the B&R 35 i 36 l trend analysis effort, to date ten suspect Trend Investiga-37 I 38 !
tion Requests have been initiated.
The HL&P trend analysis 39 i 40 l effort has resulted in the issuance of thirteen Trend Investi-49 I fl gation Requests.
We expect that in the long term the trend analysis efforts will aid in reducing the number of non-l 45 conformances,on the Project.
46 !
i 47 !
48 j l
49 l l
00 :
l 51 l
-104-
4 L
2' 3
4 5,
Q. 33 Bef re the issuance of the Order, what was the 67l procedure for directing that actions be taken to prevent o
3 recurrence of nonconformances?
.0 i A. 33 Prior to the Order the B&R Project QA Manager JL :
L2 used a form called a corrective Action Request (CAR) to L3 L4 direct an organization working on the Project to take action LS L6 to prevent recurrence of nonconformances.
The CAR would L7 identify the organization affected, the work activity and gg b3 the problem in need of correction.
When the corrective 20 Il action was taken, the affected organization would reply to 22 13 the CAR, describing the corrective action.
24 '
2D Q. 34 What were the Task Force findings about the CAR 26 27 ;
system?
23 A. 34 The Task Force found that the CAR system generally 29 30 worked well, but that in some cases it took longer than 32 necessary for an affected organization to respond to a CAR.
33 34 !
Q. 35 What is the revised procedure for recurrence 35 36 ;
control?
37 '
38 A. 35 B&R Qual.ity Engineering now has responsibility t
39 40,
for ensuring the identification of the causes of trends and 4}5 issuing CARS for recurrence control.
CARS are still issued 4
j 43 to identify for correction, significant or repetitive condi-l 45 !
tions adverse to quality and procedure inadequacies.
- However, l
46 47 CARS now have specified time limits for taking responsive 1
48 i
49 action and may constrain or " hold" work on specific tasks or 50 l
l 51 j
I
-105-l l
Li 2
3l 4i 5'
by specific crews until the identified problem is resolved.
6 After a CAR is issued, Quality Engineering determines whether 7;
8 i corrective action has been taken, and whether it was effec-0
'0 l!
tive in preventing recurrence of the problem.
Quality
.1 '
'2' Engineering may initiate a Stop Work Order if corrective
'3 L4 action is not taken within the time limits or if the inves-LU L6 tigative finding is that the corrective action has been ineffec-L7 i tive.
In addition, HL&P QA identifies problems to B&R QA for (g
{
underlying cause investigation and recurrence control 21 l via the B&R CAR system.
HL&P QA reviews all CARS for con-22 -
23 currence with the B&R disposition and performs periodic 24 !
25 follow-ups to determine the effectiveness of implementation.
25 27 i Q. 36 Why do these new procedures on reporting, trend-23 :
29 !
ing and rectifying problems represent an improvement over 30 the previous system?
32 I A. 36 With the creation of the Data Analysis Group, 33 34 !
the MRB, and the new procedures assoM ated with the noncon-l 35 i 36 i formance reporting system, our understanding of quality 37 I 38 l problems is substantially strengthened.
The Data Analysis 39 1 I
40j Group has improved the process of identifying trends adverse e
41 i 42 ;
to quality.
The MRB and Quality Engineering scrutinize NCR's to determine if recurrence control by a CAR is requirec}.
l 45 These changes,substantially increase our ability to identify 46 '
l 47 the need for recurrence control.
l 48 l
49 i
l 50 51 l
-106-i
1 i 2
3l 4t 5l Q. 37 Does the revised system for corrective action 6\\
7, comply with Appendix B to 10 CFR Part 50, applicable NRC 0
Regulatory Guides and industry standards?
LO !
A. 37 Yes.
The procedures comply with Criterion XVI L1 ;
L2 !
of 10 CFR 50 Appendix B and Section 17 of ANSI N45.2, as L3 '
14 well as other related requirements and criteria.
15,
16 Q. 38 How well have the new corrective action proce-L7 i (g !
dures been working?
L9 '
A. 38 Since the issuance of the new CAR system in 20 31 October of 1980, the procedures for processing CAR's have 22 23 been refined.
Difficulties with the system were initially 24 25,
experienced because the CAR system was used to identify 26 27,
minor problems, thus diluting its effectiveness in high-23 i 29 -
lighting information to management.
To eliminate this 30 3g problem we have initiated a new form called the Field Action 32 Request for minor procedure deficiencies.
The CAR system is i
E4 !
now reserved for its intended purpose--significant and 30 !
36 l recurring conditions.
This adjustment, together with HL&P 37 :
38 i QA's increased involvement and control, are leading to much
~
39 !
s to i more effective recurrence control.
