PRA importance measures for maintenance prioritization applications

Reliability Engineering and System Safety 43 (1994) 307-318

~ . 471. ?:

PRA importance measures for maintenance prioritization applications W. E. Vesely, M. Belhadj & J. T. Rezos SAIC, Inc., 655 Metro Place South, Suite 745, Dublin, Ohio 43017, USA (Received 2 February 1993; accepted 21 July 1993) Various importance measures are standardly calculated in a standard probabilistic risk assessment (PRA). Approaches are developed in this paper for using two of these measures---the minimal cutset contribution and the risk reduction importance---for prioritizing the risk importances of maintenances. One approach which is developed prioritizes maintenances based on the risk importance of the associated contributor. The second approach prioritizes maintenances based on the risk impact if the maintenance is ineffective. The core damage frequency is used as the risk measure for prioritization. The demonstration studies, which are carried out using the Surry Plant PRA, indicate that the two approaches give similar results if appropriate cutoff criteria are used. As an additional evaluation, risk unimportant maintenances are identified using the risk increase importance, or risk achievement worth, calculated in the PRA.

INTRODUCTION

The importance measure which determines those maintenances that are risk unimportant is based on the risk impact which occurs if the c o m p o n e n t fails. If the risk impact is negligible even when the c o m p o n e n t fails, then maintenance on the c o m p o n e n t is not important from a risk standpoint since the risk is not sensitive to the proper functioning of the component. It should be noted that cases can arise where maintenances on a c o m p o n e n t can be classified as being both important and unimportant depending on the effects of the maintenance. These cases need special consideration as discussed. The importance measures that are identified are specific cases of the standard importance measures that are usually calculated in a P R A , which allows for straightforward implementation. The standard importance measures are simply normalized in an appropriate manner for the maintenance applications. What is new in this work is the bases and rationale developed for use of the specific importance measures for maintenance applications.

Importance measures are standardly c o m p u t e d in a probabilistic risk assessment ( P R A ) to identify the important risk contributors and the important risk sensitivities. Various types of importances are calculated, including risk contribution importances, risk reduction importances and risk increase importances. These importances also go by various names such as the Birnbaum importance, the risk achievement worth, and the Fussell-Vesely importance. 1'2 (See specifically chapter 10 of Ref. 2.) This paper focuses on P R A importance measures which can be useful for maintenance prioritization applications. Specific importance measures are identified which can be used to identify risk important maintenances as well as risk unimportant maintenances. Two different measures are identified that can be used to determine risk important maintenances. One importance measure determines the risk importance of the maintenance based on the risk importance of the equipment being maintained. The other importance measure determines the importance of the maintenance based on the risk impact that would occur if the maintenance were not carried out effectively. In the demonstrations, both measures gave similar results.

DETERMINING THE RISK CONTRIBUTOR IMPORTANCE OF A MAINTENANCE One way to identify the risk importance of a maintenance is to identify the risk importance of the equipment being maintained. Risk important maintenances can be defined to be those maintenances which

Reliability Engineering and System Safety 0951-8320/94/$07.00 (~ 1994 Elsevier Science Limited, England. 307

308

W. E. Vesely, M. Belhadj, J. T. Rezos

are performed on risk important equipment. The measure of the risk importance of the equipment will thus provide the risk importance of the maintenance in terms of the importance of the risk contributions affected by the maintenance. Now, the risk contributions in a P R A are standardly prioritized by prioritizing the minimal cutset contributions. A minimal cutset is a smallest combination of basic events which if they all occur will result in core damage (or other undesired event analyzed by the PRA). For a core damage, which will be our focus for the specific applications, the combination of basic events consists of the initiating event, such as a pipe break, and a combination of component failures or other basic events which result in the loss of necessary safety functions to prevent a core damage. The core damage frequency (CDF) contribution from a given minimal cutset is quantified by multiplying the frequency of the initiating event in the minimal cutset times the unavailabilities of the basic events in the minimal cutset. The P R A identifies the minimal cutsets for core damage and ranks them in order of their C D F contribution from the largest to smallest, down to some cutoff point. To first order, as generally assumed by the P R A , the sum of the core damage minimal cutset contributions equals the CDF. The minimal cutset contributions provide a straightforward way of identifying the risk important maintenances. The risk important minimal cutsets are identified and the set of maintenances associated with the basic events in the risk important minimal cutsets can then be defined as the risk important maintenances. We will call this approach for determining the risk important basic events and associated maintenances, the minimal cutset prioritization approach. Note, that the minimal cutset approach is applicable not only for maintenance prioritization but also for all other activities and procedures associated with the basic events in the risk important minimal cutsets. A criterion is needed to define the risk important minimal cutsets, i.e. to cut off the risk important

Table 1. The minimal cutset maintenance prioritization approach 1. Rank the minimal cutsets in terms of their contribution to the CDF 2. Divide each minimal cutset contribution by the total CDF to give the relative minimal cutset contribution 3. Prepare a running sum of the ranked, relative minimal cutset contributions and cutoff at some significant percentage, such as 90%, to identify the risk important minimal cutsets 4. Identify the maintainable components and equipment associated with the basic events in the risk important minimal cutsets. The associated maintenances are then the risk important maintenances

minimal cutsets from the marginal and unimportant ones. In terms of relative contributions, the risk important minimal cutsets can be defined to be the collection of top minimal cutsets which contribute a significant percentage to the CDF, such as contributing 90%. The precise cutoff criterion will often not be critical as long as a significant percentage of the total contribution is obtained. Table 1 summarizes the minimal cutset maintenance prioritization approach. This approach is also basically the approach described in N U R E G / C R - 5 6 9 5 , 3 where it is termed the risk-focused maintenance approach.

D E T E R M I N I N G THE RISK I M P A C T I M P O R T A N C E OF A M A I N T E N A N C E Another approach for determining the risk importance of a maintenance is to determine the risk impact if the maintenance is assumed not to be carried out effectively. Risk important maintenances are then defined to be those maintenances which, if not carried out effectively, will have significant risk impacts. We need to model the impact of an ineffective maintenance on a component in order to determine the associated risk impact. We will assume as a general model that ineffective maintenance will cause the component unavailability to increase by some factor f. From NUREG/CR-55104 and as shown in Appendix 1, when the component unavailabilities increase by a general factor f the C D F increase AC is given by

AC = ~ rif + E ruf e + " ' " + i= 1

i>j

~'.

ri,...i,f*

(1)

il>"->i k

where ri is the risk reduction importance of component i; ru is the joint risk reduction importance of components i and j; r~,...i, is the joint risk reduction importance of components i ~ . - . i,; and k is the largest size of the minimal cutsets. Equation (1) is an exact equation for all values of f. Truncation of the terms on the right hand side will result in various approximations. The components in eqn (1) are those associated with the basic events defined in the P R A and in the minimal cutsets. If ineffective maintenances on different components result in different factor increases in the unavailabilities, then f can be taken as the maximum factor in which case AC is the maximum C D F increase. The individual risk reduction importances r~ are standardly tabulated in PRAs and the joint risk reductions ru. . . . . r~,...i, are extensions to multiple components. From their definitions, the individual and joint risk reductions can be straightforwardly

Importance measures for maintenance prioritization obtained from the minimal cutset contributions: r~

= t h e sum of the contributions from the minimal cutsets, each containing component i = the sum of the contributions from minimal cutsets, each containing both components i and ]

rij

~/I" "'/k

= the sum of the contributions from minimal cutsets, each containing components i,, i2 . . . . and ik.

