ML20129E608
| ML20129E608 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 10/18/1984 |
| From: | Hayes B NRC OFFICE OF INVESTIGATIONS (OI) |
| To: | Dircks W NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| Shared Package | |
| ML20129E602 | List: |
| References | |
| NUDOCS 8506060549 | |
| Download: ML20129E608 (13) | |
Text
'. 0 ENCLOSURE 2 UNITED STATIs
[
NUCLEAR REGULATORY COMMISSION gOctober
.r 18, 1984
(
MEMORANDUM FOR:
Will'
. Dircks, f
Execut we Director for Operations
- ]
n 8. Hayes, Director FROM:
fice of Investigations
SUBJECT:
INVESTIGATION OF TMI-2 POLAR CRANE ALLEGATIONS' Your September 26, 1984 memorandum to me requests that DI provide you our views on the issues of deliberateness, wi11 fulness, and/or intent arising from the subject investigation. I believe that you will find this memorandum and f ts enclosures responsive to your mquest. However, before we deal with the specifics of this matter, I think it useful to recount the events leading up i
to this effort and to explain the criteria that we used to arrive at our opinion.
On June 4,1984. Dr Henry Myers fomarded five questions on this subject to Norm Haller. O! was subsequently assigned responsibility for the answers to
{
questions 1, 3, and 5. The staff, in this case the TMI Project Office, had the responsibility for questions 2 and 4. Although both offices prepared their answers separately, we agreed to share one anothers draft answers. The 0!
response.to questions 3 and 5 caused concern to some members of the staff.
The staff's particular concern,is Ol's position that noncompliance with GPUN administrative procedures represented some degree of deliberateness. This
. issue resulted in'.several meetings between 01 and members of your staff. These negotiations did not result in any significant change in'the opinion.of either party. Consequently. 01 furnished OCA its draft answers to the questions for.
which it was responsible. When you and I talked about this matter, we agreed that I should meet with Harold Denton and Dick DeYoung for further discussion.
During that subsequent meeting, I expressed reluctance to produce a written i
document capturing the evidence in the OI report supporting the OI viewpoint l
unless you so requested it. You did, and this is why we have undertaken this project.
l I believe it important, as a first s.tep, to place OI's position in proper i
' perspective. OI's opinion - it is only an opinion, albeit an informed opinion l
- was articulated in response to specific questions raised by Dr.Myers regard-ing the staff's evaluation of OI's September 1,1983 THI-2 report. OI was essentially asked if we agreed that there was 1)"no evidence of deliberate circumvention of administrative requirements" (the staff's finding) and, 2) if such " administrative deficiencies" were more the result of confusion than deliberateness...
Deliberate means intentional or willful. Similarly, and perhaps redundantly, t
circumvention means avoidance by stratagem. Thus the issue was and is whether there is any evidence that the noncompliance was purposeful or intentional rather than accidental. The NRC Enforcement Policy also defines careless disregard of NRC requirements to be tantamount to will.ful.
8506060549 050517 fDR ADOCK 05000320 PDR
,c W. J. Dircks 2
j t
Given this O! was in a quandary. The staff had made a categetrical statement that there was "no evidence of deliberate circumvention". Thus any quantum of evidence would invalidate the staff's categorical "no evidence" statement.
Even though we had not made explicit reference to deliberateness in.either the OI report or the memorandum of transmittal, we believed that both documents contained implicit indications of deliberate circumvention of administrative requirements by TMI-2 personnel. We further believed that both documents con-tained ample infomation to illustrate that the decision to not foJ1ow the i
administrative procedures was the end result of an ordered thought process j
rather than confusion.
i In contrast, the staff asserted that the noncompliance was more the result of confusion than of deliberateness. It was our view that this statement had to be read in hamony with the "no evidence" statement. Thus we interpreted the statement to mean that the administrative deficiencies were solely the result of confusion. To interpret this statement as meaning that there was some deliberateness but an even granter amount of confusion would be to acknowledge that some evidence of deliberate circumvention exists. Such an interpretation would appear to contradict the staff's earlier "no evidence" statement, Given the fore' going, 01 saw only one option. If we believed that the 0! report j
i documented anJ evidence of deliberate rather than accidental avoidance (i.e.,
i circumvention), then OI had to disagree with the staff's position. This is how we arrived at the aforementioned impasse..
