1 Introduction
Quality is the key that satisfies customers and retains their loyalty. Long-term sales and
profitability are heavily influenced by high-quality products and services. Poor quality or
a product failure that leads to a recall effort generates negative publicity and damages
one’s reputation. One may be forced to discard faulty products and incur extra
manufacturing costs to repair them in some cases. If faulty goods are delivered to
customers, one may be required to pay for refunds and repairs, and might face legal
charges if one does not follow consumer or industry guidelines. Poor design, incorrect
manufacturing methods, lack of knowledge or experience, equipment complexity, poor
maintenance policies, organisational rigidity, and human errors can all contribute to a
system's failure (Balagurusamy, 1984).
The quality of any product, tool, device, or a system (for example a network, a power
plant or a supply chain system) whether automatic or mechanical; hardware or software;
or interdisciplinary, can be measured in terms of performance, reliability and robustness
involving some decision-making processes to deal with the vagueness in systems. On a
regular basis we hear the word ‘reliability’ and ‘reliable’ in advertisements, brand labels
and in casual discussions. The term "reliability" is often used to describe "how quality
changes over time." Dissimilarity among reliability and quality has been that, quality
relates to how fine a thing executes its intended purpose, whereas reliability relates to
how well the thing maintains its original quality through time and under various
conditions (Savage & Carr, 2002). An automobile that is fuel efficient, easy to operate
and safe can, for example, be considered high quality. This automobile can be considered
reliable if it meets these criteria for multiple years, performing good and maintaining its
safety when driven in adverse weather conditions.
1.1 Reliability: History and Definition
When the II-World War happened in 1939, the demands of modern technology ignited the
importance of reliability and quality control (Coppola, 1984). The army and navy of the
United States created a joint committee around 1943, to look into the failure of vacuum
tubes that were considered to be a major cause of problems during the war. Methods for
assessing quantitative reliability were devised and implemented. During the decade, Bell
Labs and Aeronautical Radio, Inc., two of the most notable businesses involved
1
themselves in the investigation of reliability concerns. The US Department of Defence
established the first major committee on reliability in 1950. This was eventually called the
‘Advisory Group on Reliability of Electronic Equipment (AGREE).’ Other countries, like
as the United Kingdom and Japan, began to show interest in applying reliability concepts
to their goods around the 1950s. The National Board for Quality and Reliability was
established in 1961 with the goal of promoting awareness about the significance of
quality and reliability in product design, manufacturing, and usage. In today’s business
world, the term “reliability” has become a buzzword. National and international reliability
seminars and conferences are held on a regular basis in most countries.
Reliability is clearly defined as “the chance/probability that a product, system, or service
will execute its intended function adequately, without failure, for a specified period of
time, or will operate in a specified environment (Balagurusamy, 1984).” The following
are the most significant parts of this definition to fully understand how reliability in a
product or service is defined:
(i) Probability: the chance of an operation’s success
(ii) Intended function: such as, lighting, heating, spinning, positioning, ordering, etc.
(iii) Specified time: hours, months, periods or strain rate
(iv) Specified conditions: for instance, pressure, speed or temperature
(v) Satisfactory or without failure: meet a set of criteria with an appropriate degree of
conformance.
Reliability can be represented mathematically as,
∞
R ( t )=P (T >t)=∫ f (x )dx (1)
0
where, f (x) denotes the probability density function and T the time period. The numerical
value of reliability is at all times between zero and one, i.e., R (0) =1, R (∞) = 0. It is a
non-increasing function of time between intervals [0, ∞).
1.2 Types of Reliability
(i) Test-Retest: Repeating the same test over time.
(ii) Interrator: A separate individual conducts the same test.
2
(iii) Parallel Forms: Multiple iterations of a test that are intended to be identical.
(iv) Internal consistency: Multi-item test where all the items are meant to calculate the
same variable.
1.3 System Configurations in Reliability
A system's complexity can range from simple (one or two components) to complex
(involving thousands of components). One method for analysing such systems is to divide
them into manageable subsystems, each representing a distinct function. The reliability of
such systems is then estimated and combined using certain probability laws to determine
the overall system's reliability. Reliability can be developed with various kinds of systems
(or sub-systems) (Kumar et al., 2019). They can be broadly classified into four types, as
explained in Table 1 below.