The risk reduction terms are so named because they indicate the risk reduction, i.e. the reduction in CDF, if the associated component unavailabilities are reduced to zero. If we want to identify the significant contributions to AC then we need to determine the significant contributions to each of the terms on the right hand side of eqn (1). We focus on the first term, using the other terms as checks. The first term on the right hand side of eqn (1) represents the contribution from individual component degradation impacts:

~

r~f = the contribution to AC from

i=1

individual component degradation impacts due to ineffective maintenances. To obtain the significant portion of the contribution given by this for any impact f, we can obtain the significant portion of ET=, rg, which is the sum of the individual risk reductions. This can be done by first ranking the risk reductions r~ from largest to smallest, which most PRAs already do. If we normalize each individual risk reduction by the sum of the individual risk reductions then we can accumulate the top normalized risk reductions in a running sum until the percentage reaches 90% or 95%, or some other high inclusion percentage. The value for the cutoff will again not generally be critical if it is high enough to include the significant portion of the total term. The associated components in the included set are then the risk important components from an impact standpoint. The maintenances on these components can then be defined to be the risk important maintenances which can significantly impact the C D F if not carried out effectively. The significant contributions from the second and higher order terms in eqn (1) can be evaluated in a similar manner to see if any new components and new maintenances enter. These higher order terms represent additional impacts from simultaneous degradations of multiple components due to ineffective maintenances simultaneously being performed on the multiple components. For example, the term rijf 2 represents the C D F impact if ineffective

309

Table2. The risk reduction maintenance prioritization approach 1. Determine the individual risk reductions of the basic events in the PRA 2. Rank the individual risk reductions in order of decreasing size 3. Normalize the individual risk reductions by dividing by the sum of the risk reductions 4. Prepare a running sum of the ranked, relative risk reductions and cutoff at some significant percentage to identify the risk important basic events and associated maintenances 5. The joint risk reductions can be checked to determine if any additional maintenances can be important because of interactions maintenances are performed on both component i and component j resulting in a factor f increase in each component unavailability. These higher order terms thus represent interactions among the maintenance impacts. The joint risk reductions for a given order (e.g. rij for the second order term) can be ranked and can be normalized by the respective sum again. A significant percentage (e.g. 95%) of the total sum can be taken and any new components and maintenances identified and added to the list. This can be repeated for each term. If enough significant contributors are included in the first term using the individual risk reductions ri then there will be a high likelihood that no significant, additional maintenances will be identified from these higher order terms. To help assure this condition, a high inclusion percentage can be used, such as 99% or 99-9%. As checks on any specific maintenances which are not included, the associated higher order contributions can be calculated using various values for the degradation factors [ to determine the sensitivity to these excluded maintenances. Table 2 summarizes the risk reduction approach for identifying the risk important maintenances by their risk impacts.

DETERMINING THE RISK UNIMPORTANT MAINTENANCES Maintenances can be classified into those which are risk important, those which are marginal, and those which are risk unimportant. In addition to those which are risk important, it is useful to identify those maintenances which are risk unimportant. One of the simplest ways to identify risk unimportant maintenances is to identify those components which, even if they fail, will have little risk impact. If a component fails and has little risk impact then maintenance is not important in maintaining the performance of this component.

310

W. E. Vesely, M. Belhadj, J. T. Rezos

The standard measure of the risk impact of a component when it fails is the risk increase importance of the component, which is also called the risk achievement worth and the Birnbaum importance. Using the CDF as the risk measure, the risk increase importance is standardly determined in the PRA by failing the component and determining the increase in the CDF when the component is failed. The risk increase importance is also equal to the sum of the contributions of the minimal cutsets containing the component with the component unavailability set to one, provided sufficient minimal cutsets are obtained by the PRA to contain the component. The risk increase importances can be determined for the basic events and can be ranked from largest to smallest, as is often done as standard in a PRA. Those basic events which have insignificantly small risk increase importances, such as those which are less than 1% of the CDF, can be identified as being risk unimportant. The associated maintenances done on these components can then be identified as being risk unimportant. In interpreting risk unimportant results, three issues need to be considered. The first issue involves a component which can be identified as being both risk important and risk unimportant. If a component has a very high unavailability, e.g. 0.1 or higher, then the risk increase from bringing the component down will be relatively small because of the small change in going from a very high unavailability to an unavailability of 1. However, because the component has a high unavailability it can also be identified as being a risk important component because of its high unavailability. For these components of high unavailability which can be both important and unimportant, maintenance can also be both important and unimportant. If the high unavailability can be reduced by maintenance, for example by overhaul or by other corrective actions, then the maintenance is important. However, a maintenance, such as preventative maintenance, which maintains performance and prevents deterioration from the present state, is unimportant. This shows the importance of interpreting the PRA results in terms of their specific maintenance implications. The second issue involves the total CDF increase from multiple components being simultaneously down at the same time. The total CDF increase can be significantly larger than the sum of the individual CDF increases if two or more components are in the same minimal cutset. One must thus assure that the total increase is also small, for example by limiting the components to be in different minimal cutsets or by evaluating the CDF increase by simultaneously failing all the prospective components which are deemed to be unimportant. The third issue that needs to be considered is

Table 3. The risk increase maintenance unimportance identification approach

1. Determine the individual risk increases of basic events in the PRA 2. Rank the risk increases in order of decreasing size and divide each one by the CDF 3. Identify those components and associated maintenances as being risk unimportant if their risk increase is less than a small per cent of the CDF 4. For implementations, check whether maintenances can be both unimportant and important. Also check cumulative effects and configuration effects to assure the CDF increases remain small related to the second issue and involves plant configurations which can cause unimportant components to become important. The CDF increase importance as standardly calculated in a PRA assumes average plant conditions, i.e. average component unavailabilities for the other components. If certain other components are actually down then the CDF increase from a normally, risk unimportant component can be significantly larger. The components which can cause these large effects will again be those components in the same minimal cutset as the unimportant component. Thus, there must be assurances that these adverse configurations are controlled and are avoided. The fourth issue that needs to be addressed is the potential importance of the components for other objectives including other risk measures not covered in the evaluation. For example, the component may be unimportant from a core damage frequency standpoint but may be important in reducing releases. Also, the component may be important in addressing specific regulations. These other areas need to be evaluated to assure overall unimportance. Table 3 summarizes the risk increase maintenance unimportance identification approach.

DEMONSTRATION OF THE MINIMAL CUTSET PRIORITIZATION A P P R O A C H

To demonstrate the minimal cutset approach for identifying the risk important maintenances, the NUREG 1150 PRA for the Surry plant is used. 5 The ranked minimal cutset contributions to the CDF as tabulated by the PRA are normalized by dividing by the CDF and multiplying by 100 to convert to per cent. The ranked, normalized minimal cutset contributions are then accumulated in a running sum and the unique, basic events in the cutsets are listed. The basic events which are listed are those having potential associated maintenances. (Operator errors were removed before prioritization.) These basic events can again be checked by the plant personnel to assure they have maintenance performed on them.