l
. Upon. receipt of your September.26 memorandum, I directed my staff to review
' the 0! report and 'all'do'cuments appended to or referenced by it. Thus, I i
i l
limited our activities to analysis of those documents that were available to x
l your staff during their review. In short, we revisited the material already~
relied on by the staff in fomulating its position that is set forth in SECY 84-36 and SECY 84-33. As enclosure 1 reflects, this review revealed consider-able evidence to support DI's view that the circumvention of the required administrative procedures by TMI-2 personnel was at least to some degree deliberate, and that it was primarily the result of work schedule difficulties i
rather than a lack of understanding of the requirements. In a word, the i
motivation appeared to be expediency. not confusion.
We started with an assumption of competence on the part of both GPUN and Bechtel regarding NRC requirements. Both entities have considerable nuclear experience with Bechtel in particular being involved in the engineering and construction of a large number of nuclear facilities.
The requirements in this instance are quite staightfomard. The technical specifications require promulgation and compliance with GPUN adninistrative procedures. Moreover.
I the contract between GPUN and Bechtel makes explicit reference to the need for compliance with the GPUN administrative procedures. Thus we feel this assumption to be quite reasonable. Certainly the implications of either this licensee or a major AE/ constructor of a number of nuclear facilities in the NT0L stage being perplexed by such requirements could be somewhat unsettling.
The forgoing assumption is strengthened by documentary evidence, in the fem of memoranda and quality assurance reports, that senior GPUN/Bechtel managers were aware not only of the requirements, but the fact that Bechtel was not
,-,m
-.-.m- -, - -
^.
'e W. J. Dircks 3
i
~
complying with them. Indeed, at one point d' senior GPUN avin'ager'falt coiapelled to write a memorandum to Bechtel noting that 1) Bechtel was not complying with these. procedures and 2) that they must do so. This memorandum in turn even-R L
tually engendered a Bechtel written response in which they promised to follow the procedures. The record reflects, however, that they did not. It'is perhaps noteworthy that these documents precede the allegations made by Messrs Parks, King, and Gischel.
Evidence also indicated that Bechtel felt that the adriinistratiNprocedures were too cumbersome. Indeed, a quick comparison of the Work Packages developed by Bechtel reveals them to be considerably less bulky than the packages associated with the required Engineering r,hange Memoranda. This coupled with testimonial evidence supports, in our view, our conclusion that this circum-vention was motivated primarily by expediency.
In summary. TMI-2 senior personnel were aware of the need to comply with the GPUN administrative procedures. They did not do so in all cases even though they were evidently aware that such compliance was an NRC requimment.
i As before, I wish to point out that 0I is not taking any position on any enforcement decisions which may flow from this effort. Such decisions are clearly within the authority and. responsibility of the staff.
In clos 6g, I regret that this matter has required the investment of so much.
, of our time and limited resources. On the other hand, if the effort assists 4
the NRC at arriving at an informed decision in this important matter, it no
- doubt has been sorth (hi's time.and effort.
w.
Enclosures:
As stated
}
t e
I l
l
. L.'
ANNOTATED INDEX OF RELATED DOCUMENTS / STATEMENTS 1.
Page 7 of the BNoC Contract, which requires that BNoC will develop for owner's, approval a manual dealing with such items as review / approval procedures, administrative procedures, quality assurance procedures, requirements for safety, etc.
(Article ;3.1)
.2.
GPUN Memorandum 4240-81-0170 dated Augurst 13, 1981, from Plant Engineer-ing to the QA manager, indicating that a conscious decision had been made shortly after the accident to set aside design reviews of modifications necessary to mitigate the consequences of the accident in the interest of j
expediency.
