An Extension To Fuzzy Fault Tree Analysis (FFTA) Application in Petrochemical Process Industry

Process Safety and Environmental Protection 9 3 ( 2 0 1 5 ) 7588
Contents lists available at ScienceDirect
Process Safety and Environmental Protection

journal homepage: www.elsevier.com/locate/psep
An extension to Fuzzy Fault Tree Analysis (FFTA)

application in petrochemical process industry
Seyed Miri Lavasani a, , Anousheh Zendegani a , Metin Celik b
a
b
Sciences & Research Branch, Tehran Science and Research Branch University, Hesarak, Tehran, Iran
Department of Marine Engineering, Istanbul Technical University, Tuzla, 34940 Istanbul, Turkey
a b s t r a c t
Fault Tree Analysis (FTA) is an established technique in risk management associated with identied hazards specic
to focused elds. It is a comprehensive, structured and logical analysis method aimed at identifying and assessing
hazards of complex systems. To conduct a quantitative FTA, it is essential to have sufcient data. By considering the
fact that sufcient data is not always available, the FTA method can be adopted into the problems under fuzzy environment, so called as Fuzzy Fault Tree Analysis (FFTA). This research extends FFTA methodology to petrochemical
process industry in which re, explosion and toxic gas releases are recognized as potential hazards. Specically, the
case study focuses on Deethanizer failure in petrochemical plant operations to demonstrate the proposed methodology. Consequently, the study has provided theoretical and practical values to challenge with operational data shortage
in risk assessment.
2014 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Keywords: Risk assessment; Fault tree analysis; Fuzzy sets; Petrochemical industry; Safety management; Operations
modelling
1.
Introduction
In conventional FTA, the Failure Probabilities (FP) of system

components (i.e. Basic Events (BEs)) are treated as exact values. However, for many systems, it is very difcult to estimate
the precise failure rate or probabilities of individual components or failure events in the quantitative analysis of fault
tree structures. In other word, the crisp approach has difculty in conveying imprecision or vagueness nature in system
modeling (Liang and Wang, 1993; Lavasani et al., 2012).
To remedy the gap about the mentioned inadequacy of the
conventional FTA, extensive research has been performed by
using fuzzy set theory. The pioneering work on this belongs
to Tanaka et al. (1983), which treated probabilities of BEs as
trapezoidal fuzzy numbers, and applied the fuzzy extension
principle to determine the probability of Top Event (TE). Lin
and Wang (1997) developed a hybrid method which can simultaneously deal with probability and possibility measures in a
FTA. Sawer and Rao (1994) applied -cuts to determine the FP
of the TE in mechanical systems modeling with Fuzzy Fault
Trees (FFTs). Cai et al. (1991) and Huang et al. (2004) adopted
possibility theory to analyze FFTs. Dong and Yu (2005) applied
the hybrid method to analyze FP of oil and gas transmission
pipeline. As another approach, Shu et al. (2006) used intuitionist fuzzy methods to analyze FT on a printed circuit board
assembly.
Furthermore, Ping et al. (2007) used FFTA for assessing failure of bridge construction. Toward marine accident analysis
and prevention, Celik et al. (2010) proposed an investigation
model based on FTA supported with fuzzy sets. Wang et al.
(2013) employed FFTA for re and explosion of crude oil tanks.
Recently, Liu et al. (2014) used FTA in emergency response
planning.
The main aim of this research is to extend FFTA methodology to petrochemical process industry. This section introduces
the existing applications of FFTA throughout the various
industries. The steps of research methodology including identifying BEs, obtaining FP of BEs with known failure rate, rating
state, aggregating stage, defuzzication process, transforming
Crisp Failure Possibility (CFP) of BEs into FP, calculating all
Corresponding author. Tel.: +98 912 3585034.

E-mail address: mohammadrezamirilavasani@rocketmail.com (S.M. Lavasani).
Available online 14 May 2014

http://dx.doi.org/10.1016/j.psep.2014.05.001
0957-5820/ 2014 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
76
Fig. 1 Structure of proposed methodology.

Minimal Cut sets (MCs) and FP of TE, and ranking of MCs are
provided in Section 2. In Section 3, a case study on an important section of petrochemical plant is demonstrated. The last
section emphasis the highlights of the research.
2.
Research methodology
In circumstances where the lack or incompleteness of data

exists, there is a need to incorporate expert judgment into
the FTA study. A framework proposed based on the fuzzy set
theory with the FTA method is capable of quantifying the
judgment from experts who express opinions qualitatively.
The new proposed framework is developed in eight different
stages. In the rst stage, BEs with known failure rates is
separated from those BEs with a vague failure rate. The
second stage is to obtain the FPs of BEs with known failure
rates. In the third stage, expert judgments are assigned
to the vague BEs. These ratings are fuzzy numbers. The
fourth stage is an aggregation procedure. It is performed by
aggregating of experts opinions. A defuzzication process
will then be adopted to transform the experts judgments
(fuzzy possibility) to corresponding crisp possibility values
by employing an appropriate algorithm. The sixth stage is to
convert crisp possibilities values to the FPs. MCs are identied
and quantied in the seventh stage. In the last stage, ranking
of all MCs can consequently be produced. Fig. 1 presents the
structure of proposed methodology.
2.1.
2.2.
Obtaining Failure Probability (FP) of BEs with
known failure rate
The foundation of a good analysis is the pedigree of failure
rate or event probability data that is assigned to BEs. There are
predominantly three methods that can be used to determine
the occurrence probability of an event namely (Preyssl, 1995):
1. Statistical method.
2. Extrapolation method.
3. Expert judgment method.
The statistical method involves the treatment of direct test

of experience data and the calculation of the probabilities. The
extrapolation method involves the use of model prediction
and similar condition or using standard reliability handbook.
The expert judgment method involves direct estimation of
probabilities by specialists.
A component is tested periodically with test interval. A failure may occur at any time in the test interval, but the failure
is only detected in a test. After a test/repair, the component
is assumed to be as good as new. This is a typical situation
for many safety-critical components, like sensors and safety
valves. If an event failure of a kind which can be inspected,
the component failure probability can be obtained from Eq. (1)
(Spouge, 2000; Rausand and Hoyland, 2004).
Identifying BEs
As mentioned, the rst step of the methodology is to identify hazards with known failure rates from vague hazards.
Occurrence failure rate of some hazards are available from PDS
(2010).
P(t) =
1

2
(1)
where is the component failure rate and is the inspection

interval.
77
If a component is of a kind which cannot be inspected.

