IR 05000272/1995080
| ML18101A776 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 06/08/1995 |
| From: | Eselgroth P, Shankman S NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18101A775 | List: |
| References | |
| 50-272-95-80, 50-311-95-80, NUDOCS 9506190462 | |
| Download: ML18101A776 (33) | |
Text
REPORT/DOCKET NO LICENSE NO LICENSEE:
FACILITY:
INSPECTION DATES:
INSPECTORS:
TEAM LEADER:
TEAM MANAGER:
U.S. NUCLEAR REGULATORY COMMISSION
REGION I
50-272/95-80 50-311/95-80 DPR-70 DPR-75 Public Service Electric and Gas Company P.O. Box 236 Hancocks Bridge, New Jersey 08030 Salem Nuclear Generating Station April 26 - May 12, 1995 Stephen T. Barr, Operations Engineer, DRS Michele G. Evans, Senior Resident Inspector, TMI Greg S. Galletti, Human Factors Specialist, NRR Tracy E. Walker, Senior Operations Engineer, DRS rgJj;/:fsft!r!A Technical Assistant Div of Reactor Projects
~uiyOir Div of Radiation Safety and Safeguards
EXECUTIVE SUMMARY From April 26 through May 12, 1995, a Special Inspection Team (SIT) under the management of the U.S. Nuclear Regulatory Commission (NRC) Region I Office conducted an assessment of performance at the Salem Nuclear Generating Station. The team comprised four inspectors, a team leader, and a team manager, all of whom were independent of the normal oversight of the Salem site. The purpose of this team inspection was to assess how effectively PSE&G was performing from a safety perspective in conducting day-to-day activities in the areas of problem identification and root cause analysjs, prioritizing and conducting work on plant equipment, equipment operability and configuration controls, and management oversight of plant performanc In summary, the station management team's attention is stretched between the oversight of daily emergent issues, efforts to improve on a daily basis and longer term improvement plan Problem Identification and Root Cause Analysis Salem has a complex group of problem reporting systems; but, problem reports reviewed by the team appeared to be appropriately prioritized within their individual groups. A check of Equipment Malfunction Identification System tags noted during plant tours against the Managed Maintenance Information System tracking system found the selected items to be appropriately entered into the system. Safety Review Group and Offsite Review Group reports reviewed by the team were well documente Incident reports, one of the problem reporting systems, have received increased emphasis and root cause analysis of incidents is improvin PSE&G established a Corrective Action Hit (CAT) Team to facilitate the use of and to evaluate the problem reporting systems; however, the ability of the CAT, at present, to fulfill its charter appears to be adversely effected by the continuing growth of root cause analysis wor Based on interviews with system engineering personnel, it does not appear that information available to trend/analyze outstanding action requests or the effectiveness of completed corrective actions associated with a system or component are being used effectivel Work Control The team assessed the effectiveness of PSE&G's processes for prioritizing, planning, scheduling, and controlling work on plant equipmen The team determined that the proceduralized process used to schedule non-outage work items was logical and well managed to minimize risk to plant operation When deficient conditions were identified by PSE&G, the action requests were reviewed promptly by planning to determine the planning system priority and appropriate action to address the condition. However, the team noted that repetitive tasks (particularly preventive maintenance activities) were not always scheduled to be performed before the due date for the task. Station employees indicated that it was a known problem that preventive maintenance work packages often did not meet quality expectations because, when the packages were originally prepared, standards for work planning were lowe The team noted that PSE&G made thorough work package preparations for the feed ii
- pump controller modifications that were planned for accomplishment over the weekend of May 6th. However, weaknesses in several other work packages and the use of work package feedback sheets were also note Management recently implemented actions intended to reduce the amount of unplanned emergent work being performed, by defining controls on "Sponsored Work"; however, what work required sponsorship was not clearly understood throughout the organization, including the Work Control Cente Management had also recently established a Safety Tagging Preparation Office to review work packages and prepare tagging requests in advance and this has benefitted the work control process. However, the effectiveness of this action was hindered because the changes to the work authorization process had not been clearly defined i~ writing and had not been clearly communicated to on-shift Work Control Center supervisors and fiist line maintenance supervisor Operations defined the daily priorities for the station, but these priorities were sometimes communicated poorly throughout the Salem organizatio The team noted weaknesses in the coordination of plant equipment activities associated with priority issues, such as a Unit 2 28V battery charger that was inoperable and in a seven day Technical Specification Action Statemen Configuration control I Work-arounds I Operability The Salem units historically have poor performance in the areas of configuration control, operator work-arounds and operability determinations for degraded equipment that constitutes a burden on the Salem operators to safely operate the Salem units. The team found that performance in these areas continues to be poo The team reviewed incident reports written in the month preceding the inspection and identified over 50 examples which described deficiencies in the area of configuration control. Although PSE&G has taken some compensatory measures to offset the problems in this area, the long standing and well documented configuration control problem area involving safety tagging and inaccurate P&Ids has not been significantly improve The team reviewed the current operator work-around list and determined that about 70 items remained as work-arounds from the list that was frozen at about 80 in September 199 During the inspection PSE&G indicated that the number of open work-around items was reduced currently to 3 The team noted that this reduction stemmed from splitting the list into work-arounds and plant betterment item The team concluded that the plant betterment list contained operator work-around items and that other operator work-arounds exist that have not been added to the lists since September 199 For example, the betterment list contains an item for the unreliable non-safety related charging pumps on Units 1 and 2 and states that operator attention is needed to keep them in service and that this usually results in securing them and having to run the safety related pump An item not included on either the walk-around or betterment list is the Unit 2 rod control being maintained in manual due to a rod stepping problem when in automati When station management was questioned about whether they considered the plant betterment list to be free of work-arounds they responded that they did not kno This lack of clarity regarding the true status of work-arounds detracts from management's abi"l i ty to assess the aggregate impact of work-a round The team iii
. '
concluded that the operator work-around problem area has not been substantially improved since late last summer and that additional management attention is warranted in this area in order to substantially reduce the number of operator work-arounds and their impact on plant operation The team found nine examples where the station had been operated with degraded equipment operability determinations prepared and accepted in the two months prior to this inspection that had deficient technical bases. These operability determinations inappropriately justified operability based upon equipment redundancy, the lack of Technical Specification or Updated Final Safety Analysis Report documentation, the lack of an effect on the reactor protection system, and fail-safe positionin (UNR 95-80-1)
Management Oversight The team observed an NRB meeting that critically reviewed the status of station activities and plans to improv The team identified several weaknesses in the effectiveness of management oversight and the performance indicators used by management to measure current Salem performanc The team observed the station manager's 8 AM meeting throughout the inspection. Although some safety perspective strengths were noted at these meetings, the managers were frequently not knowledgeable of the specifics for ongoing priority issues, both safety and non safety related, and demonstrated a weak safety perspective in other ways, such as not delving into a weak root cause determination for the failure of a containment airlock door leakage test that was later identified as wrong and not challenging a system manager conclusion that the service water system condition had not degrade In a number of cases, PSE&G management had available to them the data which could have been used to provide more meaningful measures of Salem performance, however, they were not routinely using this data to identify and evaluate trends, and measure performanc The performance indicators being used by PSE&G management to measure Salem performance were weak in that some indicators either were not predictive and/or had too high a threshold for inclusion of dat During recent weeks prior to the NRC inspection, PSE&G had taken action to improve upon several weaknesses with the performance indicators being used by Salem management to measure Salem performanc For some instances when trends were identified, PSE&G had not taken substantive action to address the trends in a timely manne For example, in the area of configuration control, an evaluation by PSE&G in 1994 revealed a negative P&ID deficiency trend, and this trend continues without substantive action to reverse i PSE&G has had a program for addressing individual problems related to configuration control and P&ID accuracy, as identified by the Salem incident report program, yet the team found that PSE&G had not taken the initiative based on the IR trending information to make an in-plant engineering assessment of the scope of the P&ID problem. Additionally, procedure adherence problems have been an area of deficient performance in past events at Salem and the team found that, although incident reports on procedure adherence problems had been showing an increasing trend of occurrences for the past several months, management had not taken action at the time of the inspection to assess the significance of this tren iv
..