The backlog of CARS 9
l 43, issued under the old s1 stem makes it impossible at this time 13 i g4l to present a quantitative assessment of the new system.
Its 15l strengths will, however, result in a decline in the numbers 16 -
II !
of NCRs and CARS.
%8,
i
%9 '
30 51,i l
-107-
L, 2,
3!
4, 5!
Q. 39 Item 8 of the Order required that HL&P develop 6i 7i and implement a more effective system of record controls.
8 g
How were records controlled prior to the Order?
O A. 39 Prior to the Order the QA vault, which is the 2i place where on-site QA records are stored, kept a number of 3
4' separate files for each of several phases of records process-5 6
ing.
Incomplete or inadequate documents were placed in a 7
g suspense file until the originating department was notified oi
~!
of the deficiencies and the deficiencies were corrected.
9
}
Completed records were placed in the pre-microfilm suspense file pending microfilming.
They were then kept in the 5
post-microfilm suspense file until film processing in Houston 6
7i was completed and a determination made that the microfilm 8-9 copies were acceptable.
Thereafter, the records were placed 0'
g in the permanent files.
-2' 3
Q. 40 What was the background behind the NRC Order i'
regarding the record controls?
s 6,
A. 40 That is not clear.
There was no discussion of 7j S
the record control system in the Investigation Report.
I 9I
~
0l understand that during the investigation there was an inci-l 1
2 dent in which an NRC inspector requested a record from the 3i 4
storage vault that was not located for several hours.
Based 5I on that we understood the NRC's criticism to be that it 7!
sometimes took too long to retrieve records from storage.
8!
9l 0:
1!
i I
-108-j
1 2;
3l 4I 5!
Q. 41 What were the Task Force findings about the 6l 7j record storage system?
8i g
A. 41 The Task Force found that the system of having a f0 separate file for each stage of record processing made it 12 !
difficult to locate certain files.
There were too many 13.
14 places to look.
It appeared best to reduce the time required 19 l 16 ;
for each processing stage, so that files could be quickly 17 i lg !
placed in a central file and cross indexed.
19 !
30 l Q. 42 How does the current records control system 3}9 l w rk?
2 23 A. 42 Prior to commencement of a work activity, Quality 24 35 '
Engineers specify the requirements for the QA records necessary 36 ;
37 i to substantiate the individual activities.
When work activities 28 l 29 :
are completed, Quality Engineering reviews the required 30 !
31,
quality documents prior to turnover and testing to verify 32 !
33j that the documents are complete and adequate.
Si f New microfilm equipment has been acquired to speed up 33 I
36l:
filming and to establish an on-site film processing labora-l 37 l
38 i tory devoted to the,QA Vault needs.
This new equipment has 39 i 40l reduced the time required to film documents and to develop, il i 42 !
process and verify the film.
l 13 I g4j An automated record index system now permits rapid 15 16 identification of records related in any of a variety of 17 l ta i 19l 50 j 51
-109-
1!
2 i' 3I 41 5;
Possible ways (e.g., purchase order, inspector, heat number, 6
drawing number).
The backlog of data to be put into the 8l computer data base has now been substantially reduced.
9i
'0 I There is a single filing system, and' documents are filed
.1 i L2l with a record " traveler" prepared by Quality Engineers.
The L3 '
L4 travelers identify the records required to substantiate each LS L6,
activity.
The index system records the status of the file,'
7 and the location of documents borrowed from the file, and is g
l9 !
used to identify overdue and missing records.
20 '
21 !
Q. 43 Does the records control system comply.with 22 13 '
Appendix B to 10 CFR Part 50, applicable Regulatory Guides 24 '
25 '
and industry standards?
26 27 A. 43 Yes.
Regulatory documents which define the 23 i 29 i requirements of a quality assurance record system are Appendix 30
~, Criterion XVII, and ANSI N45.2 Section 18.
These documents B
31.
32 '
define requirements for collecting, filing, indexing, storing, 33 34 !
maintaining and dispositioning of records.
35 !
36 -
HL&P STP Site Quality Assurance Procedure PSQP-A4, 37 I 33 i
" Control of Site QA Documentation" and B&R Quality Assurance 39 !
40 i Procedure ST-QAP-17.1 " Records Control Procedure," adequately 4'f' address all applicable requirements at the STP construction 43 i
site.
Prior to any submittal of quality records to record 44 45 storage facil,ities, both procedures provide for a review by 46 47 i QA personnel for completeness and adequacy.
48 j
49 i
50 i
51 l
l
-110-
1 2,
3!
4!
5l Specific procedures for control of documents filed as 6I quality records are in place for handling all records in the 7 !
a!
record vault.