3.
GPUN Letter' to the QA manager. BNoC from the QA manager. GPUN dated October 22, 1981. QAf3732, advising BNoC that their QA Manual dated August 1981 was approved conditional on BNoC recognizing therein the necessity of complying with the GPUN technical specifications for all j
design, co,nstruction and recovery activities.
4.
GPUN Letter from Recovery Operations to the BNoC Field Construction Manager at TMI II dated February 23, 1982 referencing administrative procedures and requiring SNoC to comply with GPUN written procedures and instructions until such time as BNoC replacement procedures are approved by GPUN. This letter also indicates that GPUN and BNoC had discussed this issue a " number of times in the past."
Specifically, two of the i
procedures mentioned in the attached list for BNoC compliance are AP 1043 j
and AP 1047.,
l l
5.
EPRI Memorandum dated April 26, 1982, to the GPUN Technical Planning l
Director, expressing concern that there are certain exclusions in the i
i i
Bechtel recovery plan.
Further, that the plan is not consistent with-industry QA practices.
5.
GPUN QA Assessment dated May 1982, submitted by the TMI QA manager to the 1
Vice President / Director. TMI Unit II which states (on page 3) that a OA i
5 top Work Notice was initiated on May 21. 1982 as a result of repetitive l
violations of administrative controls for conducting work activities.
l l
Construction work was apparently accomplished without approved engineer-i i
ing docu.ients which define and authorize the activities.
The TMI QA manager asserts that management at TMI II appear to have the attitude j
toward administrative control programs that it takes too long to get work authorizations approved and into the field.
}
i 7.
BNoC Letter from the Field Construction Manager to the GPUN Recovery Operations and Construction Director. June 29,1982, in response to i above, confirming that BNoC has reviewed the GPUN Adminis-I trative Procedures and have no significant concerns about applicability I
or usage.
- BNoC also states that it is their full intent to comply with j
GPUN procedures until they establish their own.
I I
l S.
GPUN Memorandum 4350-82-0417 dated July 13. 1982. from Recovery Op.
l erstions to the TM1 QA manager, outlining the limited use of " Work Packages" in conjunction with GPUN authorized documents such as i
)
~
procedures and Engineer Chr.nge Memorandums (ECM). This letter states that it represents the position of GPU management with respect to thi It specif-quick look program, et leat,t, a'nd the use of work packages.
ically identifies that the work package is not a substitute for:other
. appropriate procedures or documentation.
1 a
Quality Deficiency' Report (QDR) submitted by QA management August 9 9.
1982, reiterating that work packages cannot substitute for approved J
This QDR was forwarded to the Recovery Operations and procedures.
Construction Director as the responsible party for corrective action.
~'
GPUN QA Assessment August 1982, submitted by the TMI QA Man'ager.to the 10.
Director, TMI Unit II, and the Director. Quality Assurance.
On page 3 this report cites recent monitorings of the quick look program, which identified problems with 8echtel Work Packages.
It characterized these as not being proceduralized and "cannot be used to accomplish important to safety work activities on the unit."
- 11., 8echtel Procedure Number CDPI-20 dated September 1,1982, " Control and Documentation of Work after Release from GPU to Construction " which, in paragraph 4.0 stipulates the use of work packages to control work, replacing GPU TMI Unit 2 procedures.
(There is no indication on the procedure document that it was approved by GPUN.)
GPUNQAAssessmentOctober1982,submittedbytheTMIQAManahrtothe )
12.
En paragraph
.Dirpetor, TMI Unit II, and the Director. Quality Assurance.
(3), page 2, it states that there' continues to be confusion in deternin-Paragraph (4),onthesame ing safety classificationi for components.
page, refers *to a stop work condition in Un'it 2 involving repeated violations of QA requiiements for work being perforined prior to the engineering document / work authorization being used.
i i,
GPUN Memorandum 4420-83-0038 dated January 27, 1983, from the Manager.
t 13.