The component failure probability P, which is also called the
unreliability, is determined from Eq. (2).
P(t) = 1 et
Table 1 Weighting score of different expert.

Constitution
Classication
Score
Title
Senior academic
Junior academic
Engineer
Technician
Worker
5
4
3
2
1
Service time
30 years
2029
1019
69
5
5
4
3
2
1
Education time
PHD
Master
Bachelor
HND
School level
5
4
3
2
1
(2)
where the component failure rate and t is the relevant time

interval. Based on the Maclaren series, the above equation for
P can be obtained from Eq. (3) if t 0.1

P(t) = 1
2.3.
1+
t
n tn
2 t2
3 t3
+
+
+ +
1!
2!
3!
n!
= t
(3)
Rating state
In this stage, experts express their opinions for each BE with

respect to each subjective attribute. Expert elicitation is the
synthesis of experts opinions of a subject where there is
uncertainty due to insufcient data because of physical constraints or lack of resources. Experts elicitation is essentially
a scientic consensus methodology. Expert elicitation generally quanties uncertainty. Examples of elds that have
contributed to probability elicitation are decision analysis,
psychology, risk analysis, Bayesian statistics, mathematics
and philosophy.
Quantication of subjective probabilities is employed in a
number of circumstances (Korta et al., 1996):
Evidence is incomplete because it cannot be reasonably
obtained.
Data exists only from analogous situations (one might know
the solubility of one mineral and might use this information
to infer the solubility of another mineral).
There are conicting models or data sources.
Scaling up from experiments to target physical processes
is not direct (scaling of mean values is often much simpler
than rescaling the uncertainties).
Expert knowledge is inuenced by individual perspectives and goals (Ford and Sterman, 1998). Therefore, complete
impartiality of expert knowledge is difcult to achieve. An
important consideration is the selection of heterogeneous
expert (e.g. both scientists and workers) or homogenous group
of experts (e.g. only scientists). The effect of difference in personal experience on expert judgment is assumed to be smaller
in homogenous group compared to a heterogeneous group. A
heterogeneous group of experts can have an advantage over a
homogenous group through considering all possible opinions.
In this study, a heterogeneous group of experts is selected
for evaluating the probability of vague events. The weighting
factors of experts are determined according to Table 1.
Rating of expert judgment can be carried out by linguistic
terms which are used for soliciting expert opinions for each
BE.
2.4.
Aggregating stage
Since each expert may have a different opinion according

to his/her experiences and expertise in the relevant eld, it
is necessary to aggregate experts opinion to reach a consensus. Different types of aggregation can be used (Hsu and
Chen, 1994; Aqlan and Ali, 2014). Aqlan and Ali (2014) proposed a new method for aggregation of expert judgment by
employing triangle fuzzy numbers. As mentioned, Aqlan and
Ali (2014) model can just aggregate triangle fuzzy numbers

whilst Hsu and Chen (1994) model is able to aggregate triangle
and trapezoidal fuzzy numbers. Linguistic terms of this paper
are mixture of triangle and trapezoidal fuzzy numbers. Therefore, Hsu and Chen (1994) method of aggregation is used in
this paper.
Hsu and Chen (1994) presented an algorithm to aggregate
the linguistic opinions of a homogenous/heterogeneous group
of experts. Suppose each expert, Ek (k = 1, 2, . . ., M) expresses
his/her opinion on a particular attribute against a specic context by a predened set of linguistic variables. The linguistic
terms can be converted into corresponding fuzzy numbers.
The detailed algorithm is described as follows:
1. Calculate the degree of agreement (degree of similarity)
Suv (R u , R v ) of the opinions between each pair of experts
Eu and Ev , where Suv (R u , R v ). According to this approach,
= (a1 , a2 , a3 , a4 ) and B = (a1 , a2 , a3 , a4 ) are two standard
A
trapezoidal fuzzy numbers. Then the degree of similarity
function of S, which is dened as:
1
|ai bi |
4
4
B)
=1
S(A,
(4)
i=1
B)
[0, 1], the larger value of S(A,
B),
the greater
where (A,
and B.
similarity between two fuzzy numbers of A

2. Calculate the Average Agreement (AA) degree AA(Eu ) of the
experts.
AA(Eu ) =
1
M1
4
S(R u , R v )
(5)
u=
/ v
v=1
3. Calculate the Relative Agreement (RA) degree, RA(Eu ) of the

experts.
Eu (u = 1, 2, . . ., M) as RA(Eu ) =
AA(Eu )
M
u=1
(6)
AA(Eu )
4. Estimate the Consensus Coefcient (CC) degree, CC(Eu ) of

expert, Eu (u = 1, 2, . . ., M):
CC(Eu ) = W(Eu ) + (1 ) RA(Eu )
(7)
78
Fig. 2 Process ow diagram of Deethanizer.

where (0 1) is a relaxation factor of the proposed
method. It shows the importance W(Eu ) over RA(Eu ). When
= 0 no importance has been given to the weight of an
expert and hence a homogenous group of experts is used.
When = 1, the consensus degree of an expert is the same
as its importance weight. The consensus degree coefcient
of each expert is good measure for evaluating the relative
worthiness of each experts opinion. It is responsibility of
the decision maker to assign an appropriate value to .
5. Finally, the aggregated result of the experts judgment R AG ,
can be obtained as follows:
R AG = CC(E1 ) R 1 + CC(E2 ) R 2 + + CC(EM ) R M
2.5.
where X* is the defuzzied output, i (x) is the aggregated

membership function and x is the output variable. The above
formula can be shown as follows for triangular and trape =
zoidal fuzzy numbers. Defuzzication of fuzzy number A
(a1 , a2 , a3 ) is
a2
X =
xa1
xdx
a1 a2 a1
a2 xa1
dx
a1 a2 a1
X =
Defuzzication process

i (x)xdx
X =
i (x)dx
(9)
aa23
a3 x
a3 a2 xdx
a3 x
dx
a2 a3 a2
a2
1
(a1 + a2 + a3 )
3
2.6.
number
(10)
=
A
a
a
xa1
x
xdx + a 3 xdx + a 4 aa4a
xdx
a1 a2 a1
4
3
2
3
a2 xa1
a3
a4 a4 x
dx + a dx + a a a dx
a1 a2 a1
4
3
2
3
2
a3
Defuzzication of trapezoidal fuzzy

