See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/3267258

A Reliability-Centered Asset Maintenance Method for Assessing the Impact of

Maintenance in Power Distribution Systems

Article in Power Systems, IEEE Transactions on · March 2005

DOI: 10.1109/TPWRS.2004.840433 · Source: IEEE Xplore

CITATIONS READS
328 3,039

3 authors, including:

Lina Bertling Tjernberg

KTH Royal Institute of Technology
185 PUBLICATIONS 6,018 CITATIONS

SEE PROFILE

All content following this page was uploaded by Lina Bertling Tjernberg on 06 August 2013.

The user has requested enhancement of the downloaded file.

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 20, NO. 1, FEBRUARY 2005 75

A Reliability-Centered Asset Maintenance Method

for Assessing the Impact of Maintenance in
Power Distribution Systems
Lina Bertling, Member, IEEE, Ron Allan, Fellow, IEEE, and Roland Eriksson, Senior Member, IEEE

Abstract—This paper proposes a method for comparing the measures can impact on reliability by either improving the con-
effect of different maintenance strategies on system reliability and dition, or prolonging the lifetime of an asset. Reliability overall
cost. This method relates reliability theory with the experience can be improved by lowering either the frequency or the du-
gained from statistics and practical knowledge of component fail-
ures and maintenance measures. The approach has been applied ration of interruptions. PM activities could impact on the fre-
to rural and urban distribution systems. In particular, a functional quency by preventing the actual cause of the failure. Conse-
relationship between failure rate and maintenance measures has quently, PM is cost-effective when the reliability benefit out-
been developed for a cable component. The results show the value weighs the cost of implementing the PM measure. There is,
of using a systematic quantitative approach for investigating the therefore, a need for utilities to incorporate systematic methods
effect of different maintenance strategies.
which relate maintenance of system assets to the improvement
Index Terms—Asset management, electric power distribution in system reliability. This is part of the wider concept of asset
system, maintenance strategy, reliability evaluation, relia- management. Asset management involves making decisions to
bility-centered asset maintenance (RCAM).
allow the network business to maximize long term profits, while
delivering high service levels to the customers with acceptable
I. INTRODUCTION and manageable risks.
Reliability evaluation and maintenance planning techniques
E LECTRIC power distribution systems constitute the
greatest risk to the interruption of power supply [1]–[3].
Traditionally, however, distribution systems have received less
have separately been well developed, for example [1]–[4],
[8], [9], with reliability assessment starting in the 1930s [10].
attention than generation and transmission, evidenced by the However, few techniques relate system reliability to component
difference in the number of publications [4]. However, focus is maintenance. Furthermore, the available techniques are not
moving toward distribution as the business focus changes from generally put into practice. The reason for this, according with
consumers to customers. the authors, is the lack of suitable input data and a reluctance
Deregulation of the power system market has led to a shift to use theoretical tools to address the practical problem of
from technical to economic driving factors. The utilities that maintenance planning.
own and operate the power distribution systems now face One method for relating reliability to PM is known as relia-
various market requirements. On the one hand, customers are bility-centered maintenance (RCM). RCM is a qualitative sys-
paying for a service (delivered energy) and the authorities are tematic approach to organizing maintenance [11]–[13]. It origi-
imposing regulation, supervision, and compensation depending nated in the civil aircraft industry in the 1960s with the introduc-
on the degree to which contractual and other obligations are tion of the Boeing 747 series, and the need to lower PM costs
fulfilled, see for example Norway [5], Sweden [6], and the U.K. in attaining a certain level of reliability. The results were suc-
[7]. On the other hand, utilities must ensure that their expendi- cessful and the methodology was developed further. In 1975,
ture is cost-effective. This means that electricity utilities must the U.S. Department of Commerce defined the concept RCM
satisfy quantitative reliability requirements while at the same and declared that it should be used in all major military sys-
time minimizing their costs. tems [11]. In the 1980s, the Electric Power Research Institute
One predominant expense for a utility is the cost of main- (EPRI) introduced RCM into the nuclear power industry. Today
taining system assets, for example through adopting preventive RCM is used or being considered by an increasing number of
measures, collectively called preventive maintenance (PM). PM electrical utilities [14], [15]. The main feature of RCM is its
focus on preserving system function where critical components
for system reliability are prioritized for PM measures. However,
Manuscript received June 29, 2004. This work was supported by the Com- the method is generally not capable of showing the benefits of
petence Center in Electric Power Engineering at the Royal University of Tech- maintenance for system reliability and costs.
nology (KTH). Paper no. TPWRS-00271-2003. This paper proposes a reliability-centered asset maintenance
L. Bertling and R. Eriksson are with the Electrical Engineering Department,
Royal Institute Technology (KTH), 100 44 Stockholm, Sweden (e-mail: (RCAM) method, which provides a quantitative relationship
[email protected]; [email protected]). between PM of assets and the total maintenance cost [2].
R. Allan is with the Electrical Engineering Department, Manchester Centre The method is developed from RCM principles attempting to
for Electrical Energy, University of Manchester Institute of Science and Tech-
nology (UMIST), Manchester, U.K. (e-mail: [email protected]). relate more closely the impact of maintenance to the cost and
Digital Object Identifier 10.1109/TPWRS.2004.840433 reliability of the system. The method has been developed from
0885-8950/$20.00 © 2005 IEEE
76 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 20, NO. 1, FEBRUARY 2005