TMI-2 Safety Review Group, to the Polar Crane Task Group leader acknowl-edging that the use of the work package format is not in strict complf-ance with AP 1047, "Startup and Test Manual" guidelines, with specific reference to th2 polar crane no load test.
4410-83-M-0137 dated February 4, 1983, from the Manager.
GPUN Memorandum 14.
THI-2 Licensing to the Deputy Program Director. TMI Project Office, NRC, with an attached Procedure Change Request, addressing the limited opera-tion (no load) and full operation (no load) test of the polar crane, dated January 25,1983. The Procedure Change Request specifically identifies the test as "Important to Safety." The memorandum indicates i
that the test was documented on a " Work Package" format, which GPUN planned to implement on February 7,1983.
4240-83-111 dated February 10, 1983, from the Plant GPUN Memorandum 15.
Engineering Director to the' Site Operations Director, referencing the This memorandum advises that polar crane load test safety evaluation.
the plant engineering review of this. safety evaluation finds it techni-cally unacceptable.
e I---*
-wee--
_--,-,r-.-.my.,,
-..g
---,---__----w-
-e
%,-,m, ym-n
.-.,. - - _, _, -,,,_-.y.wmme-ww.,_- - - - -
-,,%me-
,m.,, _,.,,.
4300-83-F-002 dated February 17, 1983, from the Deputy GPUN Memorandum 16.
Manager of Recovery Programs, to the plant Engineering Director, refer-encing the pt.it engineering finding on the polar crane load test safety j
evaluationFlhis' memorandum attempts'to answer the concerns"digiressed by
. plant engineering in Attachment 14 above.
\\
6110-83-039 dated February 23, 1983, from the TMI QA GPUN Memorandum
~ 17.
Manager to the Director of THI Unit II., addressing concerns raised re-One of these concerns surfaced by the garding the polar crane load test.
QA manager is that modifications to the polar crane were not appropriately documented.
Quality Assurance Modifications / Operations Section Document Coment 18.
Sumary, apparently completed and reviewed in the March 1983 timeframe by This sumary j
TMI QA and fontarded to the Bechtel Pohr Crane Task Group.
documents a review of a Unit Work Instruction, dated February 26, 1983, titled " Load Test for Polar Crane." The sumary review is based on AP 1047, and uncovered several points of noncompliance with AP 1047.
4370-83-1019 dated February 28, 1983, from the Recovery-GPUN Memorandum 19.
Operations Manager to the Director of Site Operations, referencing resolutions to coments made regarding the polar crane load test proce-dure (the same, Unit Work Instruction mentioned in' Attachment 17 above) resulting from a previous review of the procedure by Site Operations.
Site Operations found several areas of noncompliance with administrative review and approval procedures (AP 1043 and AP 1047).
In their resolu-tions of these coments, dated February 25, 1983, the Polar Crane Task l
Group generally contends-tha.t AP 1047 does not apply, since this proca-Under Item 4 dure is a coristruct.iona.1 test, and AP 1047 is operational.
of the attached Coment Resolution form it indicatef that this procedure
- J This appears to be in was reviewed and approved by Q0ality Assurance.
f contravention of the date of review and the spirit of coments by Quality Assurance in their Document Coment Sumary as described in Attachment 17 of this sumarization.
GPUN QA Assessment Report, February 1983, submitted by the TMI QA Manager 20.
to the Director, TMI Unit II and the Director Quality Assurance, which indicates (on page 3) that there was a failure to define and reviseThis existing procedures / requirements for " turn back" to construction.
report also alludes to a continual problem in Unit 2 with implementation of administrative controls. The report states that it appears that when a new activity is initiated that doesn't exactly fit into the established controls, the tendency is to work around the program, rather t'han making necessary procedure changes to accommodate the new activity or situation, The QA Manager contends that Unit 2 management must assure that work is j
conducted in accordance with presently approved program procedures or i
revise those >rocedures to reflect new practices.
Further, it is recom-l mended that tiis be given high priority, and if appropriate action is not taken, a QA Stop Work Action will be initiated.
