ARTICLE IN PRESS

Reliability Engineering and System Safety 93 (2008) 784–792

www.elsevier.com/locate/ress

A framework for intelligent reliability centered maintenance analysis

Zhonghua Chenga,b,, Xisheng Jiab, Ping Gaoa, Su Wua, Jianzhao Wanga
a
Department of Industrial Engineering, Tsinghua University, Beijing, PR China
b
Department of Management Engineering, Mechanical Engineering College, Shijiazhuang, PR China
Received 22 March 2006; received in revised form 17 January 2007; accepted 21 March 2007
Available online 4 April 2007

Abstract

To improve the efficiency of reliability-centered maintenance (RCM) analysis, case-based reasoning (CBR), as a kind of artificial
intelligence (AI) technology, was successfully introduced into RCM analysis process, and a framework for intelligent RCM analysis
(IRCMA) was studied. The idea for IRCMA is based on the fact that the historical records of RCM analysis on similar items can be
referenced and used for the current RCM analysis of a new item. Because many common or similar items may exist in the analyzed
equipment, the repeated tasks of RCM analysis can be considerably simplified or avoided by revising the similar cases in conducting
RCM analysis. Based on the previous theory studies, an intelligent RCM analysis system (IRCMAS) prototype was developed. This
research has focused on the description of the definition, basic principles as well as a framework of IRCMA, and discussion of critical
techniques in the IRCMA. Finally, IRCMAS prototype is presented based on a case study.
r 2007 Elsevier Ltd. All rights reserved.

Keywords: Reliability-centered maintenance (RCM); Case-based reasoning (CBR); Analysis system

1. Introduction consistent understandings of RCM terms and principles;

Reliability-centered maintenance (RCM), as a procedure
to identify preventive maintenance (PM) requirements of
complex systems, has been recognized and accepted in many
industrial fields [1–6], such as steel plant, aviation, railway
network or ships maintenance. The countries applying
RCM include the United States, Britain, Japan, etc.
RCM was introduced into China in the late 1980s, and
the first RCM standard GJB1378 was published and put
into practice in 1992. Since then, RCM has been a popular
methodology widely used in China’s military to identify
PM requirements of weapon systems. Numerous RCM
programs of the in-service equipments have been devel-
oped. The major problem in the application of RCM is that
the quality of RCM program is highly dependent on the
experience and skills of the RCM analysts [16,20].
In order to ensure the proper use of RCM, two efforts
have been made in our application: (1) to strengthen the
training of RCM group to ensure that the analysts have
Corresponding author. Tel.: +86 010 62771010.
E-mail address: chengzhonghua@mail.tsinghua.edu.cn (Z. Cheng).
(2) to develop a computer aided RCM system (CARCMS)
to ensure the consistency of the RCM procedures.
Although the traditional CARCMS is a preferred tool
for RCM analysts, it is not an intelligent system and
cannot provide the similar cases for analysts. RCM
practitioners were surveyed and responded noting need to
access RCM cases of similar equipments or similar items
for their reference in conducting RCM practice. A good
RCM case reflects not only reliability data of the analyzed
equipment, but also the principles and right understanding
of RCM concepts.
Based on the above considerations, case histories of
approximately 70 RCM cases conducted in the past 10
years were compiled and analyzed. These RCM cases
covered a wide range of ground weapon systems.
RCM case includes all historical records in the RCM
process, such as the list of functionally significant items
(FSIA), the failure mode and effect analysis (FMEA)
information and the RCM decision information. Apart
from these, the basic reliability prediction information in
current industry standard is also collected. At the same
time, case-based reasoning (CBR) technology is explored