comprehensive application studies for real power distribution input data, defines the quantitative relationship be-
systems. Application studies have been made on two different tween reliability and PM measures.
distribution systems in Sweden: a rural system of overhead Stage 3 System reliability and cost/benefit analysis: puts the
power lines in southern Sweden, and an urban underground results of Stage 2 into a system perspective, and
cable system in central Stockholm, the Birka System. Both evaluates the effect of component maintenance on
studies used data for the systems in question, and were done in system reliability and the impact on cost of different
close co-operation with the operating utilities (Sydkraft AB and PM strategies.
Fortum Distribution AB (former Birka Nät AB), respectively). These three stages emphasize a central feature of the method:
More details are provided for the Birka System in Section IV. that the analysis moves from the system level to the component
level and back to the system level.
II. RCAM METHOD B. Economic Evaluation
A. Reliability Evaluation The economic evaluation brings the RCAM analysis to its
This paper addresses the effects of failure events in electric final step: to relate the benefits in costs due to the impact of
power distribution systems. These events occur randomly and maintenance on reliability. The motivation for any PM strategy
therefore models based on probability theory have been used. is that the cost of applying the PM measure should be less than
A computer code RADPOW (reliability assessment of elec- taking no action at all. If little or no PM is done, then more
trical distribution systems), based on the analytical approach, system failures are likely to occur resulting in more repair ac-
has been developed within the Competence Centre of Electrical tions being required, i.e., in more corrective maintenance (CM)
Engineering at KTH [2]. A network modeling technique and the actions. Therefore, the important issue is to compare the costs
minimal cut set (load-point-driven) approach [1] is used to de- associated with different maintenance methods, including both
duce the failure modes. RADPOW evaluates the load point in- PM and CM with the objective of minimizing the total cost of
dices, and the overall system indices. maintenance.
There are several costs that can be related to the effect of
— The load point indices are: expected failure rate
system failures. Two direct utility costs are: 1) cost of failure
, annual outage time
(CM), e.g., repair costs and losses in revenue due to nondeliv-
(unavailability) (U) [h/yr], average outage duration (r)
ered energy and 2) cost of the PM actions, e.g., planned main-
[h/int], and average energy not supplied (E) [kWh/yr].
tenance or replacement of a component in advance of failure.
— The system indices are: SAIFI [int/yr,customer],
However, the cost of failure also depends on the customer cost
SAIDI [h/yr,customer], CAIDI [h/int], and AENS
[16]. A supply interruption affects the customer, who will suffer
[kWh/yr,customer].
supply unavailability and may suffer direct costs and/or be com-
As a first step in the method, the critical components for
pensated via a penalty payment. Consequently, the proposed
the system reliability are identified from a sensitivity analysis.
cost analysis considers:
These components are further studied, focusing on the impact
of maintenance measures. The relationship between reliability • the cost of failure ;
and maintenance has been established by relating the effect of • the cost of preventive maintenance ;
PM to the causes of failures for the component being assessed. • the cost of interruption .
Two different approaches have been used. The first approach The optimal maintenance method and PM strategy is the solu-
assumes a constant reduction ratio between failure rates and the tion that minimizes the sum of these three costs. However, in
effect of PM, whereas the second approach assumes this ratio to some cases it may not be necessary to include , for example
be dependent on time. In the first case, depends only on for a simple or first-order comparison of strategies.
the effect of PM (Approach I). In the second case, is The economic evaluations have been made using fundamental
also time-dependent (Approach II), and the failure rate reduction techniques. The costs are evaluated on an annual basis with an
is a consequence of the PM actions considered for the specific assumed increase due to inflation . Furthermore, the invest-
component that is studied. ments in PM measures are spread over the remaining time of
Formulating the failure rate model for Approach II is a com- the assessment period . Finally, the present worth value of the
plicated task. This has presently been done for one component total annualized costs is evaluated. The present worth value of
type, underground cables, which was shown to be critical for the one outlay to be paid after years with the discount rate
reliability of one of the systems used in these studies. The de- , is gained by multiplying by the present worth value factor
tails of the underlying theory are too extensive to be developed .
in this paper, so only the overall principles, results and applica-
tions are included. III. STEPS IN THE RCAM METHOD
The main stages of the RCAM approach are as follows. Fig. 1 illustrates the logic for the RCAM method. This figure
Stage 1 System reliability analysis: defines the system and includes the different stages and steps in the method, and the
evaluates critical components affecting system reli- systematic process for analyzing the system components and
ability. their causes of failures. The resulting method has been imple-
Stage 2 Component reliability modeling: analyzes the com- mented in MATLAB where output from RADPOW is used as
ponents in detail and, with the support of appropriate input [2].
BERTLING et al.: RCAM METHOD FOR ASSESSING THE IMPACT OF MAINTENANCE IN POWER DISTRIBUTION SYSTEMS 77