4200-83-105 dated March 1, 1983, from the Acting Site l
21.
GPUN Memorandum Operations Director to the Startup and Test Supervisor regarding the applicability of AP 1043 and AP 1047 to the Polar Crane Test Program.
This memorandum references a meeting on February 23, 1983 held by the
4 TMI-2 Director, and a meeting of the Test Working Group on February.25, 1983 Essentially it was agreed that these procedures applied to the polar erane program.
The memorandum also affims that the requirements of AP 1043 and AP 1047 were not being complied with, attributing this
, noncompliance to a lack of familiarity with these procedures.
f.-
22.
GPUN Memorandum, dsted March 7,1983, used as a cover letter for the minutes of a Test Working Group meeting held on March 4,1983 (attached).
These minutes mention noncompliance with administrative procedures on page 2.
However, the minutes further reflect (p3) that it was agreed that an Engineering Change Memorandum (ECM) would be required for instal-lation of certain fuses in the polar crane main disconnect. The minutes i
go on to state that for the sake of expediency, documentation would be '
prepared utilizing another administrative procedure to perform this modification in advance of ECM approval.
Lastly, the minutes indicate that it was agreed that the crane load test would be in compliance with i
AP 1047.
1 23., Quality Deficiency Report (QDR), dated March 8, 1983; submitted by the TMI Office of Quality Assurance, which classifies the polar crane refur-bishment as "important to safety" and cites various deficiencies related to noncompliance with AP 1043 and AP 1047.
This QDR specifically states that all refurbishment and testing activities on the polar crane were performed to Bechtel Work Packages.
Further, that this document format was defined in an internal procedure of Recovery Operations, and was not
. approved by any other organization at TMI-2 for performing work.
.24.
GPUN Memorandum 6110-83-046, dated March 10. 1983, concerning a polar crane refurbishment review conducted by the TMI Office of Quality Assur-ance. This review found numerous deficiencies in the Bechtel Work Packages. -It concludes that although the tests were technically adequate, the test results were not always clearly documented and administrative-l controls for testing (AP 1047) were not followed.
25.' Portion of a QDR regarding the polar crane refurbishment program describ
- ing corrective action taken by the Manager, Recovery Programs.
In the section titled "Cause" there is a description of the procedures used in -
refurbishment which were understood to have administrative 1y severed the crane from.the plant placing.it. in a " turned back to construction" condition.
The report then states, "This allowed engineering direction and decision making to flow in a more streamlined fashion, since many l
required refurbishment steps could not be defined until after inspections l
and tests were performed."
It is admitted in the QDR that no such " turn i
back" was ever provided for in the TMI-2 procedures.
Therefore, all l
crane modifications should have been accomplished in accordance with AP l
1043 and AP 1047, dated March 21, 1983. Under the section titled "Cor-rective Action" this report indicates that in future, AP 1043 and AP 1047 will be utilized in work activities.
26.
Extract of Pages I-2, 3, 4, and II-3, 9, 10, 23 from the Technical Examination of Alleged Procedural and Managerial Deficiencies at Three Mile Island, Unit 2.
Section I, pages 2-4, trace the procedural require-ments involved in this investigation through the technical specifications and regulatory guidance.
In addition, this section sumarizes the. major
--,---n--,n-----,-,_.-_
n__... _,. _ _, _ _ - -. - - -.. - _ _ _ _ _ _
l allegations and the results of the technical examination of these allega-l tions.
Section II, page 3. sumarizes the findings concerning the vie of BNot Work Packages.Section II',' pages 9-10, contain a sumary of the findings >regarding BNoC review and approval processes and procedures;
.Section II, page 23 sumarizes the findings re'.ated to siodifications to l
containsient. penetrations, safety classifications. and safety evaluations.
1 l
27.