(a1 , a2 , a3 , a4 ) can be obtained by Eq. (11).
(8)
Defuzzication is the process of producing a quantiable

result in fuzzy logic. Defuzzication problems emerge from
the application of fuzzy control to the industrial processes
(Zhao and Govind, 1991). The center of area defuzzication
technique is selected here. This technique was developed by
Sugeno in 1985 (Sugeno, 1999). This method can be expressed
as:
1 (a4 + a3 ) a4 a3 (a1 + a2 ) + a1 a2
3
(a4 + a3 a1 a2 )
(11)
Transforming CFP of BEs into FP
As aforementioned, there are data available for failure rates

of some events whilst the data associated with the others
are vague. There is inconsistency between FPs of certain BEs
and CFPs of vague events. This issue can be performed by
using Eq. (12). Onisawa (1988) has proposed a function which
Table 2 BEs and failure states of Deethanizer.
1.1 Power Failure
1.1.1 Power failure from source

(Mobin)
1.1.2 Trip from Sub Station
1.Trip of P-401
1.2.1 Poor Maintenance
1.1.2.1 Breaker Failure

1.1.2.2 Transformer Failure
1.1.2.3 Human error to stop the
pump
1.2.1.1 Poor PM (planning and
control)
1.2 Equipment Failure

1.2.1.3 Poor quality of equipment
parts
1.2.2 Blockage in suction of P-401
1.3 Human error to stop the pump

from DCS/Push bottom
2.1 Failure of Instrument Air (IA)

2. Failure FV 40071 A to close
2.2. Failure/Error of FIC
1.3.1 Lack of knowledge (Training)

1.3.2 Lack of skill (experience)
1.3.3 Lack of
perception/carelessness
2.1.1 Loss of IA from source
(Mobin)
2.1.2 Human error to close the IA
valves in plant/offsite battery limit
2.2.1 Equipment Failure
2.2.2 Loose connections in IA
network
2.3 Human error in DCS to close

the FV

3. Failure of E-420
1.2.2.1 icing due to dryers

deciency
1.2.2.2 Hydrate formation
1.2.2.3 chocking of strainer
1.2.1.2.1 Lack of supervision

1.2.1.2 .2 Lack of competency of
maintenance workers
1.2.1.3.1 Procurement inadequacy
1.2.1.3.2 Lack of supervision and
inspection by asset integrity dept.
1.2.1.2 Bad/wrong installation

(Mobin)
3.1 Failure of C-501 due to poor PM

3.2 Equipment failure due to
corrosion
4. Failure of TV-40075 to open
3.2.1 Internal (Inside erosion)

3.2.2 External (Humidity)
(Mobin)
79
80
5. Failure of PV-40102 to close

4.1 Failure/Error of TIC
Human error in DCS to open the FV
Failure/Error of PIC
2.3Human error in DCS to close the
FV
Failure of TV-40104
6.2 Equipment Failure
6.3 chocking due to polymerization
7.1 Failure of external body of
T-402
7.2 Failure of supports
4.1.1 Equipment Failure

6.1.1 Failure/Error of TIC

2.3 Human error in DCS to open
the FV
6.2.1 Failure of FV 400916.2.2 6.2.3
Failure of C-501
Equipment failure due to corrosion
7.1.1 Bad manufacturing (Design)
7.1.2 Poor welding
7.1.3 Vibration
7.1.4 corrosion
7.2.1 Bad manufacturing
7.2.2 Poor welding
7.2.3 Vibration
7.2.4 Lack of re proong in case of
re in adjacent area

(Mobin)
6.1.1.1 Equipment Failure
6.1.1.2 Loose connections in IA
network
6.2.3.1 Internal corrosion
6.2.3.1 External corrosion
7.1.2.1 Poor Procedure
7.1.2.2 Poor Supervision
7.1.4.1 Internal
7.1.4.2 External
Table 2 (Continued)
81
Fig. 3 Fault tree representation of Deethanizer failure.

can be used for converting CFP to FP. This function is derived
by addressing some properties such as the proportionality of
human sensation to the logarithmic value of a physical quantity. The probability rate can be obtained from possibility rate
as follows (Onisawa, 1988, 1990, 1996; Onisawa and Nishiwaki,
1988; Lin and Wang, 1998):

FP =
2.7.
1/10K ,
CFP =
/ 0
0,
CFP = 0
K=
1 CFP
1/3
CFP
2.301
(12)
Criticality importance
Risk Reduction Worth (RRW) or Top Decrease Sensitivity
(TDS)
Calculating all MCs and FP of TE
By denition, a MC is a combination (intersection) of BEs leading to the TE. The combination is a minimal combination in
that all failures are needed for the TE to occur. TE probability
can be obtained from Eq. (13).
P(t) = P(MC1 MC2 . . . MCN ) = P(MC1 ) + P(MC2 ) +
+ P(MCN ) P(MC1 MC2 ) + P(MC1 MC3 ) +
N1
+ P(MCi MCj )) + (1)
P(MC1 MC2 . . . MCN )
(13)
Where P(MCi ) is the occurrence probability of minimal cut

set i.
2.8.
or otherwise) can be apportioned to different sources of

uncertainty in its inputs. In other words, SA can be used as
method for testing robustness of a model results in presence
of uncertainty. SA of FTs is estimated by Importance measures
techniques. The following probabilistic importance measures
are currently in use for FTA:
SA can be useful for a range of purposes, including:
Testing the robustness of the results of a model or system

in the presence of uncertainty.
Increased understanding of the relationships between input
and output variables in a system or model.
Uncertainty reduction: identifying model inputs that cause
signicant uncertainty in the output.
Searching for errors in the model (by encountering unexpected relationships between inputs and outputs).
Model simplication xing model inputs that have no
effect on the output, or identifying and removing redundant
parts of the model structure.
Ranking of MCs
One of the most important outputs of an FTA is the set of

importance measures that are calculated for the TE. Such
importance measures establish the signicance for all the MCs
in the FT in terms of their contributions to the TE probability.
Sensitivity Analysis (SA) is the study of how the uncertainty
in the output of a mathematical model or system (numerical
Fussell-Vesely Importance (F-VI) is the contribution of the

MC to the TE probability. F-VIs are determinable for every MC
modeled in the FT. This provides a numerical signicance of
all FT elements and allows them to be prioritized. The F-VI is
calculated by summing all the causes (MCs) of the TE involving
the particular event. This measure has been applied to MCs to
determine the importance of individual MC. Where Qi (t) is the
82
Table 3 Deethanizer BEs.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Deethanizer failure
Fault tree ref
BE failure rate
Power failure from source (Mobin)