2) Identify critical voltage levels and components for the

system reliability based on results from reliability analysis.
Theapproachforthesensitivityanalysisisasfollows:catego-
rize components according to their type, vary their input failure
rates for one type at a time, and evaluate the resulting indices
for the system and different load points. Perform this analysis
for different voltage levels and load points. The results provide
a prioritized list of components for PM measures.
Stage 2—Component reliability analysis.
3) Identify failure causes by failure modes analysis for each
component identified as critical and affected by PM.
• Identify causes of failures from an understanding of:
component functions, failure modes and failure events.
• Determine the percentage each cause contributes to
the total number of failures from interruption data and
expertise.
• Identify experience data for interruptions due to these
causes of failures.
• Identify possible effect of alternative PM methods.
4) Define a failure rate model.
For components model the failure rate func-
tion as follows:
a) Approach I:
Simply assume that the failure rate equals the av-
erage failure interruption, , from reliability input
data (from Step 1)

(1)

b) Approach II:
Assume that the component failure rate function can
be obtained as a sum of contributions from the different
causes of failures of type . Deduce a
model for the failure rate as a function of time, using
experience data from Step 2 for the failure rate mod-
eling, as follows:

(2)

5) Model effect of PM methods on reliability for each failure

cause.
Assume that the PM method , preventing
failure cause is applied to component number . For each
PM method define a failure rate model as follows:
a) Approach I:
• Assume that the effect of applying PM is a re-
duction of the actual failure cause with %
reduction, where and , is the per-
Fig. 1. Logic for the RCAM method (the steps that feature the asterisk (*) use centage contribution to the total failures of that
RADPOW for reliability analysis).
failure cause, and given from Step 3.
• Assume that the failure rate for the analyzed
The ten steps needed to perform the RCAM approach, as component is reduced by the same percentage.
identified in Fig. 1, are presented in more detail in this section. The resulting failure rate function can be evalu-
Stage 1—System reliability analysis ated from
1) Define reliability model and required input data.
Define input data including: network data, component re- (3)
liability data and customer data, and a reliability model.
78 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 20, NO. 1, FEBRUARY 2005

b) Approach II: b) Approach II:

• Deduce a model for functional relationship be-
tween reliability and PM activities as a function
of time. This model requires more knowledge .. .. (6)
about the component behavior and the effect of . .
applying PM with method and the impact on
specific failure causes. where we also have (6 ), which is shown at the bottom
• The resulting failure rate function can be evalu- of the page, define the resulting failure rate function.
ated from 9) Compare system reliability when applying different main-
(4) tenance methods and PM strategies.
• Perform system reliability analysis with result from
6) Deduce different plans for applying PM, and evaluate the Step 8 as input data for included components. The
resulting effect on the component failure rate. output is the system and load-point reliability indices
Note that for Approach II this requires the effect of ap- that show the different effects of the PM strategy (S)
plying PM at different times on the resulting failure rate func- on the system.
tions to be evaluated. • Compare the impact of PM strategy on system and
Stage 3—System reliability and cost/benefit analysis. load-point reliability indices.
7) Define and implement different strategies for PM. • For Approach II, an alternative is to compare the av-
A PM strategy, , for the system is defined by: erage load-point indices during the period, evaluated as
• applied PM methods denoted by: ; follows:
• proportion of the component type that are affected by
each PM method denoted by , and also for Approach (7)
II, and within the period ;
• number of times PM is applied ; and similarly for each load
• at what times PM is applied . point, , in the system model.
8) Estimate the resulting composite failure rate. • Analyze the effect of using different PM strategies on
This step implies developing the failure rate model for system reliability.
10) Identify cost effective PM strategy
the component applied with PM strategy . The resulting
• Evaluate cost functions in [cost/yr], based on those that
failure rate function provides the input data for component
were introduced in Section II:
type to the system reliability model.
• the cost of failure ;
• Define which failure causes are affected by each PM
• the cost of preventive maintenance ;
method in the strategy. Let denote the affected
• the cost of interruption
causes, and denote the nonaffected causes.
with and without PM respectively as follows:
• The resulting failure rate function captures the average
composite failure rate characteristic for the component a) Approach I:
. It is made up of several parts, depending on the PM
(8)
strategy.
a) Approach I:
where is the cost of failure for component
• Define the extent of the effect for each failure
[cost/int].
cause, affected by PM method , that is . b) Approach II:
• Evaluate the resulting composite failure rate for
component type , which is given as follows:

(9)
(5)
where is the inflation rate.

..
.

(6 )
BERTLING et al.: RCAM METHOD FOR ASSESSING THE IMPACT OF MAINTENANCE IN POWER DISTRIBUTION SYSTEMS 79

c) Approach I:

(10)

where is the cost of applying PM method for

component [cost/measure].
d) Approach II: is shown in (11), at the
bottom of the page, where the cost of applying PM, at
each PM occasion, is equally spread over the remaining
time period.
a) Approach I:
(12)
where is the customer interruption cost in
[cost/kWh].
b) Approach II:
Fig. 2. Identifying critical components for the Birka system with cases (1) base
case, (2) bus bars, (3) breakers, (4) cables, and (5) transformers (Step 2).