Affidavit of Richard D. Parks dated Marc,h 21, 1983, listed as Exhibit D-2 L
in the.01 Report of Investigation.
i Parks indicated.that he was present at Head Lift Task Force meetings a.
i and either directly or indirectly through the Site Operations Office, issued correspondence from November 1982 on, wherein it was brought to the attention of plant management that there were safety-related violations regarding engineering / review / approval of proce-i dures and noncompliance with QA requirements. According to Parks, i
nothing was done to correct this condition until the issue was forced.
This topic is discussed by Parks on pages 6, 7, 8, 9, 20, 21, 22, 25, 26, 30, 32, and 34 of the affidavit.
I b.
TFe affidavit refers to a meeting which took place on February 23, i
i 1983, during which Recovery Operations contradicted a previously stated position. According to Parks R0 had previously said there were no replacements of unlike kind in the crane refurbishment (page i
22). At this meeting, they admitted that there had been such replacements (pp25.,26.and27).
c.
Pages 6, 7 and 9 of the affidavit describe violations of AP 1043,
.AP 1047. snd sther procedures, indicating Parks' opinion that such 2
violations were the result of an intentional decision by plant
~
management, essentially for the sake of expediency.
i d.
Pages 8,17-19, 21, 22, 30, 31, 38, and 39 of the affidavit coment i
on what Parks perceives as attempts by plant management representa-tives to keep the violations / noncompliance concerns from surfacing outside plant channels and even outside specific meetings.
e.
Pages 20, 25, 27, 28, 29, 51, and 52 refer to what Parks believes were attempts to intimidate, harass, transfer or fire him. King, and Gischal, as a result of their surfacing concerns over safety-related violations / noncompliance.
j 28.
Interview of Laurence P. King dated March 2,1983, listed as Exhibit D-4 i
in the OI Report of Investigation.
a.
Pages 3,.4,* and 5 of the interview contain information addressing l
the point that plant management knew and understood that procedures l
were being violated.
1 b.
Pages 1, 2, 3, ard 5 reveal coments by King which indicate that specific procedures were not complied with in order to meet schedul-i ing dates and other. deadlines.
4 1
~, -
-,_.._-_-_-~..___-__.,,.,,...._.__._r
, + - -
Information.was also provided by King (pp 3,4, and 5) suggesting c.
that plant management personnel may have attempted to contain concerns about safety-rela'ted noncompliance within plant channels.
.3 3...
...3 d.
,In the course of providing his oral statement to an DI investigator
/ *-
. (pp 1-5), King described the events. surrounding.his being fired and
. expressed his belief that the pressure to do so on plant. management resulted from his voicing concerns over violations.
He also men-tions his perception of possible intimidation and harassment efforts against others who tried to. surface safety-related noncompliance issues.
29.
Affidavit of Edwin H. Gischel dated April 2,1983, listed as Exhibit D-5 in the 01 Report of Investigation, a.
On pages 6, 9,10,15,16, and 20. Gischel describes various times and methods by which plant management was made aware that procedures were being violated over an extended period of time, and that there were safety-related concerns which needed attention.
b.
On pages 2, 6, 7, 9, 10, 12, 16, and 20 of the affidavit, Gischel discusses specific noncompliance and attributes it essentially to an emphasis by plant management on expediency and meeting scheduling
)
requirements.
.c.
0n pages 10, 11, and 16 of the affidavit, Gischel alludes to actions by plant management personnel which could be characterized as efforts to suppress the surfacing of safety-related concerns by
. plant. employees..,. -
d.
pages 9,10,11,13,14,15, and 16 of the Gischel affidavit contain information regarding alleged intimidation and harassment of plant personnel attempting to surface safety concerns, violations and noncompliance.
30.
Sworn statement of Edwin H. Gischel dated April 8,1983, listed as l
Exhibit D-6 in the 01 Report of Investigation, which outlines an incident i
in which Bechtel allegedly bypassed the plant engineering modifications control system in an attempt.to. downgrade safety classifications, pre-sumably in order to expedite work completion.
l 31.