Breaker failure
Transformer failure
Human error to stop the pump (P-401)
Poor PM (planning and control) (P-401)
Lack of supervision (P-401)
Lack of competency of maintenance workers
Procurement inadequacy
Lack of supervision and inspection by asset integrity dept.
Icing due to dryers deciency (Suction P-401)
Hydrate formation (Suction P-401)
Chocking of strainer (Suction P-401)
Lack of knowledge (Training)
Lack of skill (experience)
Lack of perception/carelessness
Loss of IA from source (Mobin)
Human error to close the IA valves in plant/offsite battery limit
Equipment Failure (FIC of FV-40071A)
Loose connections in IA network (FIC of FV-40071A)
Human error in DCS to close the FV
Failure of C-501 due to poor PM
Internal (inside erosion) (E-420)
External (humidity) (E-420)
Equipment failure (TIC of TV-40075)
Loose connections in IA network (TIC of TV-40075)
Failure/error of PIC (PV-40102)
Equipment failure (TIC of TV-40104)
Loose connections in IA network (TIC of TV-40104)
Failure of FV-40091
Failure of C-501
Internal corrosion (E-422)
External corrosion (E-422)
Chocking due to polymerization (E-422)
Bad manufacturing (design) (external body of T-402)
Poor welding Procedure (external body of T-402)
Poor welding Supervision (external body of T-402)
Vibration (external body of T-402)
Internal corrosion (external body of T-402)
External corrosion (external body T-402)
Bad manufacturing (supports)
Poor welding (supports)
Vibration (supports)
Lack of re proong in case of re in adjacent area (supports)
1.1.1
1.1.2.1
1.1.2.2
1.1.2.3
1.2.1.1
1.2.1.2.1
1.2.1.2.2
1.2.1.3.1
1.2.1.3.2
1.2.2.1
1.2.2.2
1.2.2.3
1.3.1
1.3.2
1.3.3
2.1.1
2.1.2
2.2.1
2.2.2
2.3
3.1
3.2.1
3.2.2
4.1.1
4.1.2
5.1
6.1.1.1
6.1.1.2
6.2.1
6.2.2
6.2.3.1
6.2.3.2
6.3
7.1.1
7.1.2.1
7.1.2.2
7.1.3
7.1.4.1
7.1.4.2
7.2.1
7.2.2
7.2.3
7.2.4
Linguistic term
Failure rate
Failure rate
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Linguistic term
Failure rate
Linguistic term
Failure rate
Failure rate
Linguistic term
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Failure rate
Linguistic term
Linguistic term
Failure rate
Failure rate
Failure rate
Failure rate
Linguistic term
Failure rate
Linguistic term
contribution of MC i to failure of the system, the importance

measure can be quantied as follows (Modarres, 2006):
IiFV (t) =
Qi (t)
QS (t)
(14)
Qi (t) = probability of failure of MCi, QS (t) = probability of failure

of TE due to all MCs .Risk Reduction Worth (RRW) measures the
decrease in the probability of the TE if a given MC is assured
not to occur. This importance measure can also be called the
Top Decrease Sensitivity (TDS) (Shu et al., 2006). RRW for a MC
shows the decrease in the probability of the TE that would be
obtained if the MC did not occur. Therefore, the RRW can be

calculated by re-quantifying the FT with considering of the
probability of the given MC to 0. It thus measures the maximum reduction in the TE probability. RRW can be used to test
model robustness.
3.
Case study
A petrochemical plant is built to manufacture Medium Density

Polyethylene (MDP) and High Density Polyethylene (HDP) for
the plastics processing industry. The facility has a capacity
of 300,000 t/a based on 7920 h/y. The products are marketed
Table 4 FP of the BEs with known failure rate.

BEs
1.1.2.1
1.1.2.2
2.1.1
2.2.1
2.2.2
3.1
FP of BEs
BEs
FP of BEs
BEs
FP of BEs
BEs
FP of BEs
0.01
0.015
0.002
0.014
0.018
0.02
3.2.1
3.2.2
4.1.1
4.1.2
5.1
6.1.1.1
0.012
0.008
0.014
0.009
0.013
0.014
6.1.1.2
6.2.1
6.2.2
6.2.3.1
6.2.3.2
6.3
0.009
0.011
0.0085
0.017
0.005
0.023
7.1.1
7.1.3
7.1.4.1
7.1.4.2
7.2.1
7.2.3
0.017
0.02
0.019
0.006
0.015
0.012
83
Fig. 4 Chen and Hwang conversion scale.
Table 5 Fuzzy number of conversion scale 6.
Table 7 Expert judgment on vague BEs.
Linguistic terms
BEs
Fuzzy sets
Very low (VL)

Low (L)
Medium (M)
High (H)
Very high (VH)
(0,0,0.1,0.2)
(0.1,0.25,0.25,0.4)
(0.3,0.5,0.5,0.7)
(0.6,0.75,0.75,0.9)
(0.8,0.9,1,1)
under the brand name LUPOLEN. For easy comprehension, the

plant is divided into four main sections:
Plant Section 01: Purication (Area 700) and dosing of feedstock (Area 100)
Plant Section 02: Polymerization (Area 200) and powder handling (Area 300 & 400)
Plant Section 03: Extrusion and product handling (Area 500)
Plant section 04: Granulate handling and logistic
Additionally, plant facilities are provided in Area 600.
Section 02 is one of the most important plant sections. Deethanizer is one of the main equipment of this section. Therefore,
Deethanizer is selected as case study in this paper.
3.1.
Identifying BEs of Deethanizer
A vent system on propylene side of the E-420 and E-422 is

provided to evacuate non-condensable components contained
in propylene that could accumulate. Bottom temperature of
Deethanizer T-402 must be controlled carefully because at a
too high temperature the heavier olens in the bottom tend
to form gums and polymers, thus fouling the bottom of the
column and the reboiler. For this reason, a spare reboiler is
provided and when reboiling becomes inefcient due to fouling, operator has to switch from the operating reboiler to the
spare one. The reboiling control valve is on the steam inlet line
where it controls the condensing pressure and hence temperature. Steam is injected in order to get a maximum temperature
at Deethanizer bottom of about 82 C. A piping pot is provided
on the steam condensate line to make hydraulic guard.
1.1.1
1.1.2.3
1.2.1.1
1.2.1.2.1
1.2.1.2.2
1.2.1.3.1
1.2.1.3.2
1.2.2.1
1.2.2.2
1.2.2.3
1.3.1
1.3.2
1.3.3
2.1.2
2.3
7.1.2.1
7.1.2.2
7.2.2
7.2.4
Expert judgment on vague BEs

