Ocean Engineering 117 (2016) 238245

Contents lists available at ScienceDirect

Ocean Engineering
journal homepage: www.elsevier.com/locate/oceaneng

Life extension and repair decision-making of ageing offshore platforms

based on DHGF method
Chong Tan, Yushi Lu n, Xiaotao Zhang
Faculty of Engineering, China University of Geosciences, Wuhan, Hubei Province, China

art ic l e i nf o

a b s t r a c t

Article history:
Received 22 July 2015
Received in revised form
14 December 2015
Accepted 20 March 2016
Available online 31 March 2016

As platforms age, ensuring their continued integrity becomes increasingly. The current life extension and
repair decision-making processes within ageing platforms are typically based on the DHGF algorithm,
which was established in order to provide reasonable predictions about the lifespan of ageing platforms,
make accurate repair decisions, and reduce risks related to uncertain and complicated environments. The
algorithm contains 18 indicators based on the Delphi method, which together-build a complete evaluation system. In this study, the hierarchical structures were established by analyzing and adjusting four
dimensions - project factors, risk factors, load factors, and structural factors, and all of them affect the
ageing platform service state. The Analytic Hierarchy Process determined a weighted subset. Gray
weights were calculated using the gray model theory, and fuzzy mathematics was then applied to form
grade evaluation for the ageing platform. A complete evaluation criterion for life extension and repair
decision-making was established, and the comprehensive score was calculated by a sequence of computational steps. Analysis showed that the decision making of this platform is Major repair, fatigue
cracks, corrosion, and marine foulingthese factors must be addressed rst. The results conrm that the
proposed model accurately describes the dynamic, economic lifespan of ageing platforms.
& 2016 Elsevier Ltd. All rights reserved.

Keywords:
Ageing platforms
Life extension and repair decisions
DHGF algorithm
Gray theory

1. Introduction
An offshore platform is the most important equipment utilized
within offshore oil exploitation. They are complex in structure and
costly to build, and will take with a life of approximately 1520 years.
Ageing offshore platforms in China are widely distributed throughout
Bohai Bay to the South China Sea. The shallow sea platform in Bohai
Bay, exceeded its design life in 2009. It is estimated that more than
two-thirds of the world's ageing platforms will continue working for
5 or 10 years after their design life. Although the lengthy design life of
platforms seems rather conservative, once they have been designed
successfully, they will maintain a long-term, stable safety status
throughout the maintenance stage. As far as life extension, the most
critical issues that must be considered are the evaluation of a platform's safety conditions, and appropriate decision-making in terms of
repair grades.
Previous researchers Bea and Moan built a system of reassessment and requalication criteria for platforms, as well as
probabilistic inspection planning of the jacket structures (Bea
et al., 2000; Moan et al., 2000, 1999). Havbro et al. later proposed a
risk monitoring method, which is applicable to structural life
n

Corresponding author.
E-mail address: 16300371@qq.com (Y. Lu).

http://dx.doi.org/10.1016/j.oceaneng.2016.03.048
0029-8018/& 2016 Elsevier Ltd. All rights reserved.

extension (Havbro et al., 2005). Around the same time, the research team Galbraith et al. studied platform structure integrity
management, and proposed a life extension system for ageing
platforms (Galbraith et al., 2005). Nielsen et al. later studied operation and maintenance of offshore wind turbine components
based on risk (Nielsen et al., 2011). At present, domestic studies on
life extension of ageing platforms are typically focused on defect
assessment (Chen, 2001). Life extension and/or repair decisionmaking models have been lacking relevant research.
There is uncertainty inherent to the use, maintenance, and life
extension of ageing platforms. Researchers Chang and Kaisa stated
that there is a wide variation between measured results and actual
results due to fuzziness (Chang et al., 1994; Kaisa, 1998). Managing
this discrepancy critical within life extension and repair decisionmaking, and a careful balance must be maintained between costsaving benets and potential risk. In this study, qualitative analysis
was combined with the quantitative analysis to comprehensively
evaluate risk factors, which inform life extension, and repair decisions based on DHGF theory. The paper builds a decision-making
model of life extension and repair based on DHGF theory, which is
benecial for leaders to make decisions by quantifying methods,
improve the level of scientic decision-making and achieve the
aim for prolonging the service life of offshore platform. By the
comprehensive evaluation of risk factors, the model may offer
valuable refereces information to the leaders.