0951-8320/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ress.2007.03.037
 ARTICLE IN PRESS
Z. Cheng et al. / Reliability Engineering and System Safety 93 (2008) 784–792
and successfully introduced into RCM analysis process. A
framework for intelligent RCM analysis (IRCMA) is
studied, in which RCM cases and the basic reliability data
of electrical and mechanical items are integrated. The
(IRCMA) approach was implemented through an expert
system.
2. RCM analysis process
RCM analysis process is as follows [8,10]:
(1) Identification of the functionally significant items
(FSI). The FSIs are the items whose failure modes
have significant economic, safety or mission effects on
the system. The items may be system, subsystems,
components or parts. The meaning of identification of
the FSIs is to identify the functionally significant items
in system, subsystem, component or part levels.
(2) Failure mode and effect analysis (FMEA): The goal of
FMEA is to identify functions as well as functional
failure modes of the bottom FSI, their failure causes
and effects.
(3) RCM logic decision. Each failure cause of FSI is
decided by RCM logic decision chart. The specific PM
policy and its interval are ascertained according to
reliability data.
(4) Combining the PM policy and developing the pre-
ventive program. The optimization of the interval of
single policy cannot ensure the optimal result of general
policies. In order to ensure the efficiency, maneuver-
ability and practicability of PM, we need to combine
the different policies into suitable groups. Generally,
the combination of maintenance policies and develop-
ment of preventive program should be based on
existing maintenance systems and regular intervals
(date, week, month, season, year), see Fig. 1.
From the above RCM process, we may find that
complex equipments are generally made up of many FSIs.
If we conduct RCM analysis on them, we must input FSIs,
carry out FMEA, conduct RCM logic decision and
combine the PM policy. The characteristics of RCM
process are plenty of data, information, and many tables.
Conducting RCM analysis on complex equipments needs
many analysts at the same time. In order to conduct RCM
FSI1
Identification FMEA
FSI2
of FSI
.......
FSIn
PM program
Fig. 1. RCM analysis process.
785
analysis, plenty of information needs to be input and much
time is consumed, especially in the stages of identification
of FSIs and FMEA.
There are generally similar structures in the same
category equipment, and some common parts and mechan-
ism in the different categories equipment. Namely, there
are similar FSIs in the different equipment. By the
adaptation of RCM analysis information of these FSIs,
the repeated analysis tasks can be reduced and time is
saved. RCM analysis efficiency can be improved without
compromising the quality of the analysis [11–15].
3. Definition and basic principles of IRCMA
By introducing CBR into RCM analysis process, IRCMA
transforms traditional RCM analysis, which is completely
conducted by expert for specific equipment, into the retrieval
process of the RCM analysis result of similar equipments by
computer, and the adaptation or check process of the RCM
analysis result of the similar equipments by expert. So the
repeated analysis work is avoided or reduced, and the
efficiency of RCM analysis can be improved.
(1) The meaning of IRCMA is to conduct RCM analysis
on the specific equipments based on the existing cases.
(2) The main task of IRCMA is to identify the PM
program for the specific equipment.
(3) The goal of IRCMA is to improve RCM analysis
efficiency and to reduce the costs as well as time of
identifying the PM program.
(4) The key to realize IRCMA is computer technology.
The basic principles of IRCMA are as follows:
(1) If the structure features, function and application
environment of equipments are the same or similar,
then their failure modes and measures to prevent failure
are usually close or similar.
(2) When the analysts conduct RCM analysis on the
specific equipments, they often unknowingly refer to
historical records of RCM analysis on similar items,
and achieve the satisfactory outcomes by the adapta-
tion of the similar cases.
(3) There are usually similar structures in the same
equipment category, and some common parts and
1A1 PM1
RCM Combining
1A2 PM2
decision PM policy
....... ....
5B3 PMn
Note£
FSI1,FSI2…FSIn denoting the code of FSI
1A1,1A2…5B3 denoting the code offailure mode
PM1,PM2…PMn denoting the code of PM tasks
 ARTICLE IN PRESS