These procedures are in compliance with the 9
1l i
0 applicable regulatory requirements.
2 Q. 44 Item 9 of the Order (the last item on admin-
.3
.4 istrative controls) required EL&P to develop and implement
.5
.6 an improved audit system.
What is the purpose of an audit
.7 l
,g{
system?
o'
}l A. 44 An audit system is used to determine the adequacy 1 '
-i of, and compliance with, established procedures, instructions, 3
drawings, and other applicable documents, and the effective-a S
ness of implementation.
6, 7.
Q. 45 Which organizations perform audits at STP?
.3 !
.9 l A. 45 The HL&P QA Program requires that planned and 0!
3i periodic audits be performed to verify compliance with all I
l3 aspects of the quality program.
HL&P performs such audits d
internally as well as audits of Westinghouse Electric Corpo-O i
'7 li ration, B&R, and of others as necessary, to determine that 6
l
- 8 -
j the STP QA program,has been developed, documented and imple-9 0
mented in accordance with established requirements.
HL&P 1i 2;
and B&R both have responsibilities for implementing audit I
3i
,,g '
systems at STP.
B&R, as architect-engineer and constructor,
,5 provides all guality functions on the project within its t
,6
i
~
.8 ;
scope, including the implementation of an audit system in i
,9 !
iO il i
-111-L
1 2;
3l 4!
accordance with 10 CFR 50, Appendix B, Criterion XVIII.
5, 6i HL&P also audits B&R's performance.
The B&R audit respon-7; 8l sibility is performed by the B&R home office QA Audit Section.
9, Ol HL&P's audit responsibilities are performed by the HL&P 1,
.2 ;
corporate QA organization, which has offices in Houston.
.3
,4 Q. 46 What was the background behind the NRC Order a
,6 :
regarding the audit system?
'7 i
A. 46 The Investigation Report found that both the
,g 9
HL&P and the B&R audit staffs were relying on a review of QA
- 0 l1!
records in their audits rather than observing work; there l2 ;
01 were several occasions when audits were conducted at less
!4 g5 than the required frequency, and neither staff had been
!6g7f conducting supplemental audits of problem areas.
'Sl Q. 47 What did the Task Force find ibout the audit
{g program?
I2 A.
47 The TesK Force found that the primary causes of 13 34!
the deficiencies were lack of sufficient staffing levels on 15 l 36 j.
the audit staffs and inadequate training.
The frequency of 37 i 33 {
audits and the depth of audits had both been curtailed by a 19 10 l
shortage of qualified auditors.
It was also found that HL&P g},:
had not included the requirement for supplemental audits in 13 !
its audit procedures.
14 !
15 Q. 48 What are the principal areas in which improvements 46,
17 !
have been made to the HL&P and B&R audit programs?
48 i 49 !
50 j 51 ;
i
-112-
1 !
2, 3i 4;
5l A. 48 The principal changes to the B&R and HL&P audit 6'
7i programs were:
8i 9!
(1)
In both cases procedures have been changed to
- 0l' assure that supplemental audits are performed.
y
- .2 l (2)
Both audit staffs have been upgraded through
.3 '
.4 increased manpower and training.
.a
.6,
(3)
HL&P audit procedures have been revised to increase
.7,g l audit depth, and both HL&P and B&R procedures were revised
[9 to assure that audits cover all aspects of the QA program.
0 n'j;,
Q. 49 What are supplemental audits and what procedure changes were madt in connection with them?
l5 A. 49 supplemental audits are audits conducted in l6 ;
!7 '
addition to regularly scheduled audits.
Generally, a supple-
!S i
!9 '
mental audit is conducted when there is some reason to 10 i il !
suspect problems in a given area.
Although HL&P did conduct 12 '
supplemental audits from time to time, the criteria for 13 j i:
conducting supplemental audits were not addressed in the 2
16j
- 7,
audit procedures in effect at the time of the NRC investiga-
- 8 !
tion.
The procedur,es now explicitly provide for supplemental
,9 l
- 0 l audits.
In addition, the changes upgrading both the HL&P 1.l 2j and B&R audit staffs and the HL&P and B&R audit procedures
)
3!
..; j will result in greater audit staff resources and better 5
'l 6
control over,the scheduling of supplemental audits.
7 8!
o f
~
G i
l
- 2. l
\\
1 l
-113-l
i 1!
2, 3i 4'
l 5l An important element contributing to improvement in 6i 7l this area is the improvement in trending and identification 8:
of the causes of nonconformances which I discussed above.
9 LO l The B&R audit group and Project QA management regularly L1 :
12 !
receive the reports of the trend analysis group, thereby 13 !
14 enabling them to identify problem areas and establish the 15 '
16 l need for performing supplemental audits.