E1
E2
E3
L
VL
L
M
M
L
M
M
L
M
M
VL
L
M
L
L
M
L
L
M
L
L
H
M
L
M
M
L
H
H
L
VL
L
VL
L
H
L
M
M
M
M
H
M
M
H
M
L
H
M
M
L
L
L
M
M
VL
L
The overhead specication for the Deethanizer is the

propylene content in the hydrogenated C2 cut. Normal value is
0.3 wt% (0.4 wt% max.). The liquid reux ow rate is adjusted
by FV-40071 to reach this specication. A polymerization
inhibitor is injected by the Polymerization Inhibitor Injection
Package W-402 in the feed and at bottom of Deethanizer in
order to limit the fouling tendency. Based on Process Flow
Diagram (PFD) which is shown in Fig. 2.
Considering the working principles and process ow diagram of Deethanizer unit, BEs associated with this section are
identied and listed in Table 2.
Fault tree representation of the Deethanizer failure is illustrated in Fig. 3.
Considering the Deethanizer process ow diagram and
related failure states, it is the next issue to determine whether
the BEs are known or unknown.
Table 6 Experts proles and decision weights.

No of expert
1
2
3
Title
Senior
Senior
Junior
Service
time (Year)
1019
<5
69
Educational
level
Weighting factor
Master
PhD
Master
5 + 3 + 4 = 12
5 + 1 + 5 = 11
4 + 2 + 4 = 10
Weighting score
12/33 = 0.363
11/33 = 0.333
10/33 = 0.304
84
Table 8 Aggregation calculation for the BE of 1.1.1.

Expert 1 (E1)
Expert 2 (E2)
Expert 3 (E3)
S (E12)
S (E13)
S (E23)
RA (E1)
RA (E2)
RA (E3)
Weight of expert 1 (E 1)
Aggregation for 1.1.1
0.1
0.3
0.3
0.75
0.75
1
0.3
0.35
0.35
0.363
0.333
0.304
0.2337
0.25
0.5
0.5
0.25
0.5
0.5
AA (E1)
AA (E2)
AA (E3)
CC (E1)
CC (E2)
CC (E3)
0.75
0.875
0.875
0.3315
0.3415
0.327
0.417125
0.417125
0.4
0.7
0.7
0.60055
3.2.
Separating BEs with known failure rate from BEs
with unknown failure rate
FP of the BEs with known failure rates are calculated and

presented in Table 4.
The elements of the FT are divided into failure probability

analysis of BEs with known probabilities of occurrence and
subjective linguistic evaluations of hazards with unknown
probability rate. 43 BEs are identied for Deethanizer failure.
24 of them are BEs with known occurrence probabilities whilst
there are not historical data available for the other 19 BEs.
Table 3 presents all of the BEs associated with the proposed
FT.
3.4.
3.3.
Calculating FPs of BEs with known occurrence
probabilities
As aforementioned, the foundation of a good analysis is
the pedigree of failure rate or event probability data that is
assigned to BEs. Therefore, occurrence probabilities of hazard with known failure rate can be estimated by using Eqs.
(1)(3). For example, the rate of mechanical failure of homogenizer is 2.4 103 with 4 inspections in a year. Therefore, FP of
mechanical failure of homogenizer can be obtained by using
Eq. (1) as follows:
FPmechanical failure of homogenizer =
1
4
2.4 103
= 4 103
2
12
Rating state
In the proposed method, a numerical approximation proposed

by Chen and Hwang (1992) is used to convert linguistic term to
their corresponding fuzzy numbers. There are generic verbal
terms in the system where scale 1 contains two verbal terms
(linguistic terms) and scale 8 contains 13 verbal terms (linguistic terms). The typical estimate of a human memory capacity,
is seven plus-minus two chunks, which means that the suitable number for linguistic term selection for human beings
to make an appropriate judgment is between 5 and 9 (Miller,
1956; Nicokis and Tsuda, 1985). Therefore, conversion scale
of 6 which contains 5 verbal terms is selected for performing
the subjective assessment of hazards with unknown failure
rate. Fig. 4 presents the fuzzy linguistic scale that is used in
this paper to involve the judgments of experts with respect to
hazards with unknown failure rate.
The given linguistic terms are in the form of both triangular and trapezoidal fuzzy numbers. Table 5 represents all the
fuzzy numbers in the form of trapezoidal numbers.
As mentioned, a heterogeneous group of experts is
employed to perform the judgment for the vague events.
Table 9 Aggregation calculation for each subjective BE.
Table 10 Deffuzication process for all subjective BEs.
BEs
Aggregation of each
subjective BE
BEs
1.1.1
1.1.2.3
1.2.1.1
1.2.1.2.1
1.2.1.2.2
1.2.1.3.1
1.2.1.3.2
1.2.2.1
1.2.2.2
1.2.2.3
1.3.1
1.3.2
1.3.3
2.1.2
2.3
7.1.2.1
7.1.2.2
7.2.2
7.2.4
(0.23,0.42,0.42,0.6)
(0.13,0.24,0.27,0.42)
(0.16,0.33,0.33,0.49)
(0.5,0.67,0.67,0.83)
(0.3,0.5,0.5,0.7)
(0.16,0.33,0.33,0.49)
(0.39,0.58,0.58,0.76)
(0.3,0.5,0.5,0.7)
(0.1,0.25,0.25,0.4)
(0.5,0.67,0.67,0.83)
(0.39,0.58,0.58,0.76)
(0.13,0.24,0.27,0.42)
(0.07,0.17,0.2,0.34)
(0.17,0.33,0.33,0.5)
(0.07,0.17,0.2,0.34)
(0.16,0.33,0.33,0.49)
(0.39,0.58,0.58,0.76)
(0.07,0.17,0.2,0.34)
(0.16,0.33,0.33,0.49)
1.1.1
1.1.2.3
1.2.1.1
1.2.1.2.1
1.2.1.2.2
1.2.1.3.1
1.2.1.3.2
1.2.2.1
1.2.2.2
1.2.2.3
1.3.1
1.3.2
1.3.3
2.1.2
2.3
7.1.2.1
7.1.2.2
7.2.2
7.2.4
Aggregation of
subjective basic events
(0.23,0.42,0.42,0.6)
(0.13,0.24,0.27,0.42)
(0.16,0.33,0.33,0.49)
(0.5,0.67,0.67,0.83)
(0.3,0.5,0.5,0.7)
(0.16,0.33,0.33,0.49)
(0.39,0.58,0.58,0.76)
(0.3,0.5,0.5,0.7)
(0.1,0.25,0.25,0.4)
(0.5,0.67,0.67,0.83)
(0.39,0.58,0.58,0.76)
(0.13,0.24,0.27,0.42)
(0.07,0.17,0.2,0.34)
(0.17,0.33,0.33,0.5)
(0.07,0.17,0.2,0.34)
(0.16,0.33,0.33,0.49)
(0.39,0.58,0.58,0.76)
(0.07,0.17,0.2,0.34)
(0.16,0.33,0.33,0.49)
Defuzzication of
subjective BEs (CFP)
0.417
0.269
0.326
0.667
0.5
0.326
0.576
0.5
0.250
0.667
0.579
0.269
0.196
0.333
0.196
0.326
0.579
0.198
0.329
85
Table 11 Converting CFP into FP.