Table 1 Systems in Reliability (Types)
A. System in Series B. System in Parallel
A system with n components is stated as an
If the functional diagram indicates that the
n-unit parallel-system if and only if the
success of an operation in a system is
successful operation of any of the
dependent on the proper operation of all n
components results in the system's success.
components, then the system configuration
is of the series or chain type.
3
C. Mixed Configuration System D. “k-out-of-n” Systems
A “k-out-of-n” configuration is one that
A mixed formation system is a hybrid of
consists of n components/subsystems, of
series and parallel system configuration.
which only k must function for the system to
be successful.
1.4 Reliability Characteristics
Reliability characteristics are measures of a system’s conditional limits. They are used to
assess the reliability of an asset. The fundamental reliability characteristics are discussed
further below.
(i) Availability- The probability or the chance of the system operating satisfactorily at
any time is referred to as availability and it is determined by the system’s reliability
and maintainability.
(ii) Maintainability- The probability that system maintenance will keep the system in, or
restore it to, a specified state within a given time period is referred to as
maintainability.
4
(iii) MTTF/ (Mean-Time-to-Failure) - It is the average length of time that a non-
repairable asset operates before failing, and it is determined by dividing the entire
operating duration of the tested units by the total number of failures.
(iv) MTBF/ (Mean-Time-Between-Failure) - It denotes the average amount of time
between system failures. It is determined by dividing the total running duration by
the number of failures during a given time period.
(v) MTTR/ (Mean-Time-to-Repair) - It is the average time necessary to identify and
repair faulty equipment. It is calculated by dividing total maintenance time by total
maintenance activities performed within a certain time period.
(vi) Cost Analysis- The summarization of the cost of the entire system into its
components to aid in the comparison of the costs of the various components is
known as cost analysis. The cost of a system generally increases exponentially with
its reliability.
(vii) Sensitivity Analysis- A sensitivity analysis shows how different values of an
independent variable affect a certain dependent variable under a given set of
assumptions. Sensitivity studies, in other ways, explore how diverse sources of
uncertainty in a mathematical model contribute to the model's aggregate uncertainty.
1.5 Decision-Making (DM)
Decision-making is described as the process of selecting amongst alternatives in order to
attain a goal. It is a critical activity in the fields of science, engineering, and technology,
as well as in the business world (Edwards, 1954). DM can be formed in three ways;
(i) Using a mathematical-model
(ii) Acquiring expertise knowledge from specialists
(iii) Developing a smart/expert system
DM eschew data-driven binary computing in favour of launching systems from machine
learning to self-driving vehicles to reflect how people manage with life's uncertainty and
ambiguity. Because of its efficacy and relationship to human enhancement, fuzzy set
5
theory is purposefully used in the construction of expert and smart systems. As a result,
the Fuzzy Logic-based technique is used in DM and is referred to as ‘fuzzy decision-
making.’ Multi-criteria decision-making (MCDM) is a sub-discipline that explicitly
examines competing criteria DM in daily life. MCDM focuses on knowledge from a
variety of areas, including decision analysis, information systems, economics, software
engineering etc., (Majumder, 2015). MCDM can be majorly classified into two problems:
(i) Multiple/Multi-criteria evaluation problems
(ii)Multiple/Multi-criteria design problems (multiple-objective-mathematical
programming problems)
An MCDM problem may be expressed mathematically in the decision space (collection
of possible decisions) as follows:
max q=f ( x ) =f ( x1 , … , xn ) (2)
Subject to,
q ∈ Q={f ( x ) : x ∈ X , X ⊆ Rn }
Where, X is the feasible set and x is the n-dimensional decision variable.
1.6 Fuzzy Theory
The majority of conventional tools for formal reasoning, computation or modelling are
precise/crisp and predictable/deterministic. For example, in classical dual logic, a
proposition can be true or false, with no in-between. An element in set theory can only
belong to a set or not, and a solution in optimization can be viable or not. Zadeh, (1965)
suggested fuzzy sets as an expansion of the conventional idea of set. Fuzzy set theory has
indeed been a valuable tool for describing situations with imprecise or ambiguous data.