EXECUTIVE SUMMARY TABLE OF CONTENTS ABBREVIATIONS... INTRODUCTION TABLE OF CONTENTS PROBLEM IDENTIFICATION AND ROOT CAUSE ANALYSIS
e
- .2.4 Problem Identification Systems *.
Nuclear Safety Review Group....
Incident Reports.... *....
Corrective Action Hit {CAT) Team.
Trend Analysis of Corrective Action Information........0 WORK CONTROL
.......... Scheduling of Work.... Schedule Adherence.... Work Authorization....
- Quality of Work Packages.... Work Priorities and Status.. Corrective Maintenance Backlog....
CONFIGURATION CONTROL / WORK-AROUNDS / OPERABILITY Configuration Control... Operator Work-Arounds *... Operability Determinations ii v
vi
1
2
3
4
6
7
11
12
14 5. 0 MANAGEMENT OVERSIGHT. *.....
....
16 Nuclear Review Board. *......... *.....
16 Station Operations Review Committee Performance.
17 Management Oversight Group
. *.....
.... *..
5. 4 Management Safety Perspective... *.
......
18 Management Oversight of Emergent Issues.
....
19 Performance Indicators 20 EXIT MEETING........*
ATTACHMENT 1 - SIT Charter ATTACHMENT 2 - Exit Meeting Attendees v
AFD AR BF BOM cw DCP EOG EMIS LCD LER MMIS MSR NED NP RDS NSG OD P&ID PI PM PMDR *
PSE&G RMS SGFP STPO TRIS UFSAR UNR wee wcs WO ABBREVIATIONS Axial flux differential Action request Boiler feed Bill of materials Circulating water Design change package Emergency diesel generator Equipment malfunction identification system Limiting condition for operation Licensee event report Managed maintenance information system Moisture separator reheater Nuclear engineering department Nuclear plant reliability data system Net safety gain Operability determination Piping and instrumentation diagram Performance indicator Preventive maintenance Preventive maintenance deferral request Public Service Electric and Gas C Radiation monitoring system Steam generator feed pump Safety tagging preparation office Tagging request information system Updated final safety analysis report Unresolved item Work control center Work control supervisor Work order vi
- INTRODUCTION The most recent NRC Systematic Assessment of PSE&G's Performance (SALP) for the Salem Generating Station, Units 1 and 2, was presented to PSE&G at a management meeting on January 12, 199 This NRC assessment highlighted, in particular, problems that PSE&G has experienced in the areas of problem identification and prioritizing, and conducting work on plant equipmen In connection with these problems, the NRC noted problems with accommodating equipment problems that resulted in unnecessary challenges to operators and safety systems, a general lack of a questioning attitude by operators, and weaknesses in operability decision makin Following the above NRC SALP assessment, PSE&G developed performance improvement plans and in March, 1995, chartered an organizational assessment team to critically review Salem activities and the improvement plan In April 1995, NRC Region I determined that an SIT inspection should be conducted to assess how effectively PSE&G is currently performing from a safety perspective in the areas of problem identification and root cause analysis; prioritizing and conducting work on plant equipment; operability determinations; operator work-arounds; and management oversight of plant performanc The SIT was conducted from April 26 through May 12, 199 The focus of this inspection was how PSE&G conducted day-to-day activities at the Salem Station related to safety performance, not a review of plans to improve. PROBLEM IDENTIFICATION AND ROOT CAUSE ANALYSIS The team reviewed PSE&G's programs for problem identification and root cause analysis to determine if adequate controls were in place to identify plant deficiencies and generate appropriate corrective actions including measures to judge the adequacy and effectiveness of such corrective action Summary Salem has a complex group of problem reporting systems; but, problem reports reviewed by the team appeared to be appropriately prioritized within their individual groups. A check of Equipment Malfunction Identification System (EMIS) tags noted during plant tours against the Managed Maintenance Information System (MMIS) tracking system found the selected items to be appropriately entered into the system. Safety Review Group and Offsite Review Group reports reviewed by the team were well documente Incident reports, one of the problem reporting systems, have received increased emphasis and root cause analysis of incidents is improvin PSE&G established a Corrective Action Hit (CAT) Team to facilitate the use of and to evaluate the problem reporting systems; however, the ability of the CAT to fulfill its charter appears to be adversely effected by the continuing growth of root cause analysis wor It does not appear that available information to trend/analyze outstanding action requests or the effectiveness of completed corrective actions associated with a system or component are being utilized effectivel *
2 Problem Identification Systems PSE&G's current problem identification process is composed of a group of problem reporting systems each governed by specific administrative procedure These reporting systems are inter-woven into a complex problem identification system which appears to provide an adequate means of capturing pertinent plant deficiencies. The team reviewed approximately 50 problem reports (QA Ars, DEFs, Ors, IRs) to determine if PSE&G had adequately implemented the prioritization schemes defined in the administrative procedure Based on the sample reviewed it appeared that PSE&G had adequately prioritized problem reports in accordance with the administrative guidance, with exceptions noted for several incident reports (IRs) discussed in Section 2.3.l of this repor A check of EMIS tags noted during plant tours against the (MMIS) tracking system found the selected items to be appropriately entered into the syste.2 Nuclear Safety Review Group The team discussed the role of the Nuclear Safety Review Group (NSR) composed of the Safety Review Group (SRG) and the Offsite Review Group (ORG) with PSE&G management and reviewed a sample of recent NSR reports (SRGC 95-014, SRGC 95-018, *sRGC 95-004, OSR 95-006, NSAG 94-014).
The reports were well documented and in most cases included both evaluations of previously implemented corrective actions and recommended additional long~term corrective actions to recognized problem.3 Incident Reports The team reviewed a sample of recent IRs and observed the daily 8:30AM engineering meeting in order to judge the effectiveness of the process used to determine IR significance levels. Significance levels are generated through a consensus of the engineering supervisors attending the daily meetin Each IR is briefly discussed and dispositioned by determining the line organization responsible for the review and assigning a significance level in accordance with the definitions provided in NAP-06, "Incident Report/Reportable Event Program," Revision 8, dated 4/7/95. Although the observed consensus approach appeared to be generally consistent, the apparent lack of more specific criteria for what is safety significant, potentially safety significant, and for minimal impact on safety significance can lead to questionable classification *
The team found a few examples of IRs (!Rs95-401, 95-551,95-507)
categorized at significance level 2 which appeared to be consistent with the significance level 1 definition of ~he current NAP-06 guidance (e.g., "safety significant or potentially safety significant or repetitive failures").
Additionally, the CAT team had identified a number of !Rs which based on their evaluation were reclassified from significance level 3 to level These examples were discussed with the CAT team and QA/NSR management who agreed that the descriptions in the current NAP-06 should be revised to be more explicit in terms of what is safety significant, potentially safety significant, and have minimal impact on safety significanc *
The team reviewed a sample of recently completed root cause analyses (IRs;95-262, 95-308,95-302, 95-294,95-297, 95-301,95-287, 95-203,94-384, 95-005,94-528, 95-012,95-015, 95-052, 95~062,95-074, 95-335).