BEs
Defuzzication of
subjective BEs (CFP)
1.1.1
1.1.2.3
1.2.1.1
1.2.1.2.1
1.2.1.2.2
1.2.1.3.1
1.2.1.3.2
1.2.2.1
1.2.2.2
1.2.2.3
1.3.1
1.3.2
1.3.3
2.1.2
2.3
7.1.2.1
7.1.2.2
7.2.2
7.2.4
0.417
0.269
0.326
0.667
0.5
0.326
0.576
0.5
0.250
0.667
0.579
0.269
0.196
0.333
0.196
0.326
0.579
0.198
0.329
Table 13 Importance level of each MC.

FP of
subjective BEs
0.0027
0.0006
0.0012
0.0149
0.005
0.0012
0.0083
0.005
0.0005
0.0149
0.0085
0.0006
0.0002
0.0013
0.0002
0.0012
0.0085
0.0002
0.0012
The weights of experts are not equal. Experts weights can

be obtained based on their proles and competencies. In
this case, three experts are employed for performing the
judgments. Table 6 shows the experts proles and decision
weights.
Expert judgment on the BEs with unknown failure rates are
illustrated by Table 7.
3.5.
Aggregation of BEs
In this stage, all the ratings are aggregated under each subjective BE. As an example, detailed aggregation calculations for BE
of 1.1.1 are given in Table 8. (Relaxation factor) is considered
0.5 in aggregation calculation of subjective BEs.
These calculations contain attribute based aggregation calculation, such as average degree of agreement (AA), relative
degree of agreement of each expert (RA), etc. After the aggregation calculations, the results of all the BEs are presented in
Table 9.
No of MCs
FP of MCs
F-V IM
MCs1
MCs2
MCs3
MCs4
MCs5
MCs6
MCs7
MCs8
MCs9
MCs10
MCs11
MCs12
MCs13
MCs14
MCs15
MCs16
MCs17
MCs18
MCs19
MCs20
MCs21
MCs22
MCs23
MCs24
MCs25
MCs26
MCs27
MCs28
MCs29
MCs30
MCs31
MCs32
MCs33
MCs34
MCs35
MCs36
MCs37
MCs38
MCs39
MCs40
0.0027
0.01
0.015
0.0006
0.0012
0.0149
0.005
0.0012
0.0083
3.73e8
0.0085
0.0006
0.0002
0.002
0.0013
0.014
0.018
0.0002
0.02
0.012
0.008
0.014
0.009
0.013
0.014
0.009
0.011
0.0085
0.017
0.005
0.023
0.017
0.00001
0.02
0.019
0.006
0.015
0.0002
0.012
0.0012
0.0089
0.0331
0.0496
0.0020
0.0040
0.0493
0.0165
0.0040
0.0274
0.0000
0.0281
0.0020
0.0007
0.0066
0.0043
0.0463
0.0595
0.0007
0.0661
0.0397
0.0265
0.0463
0.0298
0.0430
0.0463
0.0298
0.0364
0.0281
0.0562
0.0165
0.0760
0.0562
0.0000
0.0661
0.0628
0.0198
0.0496
0.0007
0.0397
0.0040
Ranking of
MCs
19
12
6
23
22
7
18
22
15
26
14
23
24
20
21
8
4
24
2
10
16
8
13
9
8
13
11
14
5
18
1
5
25
2
3
17
6
24
10
22
Table 12 FP of all MCs.

MCs
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
FP
MCs
FP
1.1.1
1.1.2.1
1.1.2.2
1.1.2.3
1.2.1.1
1.2.1.2.1
1.2.1.2.2
1.2.1.3.1
1.2.1.3.2
1.2.2.1 1.2.2.2 1.2.2.3
1.3.1
1.3.2
1.3.3
2.1.1
2.1.2
2.2.1
2.2.2
2.3
3.1
3.2.1
0.0027
0.01
0.015
0.0006
0.0012
0.0149
0.005
0.0012
0.0083
3.73e8
0.0085
0.0006
0.0002
0.002
0.0013
0.014
0.018
0.0002
0.02
0.012
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
3.2.2
4.1.1
4.1.2
5.1
6.1.1.1
6.1.1.2
6.2.1
6.2.2
6.2.3.1
6.2.3.2
6.3
7.1.1
7.1.2.1 7.1.2.2
7.1.3
7.1.4.1
7.1.4.2
7.2.1
7.2.2
7.2.3
7.2.4
0.008
0.014
0.009
0.013
0.014
0.009
0.011
0.0085
0.017
0.005
0.023
0.017
0.00001
0.02
0.019
0.006
0.015
0.0002
0.012
0.0012
86
Table 14 Result of SA.