Problems are dealt with fuzzy sets by giving a degree to an object's belonging to a set. In
mathematics, fuzzy sets (also known as uncertain sets) are similar to sets whose elements
have varying degrees of membership.
6
1.7 Fuzzy Logic and Fuzzy Numbers
Lotfi Zadeh introduced Fuzzy Logic in 1965. It explains the intermediary values between
conventional analyses in relation to true and false. Fuzzy Logic's approach imitates the
way humans take decisions, which includes all in-between possibilities between the
digital values YES and NO.
Fuzzy numbers are a fuzzier subset of real numbers that are usually presented in the form
of ‘Triangular Fuzzy Numbers (TFN)’, trapezoidal fuzzy numbers, and Gaussian fuzzy
numbers (Stevic et al., 2019). According to several explanations (Dubois & Prade, 1978;
~
Kore et al., 2017) TFN is denoted as A (l, m, u) where the membership function
~
x ϵ A , μ A ( x ) : R → [ 0 ,1 ]is known by Equation 3.
{
0 ,∧x <l;
x −l
,∧l≤ x ≤ m ;
μ~A ( x )= m−l
(3)
u−x
,∧m≤ x ≤ u ;
u−m
0 ,∧x >u .
As illustrated in Figure 1, values l and u represent the bottom and upper bounds of the
~
fuzzy number A , while m represents the modal value.
Figure 1 The membership function of the ‘Triangular Fuzzy Number (TFN)’
1.8 Steps Involved in Fuzzy Decision-Making
FMCDM processing includes following steps:
7
1st: Outlining the linguistic terms/variables.
2nd: Constructing the membership-functions.
3rd: Creating knowledge or expertise based rules.
4th: Obtaining fuzzy values.
5th: Performing de-fuzzification.
1.9 Fuzzy Reliability
Fuzzy reliability theory is a new branch of reliability theory that combines reliability with
fuzzy mathematics. Failure is almost unavoidable with technology. Conventionally, the
reliability/dependability of a system/structure is described in reliability theory as, the
possibility that the system performs its specified function correctly (without fail) in fixed
time duration under pre-established conditions (Hongzhong, 1994). Fuzzy theory assumes
the uncertainty structure of failure in fuzzy reliability. That is, the success and failure of
every system are defined by their fuzzy states. As a result, the system can be handled in a
fuzzy way rather than in a precise manner.
2 Literature Review
In today's competitive world, reliability analysis is playing an important part in predicting
a component’s life (Rauzy, 2021). Adumene & Okoro, (2020) used the Markovian model
to assess the reliability performance of an offshore energy sustainable system under
extreme weather circumstances. To study the reliability under the k-out-of-n system
configuration, Devrim & Eryilmaz, (2021) considered a sustainable hybrid system
containing several modules of solar and wind turbines. A hybrid wind turbine and solar
energy system is extremely sustainable for generating electricity both during the day and
at night to offer 24 hour power. Cai et al., (1991) and Cui & Li, (2007) have put in a lot of
effort to develop models for assessing software reliability in the presence of multiple
critical errors, including redundant, non-redundant repairable, and coherent systems with
dependent components. Many researchers have examined various types of power systems
with various types of power failure (Singh & Kim, 1988; Prajapati & Mahajan, 2021).
8
Other applications in reliability engineering include failure analysis of various systems.
Ossai et al., (2016) studied the failure probability estimation using the Markov model and
Monte-Carlo techniques. Ram & Manglik, (2016) assessed a finger-print biometric
system considering its failures in terms of reliability mathematical modelling. Yingkui &
Jing, (2012) contributed in the reliability analysis and evaluation of multi-state systems
(MSS). Ursani & Corne, (2017) used the concepts of reliability engineering in
classification process involved in machine learning. The development of the fuzzy
reliability theory began in the 1990s. The edited volume by (Kacprzyk & Onisawa, 1995)
is the first specialised collection of papers in this field. Fuzzy methodology has been used
in reliability engineering and related fields since Kauffman, (1975). He proposed the
concept of ‘component failure possibility’ to substitute the idea of ‘component failure
probability.’ Fuzzy approach has lately become popular in (i) Fault diagnosis (Mustapha
et al., 2004), (ii) Software reliability (Wang & Chen, 2006), (iii) Structural reliability
(Biondini et al., 2004), (iv) Human reliability (Konstandinidou et al., 2006), (v) Quality
engineering (Liang & Weng, 2002), (vi) Safety and Risk engineering (Guimaraes & Lapa,
2007).