The sample contained IRs which represented all three significance levels per NAP-0 The team agreed with the final root cause determinations for the IRs reviewe However, the team also noted a wide variation in the level of detail between the analyses and a lack of information to support responses on the root cause evaluation form These weaknesses could limit useful comparative analyses between incident reports and the ability to reconstruct an incident during subsequent evaluations. Further, the team noted (1) a lack of root cause training for all individuals given responsibility for performing IR analyses; and (2) a lack of explicit written expectations for these review efforts in NAP-0.4 Corrective Action Hit (CAT) Team PSE&G established a Corrective Action Hit (CAT) Team to provide support in three main areas: (1) immediate support to line organizations for the review of incoming IRs; (2) review and development of causal analyses for the backlog of problem reports (IRs, DEFs, DRs), an oversight function to independently evaluate the causal determinations and proposed corrective actions developed by the cognizant line organizations for specific IRs, and (3) to evaluate the effectiveness of those corrective action The team interviewed members of the CAT team and observed the groups daily activities to determine the effectiveness of the organization. The team determined that the CAT team is attempting to focus on each of the three main area However, due to the volume of requests to assist with the follow-up to current IRs and special ass.ignments (#12 charging p.ump, safety tagging issues, and the control rod assembly mishandling event), the CAT team has not been able to completely review the existing backlog of problem reports (e.g., DR, DEFs, ARs, etc.-.. ) and has not been able to dedicate resources to fully address their oversight functio *
For example, the revised NAP-06 guidance regarding the dispositioning of incident reports was effective on April 12, 199 The revision created a graded approach to root cause analysis of IRs based on a tri-level system of incident significance. The SIT gathered approximations from station personnel on how long the analysis was taking for each of significance levels and applied these estimates to the number of IRs categorized at each level over the past month to assess the workload generate From this data, for the period 4/12/95 to the 5/11/95, the corrective action program has generated approximately 1625 person-hours of effort to review IRs and determine causal factors for eac At present it appears that the workload associated with reviewing IRs is growin This is another factor affecting the ability of the CAT Team to fulfill its charte.5 Trend Analysis of Corrective Action Information The team reviewed PSE&G's processes used to track the effectiveness of corrective actions taken in response to problems identified through the various deficiency reporting system In general, the current systems provide
for tracking the number of currently open action requests, the line organization responsible for implementing the corrective actions, and the required implementation due date However, it does not appear that available information to trend/analyze outstanding action requests and the effectiveness of completed corrective actions associated with a plant system or components has been thoroughly use *
For example, the system engineering log books provide raw IR data and specific component performance trends for use in developing corrective maintenance plan The team discussed the methods that are used to evaluate system performance with a system manager and this manager mentioned that the current MMIS database is difficult to use for generating evaluation reports in that it does not currently provide a direct method for trending problem reports or corrective actions take As a result ad-hoc reports can be generated by downloading MMIS data into other applications but is limited to a few engineers with expertise in performing such downloads. Additionally the manager receives periodic reports from the Reliability Assurance Group if system components require repeat rework (e.g., 10-15 repeat work orders). The system managers also receive copies of !Rs related to their systems from their supervisor attending the 8:30 meetin Although there appears to be a significant amount of information available to analyze specific systems, the current process does not specify what analyses should be performed by the system manager Notwithstanding the cumbersomeness of using the current MMIS to generate trending information, it can be don However, the team found that the above system manager was unable to specify any analyses produced as a result of the information contained in the !Rs and log book for his system. Also, based on discussions with Technical Department management, such trending has not been widely performed by system manager.0 WORK CONTROL The team assessed the effectiveness of PSE&G's processes for prioritizing, planning, scheduling, and controlling work on plant equipmen The team observed daily activities associated with the work control process, including schedule review and status meetings and work control center (WCC) activitie The team also reviewed procedures and other written guidance, schedules, and work orders (WOs) and held discussions with personnel involved in all aspects of the work control proces Summary When deficient conditions were identified, the incoming action requests (ARs)
were reviewed promptly to determine the priority and appropriate action to address the condition. Operations defined the daily priorities for the station, but these priorities were not always communicated clearly and effectively throughout the Salem organization. The team noted weaknesses in the coordination of activities associated with priority issues. The team determined that the proceduralized process used to schedule non-outage work items was logical and well managed to minimize risk to plant operation The team noted that, in general, the station did a good job of adhering to the
- published schedule The team noted that repetitive tasks (surveillance tests and preventive maintenance activities) were not always scheduled to be performed before the due date for the task. Management recently implemented actions intended to reduce the amount of unplanned emergent work being performed, by defining controls on "Sponsored Work".
What work required sponsorship was not clearly understood throughout the organization, including the we PSE&G had recently established a safety tagging preparation office (STPO) to review work packages and prepare tagging requests in advance. This was a good initiative to relieve some of the wee workloa However, the effectiveness of the change was hindered because the changes to the work authorization process had not been clearly defined in writing and had not been clearly communicated to on-shift wess and first line maintenance supervisor PSE&G representatives indicated that it was a known problem that preventive maintenance (PM) work packages often did not meet quality expectations because, when the packages were originally prepared, standards for work planning were lowe Weaknesses in work packages and the use of work package feedback sheets were also note.1 Scheduling of Work The team determined that the proceduralized process used to schedule non-outage work items was logical and well managed to minimize risk to plant operation Work items were scheduled based on priority, risk assessment, and opportunity within the established system window and work week schedules.
Limitation of work within a week to one bus or channel minimized the potential of inter-system operational conflicts between scheduled work activitie Net safety gain (NSG) assessments were performed to assess the risk associated with performing preventive and corrective maintenance on all risk significant system NSGs were required for both planned and unplanned maintenance activities. The only exception to the requirement for a NSG was if a deficient condition requiring unplanned corrective maintenance rendered the affected equip~ent inoperable and a Technical Specification action statement had already been entere The team determined that non-outage work schedules received multiple multi-discipl ined reviews prior to finalization of the schedule two weeks in advance of the scheduled work wee Locking in the schedules two weeks in advance allowed time for thorough review of the planned work activities and facilitated preparation of tagging requests and work packages well in advance of work performanc During the inspection, PSE&G reduced power in unit 1 to perform repairs on the SGFP governor They performed a detailed NSG assessment, identified compensatory measures to reduce the risk due to unavailability of the SGFPs, and developed a detailed schedule for all the work activities including taking the pumps out of service and performing post maintenance testin Furthermore, PSE&G performed thorough work package preparations for the feed pump controller modifications that were planned for accomplishment over the weekend of May 6t *
However, the following scheduling weaknesses were noted:
Repetitive tasks (surveillance tests and preventive maintenance activities) were not always scheduled to be performed before the due date for the task. The team identified a large number of PMs and a few surveillance tests that were scheduled to be performed in the grace period between the task due date and the overdue dat *
The team also noted that PM deferral requests (PMDRs) were not always initiated in a timely manne PSE&G management's expectation was that PMDRs would be initiated as soon as it was identified that a PM could not be scheduled for completion prior to the overdue date so that the PMDR could be resolved prior to the overdue dat The team identified several PMs that were scheduled to be performed after the overdue date that did not have open PMDR.2 Schedule Adherence The team noted that, in general, the station did a good job of adhering to the published schedule Management recently implemented actions intended to reduce the amount of unnecessary unplanned work being performe By defining controls on what was referred to as "Sponsored Work" in a memorandum from the station General Manager, any work added during the actual work week had to be approved by station managemen The team noted the following weaknesses with implementation of this policy:
The policy was not clearly defined and was not effectively communicated to the working level of the organization (i.e., the on-shift WCSs and first line supervisors).
- The definition of what work required sponsorship was not clearly understood throughout the organization. Maintenance and operations personnel both indicated confusion on whether work items that were added to a previously scheduled job required sponsorshi *
All of the WCSs that were questioned by the team understood the expectation to strictly adhere to the work schedule, but none of them were familiar with the term "sponsored work."