No of MCs
MCs1
MCs2
MCs3
MCs4
MCs5
MCs6
MCs7
MCs8
MCs9
MCs10
MCs11
MCs12
MCs13
MCs14
MCs15
MCs16
MCs17
MCs18
MCs19
MCs20
MCs21
MCs22
MCs23
MCs24
MCs25
MCs26
MCs27
MCs28
MCs29
MCs30
MCs31
MCs32
MCs33
MCs34
MCs35
MCs36
MCs37
MCs38
MCs39
MCs40
3.6.
FP of MCs
F-V IM
0.0027
0.01
0.015
0.0006
0.0012
0.0149
0.005
0.0012
0.0083
3.73e8
0.0085
0.0006
0.0002
0.002
0.0013
0.014
0.018
0.0002
0.02
0.012
0.008
0.014
0.009
0.013
0.014
0.009
0.011
0.0085
0.017
0.005
0.023
0.017
0.00001
0.02
0.019
0.006
0.015
0.0002
0.012
0.0012
0.0089
0.0331
0.0496
0.0020
0.0040
0.0493
0.0165
0.0040
0.0274
0.0000
0.0281
0.0020
0.0007
0.0066
0.0043
0.0463
0.0595
0.0007
0.0661
0.0397
0.0265
0.0463
0.0298
0.0430
0.0463
0.0298
0.0364
0.0281
0.0562
0.0165
0.0760
0.0562
0.0000
0.0661
0.0628
0.0198
0.0496
0.0007
0.0397
0.0040
MCs Rank
19
12
6
23
22
7
18
22
15
26
14
23
24
20
21
8
4
24
2
10
16
8
13
9
8
13
11
14
5
18
1
5
25
2
3
17
6
24
10
22
Defuzzication process of subjective BEs
The center of area deffuzication technique is employed to

calculate the deffuzication of all the subjective BEs. Table 10
shows the result of subjective BEs deffuzication.
3.7.
Converting CFP of BEs into FP
CFP of the subjective BEs can be transformed into the corresponding FP by using Equation 12. Table 11 presents FP of all
the subjective BEs.
Revised TE value
0.3006
0.2954
0.2918
0.3020
0.3016
0.2919
0.2989
0.3016
0.2966
0.3024
0.2965
0.3020
0.3023
0.3010
0.3015
0.2925
0.2897
0.3023
0.2882
0.2940
0.29682
0.292541
0.29611
0.293258
0.292541
0.29611
0.294687
0.296465
0.290382
0.29894
0.286024
0.290382
0.302438
0.288209
0.288935
0.298235
0.291823
0.302306
0.293973
0.301607
3.8.
RRW (TEinitial TErevised )

0.0019
0.0070
0.0106
0.0004
0.0008
0.0106
0.0035
0.0008
0.0058
0.0000
0.0060
0.0004
0.0001
0.0014
0.0009
0.0099
0.0128
0.0001
0.0142
0.0085
0.0056
0.0099
0.0063
0.0092
0.0099
0.0063
0.0078
0.0060
0.0121
0.0035
0.0164
0.0121
0.0000
0.0142
0.0135
0.0042
0.0106
0.0001
0.0085
0.0008
RRW rank
19
12
6
23
22
7
18
22
15
26
14
23
24
20
21
8
4
24
2
10
16
8
13
9
8
13
11
14
5
18
1
5
25
2
3
17
6
24
10
22
Calculating FP of TE
To quantify the probability of TE of the fault tree a probability for each BE in the fault tree must be provided. These
BE probabilities are then propagated upward to the TE using
the Boolean relationships. In other words, conventional FTA
rules are employed for TE quantication. Therefore, all BEs are
considered independent. The BE probabilities can be propagated upward using MCs. Table 12 presents the FP of all the
MCS. Furthermore, TE is obtained by using Eq. (13). The value
of FP of the TE is 0.3024 per year.
Fig. 5 Result of sensitivity analysis for revised TE.
87
Fig. 6 RRW results.
3.9.
Ranking of MCs
Table 13 presents the ranking of MCs based on their calculated

importance levels (Eq. (14)).
In a SA, an input data parameter, such as a component
FP is changed, and the resulting change in the TE probability is determined. This is repeated for a set of changes using
either different values for the same parameter or changing
different parameters, e.g. changing different FPs. Usually for a
given sensitivity evaluation, only one parameter is changed at
a time. This is called a one-at-a-time sensitivity study. This
method is employed here to validate the sensitivity of the
proposed model. RRW is employed to perform SA. The RRW
can be calculated by re-quantifying the MCs probability of the
given event set to 0. It is expected that eliminating of MCs that
have the highest contribution to the occurrence of TE should
result in reducing the occurrence rate of TE more than other
MCs. Therefore, ranking of RRW values is expected to be the
same as the initial ranking result of MCs. As shown in Table 14,
MCs31 has the highest contribution to the TE occurrence probability. Therefore, the RRW value of MCs31 is 0.0164 which is
the highest as expected. It shows the ranking result which
remains the same as the previous calculation. The proposed
model satises the aforementioned expectations.
The fourth column of Table 14 shows the value of the TE
after eliminating of MCs. Fig. 4 includes 40 bars; the red color
bar shows the TE value which is 0.3024. All new TE values (Column 5th of Table 14) are presented in blue bars. As shown in
Table 14, MC number 26 is the most critical MC of the system.
If the value of MC 26 is reduce to zero, it would be expected
the new TE value reduce more than others. Fig. 5 can conrm
the expectation.
FVI and RRW are employed for ranking of MCs in this paper.
Results of FVI measure and RRW are shown in the 3rd and 6th
columns of Table 14.
As mentioned, one of the advantages of SA is to identify
error in the model. Result of Table 14 and Fig. 5 show that the
proposed model can produce robust outcomes. As obvious in
Fig. 6, controlling the rst 11 most critical MCs would reduce
the TE rate from 0.3 to 0.15. It means that control of the determined MCs will ensure considerable safety improvement in
the Deethanizer section of the Arya Sasol Plant.
4.
Conclusion and discussions
As one of the heavy industry discipline, petrochemical plants

are required to implement effective and consistent safety
planning against potential hazards (i.e. re, explosion and
toxic gas releases) in order to ensure sustainable production. This research focused on developing a FFTA methodology
demonstrating with Deethanizer failure within petrochemical

