CRITICAL RELIABILITY ANALYSIS OF SUPERHEATER TUBES OF

COAL BASED BOILER

1
GAGAN KUMAR GUPTA, 2SOMNATH CHATTOPADHYAYA
1,2
Department of Mechanical Engineering, Indian Institute of Technology (Indian School of Mines), Dhanbad, India
E-mail: 1gagan_k_gupta@rediffmail.com

Abstract - The coal based Boiler is a vital system providing steam not only to generate power but also supply process steam
to various sections like in Steel industries, Tyre industries, Textile industries and Cement industries etc. Running of this
system constitutes several types of cost with major constituent of coal and similarly with high profit business. It consists of
several pressure parts subsystems like Superheater, Water Walls and Economiser affecting the performance of system. The
one of the most critical subsystem is Superheater, when superheater fails, entire system must be shut down to effect repairs
resulting loss of profit. Generally, reliability based preventive maintenance interval (PMI) at various reliability levels of the
subsystems are estimated, but the PMI of every subsystem is brought the entire subsystem to stop and resulting loss of profit.
This present paper investigates the methodology for determination of critical reliability based on the trend test of failure,
repair data, and the suitable distribution model. Therefore, optimal decision for Superheater can be made based on reliability
analysis.

Keywords - Critical subsystems, Distribution fitting, Reliability, TBF, TTR, Superheater

I. INTRODUCTION reliability analysis, PMI and optimizing maintenance

through the use of mathematical models.
The coal based thermal power system is represented Traditionally, optimal PM intervention schedules
as a set of three subsystems: fossil fuel boiler, steam have been obtained using models, deterministic or
turbines and three-phase alternator. If the cost is probabilistic, which involves minimization of total
considered for running thermal power plant, only one costs incurred in relation to maintenance activities.
element comes into existence is ‘Fuel’ which
constitutes a major cost factor, and is used directly in
Boiler. It is interesting to note that very few studies
have been conducted on individual equipments of
coal fired Boiler. The components/subsystems of a
coal based Boiler, however well designed, will not
perform satisfactorily unless those are maintained.
Maintenance strategies for boiler subsystems have
witness a paradigm shift over the recent decades from
breakdown maintenance to more sophisticated
strategies like condition monitoring and reliability
centred preventive maintenance (PM). Reliability of
the plant operation is directly linked with the plant
safety/loss prevention. The BOILER consists of Figure 1: Heating zone in Boiler
several subsystems like I.D. Fan, F.D. Fan, P.A. Fan,
Ball Mill, R.C. Feeder (PA Fan, Ball Mill and R.C. The general objective of maintenance of the
Feeder constitute one complete subsystem for subsystems is to make use of relevant information
BOILER), Economizer, Super heater, Boiler drum, regarding failures and repairs. Therefore, this present
Water wall tubes and High pressure valves. A work proposes the heuristic methodology for critical
simplified scheme of the Boiler is shown in Fig 1. reliability analysis and the objective of this work is
One of the most critical subsystem of BOILER is threefold. The present paper demonstrates the
Superheater. Any mode of failure in Superheater heuristic methodology considering the coal based
leads to complete stoppage of BOILER. A robust BOILER of captive power plant (CPP) of BPSCL (A
maintenance plan is necessary for the Superheater as joint venture of SAIL & DVC) Bokaro, India. The
it deals with hazardous substances, often under severe overall system is classified into subsystems so that
operating conditions. Thus, plant managers and the failures can be categorized. The classification
engineers today are faced with important PM used for the reliability analysis of the steam
decisions aimed at integrated loss prevention. generating plant is presented in Table 1.It is
Quantitative approaches which link interesting to note that very few studies have been
component/subsystem deterioration to condition conducted on individual equipments of coal fired
improvement by maintenance can help determine the Boiler. A study was performed on reliability of an
effect of maintenance on reliability. A large number experimental fluidized boiler of a coal fuelled plant to
of papers have been published on the subject of determine the major contributors to plant outage in