311

Importance measures for maintenance prioritization Table 4. Minimal cutset prioritization of the top potentially maintainable basic events for Surry Event index

Cutset rank

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

1 1 1 2 2 2 3 3 4 5 5 6 7 8 8 9 10 10 11 15 16 17 18 20 21 23 24 25 26 30 30 30 31 38 38 40 42 57 57 58 77 78 79

Event code

OEP-DGN-FS BETA-3DG RCP-LOCA-750-90M K Z R OEP-DGN-FS-DG01 OEP-DGN-FS-DG02 OEP-DGN-FS-DG03 MSS-SRV-OO-ODSRV SGTR-SGSRV-ODMD 1 BETA-2DG SGTR-SGSRV-ODMD2 LPR-MOV-FI'-1862A BETA-2MOV QS-SBO LPI-MDP-FS BETA-LPI LPI-MOV-PG-1890C AFW-PSF-FC-XCONN OEP-DGN-FR-6HDG1 OEP-DGN-FR-6HDG3 OEP-DGN-FR-6HDG2 ACC-MOV-PG-1865C ACC-MOV-PG-1865B RMT-CCF-FA-MSCAL HPI-MOV-FT LPR-MOV-Fl'-1860A LPR-MOV-FT-1890A PPS-MOV-Fr-1535 PPS-MOV-FC-1535 PPS-MOV-FC-1536 AFW-CCF-LK-STMBD MSS-SOV-OO-ODADV SGTR-SGADV-ODMD RCP-LOCA-467-150 HPI-MOV-FI'-1350 SBO-PORV-DMD PPS-SOV-OO-1456 PPS-SOV-OO-1455C HPI-CKV-FT-CV25 HPI-CKV-FT-CV225 HPI-CKV-FT-CV410

Event description

Diesel failure Beta for 3 diesels Seal failure Failure to scram Unfavorable moderator coefficient Failure to scram Diesel failure Diesel failure Diesel failure Safety relief valve failure Steam generator failure Beta for 2 diesels Steam generator failure Motor-operated valve failure Beta for 2 motor-operated valves Station blackout event Motor-driven pump failure Beta for 2 motor-driven pumps Motor-operated valve failure UNIT-2 event Diesel failure Diesel failure Diesel failure Motor-operated valve failure Motor-operated valve failure Common-cause failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Common-cause failure Solenoid-operated valve failure Steam generator rupture Seal failure Motor-operated valve failure Station blackout event Solenoid-operated valve failure Solenoid-operated valve failure Check valve failure Check valve failure Check valve failure

T h e m a i n t e n a n c e s a s s o c i a t e d with t h e basic e v e n t s in the t o p c o n t r i b u t i n g m i n i m a l cutsets, e.g. t h e t o p 9 0 % , a r e t h e n i d e n t i f i e d as the risk i m p o r t a n t maintenances. T a b l e 4 shows t h e t o p 43 basic e v e n t s so identified. Appendix 2 contains a more complete prioritization. S e l e c t e d e v e n t s c o u l d be f u r t h e r e l i m i n a t e d u p o n m o r e d e t a i l e d r e v i e w for r e f e r e n c e to m a i n t e n a n c e ; h o w e v e r , we w o r k with this t a b l e to d e m o n s t r a t e the a p p r o a c h . T h e first c o l u m n in T a b l e 4 is t h e basic e v e n t c o u n t e r a n d t h e s e c o n d c o l u m n is t h e r a n k o f the cutset c o n t a i n i n g t h e basic e v e n t . T h e t h i r d c o l u m n is t h e basic e v e n t c o d e as d e f i n e d in t h e P R A

Cutset frequency (per year)

Cumulative cutset (%)

1.17 × 10 -6 1.17 × 10 -6 1.17 × 10 6 8.43 × 10 7 8.43 x 10 -7 8.43 × 10 -7 6.21 x 10 -7 6.21 x 10 7 6.21 × 10 7 6.09 × 10 7 6.09 × 10 -7 5.77 × 10 -7 5.18 × 10 7 4.58 X 10 --7 4.58 × 10 7 4.54 X 10 -7 4.50 X 1 0 - 7 4.50 x 10 -7 4.40 X 10 -7 3.60 × 10 7 3.39 x 10 7 3.39 x 10 7 3.39 x 10 7 3.25 x 10 -7 3.25 x 10 7 3.00 x 10 7 2.64 x 10 7 2.64 x 10 7 2-64 x 10 7 2.42 X 10 -7 2-42 X 10 -7 2.42 × 10 7 2.40 × 10 7 2.21 x 10 7 2.21 X 10 -7 2.19 × 10 -7 2.02 x 10 -7 1.40 × 10 7 1.40 × 10 7 1-40 x 10 -7 1-00 x 10 7 1.00 x 10 7 1.00 X 10 -7

3-537 3-537 3.537 6.092 6.092 6.092 7.974 7.974 9.857 11.703 11.703 13.453 15.023 16.410 16.410 17.786 19.150 19.150 20-484 25.215 26.242 27.269 28.296 30.308 31.293 33.136 33.937 34.737 35-538 38.519 38.519 38.519 39.247 44.049 44.049 45.379 46.639 54.075 54-075 54-498 61.717 62.020 62.323

a n d the f o u r t h c o l u m n p r o v i d e s a g e n e r a l d e s c r i p t i o n o f the e v e n t . T h e e v e n t c o d e defines t h e d e t a i l e d e v e n t t h a t is i n v o l v e d ; for a c o m p o n e n t f a i l u r e , t h e c o d e identifies the specific s y s t e m , specific c o m p o n e n t (by its identification n u m b e r ) , a n d t h e specific failure m o d e . T h e S u r r y N U R E G 5 c o n t a i n s t h e k e y for the event c o d e . T h e next to last c o l u m n in T a b l e 4 is t h e c u t s e t f r e q u e n c y , i.e. C D F c o n t r i b u t i o n o f t h e m i n i m a l cutset c o n t a i n i n g t h e basic e v e n t . T h e last c o l u m n is the cutset c u m u l a t i v e p e r c e n t a g e c o n t r i b u t i o n , which is the r u n n i n g s u m o f t h e c u t s e t c o n t r i b u t i o n s including t h e p r e s e n t cutset. T h e r e a r e r e p e a t s in t h e

312

W. E. Vesely, M. Belhadj, J. T. Rezos

last column, as there are in the second column, since a cutset generally contains several basic events which have associated maintenances. The prioritization can be continued in the manner shown in Table 4 until a significant percentage of the total cutset contribution is obtained, such as 90%. As Appendix 2 shows, to include 90% of the total contribution, 85 basic events are identified. If the coverage is increased to 95% then 16 additional basic events are identified for a total of 101 basic events. The basic events are included in terms of their minimal cutset importance, and hence these last additional events are of much lesser importance than the first events included. For implementation, the risk important basic events and associated maintenances can be organized in various ways. If maintenance procedures are defined by type of component, then the risk important components of a similar type can be extracted, i.e. the risk important motor operated valves identified, the risk important motor driven pumps identified, etc. If maintenance procedures are defined by system, then the risk important components in given systems can be extracted. The risk important prioritization can thus be used in any way deemed most appropriate for implementation.