plant operational concept. Consequently, the research highlights the following points:
A fuzzy methodology for FT evaluation seems to be an
alternative solution to overcome the weak points of the conventional approach.
By using linguistic variables, it is possible to handle the
ambiguities in the expression of the occurrence of a BE. In
addition, the state of each BE can be described in a more
exible form, by using the concept of fuzzy sets.
Instead of using CFP, FP is used to characterize the failure
occurrence of the system events. It can efciently express
the vagueness of the nature of system phenomena and
insufcient information. Further, regardless of the complexity of the system, it is also possible to identify which BE can
inuence system FP the most.
The importance measure can provide useful information for
improving the safety performance of a system. F-VI measure index assists the analyst in identifying the critical MCs
in the system for reducing occurrence likelihood of a TE.
As further research, application of FFTA methodology to
other critical processes in petrochemical plant can be conducted. Moreover, multi attribute decision making (MADM)
techniques can be adopted into the proposed methodology to
make cost benet analysis for controlling the determined MCs.
Acknowledgement
The authors gratefully acknowledge to HSE manager (Hossein
Charkhand) of Arya Sasol Petrochemical Company (A.S.P.C.)
for technical information support in the demonstration of the
case study on Deethanizer failure in petrochemical plant operations.
References
Aqlan, F., Ali, M.E., 2014. Integrating lean principles and fuzzy
bow-tie for risk assessment in chemical industry. J. Loss
Prevent. Process Ind. 29, 3948.
Cai, K.Y., Wen, C., Zhang, M., 1991. Fuzzy states as a basis for a
theory of fuzzy reliability in the possibility context. Fuzzy Sets
Syst. 42, 1732.
Celik, M., Lavasani, S.M., Wang, J., 2010. A risk-based modelling
approach to enhance shipping accident investigation. Saf. Sci.
48, 1827.
Chen, S.J., Hwang, C.L., 1992. Fuzzy Multiple Attribute Decision
Making, 1st ed. Springer Verlag, Berlin, Heidelberg, Germany.
Dong, H.Y., Yu, D.U., 2005. Estimation of failure probability of oil
and gas transmission pipelines by fuzzy fault tree analysis. J.
Loss Prevent. Process Ind. 18, 8388.
88
Ford, D., Sterman, J.D., 1998. Expert knowledge elicitation to

improve mental and formal models. Syst. Dyn. Rev. 14,
309340.
Hsu, H.M., Chen, T.C., 1994. Aggregation of fuzzy opinion under
group decision making. Fuzzy Sets Syst. 79, 279285.
Huang, H.Z., Tong, X., Zuo, M., 2004. Posbist fault tree analysis of
coherent systems. Reliab. Eng. Syst. Saf. 84, 141148.
Korta, J.P., Lee, M.P., Eisenberg, N., Dewispelare, A., 1996. Branch
Technical Position on the Use of Expert Elicitation in the
High-level Radioactive Waste Program, 1st ed. U.S. Nuclear
Regulatory Commission, NUREG-1563.
Lavasani, S.M., Wang, J., Yang, Z., Finlay, J., 2012. Application of
MADM in a fuzzy environment for selecting the best barrier
for offshore wells. Exp. Syst. Appl. 39, 24662478.
Liang, G.S., Wang, M.J., 1993. Fuzzy fault-tree analysis using
failure possibility. Microelectron Reliab. 33, 583597.
Lin, T.C., Wang, M.J., 1997. Hybrid fault tree analysis using fuzzy
sets. Reliab. Eng. Syst. Saf. 58, 205231.
Lin, T.C., Wang, M.J., 1998. Hybrid fault tree analysis using fuzzy
sets. Reliab. Eng. Syst. Saf. 58, 205231.
Liu, Y., Fan, Z.P., Yuan, Y., Li, H., 2014. A FTA-based method for
risk decision making in emergency response. Comput. Oper.
Res. 42, 4957.
Miller, G.A., 1956. The magical number seven plus or minus two:
some limit on our capacity for processing information.
Psychol. Rev. 63, 8197.
Modarres, M., 2006. Risk Analysis in Engineering: Probabilistic
Techniques, Tools an Trends. CRC, USA.
Nicokis, J.S., Tsuda, I., 1985. Chaotic dynamics of information
processing: the magic number seven plus-minus two. Bull.
Math. Biol. 47, 343365.
Onisawa, T., 1988. An approach to human reliability in
man-machine systems using error possibility. Fuzzy Sets Syst.
27, 87103.
Onisawa, T., 1990. An application of fuzzy concepts to modeling

of reliability analysis. Fuzzy Sets Syst. 37, 266286.
Onisawa, T., 1996. Subjective analysis of system reliability and its
analyzer. Fuzzy Sets Syst. 83, 249269.
Onisawa, T., Nishiwaki, Y., 1988. Fuzzy human reliability on the
CHERNOBYL accident. Fuzzy Sets Syst. 28, 115127.
PDS, 2010. Reliability Data for Safety Instrument and System, 1st
ed. SINTEF Publisher, Trondheim, Norway.
Ping, H., Zhang, H., Zuo, M.J., 2007. Fault tree analysis based on
fuzzy logic. In: Annual Reliability and Maintainability
Symposium, vol. 10, pp. 7782.
Preyssl, C., 1995. Safety risk assessment and management the
ESA approach. Reliab. Eng. Syst. Saf. 49, 303309.
Rausand, M., Hoyland, A., 2004. System Reliability Theory:
Models, Statistical Methods and Applications, 1st ed. John
Wiley and Son, Inc., New Jersey, USA.
Sawer, J.P., Rao, S.S., 1994. Fault tree analysis of fuzzy mechanical
systems. Microelectron. Reliab. 34, 653667.
Shu, M.H., Cheng, C.H., Chang, J.R., 2006. Using intuitionistic
fuzzy fault-tree analysis on printed circuit board assembly.
Microelectron. Reliab. 46, 21392146.
Spouge, J., 2000. A Guide to Quantitative Risk Assessment for
Offshore Installation, 2nd ed. CMPT publications, Aberdeen,
UK.
Sugeno, M., 1999. Fuzzy Modelling and Control, 1st ed. CRC Press,
Florida, USA.
Tanaka, H., Fan, L., Lai, S., Toguchi, K., 1983. Fault tree analysis by
fuzzy probability. IEEE Trans. Reliab. 32, 453457.
Wang, D., Zhang, P., Chen, L., 2013. Fuzzy fault tree analysis for
re and explosion of crude oil tanks. J. Loss Prevent. Process
Ind. 26, 13901398.
Zhao, R., Govind, R., 1991. Defuzzication of fuzzy intervals.
Fuzzy Sets Syst. 43, 4550.

An Extension To Fuzzy Fault Tree Analysis (FFTA) Application in Petrochemical Process Industry

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

An Extension To Fuzzy Fault Tree Analysis (FFTA) Application in Petrochemical Process Industry

Transféré par

Droits d'auteur :

Formats disponibles

Process Safety and Environmental Protection 9 3 ( 2 0 1 5 ) 7588

Contents lists available at ScienceDirect

Process Safety and Environmental Protection