Sworn statement of Lake H. Barrett dated April 13-14, 1983, listed as Exhibit D-9 in the OI Report of Investigation, which, on pages 3-6, 9-14, i
and 15, addresses various violations of safety-related procedures.
l 32.
Report of Interview of John J. Barton dated May 11-12, 1983, listed as Exhibit D-12 in the OI Report of Investigation, details Barton's denials l
of connents attributed to him regarding procedural violations having been accomplished for the sake of expediency.
Barton does admit that proced-ures were violated.
33.
Sworn statement of Blaine E. Ballard dated April.19,1983, listed as Exhibit D-13 in the 01 Report of Investigation, which, on pages 1 and 2
corroborates that TMI QA management cited plant management for violating administrative procedures.
34.
Sworn statement of Ronald'L. Freemerman dated May 12, 1963, listed as Exhibit D-18 in the 01 Report of Investigation, which, 6n pages 1, 2, and 4 indichtes that Bechtel was aware of GPUN procedures and set up their own procedures based on savings in schedule and cost of the Bechtel system over the GPUN system.
He also de. scribes the Bechtel " black box" theory for repairing the polar crane.,.'
'35.
5 worn statement of Mervyn K. Pastor dated July 28, 1983, listed as Exhibit D-19 in the 01 Report of Investigation, describes a series of conversations Pastor had with the manager of Recovery Operations, a Bechtel employee.
He opined that this manager understood the.GPUN procedures, but would have preferred to work under Bc:hrel procedures.
- 36. Sw'orn statement of John J. Barton dated August 4,1983, listed as Exhibit D-25 in the 01 Report of Investigation, which, on pages 1-3, comments on procedural noncompliance by BNoC, and efforts by GPUN to correct the condition.
e 4
0 1
w.
e e
I i
N 4
l
SUPetARY OF DI ANALYSIS
',, q, s,,
This documents the results of our review of DI report H-83-002 of September 1, 1983. This review, which was undertaken at the request of the Executive Director.for Operations was to' identify'andilist all evidence contained in or referenced by the subject report that indicates that the circumvention of GPUN administrative procedures by TMI-2 personnel was willful. W111 fulness is 4
defined in 10 CFR 2. Appendix CIII as a "... spectrum of violations ranging from deliberate intent to violate or falsify to and including careless disre-gard for requirements." Using this definition, our review identified a -
considerable amount of evidence indicating that the circumvention of proce-l-
dures was wi11ful. Additionally, although we found some evidence of confusion, l
as described below, the weight of the evidence indicates that the circum-i vention was a deliberate decision apparently based on a sense of expediency and was largely unaffected by confusion.
l It is worthwhile to clarify two other terms. First, we use the word " evidence" in its broadest sense, i.e. that which tends to prove or disprove something.
Second, we use the term " confusion" as meaning a state of mind on the part of THI-2 personnel rather than indiuting chaos or disorder, its more precise definition. We believe this to have been the intent of the staff when it described the circumvention as being a result of confusion rather than delib-ersteness as chaos would not be a condition reasonably expected regarding a closely regulated activity such as the THI-2 cleanup.
In order.,.to demonstrate that the circumvention was willful, one must demon-strate that 1) THI-2 personnel knew of the requirement, and 2) chose not to
-comply with the requirement. With regard to the first, we need only demon-
. strate that 1) it'is'cessonable to assume such knowledge based on the back-1 grounds of the personne1' involved, 2)that such knowledge"is reasonably a co_ndition of employment, and/or 3) key personnel had actual knowledge of these requirements. Our review revealed evidence of all three qopditions.
't The administrative requirements in question are NRC requirements. Development, promulgation, and compliance with GPUN administrative procedures is essential-ly required by the licensee's technical specifications. It is not unreasonable l
to assume that the senior GPUN officials at TMI-2, most of whom have had substantial nuclear experience, would be familiar with the conditions of the license. Indeed, it would be unreasonable not to expect this as a measure of licensee competence. At the same time, Bechtel, although a nonlicensee, has extensive experience in the construction of nuclear facilities, several of I
these in the NTOL stages. It does not appear unfair to expect Bechtei thus to l
be generally aware of NRC requirements for operating plants as certainly the design and construction of plants must acconmodate the operating requirements of the plant. These inferences notwithstanding, our review also disclosed evidence of actual. knowledge.