Proceedings of WRFER International Conference, 02nd July, 2017, New Delhi, India
30
 Critical Reliability Analysis of Superheater Tubes of Coal Based Boiler

terms of equipment failures and plant management on the computational performance of their
[1]. For Product maintenance we refer to Markeset T, algorithms. Barata et al. [18] consider a rather
Kumar U. [2 & 3], Blanchard BS, Fabrycky WJ. [4]. comprehensive model to optimize maintenance
Arora and Kumar [5 ,6 & 7] carried a maintenance decisions for multi- component systems. Emphasis is
planning and management of coal crushing system in on the optimization procedure and the required
the coal – fired thermal power plant, availability computing times. System downtime for a two-
analysis of steam and power generation systems in component series system is considered for two
the thermal power plant and stochastic analysis and settings of the parameters. More advanced algorithms
maintenance planning of the ash handling system in are proposed by the following authors. Van
the thermal power plant. Horenbeek and Pintelon [19] consider a system with
There exists a large volume of literature discussing components that are economically and structurally
RAMS analysis for various types of products and inter- dependent. The entire system has to be stopped
applications under varying conditions. Coffin K [8] and a set-up cost has to be paid if preventive or
evaluated the customer support during new product corrective maintenance is per- formed on one or more
development. Bartow RE, Prochan F [9] discussed components.
the statistical theory of reliability and life testing. A rather complex policy is proposed and applied to a
Blanchard BS, Verna DF, Peterson EL. [10] single numerical example. Bouvard et al. [20] aim to
discussed the maintainability: a key to effective group maintenance actions for components of a
serviceability and maintenance. Dhillon BS. [11] system to reduce set-up costs. An advanced algorithm
analysed the engineering maintainability: how to is proposed that consists of various steps. A few
design for reliability and easy maintenance and numerical experiments are performed, but these do
Kumar, U.[12] discussed the reliability analysis of a not provide general insights. Marseguerra et al. [21]
load-haul-dump machine management. consider a continuously monitored multi-component
However, there is a scarcity of literature regarding system and use a Genetic Algorithm for determining
RAMS and pertinent risk-analyses. Warburton et al. the optimal degradation level beyond which
[13] demonstrated a methodology for predicting preventive maintenance has to be performed.
mechanical-failure characteristics. Moss [14] Clustering of maintenance activities is not taken into
described how to design to achieve least account. Wijnmalen and Hontelez [22] consider a
maintenance-expense through the use of life-cycle rather complicated system with various component
cost (LCC) analysis. Theoretical assessments of what types. The repair cost depends on the condition of the
maintenance can and cannot do, the component. The system set-up cost is reduced when
comprehensiveness of maintenance plans, and options components of various types are repaired
for equipment assessment were identified by this new simultaneously; the type set-up cost is reduced when
approach. Reliability-centred maintenance (RCM) components of the same type are repaired
provides a standard, common methodology for simultaneously. A heuristic approach based on a
assessing, ranking, and evaluating the effectiveness of decomposition of the multi-component problem into
any maintenance procedure. Nelson, Wayne [15] several single-component decision problems is
described the weibull prediction of a future number of presented. The focus is mainly on the performance of
failures. the heuristics, general insights are not provided.
The major part of the available research literature, Zhou et al. [23] develop an algorithm for
dealing with production and preventive maintenance maintenance optimization of series–parallel systems
integration, focuses on RAM investigation only. The with multi-state economically dependent components.
proposed critical reliability analysis of optimal cost Stochastic ordering theory is used to reduce the
model for overhauling time assume that preventive search space and improve the computational
maintenance interval based on reliability are known efficiency. A single system configuration is
for all critical subsystems and cost factor is to considered, with the main focus on the computational
minimized with various combinations of PM interval performance of the algorithm. Linetal. [24] proposed
collectively of all critical subsystems to ascertain the a hybrid model obtained by combining age reduction
overhauling time. Hongetal. [16] investigate the an dhazard rate increase effects. The hybrid hazard
influence of dependent stochastic degradation of rate model reduces the effective age to a certain lower
multiple components on the optimal maintenance value and tends to steepen the slope of the hazard rate
decisions. Emphasis is on the copula that are used to function. If the hazard rate function of the system is
model the dependent stochastic deterioration of the r (t) before performing an imperfect preventive
components. Songetal. [17] consider systems where maintenance task at time T, it becomes β × r (∝
each individual component may fail due to two T + t) right after the preventive maintenance task,
competing statistically dependent failure modes, and where β ≥ 1, 0 ≤∝≤ 1 and t ≥ 0. When β = 1, the
the failure processes among the components are also hybrid model reduces to an age reduction model, and
statistically dependent. Reliability analysis is it reduces to a hazard rate adjustment model when
performed for two specific series systems and one ∝ = 0. Lin et al. [25] extended this work to repairable
specific parallel system. Some studies mainly focus systems with two categories of failure modes, namely