DEMONSTRATION OF THE RISK REDUCTION PRIORITIZATION APPROACH The Surry NUREG 1150 PRA is again used to demonstrate the risk reduction approach for identifying the risk important maintenances. As was described, the risk reduction approach prioritizes maintenances in terms of their risk impacts if not carried out effectively. The ranked, individual risk reduction importances tabulated by the PRA are normalized by their sum and then a running sum of the relative values is taken to obtain the cumulative relative contribution, as was done for the previous minimal cutset approach. The listing of the associated basic events then gives a prioritization of the events in terms of their CDF impact. Table 5 shows the top 55 basic events prioritized by their risk reduction importance. Appendix 3 contains the more complete prioritization. Only those basic events were ranked which had potential associated maintenances (e.g. operator errors were excluded). These events would again be checked and further events could be removed in actual implementation. The first column in Table 5 is the event rank and the second column is again the event code as identified in the Surry PRA. The third column is the event description. The fourth column is the CDF risk reduction importance, which again is the sum of the

contributions of the minimal cutsets containing the basic event. The next to last column is the relative risk reduction, which is the individual risk reduction divided by the sum of the risk reductions and multiplied by 100 to convert to per cent. The last column is the running sum of the relative risk reductions up through the current basic event. Having constructed the listing, the risk important events can then be identified as those constituting a significant percentage, such as 95%, of the CDF impact as measured by the sum of the risk reductions. The maintenances associated with the significant risk impacting events can then be identified as the important, risk impacting maintenances.

COMPARISON OF THE MINIMAL CUTSET AND RISK REDUCTION PRIORITIZATION PROCEDURES Table 6 compares the percentage level at which a basic event enters using the risk reduction approach versus the level at which it enters using the minimal cutset approach. Figure 1 shows an analogous comparison of the cutoff percentage versus the number of basic events included for the minimal cutset and risk reduction approaches. As Table 6 and Fig. 1 show, a basic event generally enters at a higher percentage level using the risk reduction approach than using the minimal cutset approach. This behavior was also found in applications using other PRAs. The demonstrations therefore indicate that a higher percentage should be used for the risk reduction approach as for the minimal cutset approach to obtain the same set of basic events and associated maintenances. For example, Table 6 or Fig. 1 indicates that using a 99% cutoff percentage for the risk reduction approach will yield basically the same set of basic events as a 90% cutoff percentage for the minimal cutset approach. Since the risk reduction approach is generally easier to apply than the minimal cutset approach, the use of a higher cutoff percentage should cause little extra work. The risk reductions are already standardly computed and ranked in the PRA, while the minimal cutset approach involves sorting of the basic events in the cutsets. Relatively few additional components and maintenances are added using a higher cutoff percentage which should also cause relatively little extra burden for the added assurance. Finally, as previously discussed, using a higher cutoff percentage in the risk reduction approach provides assurance that interaction effects from maintenances are included in the prioritization.

Importance measures for maintenance prioritization

313

Table 5. Risk reduction prioritization of the top potentially maintainable basic events for Surry Rank

1 2 3 4 5 6 7 8 9 10

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55

Event

OEP-DGN-FS-DG01 RCP-LOCA-750-90M OEP-DGN-FS OEP-DGN-FS-DG02 OEP-DGN-FS-DG03 OEP-DGN-FR-6HDG1 QS-SBO BETA-2MOV BETA-3DG OEP-DGN-FR-6HDG3 SBO-PORV-DMD BETA-2DG OEP-DGN-FR-6HDG2 R K MCW-CCF-VF-SBO MSS-SRV-OO-ODSRV HPI-MOV-FT PPS-SOV-OO-1455C PPS-SOV-OO-1456 OEP-CRB-FI'-15H3 RCP-LOCA-467-150 AFW-PSF-FC-XCONN LPR-MOV-FT-1862A SGTR-SGSRV-ODMD1 LPI-MDP-FS BETA-LPI AFW-TDP-FR-2P6HR LPI-MOV-PG-1890C AFW-TDP-FS-FW2 AFW-CCF-LK-STMBD SGTR-SGSRV-ODMD2 OEP-CRB-FT-15J3 LPR-MOV-FT-1860A RMT-CCF-FA-MSCAL PPS-MOV-FC-1536 PPS-MOV-FC-1535 LPR-MOV-FT-1890A PPS-MOV-FF-1535 ACC-MOV-PG-1865C ACC-MOV-PG-1865B MSS-SOV-OO-ODADV SGTR-SGADV-ODMD HPI-CKV-FT-CV225 HPI-CKV-FT-CV410 HPI-CKV-FT-CV25 HPI-MOV-FT-1350 AFW-MDP-FS BETA-AFW PPS-MOV-FT-1536 AFW-TDP-FR-2P24H RCS-PORV-ODMD LPR-MOV-FT-1862B OEP - DGN- F R - D G 0 1 AFW-MDP-FS-FW3A

Event description

Diesel failure Seal failure Diesel failure Diesel failure Diesel failure Diesel failure Station blackout event Beta for 2 motor-operated valves Beta for 3 diesels Diesel failure Station blackout event Beta for 2 diesels Diesel failure Failure to scram Failure to scram Common-cause failure Safety relief valve failure Motor-operated valve failure Solenoid-operated valve failure Solenoid-operated valve failure Circuit breaker failure Seal failure UNIT-2 event Motor-operated valve failure Steam generator failure Motor-driven pump failure Beta for motor-driven pumps Turbine-driven pump failure Motor-operated valve failure Turbine-driven pump failure Common-cause failure Steam generator failure Circuit breaker failure Motor-operated valve failure Common-cause failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Solenoid-operated valve failure Steam generator failure Check valve failure Check valve failure Check valve failure Motor-operated valve failure Motor-driven pump failure Beta for motor-driven pumps Motor-operated valve failure Turbine-driven pump failure Steam generator event Motor-operated valve failure Diesel failure Motor-driven pump failure

Risk reduction (per year)

Risk reduction (%)

8.22 x 10 6 5.20 x 10 6 4-88 × 10 6 4-38 x 10 6 4.38 × 10 6 4.08 x 10 6 3.04 × 10 6 2.72 × 10 6 2.66 x 10 -6 2.32 × 10 -6 2.27 x 10 6 2-25 x 10 6 2-09 x 10 6 1.51 x 10 6 1.51 x 10 -6 1.38 × 10 6 1.25 x 10 6 1-20 × 10 6 1-20 x 10 6 1.20 x 10 6 1.06 × 10 -6 9.74 × 10 -7 8.75 × 10 -7 7.95 × 10 -7 6.77 X 10 -7 6.75 × 10 -7 6.75 x 10 7 6.60 x 10 7 6.60 × 10 -7 6.42 × 10 - 7 5-82 x 10 7 5-75 x 10 7 5-65 x 10 -7 4.58 x 10 7 4.50 x 10 7 4.31 × 10 -7 4.26 x 10 7 4.09 x 10 7 3.87 x 10 7 3.25 x 10 7 3-25 x 10 -7 2.54 × 10 -7 2.54 × 10 -7 2-10 × 10 7 2.06 x 10 7 2.06 × 10 -7 2.02 × 10 - 7 1.73 x 10 7 1.73 × 10 7 1-45 x 10 7 1.26 × 10 7 1.22 × 10 7 1-09 × 10 7 1.02 × 10 7 1.00 x 10 7