Fuzzy logic, as well as its capacity to cope with uncertainty, has found widespread
application in a wide range of engineering and scientific tasks. Fuzziness can be present
in many aspects of everyday life, including engineering (Blockley, 1981), medicine (Vila
& Delgado, 1983), meteorology (Cao & Chen, 1983), manufacturing (Mamdani, 1981),
and others. The key applications of fuzzy theory lie in the areas of automotive systems
(Yuan et al., 2021), consumer electronic goods (Manchala, 1998), domestic goods
(Ciabattoni et al., 2013) and environment control (Shepard, 2006). Fuzzy logic principles
have never been confined and have even expanded into the realm of decision making.
Ozkan-Ozen et al., (2020), used a fuzzy decision-making tool to evaluate synchronized
barriers for adoption of circular SCs and Industry 4.0. Mangla et al., (2018) have
implemented fuzzy decision-making techniques in various sustainable food supply chain
models. In today's age of smart technology, fuzzy logic has been proposed as an
integrated method in smart adaptive complex systems. Fuzzy interpretations are a natural
foundation for building smart systems all around the world. Since fuzzy theory is robust,
it helps in dealing with the uncertainty in engineering systems. In artificial intelligence
(AI), fuzzy logic is employed in natural language processing and other variety of
demanding AI technologies (Garibaldi, 2019). It is also widely used in control systems
9
like expert systems as a part of Industry 4.0 (I4.0) technologies (Huo et at., 2020).
Further, Kumar et al., (2017) have explored a wide range of sustainable and renewable
energy applications around MCDM issues.
3 Research Gaps
A thorough analysis of the literature reveals the following research gaps:
(i) Reliability evaluation and optimization of various complex systems.
(ii) Sustainability analysis of systems using decision-making techniques.
(iii) Assessment of the availability, MTTF, and cost of several advanced complex
systems.
(iv) Sensitivity computation of several advance industrial systems.
(v) Failure analysis of various industrial systems under fuzzy theory and decision-
making.
4 Proposed Research Framework and Methods
Considering the above - mentioned gap in the field of complex system reliability, we
propose the following research work:
(i) To investigate all possible types of systems and their configurations.
(ii) To evaluate the reliability, availability and MTTF of various complex industrial
systems.
(iii) To obtain a cost analysis of various complex industrial systems.
(iv) To conduct the sensitivity analysis for multi-state systems.
(v) To assess the reliability of complex systems using fuzzy reliability.
(vi) To perform failure analysis of complex systems using fuzzy decision-making.
(vii) To investigate the applications of various multi-state and multi-criteria decision-
making techniques for sustainable systems.
10
This research will make use of the following methodologies and techniques:
(i) Markov Model formulation for reliability and its measures.
(ii) Fuzzy Triangular Numbers for evaluating systems reliability.
(iii) The technique of supplementary variables.
(iv) Laplace transformations or matrix method for solution of the PDEs.
(v) Fuzzy and non-fuzzy decision-making tools and techniques.
References
1. Adumene, S., & Okoro, A. (2020). A Markovian reliability approach for offshore
wind energy system analysis in harsh environments. Engineering Reports, 2(3),
e12128.
2. Balagurusamy, E. (1984). Reliability engineering. Tata McGraw-Hill Education.
3. Biondini, F., Bontempi, F., & Malerba, P. G. (2004). Fuzzy reliability analysis of
concrete structures. Computers & Structures, 82(13-14), 1033-1052.
4. Blockley, D. I. (1981). Logical analysis of structural failure. Journal of the
Engineering Mechanics Division, 107(2), 355-365.