IV. RESULTS FROM APPLICATION STUDIES

(13)
This section provides selected results from application
• Evaluate the total annualized costs in [cost/yr]: studies of the Birka system including failure rate modeling
for the underground cables and with the effect of PM on one
a) Approach I:
failure cause (water-treeing). For each of the results presented
in figures the corresponding step in the RCAM method is noted.
Stage 1—System reliability analysis for the Birka system
(14) The disturbance data for the Stockholm city power system
(from 220-, 110-, 33-, to 11-kV level) and the period 1982–1999
b) Approach II: was surveyed [17]. The statistics showed that the 11-kV voltage
level contributed most to the number of failures and customers
affected. A system was selected to investigate this voltage level
in more detail. This system includes the 220/110-kV Bredäng
(15) station and 33/11-kV Liljeholmen station, which are connected
to each other via two parallel 110-kV cables. From the Lil-
• Evaluate present values in [cost]:
jeholmen station (LH11) there are 32 outgoing 11-kV feeders
a) Approach I: that supply the southern part of central Stockholm and 14 300
The same value as given by (14). customers. In the model, customers are represented as one av-
b) Approach II: erage 11-kV load point. The following component types were
included: bus bars, breakers, underground cables, and trans-
formers. Furthermore, these were categorized into the different
voltage levels between 220–11 kV.
The reliability of the Birka system was analyzed using input
(16) reliability data from experience and statistics and RADPOW
[18]. Fig. 2 shows results from Step 2 in the RCAM method
The cost-effective solution is the maintenance defining the critical components. For each case, a specific com-
strategy that provides the lowest total cost when ponent failure rate is assumed to be zero, and the resulting effect
comparing the total costs for PM with different on the load point indices is evaluated. Case 1 refers to the base
sets of , and with no PM, that is CM. case with no PM. The most significant reduction occurs in Case

.. .. (11)
. .
80 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 20, NO. 1, FEBRUARY 2005

Fig. 3. Process to relate underlying failure cause to reliability. (Step 3–5).

4, when cables are considered 100% reliable. This shows that

these have the greatest impact on the failure rate and the un-
availability for the average 11-kV customer. The significant rise
in average outage time is because the repair time for the domi-
nant population of cables, that is 11 kV, is much lower than the
repair times for the other components. Therefore, the average
restoration time increases when the number of short interrup-
tions is reduced. The conclusion is that the 11-kV cables are
critical components for this system.
Stage 2—Component reliability modeling.
A comprehensive failure modes analysis was made (Step 3)
using 18 years of data and 58 interruptions that were caused by
the 11–kV underground cables. The underlying causes of fail- Fig. 4. Resulting failure rate model for a water-treed cable affected by PM
measures after 11 years (Steps 4–5, Approach II).
ures for each of these interruptions were investigated. The class
of material or method made the most significant contribution
with 59% of the total failures, including the underlying failure with a cable having the same characteristics as the current cable
causes of material faults. had when new. These assumptions were motivated by two as-
1) Approach I: pects: that the water trees grow to a maximum length (that of
The information from the failure modes analysis provides the insulation thickness) and that this provides a worst-case sce-
input data for the failure rate modeling (Step 4). nario when showing the benefit of PM. However, it should be
2) Approach II: noted that for these XLPE insulated cables, a new cable would