Proceedings of WRFER International Conference, 02nd July, 2017, New Delhi, India
31
 Critical Reliability Analysis of Superheater Tubes of Coal Based Boiler

the maintainable and non-maintainable failure modes. are arranged parallelly inside the boiler. It shows how
Only the system's failure rate corresponding to the the hot gases (flue gases) move inside the boiler. The
maintainable failure mode is altered whenever an superheater tubes are of varying thickness and
imperfect preventive maintenance task is performed. material compositions with uniform outer diameters
Rather than analysing complex policies that consists of Ø 38 mm.
of several cost functions and rely on detailed
continuous preventive maintenance information, we III. METHODOLOGY & DATA COLLECTION
consider simple and direct approach using cost
involved in collective preventive maintenance
activities referred as overhauling at particular interval
of time with cost involved in individual repair of
subsystems.

II. SUPERHEATER

The superheater is a heat exchanger that overheats

(superheats) the saturated steam. By superheating
saturated steam, the temperature of the steam is
increased beyond the temperature of the saturated
steam, and thus the efficiency of the energy
production process can be raised.

The benefits of using superheated steam are:

 Zero moisture content
 No condensate in steam pipes Figure 3. Schematic arrangement of Superheater tubes inside
boiler
 Higher energy production efficiency

Figure 2: Heat Load Graph

The heat transfer in the furnace results in a phase

change of the working fluid (water to steam). The
small water/steam temperature rise is due to the fact
that the water enters the furnace slightly sub-cooled
(not saturated). These temperatures are only
examples. They can be at various levels at different
types of boiler, but the heat load graph look
practically the same. The heat load graph is shown in
figure 2. The superheater is located in the radiation &
convection zone of a boiler where heat is transferred
from flue gas to dry steam, flowing into superheater
tubes. The temperature of flue gas in this zone is
around 700 °C and the temperature of steam coming Figure 4: Methodology for Reliability Analysis
out of the superheater tubes (final superheated steam)
is around 540 ± 5 °C. Figure 3 shows one panel of 3.1 Methodology
superheater tubes, and several numbers of such panels The reliability characteristic of a piece of equipment
can be determined, respectively, by the analysis of
Proceedings of WRFER International Conference, 02nd July, 2017, New Delhi, India
32
 Critical Reliability Analysis of Superheater Tubes of Coal Based Boiler