17 l gg !
HL&P has established its own program to analyze trends 99 '
based en the B&R trending dana as supplemented by HL&P 20 21 l collected data.
The HL&P Site QA group has established a 22 23 Quality Systems group that reviews documents which record 24 !
25 r unsatisfactory conditions and thet monitors B&R trending 26 27 reports.
The HL&P audit group and the EL&P Project QA 28 i 29 !
management receive both.T&P and B&R trend analyses.
30 3;
Q. 50 What steps have been taken to upgrade the respec-3 :
tive audit staffs of HL&P and B&R?
33,>
34 A.
50 The audit staffs of HL&P and B&R have been 35 36 j upgraded through both increases in manpower and training 37 !
38 i programs directed at improving the auditing skills of the 39 !
40l respective staffs.
While HL&P and B&R have been recruiting 47 42 qualified personnel, consulting firms have been employed to 43 provide experienced nuclear auditors to augment the audit 44 45 staffs.
46 47,
48 l I
49 50 51
-114-
i 1,
2 3!
4i The function of the HL&P corporate audit group has been g
7l!
6 restructured to minimize conflicting assignments and respon-8 sibilities of auditors.
Pre.viously this staff consisted of 9l, LO l five and was responsible for document reviews and procedure L1 !
L2 j development in addition to its audit responsibilities.
Now 13 '
14 the group's primary function is to prepare for and conduct 3-
{j audits of quality-related activities and the staff has been f
increased to eight.
Auditors are continually receiving 19 I additional training in the areas of codes, standards, proce-20 !
21 !
dures, and other documents related to QA programs and audit-22 l 23 <
ing.
For example, EL&P has retained an outside consultant 24.
23 '
to conduct an inhouse training program for its auditors.
26 l Participants have included Lead Auditors, Auditors and 27 20 Auditors in Training.
When successfully completed, this 30 i program leads to certification as a Lead Auditor in accord-31 '
32 l ance with the criteria of ANSI N45.2.23, provided other 33 '
34 !
procedural requirements are met.
I 35 1 36 l Q. 51 What changes have been made in audit procedures 37 I 3g i of HL&P and B&R?
39 40 l A.
51 The HL&P corporate audit procedure has been 41 revised to require both the review of objective evidence 42 '
43 !
(records) and direct observation of work being performed to 44 45 assure adherence to procedures and compliance with quality l
46 l 47 !
requirements.
i 48 49 SO j
51 l
i
-115-
1i 2;
3I 4;5; HL&P has developed an annual audit plan, which is 6
7 reviewed and rev.ised at least once every six months.
In 0I addition to the annual audit plan, a more detailed audit 9
0l schedule is issued quarterly.
This schedule provides for 1i 2!
supplemental audits as well as required audits.
A matrix 3'
4 has been prepared delineating all B&R procedures applicable 5
6, to STP and the corresponding audit (engineering, construction, l
discip line, etc.) to which they apply.
This matrix is 9
utilized by the HL&P and B&R audit groups to assure that 0
both groups audit all quality activities within the required 3
frequency.
4 5
Q. 52 Do the audit programs of HL&P and B&R comply 6
7i with Appendix B to 10 CFR Part 50, applicable Regulatory S
g Guides and industry standards?
i A.
52 Yes.
The HL&P and B&R Audit Programs meet the 2'
requirements of 10 CFR 50, Appendix B, Criterion XVIII.
All 3;
4' audits are conducted in accordance with ANSI N45.2.12.
5i 6:
7!
Auditors are certified in accordance with ANSI N45.2.23, 8
which has been endorsed by Regulatory Guide 1.146.
9' Ol Q. 53 Does the overall Project QA Program, which you 9
?
have described in your testimony, including both the HL&P
)
1
~'
3 4,
and B&R portions of the program, comply with Appendix B to 3!
10 CFR Part 50, applicable Regulatory Guides and industry 0
l 7!
standards?
8i 9!
0 1!
t
-116-t
L 2
3' 4
5, A.
53 Yes.
As described in my previous testimony, the 6'
7l QA program for STP complies with Appendix B, applicable
,f Regulatory Guides and industry standards.
.0 l As an additional assurance that we will continue to
.1 1
.2 i remain in compliance with all requirements, HL&P is committed
.3
,4 to having an independent audit of the STP QA Program at
.5
.6 least once every 12 months.
.7,
.3 I
.9 ;
l0 l1 !
l2 '
'3 T. Hudson:ll:A 14 15 16
!7
!S 19 -
10 '
11 12 i3,
14 :
15 16 :
17 !
18 ;
19 ;
60 i l
il i 12j is !
14 1 15 66 1
17 -
18 i 19 !
10 !
il
-117-