786 Z. Cheng et al. / Reliability Engineering and System Safety 93 (2008) 784–792
Case Base
user Problem Case
Description Retrieval
Fig. 2. CBR process.
mechanism in the different categories equipment. By
the adaptation of RCM analysis cases of these items,
the repeated analysis tasks are reduced and RCM
analysis efficiency can be improved.
(4) CBR may aid to complete the intelligent retrieval,
adaptation and learning of the similar RCM analysis
cases.
4. Technology foundations of IRCMA–CBR
CBR is a methodology of problem solving in artificial
intelligence (AI), whose fundamental idea is that a solution
is obtained through reference to a catalogue of previous
problem solving experience.
The general CBR process includes case retrieval,
adaptation, evaluation and learning [9]. The following
chart may describe CBR cycle (see Fig. 2):
(1) Description of the problem to be solved.
(2) Retrieval of the best matching case(s) from the
collection of past cases held in the case base.
(3) Adaptation of the selected case to match the new
problem situation, as the selected case will not be, in
general, directly applied to the new situation.
(4) Evaluation of a proposed solution fulfilling the
requirements.
(5) Learning from both successful and unsuccessful pro-
posed solutions: if the proposed solution is successful
then it is added to the case base; if it fails then an
attempt is made to explain why it failed.
5. A framework for IRCMA
The framework is made up of function modules to
realize the IRCMA.
The basic input of the framework is the description of
equipment requiring RCM analysis. The output is the PM
requirements of the equipment. IRCMA is driven by a
reasoning process based on CBR.
A framework for IRCMA is shown in Fig. 3, which
depicts the concept of the IRCMA.
From Fig. 3, we can find that according to user’s
description of RCM analysis requirement, IRCMA can
quickly and accurately conduct the RCM analysis on
equipment by automatically retrieving the similar cases and
adapting it. The adaptation of similar cases may be
completed by identification of FSI, FMEA and RCM
logic decision. If the adapted similar cases are evaluated
and proved to be successful in practice, the successful
Case
Learning
Case Case Outcome User
Adaptation Evaluation Output
RCM analysis outcomes will be added to the case base as a
new case. Strictly speaking, the critical realization techni-
ques of the IRCMA include case acquisition and repre-
sentation, formation of query case, case retrieval,
adaptation and learning. These techniques will be discussed
in detail in the following sections.
5.1. Case acquisition and representation
The scale of case base and the method of case
representation have greater influence on the efficiency of
reasoning [9]. At present we have gathered and captured
RCM analysis resources and records of about 70 kinds of
equipment, including artillery, armor, etc. Those data are
arranged, classified and integrated to form unique case
base by knowledge engineers and equipment maintenance
experts (see Fig. 4).
A whole RCM analysis case ¼ basic features informa-
tion+RCM analysis information.
Some RCM analysis cases and their contents are as
follows (see Table 1).
There are many representation methods for case, such as
first-order logic, predicates, semantic nets, frames, and
object oriented knowledge representation etc. [7]. Because
frame representation can combine basic features informa-
tion with RCM analysis of case, frame representation is
applied in this paper.
Frame is made up of frame name and slots, which
represent the features of case and include facets [18]. Frame
representation of equipment RCM analysis case is shown
in Table 2.
5.2. Formation of query case
Query case is the collection made up of basic features
information of the equipment to be analyzed. Formation of
query case is to abstract the basic features information of
the analyzed equipment (system) by user and computer,
put them into dynamic database and form the equipment
(system) query case. According to these features informa-
tion, the cases in case-base matching query case may be
retrieved quickly.
The basic features information abstraction of the
analyzed equipment is to get the name, application
environment, category and main structures of equipment.
5.3. Case retrieval
Case retrieval is the process of finding the case(s)
that are closest to the current query case in the
 ARTICLE IN PRESS
Z. Cheng et al. / Reliability Engineering and System Safety 93 (2008) 784–792 787