5. Cai, K.Y., Wen, C.Y. and Zhang, M.L. (1991); A Critical review on software
Reliability modelling. Reliability Engineering and System Safety, Vol. 32, pp.357-
371.
6. Cao, H., & Chen, G. (1983). Some applications of fuzzy sets to meteorological
forecasting. Fuzzy Sets and Systems, 9(1-3), 1-12.
7. Chen, L. H., & Weng, M. C. (2002). Using fuzzy approaches to evaluate quality
improvement alternative based on quality costs. International Journal of Quality
& Reliability Management.
8. Ciabattoni, L., Grisostomi, M., Ippoliti, G., & Longhi, S. (2013, November). A
fuzzy logic tool for household electrical consumption modelling. In IECON 2013-
39th Annual Conference of the IEEE Industrial Electronics Society (pp. 8022-
8027). IEEE
11
9. Coppola, A. (1984). Reliability engineering of electronic equipment a historical
perspective. IEEE Transactions on Reliability, 33(1), 29-35.
10. Cui, L. and Li, H. (2007); Analytical method for reliability and MTTF assessment
of coherent systems with dependent components, Reliability Engineering &
System Safety, Vol.92 (3), pp- 300-307.
11. Devrim, Y., & Eryilmaz, S. (2021). Reliability-based evaluation of hybrid wind–
solar energy system. Proceedings of the Institution of Mechanical Engineers, Part
O: Journal of Risk and Reliability, 235(1), 136-143.
12. Dubois, D., & Prade, H. (1978). Operations on fuzzy numbers. International
Journal of Systems Science, 9(6), 613-626.
13. Edwards, W. (1954). The theory of decision making. Psychological
bulletin, 51(4), 380.
14. Garibaldi, J. M. (2019). The need for fuzzy AI. IEEE/CAA Journal of Automatica
Sinica, 6(3), 610-622.
15. Goševa-Popstojanova, K., & Trivedi, K. S. (2001). Architecture-based approach
to reliability assessment of software systems. Performance Evaluation, 45(2-3),
179-204.
16. Guimaraes, A. C. F., & Lapa, C. M. F. (2007). Fuzzy inference to risk assessment
on nuclear engineering systems. Applied Soft Computing, 7(1), 17-28.
17. Hongzhong, H. (1994). A critical review on conventional reliability
theory. Machine Design, 11(3), 1-5.
18. Huo, J., Chan, F. T., Lee, C. K., Strandhagen, J. O., & Niu, B. (2020). Smart
control of the assembly process with a fuzzy control system in the context of
Industry 4.0. Advanced Engineering Informatics, 43, 101031.
19. Kacprzyk, J., & Onisawa, T. (1995). Reliability and safety analyses under
fuzziness. Heidelberg: Physica-Verlag.
20. Kauffman, A. (1975). Introduction to the theory of fuzzy subsets. Academic Press,
New York, 1.
21. Konstandinidou, M., Nivolianitou, Z., Kiranoudis, C., & Markatos, N. (2006). A
fuzzy modeling application of CREAM methodology for human reliability
analysis. Reliability Engineering & System Safety, 91(6), 706-716.
12
22. Kore, M. N. B., Ravi, K., & Patil, A. P. M. S. (2017). A simplified description of
fuzzy TOPSIS method for multi criteria decision making. International Research
Journal of Engineering and Technology, 4(5), 2047-2050.
23. Kumar, A., Ram, M., & Joshi, L. K. (2019, October). Fuzzy Reliability of Series-
Parallel System Using UGF and Hesitant Fuzzy Set. In 2019 International
Conference on Innovative Sustainable Computational Technologies (CISCT) (pp.
1-4). IEEE.
24. Kumar, A., Sah, B., Singh, A. R., Deng, Y., He, X., Kumar, P., & Bansal, R. C.
(2017). A review of multi criteria decision making (MCDM) towards sustainable
renewable energy development. Renewable and Sustainable Energy Reviews, 69,
596-609.
25. Lin, C. T., & Yang, S. Y. (2003). Forecast of the output value of Taiwan's opto-
electronics industry using the Grey forecasting model. Technological Forecasting
and Social Change, 70(2), 177-186.