- The team was concerned that, because sponsorship was not a formal process and that the policy was not well understood at the working level of the organization, the process could be unintentionally or inadvertently bypassed through the scheduling proces The schedules in the POD were discussed at the 6:30 a.m. supervisors meeting, but there was no line by line review of the schedule that would identify work items that had been added to the schedule without sponsorshi....
7 Work Authorization PSE&G had recently established a safety tagging preparation office (STPO) to review work packages and prepare tagging requests in advance. This was a good initiative that relieved some of the wee workloa However, the team noted that the work authorization process was still in a state of flux and identified weaknesses and inconsistencies in its implementatio For example:
The effectiveness of the change was hindered because the changes to the work authorization process had not been clearly defined in writing and had not been clearly communicated to on-shift wess and first line maintenance supervisor *
No specific training was provided to the on-shift WCSs on the function of the STPO and the changes in the division of responsibilities for work package review and work authorizatio *
In some cases the STPO pre-approved WOs for emergent work, and in other cases the packages were reviewed and the tagging requests were prepared in the we These differences often led to confusion on responsibilities and status of work package *
The team observed several instances in. which the combination of lack of clear understanding of responsibilities, inconsistent implementation of the process, and poor communications led to confusion in the wee. For example, on the morning of May 5, one of the wess spent a considerable amount of time assisting a maintenance supervisor with a WO for work on 138 CW pum The maintenance supervisor was confused about the status of the tagging request for the job. The work had been previously scheduled for the following week, but had been added to the schedule for May 5 and sponsored by the General Manage The STPO had pre~approved the WO and the required tags had already been added to the tagging request, but this was not communicated to the maintenance supervisor or the wcs in the we.4 Quality of Work Packages PSE&G used maintenance feedback sheets in order to improve the content of work packages and to identify and correct the reasons for lost time during maintenance work. The following weaknesses were noted in the use of work packages and feedback sheets:
Records of the feedback sheets and applicable corrective actions were not being consistently maintained by the maintenance organization to ensure that all feedback is acted upon and to enable valid trending of the informatio *
PSE&G representatives indicated that it was a known problem that PM work packages often did not meet quality expectations because, when the packages were originally prepared, standards for work planning were
.
'
- lowe At the time of the inspection, the planning department was relying on the feedback process to improve the work packages for repetitive tasks rather than correction during the planning proces The team observed that problems in work packages / planning were still found during performance of the job in spite of the improvements in the planning and review processe For example, The planning process did not identify that maintenance on the 18 diesel generator required running the other diesels. The planning process identified preventive and corrective maintenance work to be performed on the 18 emergency diesel generator (EDG), yet failed to plan for or schedule the Technical Specification required surveillance running of the other two Unit 1 EDG It was only when complications arose during the performance of the work that operators decided they needed to run the other Unit 1 EDG When the operators reached that conclusion, the other two EDGs were operated concurrently due to time constraints in the original Technical Specification allowed outage tim The noted that running the two EDGs concurrently was not consistent with the safety perspective of NRC Regulatory Guide 1.108, "Periodic Testing of Diesel Generator Units Used as Onsite Electric Power Systems at Nuclear Power Plants." The team concluded that the situation of running the two other EDGs during maintenance on the 18 EDG could have been avoided had the original corrective maintenance work package included the proper scheduling of the required EDG surveillance run One of the work package problems regularly identified *on maintenance feedback sheets was that bills of material (BOMs) are often out of dat Planning reviews and updates the BOMs for planned work, but when the job scope increases due to unanticipated problems, the broader problem of out of date BOMs has become a delay factor in returning equipment to servic At the time of the inspection, no program was in place to address the larger BOM problem; but, the team was told that the station plans to implement on Work Priorities and Status The team determined that the established general criteria for prioritization of work orders in the MMIS was adhered to consistentl When deficient conditions were identified, the incoming ARs were reviewed promptly to determine the priority and appropriate action to address the conditio *significant problems that impacted plant equipment were also communicated directly to the shift cre The senior schedulers and senior work control supervisors interfaced often throughout the da The criteria for higher (A and B) and lower (2, 3, and 4) priority work item categories were fairly well defined and discriminated based on safety significance and plant impac The middle prioritization category (priority 1) covered a broad range of items, but the process for evaluation and scheduling of work resulted in further prioritization of jobs based on factors including safety significance, plant impact, and age of the work ite Operations defined the daily priorities for the station, but these priorities were not always communicated clearly and effectively throughout the Salem organization. During the second week of the inspection, operations management attempted to provide some consistency in communication of priorities between the various daily meetings by defining critical evolutions and operations priorities for each unit and presenting these issues verbally at the 8:00 a.m, 8:30 a.m., and 1:00 p.m. meeting The effectiveness of this effort was limited by the verbal presentation, the inconsistencies in the written information provided at the meetings, and the timing of the presentatio In addition to the inconsistencies between the various meetings, communications related to priority issues were often hindered when personnel with key status information were not present at the status meetings or were not knowledgeable of the status of the issu In some cases, the individual running the meeting did not appear to have a clear understanding of the priority issues. The team also noted a number of instances in which personnel from different groups had a different understanding of the status of a priority issue. This confusion was rarely resolved at the meeting The team identified the following examples of poor communication on priority issu~s at the daily meetings:
At the 8:30 meeting on May 4, 1995, the radiation monitoring system (RMS) central processing unit (CPU) was identified as the number one operational concern, but representatives at the meeting did not know the status of the wor *
At the 8:30 a.m. meeting on May 8, 1995, the unit 2 control rods in manual was identified as a daily operations priority. This issue was not presented as an operations priority at the 8:00 a.m. managers'
meeting on the same da *
On May 9, 1995, the 11 reactor coolant pump seal leak-off trend was
. presented as a daily operations priority and was listed as an operational concern on the daily status shee However, there was confusion during the 8:30 a.m. meeting concerning whether information known by engineering, that the other three reactor coolant pump seal leak-offs were also trending up, had been communicated to operation *
On May 9, 1995, only one of the operational concerns listed on the unit 1 daily status sheet was listed as an operations priority on the PO None of the daily operations priorities provided verbally at the BAM meeting were listed on the operations priority list on the PO *
At the 8:30 meeting on May 4, 1995, there was confusion concerning the status of the operability determination for the unit 1 axial flux differential (AFD) monitor which was listed as an operational concern on the daily status sheet. The holdup on the issue was identified on the daily status sheet as the design change package (DCP), but some individuals at the meeting believed that the holdup was due to a parts proble The AFD monitor was added to the operations priority list in the POD on May 5, 1995, but was deleted from the list of operational
....
concerns on May 8 (over the weekend}.