Data from the statistics (Step 3) were complemented with not have the same characteristics due to changes in the manu-
practical experience. From discussions with maintenance per- facturing techniques. Nevertheless, a changed characteristic can
sonnel a list of underlying causes of cable faults was defined. be included quite readily.
One of these causes was water treeing. This is a tree-like phe- In practice, PM procedures are likely to be performed several
nomenon that involves water penetration through the insula- times during the lifetime of a particular component, in which
tion, occurring primarily in the early produced (mid-1970s) case the characteristic shown in Fig. 4 would have a series of
XLPE insulation cables. Data related to this failure were col- decrements similar to that shown. The number of occasions and
lected and selected. These include disturbance statistics [19], their timing should depend on the cost of performing the PM
measurements and modeling of the cable condition [20], and actions and the cost-benefit of doing so. The RCAM approach
PM of cables [21]. One effective method for preventing fail- described in this paper allows this to be assessed objectively.
ures of water-treed cables is the rehabilitation method [21], The resulting cable failure rate model was used for the Birka
[22]. This involves injecting a silicon-based liquid between system. The characteristics of the XLPE cables in this system
the individual wires of the conductor, which stops the growth are consequently assumed to follow those of the XLPE cables
of the current water trees. The water trees, on the other hand, with insulation degradation due to water treeing. (It should be
impact on the breakdown strength of the cable, which can be stressed that this assumption enabled complete demonstration
measured with diagnostic methods. Based on the experience of the RCAM method, rather than providing a true picture of
data and the logic shown in Fig. 3, a failure rate model (Step the cables in the Birka system.) To obtain the composite failure
4) and a functional relationship between the failure rate and rate for the cable, it was assumed that the total failure causes
the effect of PM measures (Step 5) were defined [2]. were due to water trees and other causes. The resulting input
Three different maintenance activities were considered data for the component then consisted of the developed failure
for these studies: no PM activities, PM by the rehabilitation rate model for failures due to water trees, and the average failure
method and PM by replacing cables systematically before they rate for the 11-kV cable in the Birka system due to other causes
failed (the replacement method) with notations: org, si, and rp, (Step 6.).
respectively. Stage 3—System Reliability and Cost/Benefit Analysis:
Fig. 4 shows the final result for modeling the failure rate, as- 1) Approach I:
suming one PM action on each cable. The initial value for the Results from the survey of statistics provided input data for
cable failure rate is relatively small but not zero, as the figure modeling the relationship between PM and reliability using
indicates. The failure rate characteristic with no PM is the re- Approach I. Sensitivity studies were made to see the effect
sulting approximation of a function obtained from experience at the system level if each of these causes of failures were
data [2]. The data is assessed from a complete population of ca- decreased individually or in combination. The different cases
bles over a 13-year aging period. It was assumed that the failure are as follows:
rate, after this time and due to this specific failure cause, is con- 1) base case;
stant. Furthermore, it was assumed that replacement is made 2) %;
BERTLING et al.: RCAM METHOD FOR ASSESSING THE IMPACT OF MAINTENANCE IN POWER DISTRIBUTION SYSTEMS 81