11.000 6-959 6.531 5.861 5.861 5.460 4-068 3.640 3.560 3.105 3-038 3.011 2.797 2.021 2.021 1.847 1.673 1.606 1.606 1.606 1-419 1-303 1.171 1"064 0"906 0-903 0-903 0-883 0.883 0.859 0.779 0.769 0.756 0.613 0.602 0-577 0.570 0.547 0-518 0-435 0.435 0.340 0.340 0.281 0.276 0-276 0.270 0.232 0.232 0.194 0.169 0.163 0.146 0-136 0.134

Cumulative

% contribution 11"000 17"959 24-490 30-351 36-212 41 "672 45"741 49"381 52"940 56"045 59"083 62"094 64"891 66"911 68"932 70"779 72"452 74"058 75-663 77"269 78.688 79.991 81-162 82.226 83" 132 84.035 84-939 85-822 86-705 87-564 88"343 89"113 89-869 90-482 91.084 91.661 92.231 92.778 93.296 93-731 94-166 94"506 94.846 95-127 95.402 95.678 95.948 96.180 96.411 96-605 96.774 96-937 97.083 97.220 97.353

W. E. Vesely, M. Belhadj, J. T. Rezos

314

DEMONSTRATION OF THE RISK INCREASE A P P R O A C H FOR DETERMINING UNIMPORTANCES

Table 6. Comparison of the top risk reduction and minimal cutset prioritizations for Surry Rank based on risk reduction 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Event code

Cumulative % contribution (risk reduction)

Cutset % at which event enters

11.000 17.959 24.490 30.351 36-212 41.672 45.741 49.381 52.940 56.045 59.083 62.094 64.891 66-911 68.932 70.779 72.452 74.058 75.663 77.269 78.688 79-991 81-162 82-226 83.132 84.035 84.939 85.822 86.705 87.564 88.343 89.113 89.869 90.482 91.084 91-661 92.231 92.778 93.296 93.731 94.166 94-506 94.846 95.127 95-402

7.974 3.537 3.537 7.974 9.857 26.242 17.786 16.410 3.537 27.269 54.075 13.453 28.296 6.092 6.092 64.870 11.703 33.937 54.498 54.075 65.384 45.379 25-215 16-410 11-703 19-150 19.150 74-766 20.484 72.270 39.247 15.023 65.897 34.737 33.136 38.519 38.519 35.538 38.519 30.308 31.293 44.049 44.049 62.020 62.323

OEP-DGN-FS-DGOI RCP-LOCA-750-90M OEP-DGN-FS OEP-DGN-FS-DG02 OEP-DGN-FS-DG(13 OEP-DGN-FR-6HDG1 QS-SBO BETA-2MOV BETA-3DG OEP-DGN-FR-6HDG3 SBO-PORV-DMD BETA-2DG OEP-DGN-FR-6HDG2 R K MCW-CCF-VF-SBO MSS-SRV-OO-ODSRV HPI-MOV-FT PPS-SOV-OO-1455C PPS-SOV-OO- 1456 OEP-CRB-FT-15H3 RCP-LOCA-467-150 AFW-PSF-FC-XCONN LPR-MOV-FT-1862A SGTR-SGSRV-ODMDI LPI-MDP-FS BETA-LPI AFW-TDP-FR-2P6HR LPI-MOV-PG-1890C AFW-TDP-FS-FW2 AFW-CCF-LK-STMBD SGTR-SGSRV-ODMD2 OEP-CRB-FT-15J3 LPR-MOV-FT- 1860A RMT-CCF-FA-MSCAL PPS-MOV-FC-1536 PPS-MOV-FC-1535 LPR-MOV-FT- 1890A PPS-MOV-FT-1535 ACC-MOV-PG-1865C ACC-MOV-PG-1865B MSS-SOV-OO-ODADV SGTR-SGADV-ODMD HPI-CKV-FT-CV225 HPI-CKV-FT-CV410

To determine the risk unimportant maintenances, the Surry 1150 PRA is used and the ranked, risk increases with regard to the CDF are normalized by the CDF. Table 7 shows the basic events with the smallest risk increases. The maintenances associated with the basic events having insignificant risk increases (e.g. less than 10% or less than 5%) can be tentatively identified as being risk unimportant maintenances. However, among the checks that need to be made, it is noted that steam generator tube rupture associated events (SGTR-SGADV-ODMD, etc.) and safety relief valve failure (MSS-SRV-OO-ODSRV) are identified as being both unimportant and important and appear in both lists. As previously discussed, the effects of maintenance need to be especially determined for these events. Total risk effects, configuration effects and importances for other objectives need also to be evaluated for the risk unimportant events as was previously discussed.

CONCLUSION

Using either the minimal cutsets or the risk reduction importances, the basic events and their associated maintenances can be prioritized for their risk importances. Both the minimal cutsets and the risk reductions are standardly tabulated in a PRA so application of the approaches is straightforward. The risk reduction approach is somewhat simpler to use as it does not involve sorting out the basic events as required in using the minimal cutset approach. However, the minimal cutset approach is still reasonable to implement. In the demonstrations, both the minimal cutset approach and the risk reduction approach give similar

Cutoff Percentage 99.9990 99.996 - ~_ ~ Minimal Cursor 99.993-

99.990 -

~ ~ ' /

/

I

' ' '

'99.9990 "99.996

-99.993 - 99.990

99.96 -

•9 9 . 9 6

99.93 -

"99.93

99.90 -

- 99.90

99.6 -

- 99.6

99.3 -

- 99.3

99.0 96 93" 9060300 5

13

21

29

37

45

53

61

69

77

85

93

101 109 117 125

Number of Basic Events

Fig. l. Cutoff percentage versus number of included basic events for Surry.