Problems Inputting information of Storing Data

input equipment to beanalyzed Dynamic database

Formation of
equipment query case

Equipment maintenance expert and knowledge engineer

Case
Providing Case acquisition
Retrieval of the similar
bases and
Interface

equipment RCManalysis
equipment case representation
mechanism

Identification ofFSI
Intelligent

Adaptation of the similar FMEA

Reason

equipment RCManalysis
User

RCM logic decision

Evaluation andLearning Retaining

of RCM analysis case
equipment case

Outcomes Acquiring satisfactory Storing RCM RCM Analysis

output RCM analysisoutcomes outcome Bases
Outcomes

Fig. 3. A framework for IRCMA.

RCM analysis case bases Table 1

Categories and contents of case

Case category Basic features RCM analysis

information information
Airplane Armor Artillery Naval ship Others
Equipment RCM Name, application Functionally
analysis case environment, category significant items (FSI)
and main structures of structure trees of
equipment equipment
artillery RCM analysis case
122 MM howitzer type xxx

At present, there are three kinds of retrieval policies of

122 MM multile rocket
type xxx analysis case

xxx type antiaircraft

RCM analysiscase

RCM analysiscase

case in AI: nearest neighbor (NN), induction and knowl-

xxx type cannon

edge-guided induction. Here, knowledge-guided induction

©© and induction are used [18,19].

5.3.1. Case retrieval process

The procedure of case retrieval is as follows (see Fig. 5):

(1) Formation of query case E, E ¼ (the name, application

Fig. 4. Structures of RCM analysis case bases. environment, category, and main structures of equip-
ment).
equipment case base. The selection of retrieval policy has (2) Identifying category (E, CL) to which query case E
significant influence on the efficiency and accuracy of RCM belongs, where CL ¼ (artillery category, armor cate-
analysis. gory,y, other categories) and E is query case.
 ARTICLE IN PRESS
788 Z. Cheng et al. / Reliability Engineering and System Safety 93 (2008) 784–792

(3) Retrieving the case base according to equipment
category (such as artillery category). By doing this,
the similar cases are limited within the same equipment
category (artillery category).
(4) Retrieving the similar case in the same equipment
category (artillery category) according to case name. So,
quantities of the similar case name are reduced further.
(5) Nearest neighbor(NN) method. The NN matching
function is a non-parametric classification algorithm
based on assumptions of the independence of attributes
in previous cases and the availability of rules and
procedures for matching. The NN techniques provide a
measure of how similar a previous case is with a given
problem [7].
The overall similarity determined by a weighed NN
matching function is mathematically represented as
follows [9]:
Pn
wi  simðQi ; T i Þ
SIMðQ:T:Þ ¼ i¼1 Pn , (1)
i¼1 wi
where wi is the weight of feature i, Q is the target case, T is
the source case, n is the number of feature in each case and
i is an independent feature from 1 to n.
Table 2
Frame representation of equipment RCM case
Setting the weights in the similarity function
appropriately can improve the performance of the NN
algorithms. Intuitively, more important feature should be
assigned larger weights than less important features
while totally irrelevant features should be assigned zero
weight.
By calculating the feature similarity and weight sums
of matched features of main structures of the similar
cases and the new query case according to similar
algorithm, the RCM analysis cases with maximal similarity
may be found.
5.3.2. Practice case example of case retrieval
Take 122 mm howitzer-type xxx equipment RCM
analysis case as an example, we discuss similarity retrieval
process of cases based on NN matching function.
Firstly structures and their pattern of equipment are
presented and query case is formed. Based on the
experience of equipment expert, weight of main structure
name is given 0.3 and weight of their pattern is given 0.7
[17]. Structure and their weight of 122 mm howitzer-type
xxx (query case) are shown in Table 3.
Table 3
Query case of 122 mm howitzer-type xxx
Main structure Pattern

Case number: Name Weight Name Weight

Frame Name: the name of equipment RCM analysis case Barrel 0.3 Monoblock 0.7
Slot1: the description about equipment category construction
facet1: equipment category (artillery category, armor Barrel lock Wedge
category, y, other categories) Counter-recoil Combination
Slot2: the description of used environment of equipment mechanism
facet1: application environment of equipment (island, Elevating gear Tooth sector type
desert y) Deflection mechanism Worm-and-wheel
Slot3: the description about main structures of equipment Wheel buffer unit Torsion bar type
facet1: main structure 1(feature 1, feature value1, weight 1; Wheel brake mechanism Both sides braking
feature 2, feature value2, weight 2y) Braking control device Gas operated
facet2: main structure 2 (feature 1, feature value1, weight 1; Aiming mechanism Independent line of
feature 2, feature value2, weight 2 y) sight type
y Balancing machine Barometric
facet n: main structure n (feature 1, feature value1, weight 1; False cradle Trough
feature 2, feature value2, weight 2 y) Top carriage Toe
Slot4: collection of RCM analysis information Bottom carriage Box type
facet1: FSI structure trees of equipment Carriage side Detached box type

Formation of Identifying the Retrieving

equipment query category of query according to
case case equipment category

Outputting the Retrieving

Nearest
similar RCM according tocase
neighbor
analysis case name

Fig. 5. Case retrieval procedure.