For instance, the contract between GPUN and Bechtel specifies that compliance i
with GPUN administrative procedures is mandatory. Additionally, GPUN advised l
8echtel in a February 23, 1983 memorandum that compliance with GPUN procedures was required by t-th the license and the contract in regard to coordingl. ion, review, and approval of modifications. Bechtel signified their understa? ding cf this and agreed to follow the procedures in a June 29, 1982 written i I
).
m
t.
Sunnary of. OI Analysis 2
response to GPUN. Notwithstanding this agreement, senior THI-2 management was repeatedly advised that Administrative Procedures (AP) 1043 and 1047 were being circunniented. This is pointed out in at least three quality assurance
- reports which were given wide site distribution. Note also that Messrs Parks.
. King, and Gischel had repeatedly pointed out,the need to comply with these
. procedures, but their attempts to correct the condition were rebuffed.
Indeed, there is considerable evidence tha't employees who attempted to raise these concerns were subjected to harassment, transferred, or other. wise pres-sured by management.
There is also evidence that there was a conscious decision by THI-2 officials to circumvent these procedures. For instance, an August 13,1g81 GPUN memoran-i dum indicates that a decision had been made shortly after the accident to set aside design reviews of modifications necessary to mitigate the consequences in the interest of expediency. The theme of expediency is touched upon also in the minutes of a Test Working Group meeting held on March 4, 1983. These
. document a consensus regarding the applicability of AP 1047.
However, the minutes further indicate that a modification of the polar crane would take place in advance of Engineering Change Memorandum (ECM) approval - for the sake of expediency.
There is evidence that there was some confusion regarding the proper require-ments at some levels at TMI-2. Some of the aforementioned QA reports indicate that some Bechtel personnel believed that the administrative procedures did
.not pertain to activities such'as the polar crane that were " turned over to
. construction".. A March J.1g83 GPUN memorandum reports that AP 1043 and 1047 l
were not being complied with' primarily due to a lack of familiarity with thenr.
Thlre are other documents such as Bechtel procedure CDPI-20 that could be,
e.
considered evidence of confusion or of deliberate circumvention.
Essentially, procedure CDPI-20 stipulates that Work Packages would replace the GPUN admini-strative procedures for work released by GPUN to construction. This could suggest that the subsequent use of these work packages was thus a good faith effort by Bechtel personnel to comply with CDPI-20, which they assumed was valid in the absence of objections from either GPUN or the NRC.
- However, it is important to note that CDPI-20.apparently was not app' roved by GPUN.
Further, it clearly was in conflict with the GPUN administrative procedures that Bechtel had previously agreed to follow.
Even if we assume that Bechtel was confused, there is ample evidence that GPUN was not. By memoranda and QA reports, SPUN officials indicated their awareness that the administrative proceures were not being followed, and so advised senior TMI-2 management. Yet this circumvention continued even after the initiation of the investigation of the Parks-King-Gischel allegations. In view of this, as well as the previously cited evidence of willfulness, it does not appear reasonable to consider confusion as having been a significant factor in the failure to adhere to the required procedures.
,w,.,,- - -., -., -
.-n.,
-,.--.-,-n--
i
., : 4).:...
o-l
. Summary of 0! Analysis 3
The above discussion highlights the evidence that in our vies speaks to willful avoidance of GPUN administrative procedures, and by extension, NRC r
requirements. In arriving at this overview, we reviewed all documents and statements appended to or referenced by the report of investigation. An annotated index of all of these documents is appended as enclosure 1 to allow l
the reader to view the same evidence that we relied on in arriving at our position.
i
Enclosure:
i Annotated Index with cited documents attached l
i e
6 6
- ..e e
~
e e
.