315

Importance measures for maintenance prioritization Table 7. Lowest risk increases for the basic events for Surly Rank 94 95 96 97 98 99

100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146

Event HPI-MOV-FT-1115E HPI-MOV-FT-1115D HPI-MOV-FT-1115C AFW-CKV-FT-CVI57 AFW-CKV-FT-CVI72 AFW-XVM-PG-XV168 AFW-XVM-PG-XV183 PPS-SOV-FT RCP-LOCA-750-90M LPI-CKV-OO-CV58 LPI-CKV-OO-CV50 LPR-MOV-FT-1890B MSS-CKV-FT-SGDHR HPI-MDP-FR-1A6HR BETA-LPI HPI-MOV-FT-1867D RCP-LOCA-183-90 RCP-LOCA-183-150 RCP-LOCA-183-210 PPS-MOV-FF-1536 SGTR-SGSRV-ODMD2 BETA-AFW CPC-MDP-FR-CCA24 SBO-PORV-DMD RMT-ACT-FA-RMTSA RMT-ACT-FA-RMTSB PPS-MOV-FC-OPER CPC-CKV-OO-CVll3 AFW-TDP-FS-U2FW2 PPS-MOV-OO-1536 PPS-MOV-OO-1535 AFW-TDP-FR-6HRU2 CPC-MDP-FR-SWA24 PPS-MOV-FF PPS-MOV-FC-1536 PPS-MOV-FC-1535 AFW-TDP-FR-2P24H HPI-MOV-FT-1867C CPC-MDP-FS-SWl0B CON-VFC-RP-COREM CPC-MDP-FS-CC2B HPI-MDP-FS OEP-DGN-FC-DG3U2 CPC-MDP-FR-SWB24 RCS-PORV-ODMD BETA-STR BETA-SRV UNIT2-LOW-POWER BETA-HPI SGTR-SGADV-ODMD SGTR-SGSRV-ODMD1 MSS-SRV-OO-ODSRV MSS-SOV-OO-ODADV

Event description Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Check valve failure Check valve failure Manual valve failure Manual valve failure Solenoid-operated valve failure Seal failure Check valve failure Check valve failure Motor-operated valve failure Check valve failure Motor-driven pump failure Beta for motor-driven pumps Motor-operated valve failure Seal failure Seal failure Seal failure Motor-operated valve failure Steam generator failure Beta for motor-driven pumps Motor-driven pump failure Station blackout event Actuator failure Actuator failure Motor-operated valve failure Check valve failure Turbine-driven pump failure Motor-operated valve failure Motor-operated valve failure Turbine-driven pump failure Motor-driven pump failure Motor-operated valve failure Motor-operated valve failure Motor-operated valve failure Turbine-driven pump failure Motor-operated valve failure Motor-driven pump failure Containment event Motor-driven pump failure Motor-driven pump failure Diesel failure Motor-driven pump failure Steam generator event Beta for strainers Beta for safety relief valves UNIT-2 event Beta for motor-driven pumps Steam generator failure Steam generator failure Safety relief valve failure Solenoid-operated valve failure

results p r o v i d i n g a higher cutoff is used for the risk r e d u c t i o n approach. Since the risk r e d u c t i o n i m p o r tance involves a s u m m a t i o n of p e r t i n e n t m i n i m a l cutset c o n t r i b u t i o n s , it is n o t surprising that b o t h give consistent results. T h e i n t e r a c t i o n t e r m s in the risk r e d u c t i o n a p p r o a c h can be used to check for a d d i t i o n a l impacts from ineffective m a i n t e n a n c e s

Risk increase 5.63 × 5.63 x 5.63 × 5-40 × 5.40 x 5.40× 5.40 >( 4-71 x 4.61 × 4.50 X 4.50 x 4.49 x 4.05 × 4.00 x 3.83 x 3.57 × 3.49 × 3-49 × 3.49 × 3.48 × 3.26 × 2.92 x 2-86 x 2-77 x 2.73 x 2.73 × 2.41 × 2-17 × 1-96 x 1.39 x 1.39 × 1.16 × 1.04 x 1-01 × 1.01 × 9.93 × 9.21 × 7.42 × 7.16 x 4.93 x 4.11 x 1.91 x 1-54 x 1.52 x 1.22 x 9.36 × 6.26 x 2.40 × 2.89 × 0.00 0.00 0-00 0-00

10 6 10 6 10 -° 10 -6

10 6 10 6 10 - 6

10 6 10 6 10 - 6

10 6 10 -6 10 -6 10 6 10 6 10-6 10 6 10 6 10 6 10 6 10 6 10 6 10 6 10 6 10 6 10 6 10 6 10 -6 10 6 10 6 10 -6

10 -6 10 6 10 6 10 -6

10 -7 10 ~ 10 ~ 10 7 10 ~ 10 -v 10 7 10 7 10 7 10 7 10 ~ 10 ~ 10 8 10 ~

Risk increase (%) 14-040 14-040 14.040 13.466 13.466 13.466 13-466 11-746 11.496 11.222 11-222 11.197 10.100 9.975 9.551 8.903 8-703 8.703 8.703 8.678 8-130 7.282 7.132 6.908 6.808 6-808 6-010 5.411 4.888 3-466 3.466 2.893 2.594 2.519 2.519 2.476 2-297 1.850 1.786 1.229 1.025 0-476 0.384 0.379 0-304 0.233 0.156 0.060 0-007 0-000 0.000 0-000 0.000

being p e r f o r m e d o n m u l t i p l e c o m p o n e n t s . Using the risk increase i m p o r t a n c e s t a b u l a t e d as s t a n d a r d in the P R A , risk u n i m p o r t a n t basic e v e n t s a n d u n i m p o r t a n t m a i n t e n a n c e s can also be identified. In i m p l e m e n t a t i o n s i n v o l v i n g the risk u n i m p o r t a n t findings, a s s u r a n c e s are n e e d e d that c u m u l a t i v e effects and configuration effects are c o n t r o l l e d .

W. E. Vesely, M. Belhadj, J. T. Rezos

316

original unavailability, where f is a general factor increase. The new cutset contribution Q~ is thus

REFERENCES 1. Vesely, W. E. & Davis, T. C., Evaluations and utilizations of risk importances. NUREG/CR-4377, August 1985. 2. Henley, E. J. & Kumamoto, H., Probabilistic Risk Assessment. IEEE Press, New York, 1992. 3. Lofgren, E. V., Cooper, S. E., Kurth, R. E. & Phillips, L. B., A process for risk-focused maintenance. NUREG/CR-5695, February 1991. 4. Vesely, W. E., Kurth, R. E. & Scalzo, S. M., Evaluations of core melt frequency effects due to component aging and maintenance. NUREG/CR-5510, June 1990. 5. Bertucio, R. C. & Julius, J. A., Analysis of core damage frequency: Surry, Unit 1. NUREG/CR-4550, April 1990.

APPENDIX 1 THE GENERAL RISK SENSITIVITY FORMULA

Q; = Q;(1 + f ) " '

where n~ is the number of maintainable components in cutset i. Expanding eqn (A1.2) as a power series gives

Q:= Q i ( l + n ~ f +

(2i)f2+'' +fn*)

(A1.3)

Now, when Q~ is summed over all the cutsets to obtain the new core damage frequency using eqn (AI.1), the first term on the right hand side of eqn (A1.3) gives the original value C. For each maintainable component, the second term will give Qif for each cutset containing the component. For each pair of maintainable components, the third term will yield Qif2 for each cutset containing the pair. Hence we may write the expression for the new core damage frequency C' as

The general minimal cutset formula for the core damage frequency C (or for any other appropriate risk result) is

C ' = C + ~ rif + E r , , f ~ + ' ' ' i>j

i=1

+

N

C = ~ Qi

(A1.2)

(AI.1)

~

ri,...iJ k

(A1.4)

il>...>itc

i=1 or

where Q~ is the ith minimal cutset contribution, and where there are N total minimal cutsets. As previously discussed, the minimal cutset contribution Q, generally consists of the frequency of an initiating event times the product of the unavailabilities of the basic events in the cutsets. Assume that the individual unavailabilities of the maintainable components are increased because of ineffective maintenances. To be general, assume that the new unavailability is a factor of 1 + f times the

Ac=

r,y+Zri, f + - . . i-- 1

+

i>j

E

ri,...,kf k

(A1.5)

il>--.>-ik

where ri is the sum of the minimal cutset contributions containing component i, ro is the sum of minimal cutset contributions each containing components i and j, etc.