 ARTICLE IN PRESS
Z. Cheng et al. / Reliability Engineering and System Safety 93 (2008) 784–792 789

According to 122 mm howitzer-type xxx query case,
equipment RCM analysis cases retrieved is shown in
Table 4.
Similarity of 122 mm howitzer-type xxx query case
and 122 mm howitzer year xxx is firstly calculated.
According to feature values in table, calculation process
is as follows:
 Match-Value(Feature1) ¼ Match-Value(barrel/bar-
rel)+Match-Value(monoblock construction/mono-
block construction) ¼ 0.3+0.7 ¼ 1
 Match-Value(Feature2) ¼ Match-Value(barrel lock/
barrel lock)+Match-Value (wedge/
whorl) ¼ 0.3+0 ¼ 0.3
 Match-Value(Feature3) ¼ Match-Value(counter recoil
mechanism/counter recoil mechanism)+Match-Value
(combination/combination) ¼ 0.3+0.7 ¼ 1
 Match-Value(Feature4) ¼ Match-Value(elevating gear/
elevating gear)+Match-Value (tooth sector-type/tooth
sector-type) ¼ 0.3+0.7 ¼ 1
 Match-Value(Feature5) ¼ Match-Value(deflection me-
chanism/deflection mechanism)+Match-Value (worm-
and-wheel/screw) ¼ 0.3+0 ¼ 0.3
 Match-Value(Feature6) ¼ Match-Value(wheel buffer
unit/wheel buffer unit)+Match-Value(torsion bar
type/fold reed) ¼ 0.3+0 ¼ 0.3
 Match-Value(Feature7) ¼ Match-Value(wheel brake
mechanism/wheel brake mechanism)+Match-Value
(both sides braking/both sides braking) ¼ 0.3+0.7 ¼ 1
 Match-Value(Feature8) ¼ Match-Value(braking con-
trol device/braking control device)+Match-Value(gas
operated/hand operated) ¼ 0.3+0 ¼ 0.3
 Match-Value(Feature9) ¼ Match-Value (aiming me-
chanism /aiming mechanism ) +Match-Value(indepen-
dent line of sight type/half dependent line of sight
type) ¼ 0.3+0 ¼ 0.3
 Match-Value(Feature10) ¼ Match-Value(balancing ma-
chine/balancing machine)+Match-Value(barometric/
spring) ¼ 0.3+0 ¼ 0.3
 Match-Value(Feature11) ¼ Match-Value(false cradle/false
cradle)+Match-Value(trough/trough) ¼ 0.3+0.7 ¼ 1
 Match-Value(Feature12) ¼ Match-Value(top carriage/
top carriage)+Match-Value (toe/toe) ¼ 0.3+0.7 ¼ 1
 Match-Value(Feature13) ¼ Match-Value(bottom car-
riage/bottom carriage)+Match-Value(box-type/de-
tached box-type) ¼ 0.3+0.7 ¼ 1
 Match-Value(Feature14) ¼ Match-Value(carriage side/
carriage side)+Match-Value(detached box-type/detached
box-type) ¼ 1.
According to formula (1), similarity of 122 mm howitzer-
type xxx query case and 122 mm howitzer year xxx RCM
case is
P4
Match  Valueðfeature iÞ
SIM ¼ i¼1 P14  100%
i¼1 WeightðiÞ
9:8
¼  100% ¼ 70%.
14
In the same way, similarity of 122 mm howitzer-type xxx
query case and 152 mm howitzer year xxx RCM case can
be calculated, which is 85%. So, 152 mm cannon howitzer
year xxx RCM case is more similar than 122 mm howitzer
year xxx, from which RCM analysis outcomes may be
acquired.
In intelligent RCM analysis system (IRCMAS) proto-
type, retrieval process of the similar equipment case based
on NN method is shown in Fig. 7.