Fig. 5. Effect on system reliability for different maintenance strategies using

Approach I for the Birka system (Step 9). Fig. 6. Impact of maintenance methods and PM strategies on cost of failure
for the Birka system (Step 10, Approach II).

TABLE I
RELIABILITY RESULTS APPLYING DIFFERENT MAINTENANCE METHODS

3) %;
4) %;
5) total of – %;
6) total for %.
The difference in percentages between cases 5 and 6 (25%)
relates to those causes that were reported as included in ma- Fig. 7. Impact of different maintenance methods on the total annual costs of
terial and method, but with no further detailed level of clas- applying a PM strategy for the Birka system. Results are shown for the case with
sification. Fig. 5 shows the benefit of these different cases on the interest rate d = 2% (Step 10, Approach II).
the system indices. It has been assumed for each case that the
causes of failures can be eliminated by the PM activities. Thus Fig. 6 shows one result from the economic evaluation ac-
the corresponding failures would be eliminated and the relia- cording to the RCAM method. Input data for the economic as-
bility indices influenced. The results show that PM measures sessment was provided by the utility, and from the Swedish cus-
to reduce individual causes of failures for a critical compo- tomer interruption costs included in [23]. It is seen that the cost
nent in the system can significantly improve the system reli- of failures is decreased for the Birka system, when the 11-kV
ability. The cases represent different maintenance strategies cables are affected by PM measures. Furthermore, it is seen that
for the RCAM method with Approach I (Step 7). the most significant decrease in cost of failures is achieved with
2) Approach II: the replacement method.
A system analysis is performed for the Birka system in- The final step in the RCAM analysis is to evaluate the present
cluding two strategies for applying the PM with either reha- worth values of the annualised total costs of maintenance. Fig. 7
bilitation or replacement . Both of these in- presents annual costs for the different maintenance methods
volve PM applied on three occasions (years ), using PM strategy S1. It can be seen directly from the annual
and with the following proportions of cables subject to PM costs that PM is a dominating cost. Furthermore, it is clearly
per occasion: 10% for and 30% for (Step 7). The re- more cost-effective to rehabilitate the cable than to replace
sults from the system reliability analysis, as shown in Table I it, since the greater benefit in reliability by the replacement
(Step 9), show consistently that the best reliability is achieved method is offset by the higher investment cost. Consequently,
with PM by replacement and with as much as possible of the the cost-effective solution is not to carry out PM in this case, but
component replaced, that is . if PM is carried out, rehabilitation is better than replacement.
82 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 20, NO. 1, FEBRUARY 2005