time to failure data and the time taken to repair it. The 3.4 Current Operational & Maintenance
methodology which can be followed for the reliability Information
analysis of the superheater, comprises: Collection of quality failure data is usually necessary
 Understanding of the system and identification in system reliability analysis for getting reliable and
and coding of subsystems and fault therein, accurate results. Data collected from the field are
 collection, sorting and classification of TBF data assumed to be the best. Field data are, however,
for each subsystem and fault, expensive and time consuming to collect. Again, data
 data analysis for verification of the identically are required to be collected over a period of time for
and independently distributed assumption, providing satisfactory representation of the true
 fitting of the TBF and TTR data for subsystems failure characterization of the mining system. Data
and faults with a theoretical probability used in recent studies have been collected over a long
distribution, period by using daily report and maintenance report.
 estimation of the reliability parameters of each Failure behaviour of critical subsystems of a steam
subsystem with a best-fit distribution, generation system has an influence on availability or
 Identification of critical subsystems and faults the failure pattern of the system as a whole. Times
and formulation of a better maintenance policy to between failure (TBF) and Time to Repair (TTR) data
improve reliability. of subsystems are arranged in a chronological order
for using statistical analysis to determine the trend of
The detailed heuristic methodology is shown in figure failure.
4. The next step after collection, sorting and
classification of the data should be validation of the
3.2 Data Collection independent and identically distributed (iid) nature of
The centre of collecting the data has been considered the TBF and TTR data of each subsystem of steam
‘Bokaro Power Supply Co. Ltd. (A Joint Venture of generation plant. The meaning of sample
SAIL & DVC) Bokaro, India. Data collections have independence is that the data are free of trends and
been carried out over a long period of time for true that each failure is independent of the preceding or
failure/repair characterization. In the present study succeeding failure. Identically distributed data means
maintenance data is collected from the plant’s that all the data in the sample are obtained from the
maintenance logbook records over a long period, same probability distribution.
which are sorted and classified for analysis. Three Verification of the assumption that the failures/repairs
basic steps must be performed before the data can be are independent and identically distributed is critical.
analyzed to determine reliability characteristics. If the assumption that the data are independent is not
valid, than classical statistical techniques for
3.2.1 These are data collection from a computerized reliability analysis may not be appropriate. Two
equipment maintenance system (database) or common graphical methods used to validate the iid
logbook, assumption are the trend test and the serial correlative
3.2.2 Sorting of the data required for analysis and test. The trend test involves plotting the cumulative
3.2.3 Data classification in the form required for the failure number against cumulative time between
analysis (i.e. TBF, TTR, frequency, total breakdown failures. The serial correlation test is a plot of data
hours, total working hours, total maintenance hours, pairs (Xi, Xi–1) for i = 1, 2... n, where n is the total
etc). number of failures. If X is dependent or correlated,
the points should lie along a line. Trend tests and
3.3 Data Evaluation serial correlation tests were carried out on TBF data
There are many sources of data in steam generation of the sub-system.
equipment, which are of relevance to reliability
modelling of equipment. In addition to the IV. ANALYSIS OF DATA
information generated by maintenance and
production functions in the form of reports, much of Validity of the assumptions of independent and
the raw data upon which these reports are based must identical distribution (idd) of Time between Failure
also be accessible in order to achieve successful (TBF) and Time To Repair (TTR) data are verified
reliability modelling. Hence reliability modelling can using Trend Test and Serial Correlation Test before
be viewed as an integral part of a unified ‘‘analysis’’ fitting to the statistical distributions.
function, dealing with a myriad of information flows
including: 4.1 Trend Test
 Data from sensors on equipment. The Trend Test for TBF/TTR data is done graphically
by plotting the cumulative frequency of failure/repair
 Data and information from operator interfaces
against the cumulative TBF/TTR respectively as
on-board equipment.
shown in figure 5 & 7. In case of Trend Test of TBF
 Historical operational and maintenance data, concave upward curve indicates that the system
information.
is deteriorating and concave downward curve

Proceedings of WRFER International Conference, 02nd July, 2017, New Delhi, India
33
 Critical Reliability Analysis of Superheater Tubes of Coal Based Boiler

indicates the system is improving. But in case of TTR

100
data, concave upward curve means that TTR is
decreasing and concave downward curve indicates 80
the TTR is increasing. If the curve is approximately a
straight line, then the data is identically distributed 60

i th TTR
and free from trends.
40
4.2 Serial Correlation Test
20
The serial correlation test can be performed
graphically by plotting the ith TBF/TTR against (i-1)th 0
TBF/TTR for i = 1,2,3,........,n, where n is the total
number of failures as shown in figures 6 & 8. 0 50 100
(i-1)th TTR
10
Figure 8: Serial Correlation Test for TTR Data
8
CumulativeFrequency