Table 4
The retrieved equipment RCM analysis cases

Name pattern 122 mm howitzer-type xxx query case 122 mm howitzer year xxx RCM 152 mm cannon howitzer year xxx
analysis case RCM analysis case

Feature 1 Barrel (name) monoblock Barrel monoblock construction Barrel monoblock construction
Feature 2 Barrel lock wedge Barrel lock whorl Barrel lock wedge
Feature 3 Counter-recoil mechanism combination Counter-recoil mechanism Counter-recoil mechanism
combination combination
Feature 4 Elevating gear tooth sector type Elevating gear tooth sector type Elevating gear tooth sector type
Feature 5 Deflection mechanism worm-and-wheel Deflection mechanism screw Deflection mechanism screw
Feature 6 Wheel buffer unit torsion bar type Wheel buffer unit fold reed Wheel buffer unit torsion bar type
Feature 7 Wheel brake mechanism both sides braking Wheel brake mechanism both sides Wheel brake mechanism single side
braking braking
Feature 8 Braking control device gas operated Braking control device hand operated Braking control device gas operated
Feature 9 Aiming mechanism independent line of sight Aiming mechanism half dependent Aiming mechanism independent line
type line of sight type of sight type
Feature 10 Balancing machine barometric Balancing machine spring Balancing machine barometric
Feature 11 False cradle trough False cradle trough False cradle canister
Feature 12 Top carriage toe Top carriage to e Top carriage toe
Feature 13 Bottom carriage box-type Bottom carriage box-type Bottom carriage box-type
Feature 14 Carriage side detached box-type Carriage side detached box-type Carriage side detached box-type
 ARTICLE IN PRESS
790 Z. Cheng et al. / Reliability Engineering and System Safety 93 (2008) 784–792
5.4. Adaptation and learning of case
Case adaptation is a process of transforming a retrieved
RCM analysis solution into the solution appropriate for
the current problem, which is completed by rule-based
reasoning and rules in rule base. For those cases that are
found to be similar, after case adaptation, information
from those ‘‘generic’’ cases is applied in the specific
situation being analyzed.
There are two methods of case adaptation, namely
automatic adaptation and manual adaptation. Significa-
tion of automatic adaptation is that the retrieved similar
cases are automatically adapted and replaced by computer
according to user question. While signification of manual
adaptation is that the retrieved similar cases are adapted
according to the experience of analysts.
If one generic case might be very similar with situation
being analyzed, but it might be different that calls for
different strategies and intervals, then those maintenance
strategies and intervals need to be recomputed and
readapted by automatic adaptation or manual adaptation.
In this paper, manual adaptation is mainly applied.
The technique used for case adaptation is rule-based
reasoning, which solves the problem by rules in rule base.
Rule consists of a condition part and an action part. If all
expressions within the condition part are evaluated to be true,
the action portion will be executed. The general form of the
rules is as follow: IF /ConditionS THEN /ConclusionS.
The adaptation for similar case will cost many steps of
reasoning. For example:
Initialized The loss of function caused by this failure
fact: mode on its own would not become evident
to the operating crew under normal