Importance measures for maintenance prioritization

317

APPENDIX 2 MINIMAL CUTSET PRIORITIZATION OF THE POTENTIALLY MAINTAINABLE BASIC EVENTS FOR SURRY

Event code

10EP-DGN-FS 2 BETA-3DG 3 RCP-LOCA-750-90M 4 K 5 Z 6 R 70EP-DGN-FS-DG01 80EP-DGN-FS-DG02 9 OEP-DGN-FS-DG03 10 MSS-SRV-OO-ODSRV 11 S G T R - S G S R V - O D M D I 12 BETA-2DG 13 SGTR-SGSRV-ODMD2 14 LPR-MOV-FT-1862A 15 BETA-2MOV 16 QS-SBO 17 LPI-MDP°FS 18 BETA-LP! 19 LPI-MOV-PG-1890C 20 AFW-PSF-FC-XCONN 21 O E P - D G N - F R - 6 H D G 1 22 O E P - D G N - F R - 6 H D G 3 23 O E P - D G N - F R - 6 H D G 2 24 ACC-MOV-PG-1865C 25 ACC-MOV-PG-1865B 26 RMT-CCF-FA-MSCAL 27 HPI-MOV-FT 28 LPR-MOV-FF-1860A 29 LPR-MOV-FT-1890A 30 PPS-MOV-FI'-I535 31 PPS-MOV-FC-1535 32 PPS-MOV-FC-1536 33 AFW-CCF-LK-STMBD 34 M S S - S O V - O O - O D A D V 35 S G T R - S G A D V - O D M D 36 RCP-LOCA-467-150 37 HPI-MOV-FI'-I350 38 S B O - P O R V - D M D 39 PPS-SOV-OO-1456 40 PPS-SOV-OO-1455C 41 HPI-CKV-FT-CV25 42 HPI-CKV-FF-CV225 43 HPI-CKV-FT-CV410

Cumulative % 3.537 3.537 3.537 6.092 6-092 6.092 7.974 7.974 9'857 11.703 11.703 13.453 15.023 16-410 16-410 17.786 19.150 19.150 20-484 25.215 26.242 27.269 28.296 30.308 31.293 33.136 33-937 34.737 35.538 38-519 38"519 38.519 39-247 44.049 44.049 45.379 46.639 54.075 54-075 54.498 61.717 62-020 62.323

44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

Event code

Cumulative %

MCW-CCF-VF-SBO OEP-CRB-FT-15H3 OEP-CRB-FT-15J3 O BETA-AFW AFW-TDP-FR-2P24H AFW-MDP-FS PPS-MOV-FT-1536 AFW-TDP-FS-FW2 ACC-CKV-FT-CVI30 ACC-CKV-FT-CVI47 ACC-CKV-FT-CVI28 LPR-CCF-PG-SUMP ACC-CKV-IzT-CV145 AFW-TDP-FR-2P6HR HPI-XVM-PG-XV24 RWT-TNK-LF-RWST AFW-MDP-FS-FW3B AFW-MDP-FS-FW3A LPR-MOV-FT-1862B AFW-CKV-OO-CVI42 PPS-MOV-OO-1536 RCS-PORV-ODMD PPS-MOV-OO-1535 BETA-STR CPC-STR-PG-3HR PPS-MOV-17F HPI-MDP-FR-IA24H HPI-CKV-OO-CV258 LPI-MDP-FS-SI1B LPI-MDP-FS-SI1A LPR-MOV-FI'-I860B AFW-CKV-OO-CVI57 AFW-CKV-OO-CV172 IAS-CCF-LF-INAIR RCP-LOCA-1440-90 RCP-LOCA-183-210 RCP-LOCA-183-150 OEP-DGN-FR-DG01 OEP-DGN-FR-DG03 OEP-DGN-FR-DG02 RCP-LOCA-183-90 SIS-ACT-FA-SISA

64.870 65.384 65.897 70-049 70-434 70-434 70.434 71.753 72-270 74.009 74.161 74.312 74.464 74.616 74.766 77.664 79.696 80.006 80-110 82-706 83.278 83.516 83.516 83.595 84.341 84.341 85.169 85-295 85.295 87.789 87-837 87.884 88.302 88.347 88.481 88-525 89-559 89.598 89.636 89.675 89.790 90.848 91.270

Event code

87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129

SIS-ACT-FA-SISB HPI-MOV-FT-1115E HPI-MOV-FI~-lll5C LPR-MOV-FI'-I890B HPI-MOV-FT-Ill5D HPI-MOV-FT-II15B MSS-CKV-FT-SGDHR CPC-STR-PG-24H RCP-LOCA-561-150 ACP-BAC-ST-1H1-2 ACP-BAC-ST-4KVIH ACP-BAC-ST-1HI PPS-SOV-171'-I455C PPS-SOV-FI'-1456 PPS-SOV-FF BETA-SRV OEP-CRB-FT-25H3 AFW-TDP-FR-6HRU2 CPC-MDP-FS-SWIOB CPC-MDP-FR-SWA3H HPI-MOV-Flr-1867D LPI-MDP-FR-B21HR LPI-MDP-FR-A21HR AFW-ACT-FA-PMP3A AFW-ACT-FA-PMP3B CON-VFC-RP-COREM SWS-CCF-FI'-3ABCD LPI-CKV-OO-CV50 LPI-CKV-OO-CV58 ACP-TFM-NO-1HI HPI-MOV-PG-1350 RMT-ACT-FA-RMTSB RMT-ACT-FA-RMTSA OEP-DGN-FC-DG3U2 AFW-TNK-VF-CST DCP-BDC-ST-BUS1B DCP-BDC-ST-BUS1A CPC-CKV-OO-CV113 CPC-MDP-FR-SWA24 CVC-MDP-FR-2A1HR LPI-MDP-FR-B24HR LPI-MDP-FR-A24HR PPS-MOV-FC-OPER

Cumulative % 91.270 92-016 92-016 92.043 92.152 92.152 92.278 93.079 93.214 94'189 94-207 94.226 94.976 94.991 95-241 95.241 95.802 95.878 95.938 95.938 96.192 96.501 96.511 96'530 96.540 96.804 96.804 96-831 96.841 97.116 97.125 97.211 97.211 97.257 97.317 97.417 97.424 97-546 97-546 97.609 97.761 97.767 97.877