At the 8:30 a.m. meeting on May 8, no one knew if the AFD monitor had been repaired or if work was planned for that da The team also noted weaknesses in the coordination of activities associated with priority issue Even when jobs were clearly identified as high priority, problems with communications and work quality often hindered accomplishment of the wor The following example of how work on the unit 2 28V DC battery chargers progressed illustrates the types of problems observed by the tea *
At 4:49 p.m. on May 1, 1995, a seven day Technical Specification action statement was entered for the 2Al 28V DC battery charger which was*
declared inoperable due to an installed ground detection system that was not *per system desig The battery charger issue was first identified as a priority issue at the daily meetings on May 2, 199 On May 3, operations was waiting for a work package to remove the ground detection system and the DCP needed to be writte *
On the morning of May 4, operations was not satisfied with the operability determination provided by engineering and the DCP was returned for revision. Later in the day, the General Manager identified that the post modification test was inadequate and the DCP was again returned to engineering for revision. The DCP was finally approved by the General Manager at 9:30 p.m. on May *
Maintenance received the work package at 2:00 a.m. on May 5, but did not commence work because the procedure required contacting installation and test personnel and none of the maintenance personnel knew how to contact the appropriate personnel. They tried to resolve the problem until 4:00 a.m. then decided to wait until day-shif *
On day-shift on May 5, the on-shift WCS in the wee was told that 28V DC battery charger work (on both units) was the highest priority work, but he was not provided with any details on the nature or plan for the wor Tagging requests had been prepared in advance in the STPO and delivered to the wee, but this was not known by anyone working in wee at the tim A plan for the work had been established, but was not communicated to the WCS in the night order book or through turnove *
The electrical maintenance supervisor brought two WOs to the WCC and didn't clearly communicate what he needed to the WC One of the WOs was for a battery charger that was in service (2A2) at the time which caused the WCS to question the wor The confusion caused by the maintenance supervisor's failure to communicate and fa i 1 ure to inform the WCS of the plan for the work resulted in delay in authorization of 28V DC battery charger work (on 281) and impacted other high priority jobs on both units including tagging of the 12 steam generator feed pump (SGFP) *
- 11 Corrective Maintenance Backlog The team noted that the trend of corrective maintenance backlog was not decreasing and that the backlog had increased over the last yea PSE&G had informal goals for timeliness of completion of work based on priority. These expectations were not defined in writing, but there was informal agreement between the operations and planning departments on the goals. There was also consistent understanding of the expectations throughout the planning and scheduling organization. There were no formal performance indicators to track achievement of these timeliness expectations, but the planning department did track the backlog of outstanding corrective maintenance work items based on ag *
As.of April 24, 1995, approximately 40% of the outstanding priority A, B, 1, and 2 non-outage corrective maintenance work items for the station were 4 months old or olde The goal for completion of priority 2 work items was 3 months with shorter time frames for completion of the higher priority items. This indicated that PSE&G did not meet their informal goals for completion of work for at least 40% of the identified non-outage wor.0 CONFIGURATION CONTROL / WORK-AROUNDS / OPERABILITY Configuration control and knowledge of plant construction has been a historical problem at Sale Partially in order to address the problem, Salem Operations created a new tagging office in early April 199 The concept of Salem operators implementing work-arounds to compensate for equipment or systems that were not performing as designed, is also an historical proble The problem was highlighted during the grass intrusion event that resulted in a reactor trip and multiple safety injections last Apri Due to long-standing problems in the performance of the systems, operators were operating the plant with the reactor control rods in manual control and, as a result of controller reset wind-up problems, manual action had to be taken for the proper operation of the steam generator atmospheric relief valves. Complications in the operation of those two systems directly contributed to the April 1994 even As a follow-up to that event, PSE&G consolidated a list of operator work-arounds in August 199 In September 1994, the list of work-arounds was frozen at about 80 problem statements, with problem initiators and owners identified for problem resolutio On September 28, 1994, an NRC special inspection team reported a lack of clear guidance at the station for making operability determinations (ODs).
Salem Operations implemented a process by which operating crew shift supervision would make ODs for degraded plant components or systems. This process was implemented via an entry in the operating crew Night Order Book and an accompanying flow chart to direct the operators while making these determinations.
- '
Summary The team found that performance in each of these three are.as continues to be poor and constitutes a burden on the Salem operators' effort to safely operate the Salem units. Specifically, configuration control problems involving safety tagging and inaccurate P&IDs continue to exist; the station operability work-around problem area has not been significantly improved and the team found that the station has been operated with degraded equipment and associated operability determinations that had deficient technical base.1 Configuration Control Due to the large number of configuration control discrepancies, the tagging office has required equipment operators to walk down work order packages in the field to verify the accuracy of P&IDs and work package boundaries for tagging orders. Other indicators of this problem are:
The team reviewed all of the incident reports written in the month preceding the inspection and identified over 50 examples that described deficiencies in the areas of P&ID accuracy, component labeling and Tagging Request Information System (TRIS) accurac *
The team reviewed an incident report in early April 1995 that documented a backlog of approximately 2000 items to be entered into TRI Further, these items had been thoroughly reviewed for safety significanc The team also observed that, while the new tagging office was making progress in the area of tagging releases and identifying configuration discrepancies, tagging errors continued to occur at the plan Four examples which occurred while the team was on site included:
a service water system sump pump breaker was found in the closed position, despite the presence of a shift supervisor administrative tag that required shift supervisor permission to close the breaker, in order to prevent an unmonitored chemical release to the river;
a work order for work on the No. 13 service water pump failed to include the tagout of the pump motor thermocouple, resulting in a potential equipment and personnel hazard;
an operator walk-down of a tagging request identified that the incorrect
"B" building air cooled condenser was listed on the tagging request, resulting in the determination that both "B" building condensers were identically labeled; and
the tagging request for the work on the No. 12 steam generator feed pump governor replacement contained an improper valve position on the TRIS lineup sheet.
I
-*
In summary, the team observed that configuration control and, to a lesse_r extent, safety tagging errors continue to occur at Sale The team determined that the breadth and scope of the problem constituted a burden on the Salem operators' effort to safely operate the Salem unit.2 Operator Work-Arounds During this inspection, the team reviewed the current operator work-around list and determined that about 70 items remained as work-arounds from th~ list that was frozen at about 80 in September 199 *
The team observed that additional work-arounds existed in the plant at the time of inspection, yet were not being tracked as such, but continue to cause operators to compensate for degraded equipment performanc Examples included: the operators dealing with numerous spurious self-test failures on the Unit 1 safeguards equipment cabinet; the Unit 2 rod control being maintained in manual due to rods inadvertently stepping as a result of a problem in the automatic control temperature comparato Neither of these examples were being tracked by PSE&G as operator work-around *
While the team was on site, PSE&G indicated that the number of open work-around items was reduced to 3 Upon further review, the team found that a number of the items from the original work-around list had been placed on a plant betterment issues list on the basis that some items may ~ave temporary fixes that keep the item from being a daily operator work-around, but could return as a work-around at any time until a plant equipment improvement is mad However, when station management was questioned about whether all items on the betterment list met this definition, station management responded that they did not kno This lack of clarity regarding the true status of work-arounds detracts from managements ability to assess the aggregate impact of work-around The team concluded that the work-around problem area has not been substantially improved since late last summe *
In an area related to work-arounds, the team reviewed the use of shift supervisor administrative safety tags. These tags are authorized by operating crew shift supervisors to control the use of certain equipment in the plant in that shift supervisor permission is required prior to operation of any component with one of these tags attache An example of this type of usage is cited in Section 4.2, above, where a shift supervisor administrative tag had been hung on the breaker for a service water sump pump to control operation of the sump pump to prevent inadvertent use and an uncontrolled environmental discharge from the service water sum In the above instance, the instructions of the tag had not been complied with, and the pump breaker was incorrectly left in the closed positio The team determined that approximately 150 tagging orders were~in effect for the use of this type of tag at the time of the inspectio The team concluded that this use of safety tags was not directly provided for in the controlling procedure for the use of safety tags, NC.NA-AP.ZZ-0015(Q},"Safety Tagging Program," and that the use of these tags constituted a form of work-around in that the tags were used
---
'
- '
to accomplish a purpose for which more appropriate means existed. The team further concluded that the use of shift supervisor administrative tags had the potential of desensitizing the plant staff to the operation of equipment with safety tags attached. Operations management informed the team that PSE&G had plans to reduce the use of these tags and implement other means such as procedures and operator aids in their plac The team determined that a large number of operator work-arounds still exist at Salem and place an additional burden on the safe operation of the plan PSE&G has developed or is developing additional ways of addressing the issue, yet by PSE&G's own performance indicators, the average age of work-arounds at Salem was 21.3 months as of April 27, 199 The team concluded that additional management attention is warranted in this area in order to substantially reduce the number of operator work-around.3 Operability Determinations The team determined through the review of operability determinations (ODs)
that PSE&G has been making some ODs without the proper regard for safety and regulatory requirement PSE&G has used equipment redundancy, the lack of Technical Specification or Updated Final Safety Analysis Report (UFSAR)