circumstance.
Target Adapting PM type of the similar equipment
condition: case retrieved from scheduled on-condition
task to scheduled restoration task.
The adaptation rules and their orders are: Rule1-Rule2-
Rule3.
Rule1: IF (the loss of function caused by this failure
mode on its own would not become evident
to operating crew under normal
circumstance).
THEN (failure consequences is hidden
consequences).
Rule2: IF (scheduled on-condition task to detect
whether the failure is occurring or about to
occur is technically infeasible and unworthy
of doing).
THEN (PM task selected is not scheduled
on-condition).
Rule3: IF (scheduled restoration task to reduce the
failure rate is technically feasible and worthy
of doing).
THEN (PM task selected is scheduled
restoration task).
Case learning is a process of adding the new case into
case base, which is also named as case retaining. If the
adapted case is evaluated to be true and feasible, and there
are not cases in case base whose similarity measure exceeds
the threshold provided in advance, such as 0.9, then the
case will be added into case base. Otherwise, that case is
discarded. With the accumulation of new RCM cases, the
case base will become much larger than before and the
ability to aid RCM analysis will boost up gradually.
6. A case study: a practical application of IRCMAS
Based on the above theories and techniques study, an
IRCMAS prototype has been developed and a case is
studied.
6.1. Function of the system
The main functions of IRCMA include project manage-
ment (new creation project, open project, delete project and
shut project), analysis tools (the FSIs, FMEA and RCM
logic decision) and case reasoning (case retrieval, case
adaptation, case evaluation and case learning), see Fig. 6.
6.2. Working approach in a practical application of the
system
The working approach and steps of applying IRCMAS
to conducting RCM analysis for 122 mm howitzer-type xxx
are as follows:
(1) Identification of FSIs. Identification of FSI is the first
step of RCM analysis. Complex equipments are usually
made up of many FSIs. When FSIs of the analyzed
equipment are identified, standard RCM procedure
needs input FSIs one by one, and plenty of information
needs to be input and much time is consumed. But the
IRCMAS firstly retrieves the structure tree of FSIs of
the similar equipment case from equipment case base.
The FSIs of the analyzed equipment are identified by
adaptation of the structure trees of FSI of the similar
equipment, and the actual savings are achieved, which
is shown in Figs. 7 and 8.
(2) FMEA. Based on standard RCM procedure, the
FMEA of the analyzed FSI needs to be completed by
identification of the functions, failure modes, failure
causes and failure effect of the selected FSI. Plenty of
data needs to be input. But based on the IRCMAS, the
FMEA of the analyzed FSI can be completed by the
adaptation of FMEA information of the similar
equipment case, as is shown in Fig. 8. The above steps
are repeated until all the FMEA is conducted for each
FSI of the analyzed equipment.
(3) RCM logic decision analysis: RCM logic decision
analysis ascertains the specific PM tasks and their
intervals for each failure causes of the FSI. If
RCM logic decision is conducted by standard RCM
 ARTICLE IN PRESS
Z. Cheng et al. / Reliability Engineering and System Safety 93 (2008) 784–792 791