This is, however, a constructed example considering only one [10] R. Billinton, “Bibliography on the application of probability methods in
type of component and does not provide the complete result power system reliability evaluation,” IEEE Trans. Power App. Syst., vol.
PAS-91, Mar./Apr. 1972.
for the Birka system. It is also important to note that cables [11] F. S. Nowlan and H. F. Heap, Reliability-Centered Mainte-
compared with other components in a power system involve nance. Springfield, VA: National Technical Information Service,
extremely high PM costs with relatively few possible PM U.S. Dept. of Commerce, 1978.
[12] A. M. Smith, Reliability-Centered Maintenance. New York: McGraw-
actions. It is, however, of significant importance for efficient Hill, 1993.
maintenance planning to evaluate the relative values of im- [13] J. Moubray, Reliability-centred Maintenance. Oxford, U.K.: Butter-
plementing different maintenance strategies, as shown in this worth-Heinemann, 1995.
[14] A. B. Swedenergy, RCM for electrical distribution systems—A sim-
application example. plified decision model for maintenance planning part I, in RCM För
Elnät En förenklad beslutsmetod för underhållsplanering—Del 1 An-
V. CONCLUSION vändningsområden och arbetssätt, 2001.
[15] International Council on Large Electric Systems, “Life Management of
A RCAM has been presented which includes establishing a Circuit-Breakers,” Cigré Working Group 13.08, Paris, France, Report
165, 2000.
quantitative relationship between system reliability and main- [16] K. K. Kariuki and R. N. Allan, “Application of customer outage costs in
tenance effort. Results from application studies show how the system planning, design and operation,” Proc. Inst. Elect. Eng. Genera-
RCAM method can be used to compare different maintenance tion, Transmission, and Distribution., vol. 143, no. 2, Mar. 1996.
[17] L. Bertling, R. Eriksson, R. N. Allan, L. Å. Gustafsson, and M. Åhlén,
methods and PM strategies based on the total cost of mainte- “Survey of causes of failures based on statistics and practice for improve-
nance, which includes the impact of the PM measure on the ments of preventive maintenance plans,” in Proc. 14th PSCC, Sevilla,
system reliability. Furthermore, the application study shows that Spain, Jun. 2002.
[18] L. Bertling, R. Eriksson, and R. N. Allan, “Relation between preventive
the RCAM method can be performed and supported by real maintenance and reliability for a cost- effective distribution systems,” in
input data. Relating maintenance effort and reliability improve- Proc. IEEE PowerTech’01, vol. 4, Sep. 2001.
ment is, however, a complex problem, and substantial input data [19] A. B. Swedenergy, The Lifetime and Usefulness of XLPE Cables, 1990.
PEX-kablar livslängd och användbarhet, Swedish.
is required to support the method, which may need significant [20] P. Werelius, P. Thärning, R. Eriksson, B. Holmgren, and U. Gäfvert,
updates of relevant data bases. “Dielectric spectroscopy for diagnostics of water tree deteriorated XLPE
cables,” IEEE Trans. Dielect. Elect. Insulation, vol. 8, no. 1, Feb. 2001.
[21] H. Faremo, “Report: Rehabilitation of XLPE Cables with long Water-
ACKNOWLEDGMENT trees,” SINTEF, Trondheim, Norway, 1997. Energiforsyningens Forskn-
ingsinstitutt (EFI), EFI TR A 4512, Norwegian.
The authors express their gratitude to those people who made [22] J. Pilling and G. Bertini, “Incorporating cablecure injection into a