26. Majumder, M. (2015). Multi criteria decision making. In Impact of urbanization
on water shortage in face of climatic aberrations (pp. 35-47). Springer,
Singapore.
27. Mamdani, E. H. (1981, December). A fuzzy rule-based method of controlling
dynamic processes. In 1981 20th IEEE Conference on Decision and Control
including the Symposium on Adaptive Processes (pp. 1098-1103). IEEE.
28. Manchala, D. W. (1998, May). Trust metrics, models and protocols for electronic
commerce transactions. In Proceedings. 18th International Conference on
Distributed Computing Systems (Cat. No. 98CB36183) (pp. 312-321). IEEE.
29. Mangla, S. K., Luthra, S., Rich, N., Kumar, D., Rana, N. P., & Dwivedi, Y. K.
(2018). Enablers to implement sustainable initiatives in agri-food supply
chains. International Journal of Production Economics, 203, 379-393.
30. Mustapha, F., Sapuan, S. M., Ismail, N., & Mokhtar, A. S. (2004). A computer‐
based intelligent system for fault diagnosis of an aircraft engine. Engineering
Computations.
31. Ossai, C. I., Boswell, B., & Davies, I. J. (2016). Application of markov modelling
and monte carlo simulation technique in failure probability estimation—A
13
consideration of corrosion defects of internally corroded pipelines. Engineering
Failure Analysis, 68, 159-171.
32. Ozkan-Ozen, Y. D., Kazancoglu, Y., & Mangla, S. K. (2020). Synchronized
barriers for circular supply chains in industry 3.5/industry 4.0 transition for
sustainable resource management. Resources, Conservation and Recycling, 161,
104986.
33. Prajapati, V. K., & Mahajan, V. (2021). Reliability assessment and congestion
management of power system with energy storage system and uncertain
renewable resources. Energy, 215, 119134.
34. Ram, M., & Manglik, M. (2016). Stochastic biometric system modelling with
rework strategy. International Journal of Mathematical, Engineering and
Management Sciences, 1(1), 1-17.
35. Rauzy, A. (2021). New Challenges and Opportunities in Reliability Engineering
of Complex Technical Systems.
36. Savage, G. J., & Carr, S. M. (2002). Interrelating quality and reliability in
engineering systems. Quality Engineering, 14(1), 137-152.
37. Shepard, R. B. (2006). Quantifying environmental impact assessments using fuzzy
logic. Springer Science & Business Media.
38. Singh, C., & Kim, Y. (1988). An efficient technique for reliability analysis of
power systems including time dependent sources. IEEE Transactions on Power
Systems, 3(3), 1090-1096.
39. Stević, Ž., Vasiljević, M., Puška, A., Tanackov, I., Junevičius, R., & Vesković, S.
(2019). Evaluation of suppliers under uncertainty: a multiphase approach based on
fuzzy AHP and fuzzy EDAS. Transport, 34(1), 52-66.
40. Ursani, Z., & Corne, D. W. (2017, November). Use of reliability engineering
concepts in machine learning for classification. In 2017 IEEE 4th International
Conference on Soft Computing & Machine Intelligence (ISCMI) (pp. 30-34).
IEEE.
41. Van Laarhoven, P. J., & Pedrycz, W. (1983). A fuzzy extension of Saaty's priority
theory. Fuzzy sets and Systems, 11(1-3), 229-241.
42. Vila, M. A., & Delgado, M. (1983). Problems of classification in a fuzzy
environment. Fuzzy Sets and Systems, 9(1-3), 229-235.
14
43. Yuan, H., Dai, H., Wu, W., Xie, J., Shen, J., & Wei, X. (2021). A fuzzy logic PI
control with feedforward compensation for hydrogen pressure in vehicular fuel
cell system. International Journal of Hydrogen Energy, 46(7), 5714-5728.
44. Yingkui, G., & Jing, L. (2012). Multi-state system reliability: A new and
systematic review. Procedia Engineering, 29, 531-536.
45. Zadeh, L. A. (1965). Electrical engineering at the crossroads. IEEE Transactions
on Education, 8(2), 30-33.
15