6 4.3 Result of Trend & Serial Correlation Test

The trend tests and Serial Correlation tests are carried
4 out on TTR and TBF data sets of the Superheater. It
is observed from the serial correlation plots Fig (5)
2 that the data points are randomly scattered without
any clear pattern, which indicates that the TBF data
0 sets are independent or free from serial correlation.
0 10000 20000 30000 Therefore, the TBF data sets can be assumed to be
independent and identically distributed (iid).
Cumulative TBF
The trend plots for TTR data sets of the subsystems
exhibit approximate straight line Fig (7). Again, it is
Figure 5: Trend Test for TBF Data
observed that the serial correlation plots Fig (8) that
the data points (TTR) are randomly scattered without
8000 any clear pattern, which indicates that the TTR data
7000 sets are independent or free from serial correlation.
6000 Therefore, the TTR data sets can be assumed to be
5000 independent and identically distributed (iid).
i th TBF

4000
3000 V. GOODNESS-OF-FIT TEST & ESTIMATION
2000 OF PARAMETERS
1000
0 Kolmogorov – Smirnov test are used for goodness –
of – fit test of iid data sets using EasyFit 5.6
0 2000 4000 6000 8000 Software.
(i-1) th TBF The test statistic for Chi – Square is
( )
Figure 6: Serial Correlation Test for TBF Data x = (1)
Where k = number of class
10 oi =observed number of failures (or repairs) in the ith
9 class
8 Ei = np = expected number of failures (or repairs) in
7 the ith class
Cumulative Frequency

6 n = sample size
5 pi=probability of failure occurring in the ith class if
4 Hypothesis (H0) is true
3 = F(ai) – F(ai-1)
2 If x <x , . , then hypothesis is true
1 (2)
0 The, test statistic, x , has a chi – square distribution
with degrees of freedom equal to k-1-number of
0 200 400 600
estimated parameters. To ensure an approximate chi –
Cumulative TTR square distribution, npi should be al least 5 for all i.
Therefore the sample size should be large enough to
Figure 7: Trend Test for TTR data achieve this result.

Proceedings of WRFER International Conference, 02nd July, 2017, New Delhi, India
34
 Critical Reliability Analysis of Superheater Tubes of Coal Based Boiler

For Kolmogorov- Smirnov, the test statistic is Probability Density Function

Dn = max{D1,D2},
0.018
Where D1 = max Φ − (3)
0.016
D2 = max −Φ (4) 0.014

∑ 0.012
t=∑ and s2= (5) 0.01

f (x )
If Dn < Dcrit then accept Hypothesis. 0.008
The assumption of distribution fitting for 0.006
TBF and TTR data of Superheater are checked with
0.004
Chi – Squared and Kolmogorov – Smirnov test. The
0.002
results are shown in the form of histogram graph and
0
probability distribution graph with using Easyfit 5.6. 0 20 40 60 80 100 120
Probability Density Function x
0.52
Weibull (3.2831; 67.587)
0.48

0.44
Figure 10: Histogram & Probability Distribution for TTR of
0.4
SPH
0.36

0.32
The results of best fit distributions and their estimated
0.28 parameters for TBF & TTR are shown in Table 1 and
f (x )

0.24 Table 2, & figures 9,10 respectively.

0.2

0.16
Subsystem Best fit Test Parameters
0.12

0.08
distribution
0.04
SPH Lognormal K– s = 1.08,
0
(2P) S tmed =
800 1600 2400 3200 4000 4800 5600 6400
x 1791.72 hrs
Histogram Lognormal Table 1 Estimation of Parameters of TBF Data

Probability Density Function

Subsystem Best fit Test Parameters
3.6E-4 distribution
3.2E-4 SPH Weibull (2P) K–S β = 3.2831, θ
2.8E-4 = 67.587 hrs
Table 2 Estimation of Parameters of TTR Data
2.4E-4