the application studies possible, particularly to staff at Fortum cost-effective reliability program,” IEEE Ind. Appl. Mag,, vol. 3333,
Distribution and Fortum Service involved in the Birka system no. 208 333, Sep./Oct. 2000.
[23] Methods to Consider Customer Interruption Costs in Power System
case study. A special thanks to Dr. J. Endrenyi for contributions Analysis. Paris, France: Cigré Task Force 38-06-01, 2001.
during final discussions. The financial support from the Compe-
tence Center in Electric Power Engineering at KTH is gratefully
acknowledged, as well as the input from the associated reference
group. Lina Bertling (S’98-M’02) received the Ph.D. degree in electric power systems
in 2002 from the Department of Electrical Engineering and the M.Sc. degree in
systems engineering in 1997, both at the Royal Institute of Technology (KTH),
REFERENCES Stockholm, Sweden.
She is currently a visiting postdoctoral student at the University of Toronto,
[1] R. Billinton and R. N. Allan, Reliability Evaluation of Power Systems, Toronto, ON, Canada, associated with Kinectrics Inc. She is also engaged at
2nd ed. New York: Plenum, 1996. KTH as Research Associate and Project Leader of the research program on asset
[2] L. Bertling, “Reliability centred maintenance for electric power distribu- management in power systems. Her research interests are in reliability evalua-
tion systems,” Ph.D. dissertation, Dept. Elect. Power Engineering, KTH, tion of power systems and development of methods for maintenance optimiza-
Stockholm, Sweden, 2002. tion.
[3] R. E. Brown, Electric Power Distribution Reliability. New York:
Marcel Dekker, 2002.
[4] R. Billinton, M. Fotuhi-Firuzabad, and L. Bertling, “Bibliography on the
application of probability methods in power system reliability evaluation
1996–1999,” IEEE Trans. Power Syst., vol. 16, no. 4, pp. 595–602, Nov. Ron Allan (F’88) is an Emiritus Professor of Electrical Energy Systems at the
2001. University of Manchester Institite of Science and Technology, Manchester, U.K.
[5] G. H. Kjølle, A. T. Holen, K. Samdal, and G. Solum, “Adequate inter- He was previously a Visiting Professor at the Royal Institute of Technology
ruption cost assessment in a quality based regulation regime,” in Proc. (KTH), Stockholm, Sweden (during the time these studies were done). His re-
IEEE PowerTech’01, vol. 3, Porto, Portugal, Sep. 2001. search interests include power system reliability and customer outage costs, on
[6] Swedish National Energy Administration, “Nätnyttomodellen,” which he has published numerous papers and books.
Distribution System Utility Effectivity Model, News Report,
http://www.stem.se/, 2002. in Swedish.
[7] OFGEM, Report on Services for Electricity Customers, Office of Gas
and Electricity Markets, published annually.
[8] C. Singh, M. Schwan, and W. H. Wellssow, “Reliability in liberalized Roland Eriksson (SM’89) received the M.Sc. and Ph.D. degrees in electrical
electric power markets—From analysis to risk management,” in Proc. engineering from the Royal Institute of Technology (KTH), Stockholm,Sweden,
14th PSCC, Sevilla, Spain, Jun. 2002. in 1969 and 1975, respectively.
[9] J. Endrenyi et al., “The present status of maintenance strategies and the Since 1988, he has been a Professor in the Department of Electrical Engi-
impact of maintenance on reliability,” IEEE Trans. Power Syst., vol. 16, neering, KTH. His research interests include condition-based maintenance and
no. 4, pp. 638–646, Nov. 2001. electrical insulation diagnostics.

View publication stats