New creation project

Open project
RCM project management

Delete project

Shut project

FSI input

Analysis tool FMEA Fig. 8. Adaptation of FMEA information of the similar equipment case.
Equipment RCM
analysts
RCM logic decision

Case retrieval

Case adaptation

Case Reasoning

Case evaluation

Case learning

Fig. 6. Functions of IRCMAS.

Fig. 9. Adaptation of RCM logic decision information of the similar

equipment case.

Fig. 7. Retrieval of the similar equipment case.

procedure, then each failure cause needs to be judged

by failure consequence, and to be identified PM tasks.
But based on the IRCMAS, the RCM logic decision of
the analyzed failure causes can be completed by the
adaptation of RCM logic decision information of the
similar equipment case, and so tasks are simplified (see
Fig. 9). The above steps are repeated until RCM logic
decision is conducted for each failure causes.
(4) Combining PM tasks to form a preventive program:
When RCM analysis case of the equipment, which has Fig. 10. Formation of preventive maintenance program.
 ARTICLE IN PRESS
792 Z. Cheng et al. / Reliability Engineering and System Safety 93 (2008) 784–792
Table 5
A segment of prevent maintenance program of 122 mm howitzer-type xxx
Number Item Type of Interval
maintenance
task
1 Counter-recoil Operator Before shooting
mechanism monitoring
2 Scheduled 4 years
restoration
3 Restoration 8 years
been conducted the above three steps, is evaluated to be
true, the new case will be retained to case base by
learning mechanism, and be applied to RCM analysis
of the next equipment. At the same tine, a new RCM
analysis outcome is output in the form of PM program.
See Fig. 10 and Table 5.
7. Conclusion
RCM analysis is a systematic engineering methodology
to identify preventive maintenance (PM) requirement for
equipments in many countries. RCM analysis process is a
much repeated task, and dependent on RCM analysis
experience. The IRCMA is feasible to improve efficiency
and quality of RCM analysis. The study outcomes in this
paper will further improve the accuracy and validity of
RCM analysis, and have great significance on RCM
popularization and application on military equipment
and civilian facility. The next tasks of ours are to further
perfect the IRCMAS.
We believe that the IRCMAS is a powerful tool for the
development of RCM programs of physical assets, and has
a potential future in the RCM market. The IRCMAS is
substituting the traditional computer aided RCM system
(CARCMS) within China’s military industry, and is
becoming the new generation of RCM analysis tool for
weapon systems under development.
References
[1] Carretero J, Perez JM, Garcia-Carballeira F, et al. Applying RCM in
large scale systems: a case study with railway networks. Reliab Eng
Syst Safety 2003;3(82):257–73.
[2] Deshpande VS, Modak JP. Application of RCM for safety
considerations in a steel plant. Reliab Eng Syst Safety
2002;3(78):325–34.
Description of maintenance job Maintenance
level
Monitor the quantity of liquid in counter-recoil Operator unit
according to standard. Otherwise, Operator unit
shooting add quantity of liquid in time
Check the adjuster and valve device for liquid Repair unit
quantity, and restore their technical states
Replace counter-recoil liquid of type four Repair unit
[3] Deshpande VS, Modak JP. Application of RCM to a medium scale
industry. Reliab Eng Syst Safety 2002;1(77):31–43.
[4] Deshpande VS, Modak JP. Modeling of uncertainties in reliability
centered maintenance: probabilistic approach. Reliab Eng Syst Safety
2001;2(71):159–64.
[5] Martorell S, Sanchez A. The use of maintenance indicators to
evaluate the effects of maintenance programs on NPP performance
and safety. Reliab Eng Syst Safety 1999;2(65):85–94.
[6] Rausand M. Reliability centered maintenance. Reliab Eng Syst
Safety 1998;2(60):121–32.
[7] Shi Y, et al. Application of case-based reasoning technique on gun
decision. Comput Mech 1999;9(25):55–9.
[8] Abdul-Nour G, et al. A reliability based maintenance policy: a case
study. Comput Ind Eng 1998;35(3/4):591–4.
[9] Watson I. Case-based reasoning is a methodology, not a technology.
Knowl-Based Syst 1999;12:303–8.
[10] Moubray J. Reliability centered maintenance. London: Butterworth-
Heinnemann; 1997.
[11] Carretero J, Garcia F, Perez JM, Perez M, Cotaina N, Prete P. Study
of existing reliability centered maintenance (RCM) approaches used
in different industries. Technical report FIM/110.1/DATSI/00,
Universidad Politécnica de Madrid, Spain, 2000.
[12] NAVAIR, Guidelines for the naval aviation reliability-centered
maintenance. Direction of Commander, Naval Air Systems Com-
mand, USA, 1996.
[13] Schlkins N. Application of the reliability centered maintenance
structures methods to ships and submarines. Maintenance 1996;11(3).
[14] Hall R. Optimizing preventive maintenance using RCM. Main-
tenance 1992;7(4).
[15] Ben-Daya M. You may need RCM to enhance TPM implementation.
J Qlty of Maintenance Eng 2002;6(2).
[16] Jia X, et al. The overview on reliability-centered maintenance. J Mech
Eng Coll 2003;2(14):29–32.
[17] Gao P, Wu S, Cheng Z. The criticality model for equipment based on
monte carlo method: The 53rd annual reliability and maintainability
symposium, 2006.
[18] Nilsson NJ. Artificial intelligence—a new synthesis. China Machine
Press; 2000.
[19] Russell S, Norvig P. Artificial intelligence: a modern approach. San
Francisco, USA: Morgan Kaufmann; 1995.
[20] Cheng Z. Intelligent RCM analysis. China: PhD’s degree thesis of
Mechanical Engineering College, 2006.