2E-4
f(x )

VI. ESTIMATION OF RELIABILITY

1.6E-4 ANALYSIS
1.2E-4

8E-5 The reliability of subsystems using Lognormal (2

4E-5 parameters) distribution for TBF data is calculated
0
with the equations no. Respectively (Ebeling 2008)
0 2000 4000 6000 8000 For Lognormal (2 Parameters)
x
R(t) = 1 − ∅ ln (6)
Lognormal (1.0816; 7.4892)
Where, Φ is the probability density function, tmed is
Figure 9: Histogram & Probability Distribution for TBF of median time to failure, s is shape parameter.
SPH
Probability Density Function
Reliability curves for Superheater is shown in figure
0.36 11, which indicates that the very poor reliability.
0.32

0.28 1
Reliability

0.24

0.2
0.5
f(x )

0.16

0.12
0
0.08
0 50000 100000
0.04

0
Running Hours
36 40 44 48 52 56 60 64 68 72 76 80 84
x

Histogram Weibull Figure 11: Reliability curve of Superheater

Proceedings of WRFER International Conference, 02nd July, 2017, New Delhi, India
35
 Critical Reliability Analysis of Superheater Tubes of Coal Based Boiler

From Table (3), it is further investigated that the [7]

Arora, N. & Kumar, D., “Stochastic analysis and
maintenance planning of the ash handling system in the
Superheater is critical from reliability point of view
thermal power plant”. Microelectronics and Reliability, vol.
as their reliability approaches to zero within 15000, 37 (5), pp. 819 – 824, 1997.
15000, 25000, 35000 and 48000 hours respectively, [8] Coffin K., “Evaluating customer support during new product
before other subsystems. Therefore, suitable actions development: an explorative study”. Int J Oper Prod Manage,
vol. 21(3), pp. 21–222, 1998.
in terms of replacement and preventive maintenance [9] Bartow RE, Prochan F., “Statistical theory of reliability and
with respect to reliability should be taken for life testing”. NY (USA): Holt, Rinehart and Winston Inc.
improvement of their reliability before other 1975.
subsystems for maximum improvement in availability [10] Blanchard B.S., Verna D.F., Peterson E.L., “Maintainability:
of Boiler. a key to effective serviceability and maintenance
management”. John Wiley and sons, New York (USA), 1995.
[11] Dhillon B.S., “Engineering maintainability: how to design for
T 500 1500 2500 4000 8000 15000 25000reliability and easy maintenance”. Houston (TX, USA): Gulf
(H) Publishing, 1999.
[12] Kumar, U., “Reliability analysis of a load-haul-dump
R(t) 0.881 0.881 0.881 0.881 0.881 0.881 0.881machine”. PhD Thesis, Lulea University of Technology,
Table 3 Reliability of all subsystems of Superheater at various Lulea, 1990.
running hours [13] Barringer H.P., “Practical reliability for refinery and chemical
plants”. H. Paul Barringer & Associates. P.O. Box 39855,
CONCLUSIONS Humble, TX, 7734, 1999.
[14] Warburton D, Strutt J.E., Allsop K., “Reliability-prediction
procedures for mechanical components at design stage”. Proc
In the present work, an attempt has been made to Inst Mech Eng , vo. 212, pp. 213–24, 1998.
decide to run the Boiler at which reliability label with [15] Nelson, Wayne, "Weibull prediction of a future number of
respect to cost benefit. Superheater is very critical failures". Quality and Reliability Engineering International,
from reliability point of view and to restore the vol. 16, pp. 23-26, 2000.
[16] Hong H.P., Zhou W. Zhang S. Ye W., “Optimal condition-
maximum reliability, there is only one way to replace based maintenance decisions for systems with dependent
the Superheater. These results are robust and the stochastic degradation of components”. Reliability
present methodology may be implemented for every Engineering and System Safety, vol. 121, pp. 276–88, 2014.