W. E. Vesely, M. Belhadj, J. T. Rezos

318

APPENDIX 3 RISK REDUCTION PRIORITIZATION OF THE POTENTIALLY MAINTAINABLE BASIC EVENTS FOR SURRY Event

10EP-DGN-FS-DG01 2 RCP-LOCA-750-90M 30EP-DGN-FS 4 OEP-DGN-FS-DG02 50EP-DGN-FS-DG03 60EP-DGN-FR-6HDGI 7 QS-SBO 8 BETA-2MOV 9 BETA-3DG 10 O E P - D G N - F R - 6 H D G 3 11 S B O - P O R V - D M D 12 B E T A - 2 D G 13 O E P - D G N - F R - 6 H D G 2 14 R 15 K 16 MCW-CCF-VF-SBO 17 MSS-SRV-OO-ODSRV 18 HP1-MOV-FT 19 PPS-SOV-OO-1455C 20 PPS-SOV-OO-1456 21 OEP-CRB-b'T-15H3 22 RCP-LOCA-467-150 23 AFW-PSF-FC-XCONN 24 LPR-MOV-Fr-1862A 25 S G T R - S G S R V - O D M D I 26 LPI-MDP-FS 27 BETA-LPI 28 AFW-TDP-FR-2P6HR 29 LPI-MOV-PG-1890C 30 AFW-TDP-FS-FW2 31 AFW-CCF-LK-STMBD 32 S G T R - S G S R V - O D M D 2 33 OEP-CRB-171'-I5J3 34 LPR-MOV-FT-1860A 35 RMT-CCF-FA-MSCAL 36 PPS-MOV-FC-1536 37 PPS-MOV-FCoI535 38 LPR-MOV-FT-1890A 39 PPS-MOV-FF-1535 40 ACC-MOV-PG-1865C 41 ACC-MOV-PG-1865B 42 M S S - S O V - O O - O D A D V 43 S G T R - S G A D V - O D M D 44 HPI-CKV-F'I'-CV225 45 HPI-CKV-FF-CV410 46 HPI-CKV-FT-CV25 47 HPI-MOV-FT-1350 48 AFW-MDP-FS 49 B E T A - A F W 50 PPS-MOV-FF-1536 51 AFW-TDP-FR-2P24H 52 R C S - P O R V - O D M D 53 LPR-MOV-FT-1862B

Cumulative % 11-0130 17-959 24.490 30.351 36.212 41'672 45.741 49.381 52-940 56.045 59.083 62-094 64.891 66-911 68.932 70.779 72.452 74"058 75-663 77.269 78-688 79.991 81.162 82.226 83.132 84.035 84.939 85.822 86.705 87.564 88.343 89.113 89.869 90.482 91.084 91.661 92-231 92.778 93.296 93.731 94-166 94.506 94.846 95.127 95.402 95.678 95-948 96-180 96-411 96.605 96-774 96-937 97.083

Event

54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106

OEP-DGN-FR-DG01 AFW-MDP-FS-FW3A AFW-MDP-FS-FW3B HPI-XVM-PG-XV24 LPR-CCFoPG-SUMP LPI-MDP-FS-SI1B LPI-MDP-FS-SIlA RCP-LOCA-1440-90 LPR-MOV-FT-1860B AFW-CKV-OO-CVI42 PPS-MOV-OO-1536 PPS-MOV-OO-1535 RCP-LOCA-183-210 RCP-LOCA-183-150 RWT-TNK-LF-RWST OEP-DGN-FR-DG02 AFW-CKV-OO-CV172 OEP-DGN-FR-DG03 ACC-CKV-FI'-CV145 ACC-CKV-I~I'-CV128 ACC-CKV-Fr-cvI30 ACC-CKV-17I'-CVI47 RCP-LOCA-183-90 PPS-MOV-FT IAS-CCF-LF-INAIR AFW-TDP-FR-6HRU2 BETA-STR RCP-LOCA-561-150 SIS-ACT-FA-SISA SIS-ACT-FA-SISB AFW-CKV-OO-CV157 OEP-CRB-FT-25H3 CPC-STR-PG-3HR HPI-CKV-OO-CV258 AFW-TDP-FS-U2FW2 HPI-MDP-FR-IA24H HPI-MOV-FT-1115B HP1-MOV-FT-1115E HPI-MOV-FT-II15D HPI-MOV-FT-III5C LPR-MOV-FT-1890B UNIT2-LOW-POWER ACP-BAC-ST-4KVIH HPI-MOV-FT-1867D PPS-SOV-FF-1456 PPS-SOV-FT-1455C ACP-BAC-ST-IHI CON-VFC-RP-COREM LP1-MDP-FR-B21HR LPI-MDP-FR-A21HR MSS-CKV-FI'-SGDHR CPC-STR-PG-24H SWS-CCF-FT-3ABCD

Cumulativc % 97-220 97.353 97-486 97.596 97.700 97.799 97-899 97-984 98.068 98.149 98.227 98.304 98-380 98-457 98.527 98.596 98.664 98.732 98-798 98-865 98-932 98.999 99-066 99-122 99.172 99.220 99.264 99.308 99.346 99-385 99.419 99.452 99-484 99.514 99.543 99.572 99.597 99.620 99.643 99.665 99.683 99.701 99.716 99.731 99.745 99.759 99.773 99.787 99.798 99.809 99.820 99-830 99.840

Event

107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146

PPS-MOV-FC-OPER ACP-BAC-ST-1HI-2 CPC-MDP-FS-SWIOB AFW-ACT-FA-PMP3B AFW-ACT-FA-PMP3A OEP-DGN-FC-DG3U2 CPC-MDP-FR-SWA3H LPI-MDP-FR-B24HR LPI-MDP-FR-A24HR BETA-SRV PPS-SOV-FT LPI-CKV-OO-CV58 LPI-CKV-OO-CV50 RMT-ACT-FA-RMTSB RMT-ACT-FA-RMTSA ACP-TFM-NO-1H1 CPC-MDP-FR-SWA24 DCP-BDC-ST-BUSIB DCP-BDC-ST-BUS1A AFW-TNK-VF-CST HPI-MOV-PG-1350 HPI-MOV-FT-1867C CPC-CKV-OO-CVI13 CPC-MDP-FR-CCA24 CVC-MDP-FR-2AIHR AFW-MDP-FR-3B6HR AFW-MDP-FR-3A6HR CPC-STR-PG-6HR HPI-MDP-FR-1A6HR CPC-MDP-FS-CC2B ACP-BAC-ST-4KV1J BETA-HPI HPI-MDP-FS CPC-STR-PG-2A3HR CPC-MDP-FR-SWB24 AFW-CKV-FT-CVI57 AFW-CKV-FI'-CVI72 CPC-STR-PG-IHR AFW-XVM-PG-XV183 AFW-XVM-PG-XV168

Cumulativc % 99.849 99.857 99.865 99.872 99.88(I 99.887 99.893 99-900 99.906 99.912 99-919 99.925 99.931 99.937 99.942 99.948 99.953 99.958 99.963 99.967 99.970 99.973 99-976 99.979 99.982 99.984 99-986 99.988 99-990 99-992 99.993 99-994 99.996 99.996 99-997 99-998 99-999 99-999 100-000 100.000