documentation, the lack of an effect on the reactor protection system, and fail-safe positioning to improperly justify operabilit Examples of these inappropriate bases for ODs prepared and accepted in the past two months included:
a 125vdc battery OD for cell seal degradation used a 10CFR50.59 review that cited for its basis the fact that the battery cell seal was not specifically mentioned in the Salem UFSAR, and therefore a safety evaluation was not required;
a No. 22 residual heat removal room cooler OD for the use of a non-qualified fan motor cited the fact that there was a redundant residual heat removal train available, therefore the use of the motor was acceptable;
an erratically-performing steam flow channel OD cited that maintenance had done all they could to repair the channel, therefore it was operable, but advised operators to be conservative and use an other channel;
a reactor coolant pump sampling valve OD for a faulty supply breaker cited the fact that the valves failed in the safe closed position, and concluded the breaker and valves were operable without assessing the ability of the components to perform their design function;
an RMS 23 and RMS 80 radiation monitor channels OD acknowledged that a design change package had inappropriately removed the annunciator reflash capability and that the Technical Specification intention of the channels was not being met, yet
concluded the channels should be considered operable until an alarm is received, only making the other channels inoperable at that time; a reactor coolant loop flow channel that had shown erratic behavior twice within a week in January 1995 did not have an OD performed until April 1995, when the channel was declared operable after a contractor was not able to repeat the behavior; an emergency diesel generator OD written to address fuses that were identified as being different than those specified justified operability via a problem report which discussed functionality without considering the basis for the apparent design change (i.e., with no lOCFRS0.59 review); and a feedwater flow control channel OD that was written after the channel exhibited improper transition from automatic to manual control justified operability based on the fact that feedwater flow was not part of the reactor protection system and that the reactor protection system was operating satisfactoril While the team was on site, operations management began to provide closer oversight of the OD proces As a result, operating crew shift supervisors improved their level of knowledge concerning operability and became more thorough in their reviews of OD The team observed, however, continuing cases of system engineering supplying justifications with improper technical base Examples of the improper OD bases proposed by system engineering that shift supervision rejected while the team was on site included:
after a discrepancy in the ground detection circuit of the 2Al 28vdc battery charger and 2A vital 28vdc bus was identified, system engineering supplied a memo to Operations attempting to justify system operability based on the availability of the redundant vital bus;
after the Unit 1 containment personnel access airlock failed a leakage test for the second time within a week, system engineering supplied a memo to Operations which concluded the cause of the failures was dirt on the airlock seal and the airlock was not degrade When Operations rejected the memo and its assumption of root cause, further investigation revealed that the airlock had, in fact, a mechanically degraded seal; and
after a discrepancy in the configuration of an air supply valve to the auxiliary feedwater system was identified in an incident report, system engineering supplied a memo to Operations justifying system operability based on the availability of a redundant air suppl Subsequent to the departure of the SIT from the site, PSE&G committed to completing a review of all active ODs by May 19, 199 During the week of May 22nd, NRC Region I obtained the results of this PSE&G review and undertook an
evaluation of this informatio Pending completion of this evaluation, the adequacy of the bases for equipment ODs is an UNRESOLVED ITEM. (UNR 95-80-1) MANAGEMENT OVERSIGHT During the period April 26 through May 11, 1995, the team assessed the effectiveness of management oversight and the performance indicators used by management to measure current Salem performance. This assessment was based upon observations of meetings, of work activities, discussions and interviews with management and other PSE&G personne Summary The team observed an NRB meeting that critically reviewed the status of station activities and plans to improv The team identified several weaknesses in the effectiveness of management oversight and the performance indicators used by management to measure current Salem performanc Management demonstrated a weak safety perspective at the station manager's meeting on a number of occasions. Also, at the meeting, the managers were not always knowledgeable of the specifics for the ongoing priority issues, both safety and non-safety related. Although the SORC identified and properly dispositioned several issues at the routine SORC meeting attended by the team, the identification of these weaknesses came solely from the Operations Manager, as opposed to the SORC as a whol The other SORC members contributed minimally to the critical aspects of the conversation. Overall, the MOG effort was a good initiative. However, the MOG failed to identify the significant team finding of inadequate OD base The performance indicators being used by PSE&G management to measure Salem performance were weak in that some indicators either were not predictive and/or had too high a threshold for inclusion of dat In many cases, PSE&G management had available to them the data that could have been used to provide more meaningful measures of Salem performance; however, they were not routinely using this data to identify and evaluate trends, and measure performanc And for those instances when a trend was identified, PSE&G did not take substantive action to reverse the trend in a timely manne The team concluded that between the oversight of planned station work, daily emergent issues, efforts to improve on a daily basis and longer term improvement plans. the station management team's attention is stretche.1 Nuclear Review Board The team attended a Nuclear Review Board (NRB) meeting on April 27th that was-devoted to briefings by the Salem management team, including support services, on current station activities and improvement plans. The NRB appeared to find that improvement plans were getting better in content, but needed further wor The team considered the NRB members were probing in their questioning of the manager In particular, in reviewing the improvement plans the NRB probed each of the managers on their basis for completeness of the identified areas for improvement and how they intended to measure their performance under
. *'
their plans. Questioning by the NRB appeared to leave many of the managers with additional work in order to fully address the questions raised by the NR. Station Operations Review Conunittee Performance On May 3, 1995 the team attended a Station Operations Review Committee (SORC)
business meeting at which the General Manager and SORC members discussed the SORC's role and expectations. For the two technical documents presented at the meeting, the SORC noted the following:
Question 4.6 of the IOCFR50.59 Safety Evaluation for DCP IEC-3387, "EOG IA Air Receiver Low Pressure Alarm," was answered inappropriatel Specifically, for the question "Does the proposal reduce the margin of safety as defined in the basis for any Tech Spec?" the engineer stated in part that the margin of safety is not reduced because the diesel generator air start system and its components are not the subject of any Technical Specification or Technical Specification basis. The SORC pointed out that this was an inappropriate answer, because the diesel generators are discussed in the Technical Specification and without the air start system the diesel generators would be considered inoperabl *
For Temporary Modification 95-028, "Boiler Feed Iron Sampling Modification," documentation did not appear to address a personnel safety concern related to removing a drain line cap at power and a nuclear safety concern regarding the potential of spraying sensitive instruments if tubing faile Although the SORC did identify these issues and disposition them properly, the team noted that the identification of these weaknesses came solely from the Operations Manager, as opposed to the SORC as a whol Follow up on the EOG Air Receiver DCP issue with one of the SORC members showed that the individual did not understand why saying that the margin of safety is not reduced because the diesel generator air start system and its components are not the subject of any Technical Specification or Technical Specification basis, was an inappropriate answe.3 Management Oversight Group Following the April 7, 1994 event, PSE&G established a Management Oversight Group (MOG) to provide additional, independent management oversight of plant operation Overall the team found that the MOG effort was a good initiative. They identified some good findings and provided real time observations of shift activities, helping to make improvements in such areas as control room command-and-control and communication However, they failed to identify the significant team finding of inadequate OD base *
lB Management Safety Perspective During the inspection, the team noted that management demonstrated a good safety perspective on a number of occasion For example, at the management team meeting on April 27, 1995, a tagging problem, from the previous day involving a safety tag on a service water sump pump breaker that had been violated was responded to quickly by the General Manager and set in motion a series of steps to sensitize station personnel to the importance of complying with the safety tagging system. Also, at the managers' meeting on May 11, 1995, several examples of good safety perspective were note For example, regarding debris found in the #2 station air compressor cooler, the operations manager cautioned the other managers to not prematurely focus on the debris as being the root cause of the elevated temperatures seen in the air compressor, without considering all other possibilitie However, notwithstanding the noted improvement seen particularly at the end of the inspection; overall, PSE&G management did not exhibit a strong safety perspective on daily emergent issues. The team observed several instances in which management did not demonstrate a strong safety perspectiv For example:
On April 27, 1995, in response to discussions at the BAM managers meeting on problems involving recent corrective maintenance on the service water system such as inoperable traveling screens, incorrect bolting, water in lube oil and inoperable strainers, the Station Manager requested that the system engineer/manager brief the management team at the following day's BAM meetin On April 2B, the system engineer told the managers that he did not believe that the condition of the service water system had degrade He stated that the increase in degraded conditions was the result of increased station emphasis on documenting, investigating and correcting degraded conditions. The management team did not question the engineer's conclusion, or ask any questions about what the history of corrective maintenance was for the system or whether the evaluation of this history supported his conclusion that the system was not degradin Follow-up questioning of the engineer by the NRC resident staff revealed that the engineer had not made comparisons of equipment history records to support the conclusion that the system had not degrade *
On May 2, 1995, at the BAM managers meeting, the Operations Manager mentioned that a maintenance worker had begun to do work to alter the configuration of a 2B volt battery charger using a work order, without a design change package (DCP).