steam producing organisation. Maintenance policy for [17] Song S, Coit DW, Feng Q, Peng H,. “Reliability analysis for
multi-component systems subject to multiple dependent
Superheater is little bit tricky and related to major competing failure processes”. IEEE Trans Reliab, vol. 63(1),
down time for inspection and subsequent steps for pp. 331–45, 2014.
maintenance. For the purpose of maintenance and [18] Barata J,Guedes Soares C, Marseguerra M, Zio E.
replacement of pressure parts for improvisation of “Simulation modelling of repairable multi-component
deteriorating systems for ‘on condition’ maintenance
reliability, the complete BOILER is brought to stop optimisation”. Reliability Engineering and System Safety,
which results in huge production loss. Hence, with vol. 76(3), pp. 255–64, 2002.
respect to improved reliability, a suitable replacement [19] van Horenbeek A, Pintelon L., “A dynamic predictive
time and inspection time should be planned so that maintenance policy for complex multi-component systems”.
Reliability Engineering and System Safety, vol. 120, pp. 39–
overall reliability of BOILER may be increased. 50, 2013.
Every organisation requires benefit in terms of cost; [20] Bouvard K, Artus S, Bérenguer C, Cocquempot V.,
hence a bridge between the cost benefit and reliability “Condition-based dynamic maintenance operations planning
is necessary. & grouping”. Reliability Engineering and System Safety, vol.
96 (6), pp. 601–10, 2011.
[21] Marseguerra M, Zio E, Podofillini L., “Condition-based
REFERENCES maintenance optimization by means of genetic algorithms
and monte carlo simulation”. Reliability Engineering and
[1] Woo, Thomas G., “Reliability analysis of fluidized – bed System Safety, vol. 77, pp.151–66, 2002.
boiler for a coal – fuelled power plant”. IEEE Transactions [22] Wijnmalen DJD, Hontelez JAM. “Coordinated condition-
on Reliability, vol. 29 (5), pp. 422-424, 1980. based repair strategies for components of a multi-component
[2] Markeset T, Kumar U., “Integration of RAMS information in maintenance system with discounts”. Eur J Oper Res, vol. 98
design processes – a symposium”. Applied Energy, vol. 84, (1), pp. 52–63, 1997.
pp 202–221, 2007. [23] Zhou Y, Zhang Z, Ran Lin T, Ma L., “Maintenance
[3] Markeset T, Kumar U., “RAMS and risk-analysis tools in optimisation of a multi-state series–parallel system
product design to reduce life-cost”. Annual reliability- considering economic dependence and state- dependent
centred maintainability symposium, pp. 116–22, 2001. inspection intervals”. Reliability Engineering and System
[4] Blanchard BS, Fabrycky W.J. “Systems engineering and Safety, vol. 111, pp. 248–59, 2013.
analysis”. Third. ed. Upper Saddle River, NJ: Prentice- Hall, [24] Lin D, Zuo MJ, Yam RCM., “General sequential imperfect
1998. preventive maintenance models”. Int J Reliab Qual Saf Eng,
[5] Arora, N. & Kumar, D., “Maintenance planning and vol. 7(3), pp. 253–266, 2000.
management of coal crushing system in the coal – fired [25] Lin D, Zuo MJ, Yam RCM., “Sequential imperfect
thermal power plant”. Proceedings of the ORSI 26th preventive maintenance models with two categories of
convocation. Bhubaneshwar, India,1998. failures modes”. Nav Res Logist, vol. 48(2), pp. 172–183,
[6] Arora, N. & Kumar, D., “Availability analysis of steam and 2001.
power generation systems in the thermal power plant”. [26] Charles E. Ebeling., “An introduction to reliability and
Microelectronics Reliability, vol. 37 (5), pp. 795 – 799, 1997. maintainability engineering” , 2000.



Proceedings of WRFER International Conference, 02nd July, 2017, New Delhi, India
36