The unacceptability of performing a design change on safety related equipment without the required reviews and documentation of a DCP was not raised as an issue at the meeting and no one was designated to provide follow-up to the managers at the next day's meetin *
On May 8, 1995, at the BAM managers meeting, the Operations Manager's representative discussed an incident report regarding the containment airlock door leakage test failure. It was the second time in a week that the test had failed. It was mentioned that dirt had been found on
- the seal both times. A manager questioned the operability of the containment air lock door and the root cause for the door failing its leakage test twice within one wee The manager was demonstrating a good questioning attitude. However, he was quickly stifled by other managers, and he backed away from his concern The other managers lacked a questioning attitude and accepted the easy answer of dirt as the caus Later in the meeting, operability of the door and apparent lack of an identified root cause was questioned, and one manager inferred that the identification of dirt on the airlock door seal constituted an adequate root cause analysi After further evaluation and testing, PSE&G determined that the airlock door had failed the leakage test because the seal was degrade *
In connection with the operability determination problem discussed in Section 4.3, the team also identified a management safety perspective weaknes While the team was still on site, PSE&G indicated their intent to follow-up with a review of active ODs in the control roo During the week following the team's departure from the site, the team made an inquiry of Salem station management as to the operability status of those systems and components identified by the team as having inadequate ODs as well as other active ODs not specifically reviewed by the tea PSE&G had not yet established valid ODs for those in question and had no timetable in place for the completion of the reviews of all active OD Station management committed to get back with the NRC the following da The next day, PSE&G committed to completing such a-review by May 19, 199 PSE&G's slow response to this issue once initially informed of the OD bases problem, is an example of weak management safety perspectiv.5 Management Oversight of Emergent Issues At the BAM managers' meetings, the team noted that the managers were not always knowledgeable of the specifics for the ongoing emergent issues, both safety and non-safety related. The managers were unable to answer questions from the General Manager involving schedule issues, when degraded safety equipment had entered its technical specification action statement, and the status of parts needed to return degraded equipment to service. The managers also were not knowledgeable in some cases of work delays and the reasons for those delay For example,
On May 5, the managers did not know the 181 battery charger LCO start and stop times, and therefore that they were in day 3 of a 7-day LCO;
On May 10, the managers did not know the status of parts for the 21 MSR drain tank level indication which had been a priority issue for over 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and was the reason for a 10% power reductio *
On May 5, the managers did not understand the reasons for delays in the 28 volt battery charger work, which had been a priority job for at least three day *
- On May 5, the General Manager described activities associated with the 28 volt battery chargers which showed that he had followed aspects of the job and had visited the work sit Upon questioning by him, the managers were unable to provide status information or reasons for delay In addition, prior to approving the battery charger DCP on May 4, the General Manager identified weaknesses in the DCP post modification testing of the battery charger *
On May 10, the General Manager pointed out and took issue with what appeared to be communication problems related to the previous days priority job.6 Performance Indicators PSE&G had identified seven performance indicators in their response to the Notice of Violation and Civil Penalty associated with the April 7, 1994 even These were Event Free Operations, Unplanned Automatic Scrams, Personnel Performance Incident Reports, Repeat Equipment Problems, Corrective Maintenance Backlog, Preventive Maintenance Overdue, and Repeat Cause Incident Report PSE&G stated that these performance indicators would be used to determine the effectiveness of the actions they had identified for improving performanc PSE&G had also identified about 18 other performance indicators used to monitor performance. These included Safety System Unavailability, Control Room Indicators Out of Service, LERs, and NRC Violations. All of these indicators are published quarterly and distributed to the department managers and abov During recent weeks prior to the NRC inspection, Salem took action to improve upon several weaknesses with the performance indicators being used by PSE&G management to measure Salem performanc In some cases the indicators either had not been predictive and/or had too high a threshold for inclusion of dat The Safety Review Group completed a review to analyze and validate several of the performance indicators in early April 1995 and provided the results to the General Manage The indicators for April 1995, to be published in May, were changed to reflect some of the findings of the SRG revie For example,
The Repeat Equipment Problem indicator was changed to recurring corrective maintenance work orders to better reflect the condition of equipment at Salem that could adversely affect the operation of the stations. For Preventive Maintenance Overdue, the indicator was changed to include preventive maintenance tasks not completed by their overdue date and deferred by the system enginee In the past, these preventive -
maintenance tasks were not included in the indicator which provided a more positive indication of the health of the preventive maintenance progra *
For the indicators designed to track human performance, Personnel Performance Incident Reports and Repeat Cause Incident Reports, the indicators. were not completely based on root cause investigations but instead on preliminary dat As a result, the actual number of personnel performance or repeat events could be substantially different
from those in the performance indicato (A Safety Review Group review of 29 incident reports with completed root cause investigations, found that 37% of those attributed initially to equipment problems were actually human performance or management problems.} Also, for these indicators, PSE&G only included incident reports that reached the threshold for reporting per 10 CFR 50.72 and 50.73, or the threshold for being designated significant by the NRG, noteworthy or significant by the industry, or the threshold for resulting in an internal Significant Event Response Tea The causal factors for incident reports that do not meet the threshold criteria are not included in this performance indicato Per the SRG review, industry experience has shown that the causal factors are the same for both inconsequential and consequential incident As a result, these indicators have not been used in a way that identified trends that in-turn may lead to more significant problems, if not correcte For Repeat Equipment Problems, the original list of monitored equipment was limited to the Nuclear Plant Reliability Data System (NPRDS} listing which did not cover some of the specific equipment that has adversely affected the operation of the stations, such as the service water strainer In approximately the last two years the valves had been out of service eight times for corrective maintenance, but they were not included in this indicato In addition, problems that resulted in degraded equipment but did not make the equipment inoperable were not reported to NPRDS, such as the problems associated with feedwater regulating valve 11BF19. This valve was worked on five times in the last 18 months for erratic operation or loose parts, but no NPRDS reports were made because the valve was not considered inoperabl The team found the following weaknesses in its review of Pis, including instances where management should have demonstrated a stronger safety perspective by acting on other data available to them:
The indicator for Preventive Maintenance Overdue included the number of preventive maintenance tasks that were open and past their overdue dat A predictive indicator would include preventive maintenance tasks past their due date instead of overdu *
In the area of configuration control, as discussed above, an extensive evaluation of the P&ID deficiency trend was conducted in 1994 and the trend continue However, PSE&G has not taken substantive action to reverse the tren NED has had a program for addressing individual problems related to configuration control and P&ID accuracy, as identified by the Salem incident report program, yet has not taken the initiative based on the IR trending information to make an in-plant engineering assessment of the scope of the P&ID problem through selected system(s) walk-downs dedicated to this purpos *
The team found that although the IR trend on procedure adherence problems had been showing an increasing trend of occurrences for the past several months, management had not taken action to date to assess the significance of this tren.:,
During the period April and May 1995, Operations management developed new indicators for mispositioning events, procedure events, tagging events, and unplanned technical specification action statement entrie The first three of these indicators showed an increasing (negative}
trend since January 199 But at the end of this inspection, PSE&G had not yet evaluated these trends to determine what actions to tak.0 EXIT MEETING On May 11, 1995, the SIT met with station management to present the preliminary findings of this inspectio On May 26th, the SIT Team Manager and Team Leader met with PSE&G management to discuss the findings from this inspection. At the May 11th and May 26th meetings, PSE&G questions were resolved and the licensee took no issue with the findings of the SI Further, the Chief Nuclear Officer stated that Salem Unit 1, shut down on May 16th for switchgear ventilation operability concerns, would remain shut down until PSE&G had completed an operations readiness revie On May 30th, the Salem General Manager called the SIT Manager and Leader to clarify PSE&G's bases for establishing the plant betterment subset of operator work-arounds and agreed with the SIT conclusion that the operator work-around problem at Salem* station had not substantially improved since the Fall of 199 Based on Region I review of this report, and discussions with PSE&G, it was determined that this report does not contain safeguards or proprietary information. A list of the May 26th exit meeting attendees is enclose *
Attachment 1 NRC SIT 50-272/95-80 and 50-311/95-80 ATTACHMENT 1 NRC SPECIAL INSPECTION TEAM REPORT NOS. 50-272/95-80 AND 50-311/95-80 April 19, 1995 MEMORANDUM TO:
FROM:
SUBJECT:
Richard W. Cooper,II, Director Division of Reactor Projects Charles W. Hehl, Director, Division of Radiation Safety & Safeguards Thomas T. Martin ORIGINAL SIGNED BY:
Regional Administrator Region I SPECIAL INSPECTION TEAM CHARTER FOR REVIEW OF CURRENT PROBLEM IDENTIFICATION AND WORK CONTROL ACTIVITIES AT SALEM The most recent NRC Systematic Assessment of Licensee Performance for the Salem Generating Station, Units 1 and 2, was presented to the licensee at a management meeting on January 12, 1995. This NRC assessment highlighted, in particular, problems that the licensee has experienced in the areas of problem identification and prioritizing and conducting work on plant equipmen In connection with these problems, the NRC noted problems with accommodating equipment problems that resulted in unnecessary challenges to operators and safety systems; a general lack of a questioning attitude by operators; weaknesses in operability decision makin I have determined that a Special Inspection Team (SIT) inspection should be conducted to assess how effectively the licensee is currently performing from a safety perspective in the areas of problem identification; prioritizing and conducting work on plant equipment; and management oversight of plant performanc The focus of this inspection will be on how the licensee is conducting day-to-day activities related to safety performance, in lieu of their plans to improv The Division of Reactor Projects (DRP) is assigned responsibility for the overall conduct of this special inspectio Peter Eselgroth, DRP, is appointed as the SIT Leader, reporting to Susan Shankman, SIT Manage DRP is assigned the responsibility for resident and clerical support, as necessary; and the coordination with other NRC offices, as appropriat Further, DRP is responsible for the timely issuance of the inspection report and the identification and processing of any potentially generic issues as a result of the team's revie Attachment 1 represents the charter for the SIT and defines the scope of the inspectio The complete list of SIT members.is identified in Attachment 2.
Docket Nos. 50-272; 50-311 Attachments:
As Stated
- Attachment 1 NRC SIT 50-272/95-80 and 50-311/95-80 cc w/atts:
J. Taylor, EDO J. Milhoan, DEDR L. Olshan, NRR J. Stolz, PDl-2, NRR W. Russell, NRR R. Zimmerman, NRR W. Kane, RI S. Shankman, RI J. Wiggins, RI R. Blough, RI W. Lanning, RI J. White, RI C. Marschall, SRI, RI P. Eselgroth, RI
Attachment 1 NRC SIT 50-272/95-80 and 50-311/95-80 ATTACHMENT 1 Salem Generating Station Review of Current Safety Performance in Selected Areas of Concern The general objective of this SIT is to:
assess the licensee's effectivene~s at problem identification, prioritizing and conducting work on plant equipment
assess the licensee's management oversight programs and performance indicators used by management to assess current performance for the purpose of improving the reliability of equipment and the material condition of the plant so as to decrease challenges to safe plant operatio The ~pecific objectives of this SIT are to assess the following activities from a safety perspective:
The current effectiveness of licensee programs for problem identification and root cause determinations. This will include plant tours, reviews of equipment history rec.ords and interviews with working level personne The licensee's day-to-day programs in the areas of work prioritization, planning, scheduling and controls. This will cover the day-to-day mechanism for determining what plant deficiencies from various reporting processes are to be addressed and how work is integrated into plant operations from a safety perspectiv The adequacy of operability determinations, including an assessment of performing such determinations when needed and the extent of engineering involvemen Steps being taken to reduce operational work-around The effectiveness of preventive and corrective maintenanc The effectiveness of management oversight programs and performance indicators used by management to assess current performanc This inspection will be conducted over approximately a four week period and the number of team members involved in inspection activities at any one time will typically be three or less, so as to minimize the impact on the licensee of this inspectio.J) "'l..
~,_,
'-.I Attachment 1
.
NRC SIT 50-272/95-80 and 50-311/95-80 ATTACHMENT 2 SALEM SIT MEMBERSHIP Susan Shankman, SIT Manager, Division of Radiation Safety & Safeguards, Region I(RI)
Peter Eselgroth, SIT Leader, Division of Reactor Projects (DRP), RI Michele Evans, Senior Resident Inspector, TMI, DRP, RI Tracy Walker, Senior Operations Engineer, Division of Reactor Safety (DRS), RI Greg Galletti, Human Factors Engineer, Nuclear Reactor Regulation Stephen Barr, Operations Engineer, DRS, RI
NAME Leon R. Eliason Jeff Benjamin Joseph J. Hagan Stan LaBruna John C. Summers Mark Reddemann Frank Thomson Bruce Preston Michael Morroni Gil Madsen Peter A. Mozllea Ernest J. Hackness Greg Suey W. F. Metcalf, S B. B. Burricelli Victor W. Lowenstein, S ATTACHMENT 2 NRC SPECIAL INSPECTION TEAM EXIT MEETING MAY 26, 1995 TITLE President & CNO NEU Director QA&NSR Vice President - Nuclear Operations Vice President Engr & Support General Manager - Salem Operations General Manager - Hope Creek Operations Mgr - Licensing & Receiving Mgr - Salem Engr Mgr - Maintenance Controls Asst. to Mgr - Engineering Support Licensing Re Station Planning Chemistry Manager Maint. - Mgr. Mechanical Director, External Affairs Senior Operations Supervisor Nuclear Regulatory Commission (NRCl Susan F. Shankman NRC Special Inspection Team - Manager NRC Special Inspection Team - Leader Peter W. Eselgroth Other Phil Duca Robbie Kankus Joe Janoch Mike Sesok Delmarva Site Rep PECO Joint Owners Affairs Atlantic Electric *
Atlantic Electric Site Representative