C. Tan et al. / Ocean Engineering 117 (2016) 238245

2. Basic thought and superiority of DHGF theory

The theoretical foundation of the DHGF comes from two
sources: one is a synthesized method from qualication to quantication presented by China's keystone space scientist Qian
Xuesen (Qian, 1954); another is WuliShiliRenli's (WSR) approach
presented by Professor Gu Jifa (Gu, 1998, 2011). Xu Wei-xiang
constructed a DH model, and presented the GF algorithm in order
to make comprehensive estimation for system according to the
WSR approach (Xu, 2000). The DHGF algorithm is based on a
combination of Delphi, Analytic Hierarchy Process (AHP), grey
relational analysis, and fuzzy comprehensive assessment methods,
each of which have a distinct set of disadvantages. DHGF is the
combination of practical experience and scientic theory, and is a
mathematics method from qualitative and quantitative view. It
uses the Delphi method to construct evaluation index system, uses
AHP method to obtain weighted matrix, uses Grey Interconnect to
count scores of experts, and uses Fuzzy Evaluating to obtain the
evaluation results (Xu et al., 2001).
The DHGF model is widely involved in various elds, including
Ma Zhi-qiang's application for it in performance evaluation of
university teachers (Ma and Bo, 2011); Feng You-ling's application
for it in international inuence of the Shanghai World Exposition
in quantitative analysis (Feng et al., 2011); Tang Qiu-sheng's application for it in risk evaluation of international logistics park
(Tang et al., 2010); Zhang Jian's application for it in comprehensive
evaluation of ship maneuverability (Zhang et al., 2010); Wang
Wen's application for it in command and control performance
evaluation of armored forces (Wang et al., 2009); Liu Kai's application for it in supportability evaluation of torpedo power system
(Liu et al., 2009).
Advantages and disadvantages of each method are presented as
follows (Table 1).
The evaluation method rst sets up a bridge between the
qualitative and quantitative analysis, and then abstract description
of the system uncertainty by using scientic calculation methods
in order to reveal the nature of things and laws. Its expert groups,
the data along with a variety of information and computer is
combined organically to mesh scientic theoretical knowledge
and practical experience together to play their respective advantages and the comprehensive positive effect.

3. Modeling procedure
General uncertainty regarding the use, maintenance, and risks
inherent to the life extension and repair of the existing ageing

239

platforms poses a serious problem. Research must focus on

managing these complex factors and more effectively gathering
and analyzing relevant data. Essentially, life extension and repair
decisions for ageing platforms are very precarious, and require
careful balance between cost-saving benets and risk. This study
adopted the DHGF theory to build a risk management system for
ageing platforms, as mentioned above.
The Delphi method establishes the risk assessment system's
index, the AHP method weights each index appropriately, then the
expert evaluation results undergo grey relational analysis, and nally overall conclusions are obtained by fuzzy comprehensive
assessment method. Basically, the algorithm utilizes a panel of
experts, relevant data theories, and practice to build a comprehensive evaluation system with all possible advantages, in which
all four methods complement each other.
The main steps in the decision-making process are as follows:
Step 1: Determine the evaluation index set.
According to WSR thought, one must use the Delphi method,
invite a panel of experts to participate in the project team, each
expert will then put forward to a series of evaluation indices independently while collecting and analyzing data, and then screen
out some unimportant indices (Gu, 1998, 2011). In this step, a
scientic and reasonable evaluation index system is built.
O = [O1, O2, , On] (n is the number of index)
Step 2: Determine the weighted subset using AHP method.
The Analytic Hierarchy Process (AHP) is a structured multi-attribute decision-making method (Satty, 1994). The main goal of the AHP
is to decompose a complex system into goals, principles, and programs, thus making quantitative and qualitative decisions. The primary advantage of this method is its capability to identify and reduce
inconsistencies within expert judgments. The AHP has been used extensively to solve problems that have multiple criteria (Satty, 2005).
In general, the AHP contains the following several steps:
(1) Dene problems.
(2) Construct hierarchical structure. Construct decisions that can
be decomposed into independent elements within a hierarchy
comprised of goals, criteria, sub-criteria, and alternatives.
(3) Construct judgment matrix and invite experts ll in it. Determine the importance of attributes, sub-attributes, and pairs
of attributes, and evaluate them on a nine-point scale (Table 2). Place each element in its corresponding level and
calibrate them on the numerical scale; from this, the judgment
matrix A = (aij )m n can be obtained.

Table 1
Comparison in the advantage and disadvantage of different evaluation models.
Methods

Advantage

Disadvantage

Delphi method

1. Simplicity
2. It has broadly representative.

AHP method

1. It has theoretic foundation.

2. Plan sort can be obtained.

Grey theory

1.
2.
3.
1.

The request of sample size is small.

Computation complexity is small.
It can reduce subjectivity.
It offered the quantitative description
phenomenon.

1.
2.
3.
1.
2.
3.
1.
2.

1.
2.
3.
4.

It can reduce subjectivity.

Evaluation results are more accurate.
Plan sort can be obtained.
Plan can be graded.

Fuzzy evaluation

DHGF model

method

for

obscure

Subjectivity is too strong.

Responses are too hasty.
Time of consultation is longer.
Subjectivity is strong.
The process of comparison and judgment is rough.
Cannot do hierarchical plans.
It is difcult to completely analyze question due to lower-resolution.
Do not take into account qualitative indices.

1. Subjectivity is strong.
2. There are problems with judgment is inexact or results are not
comparable.
1. The request of sample size is more.
2. A great deal of calculation is needed.

240

C. Tan et al. / Ocean Engineering 117 (2016) 238245

[d1, ), its whitened function is expressed as follows:

Table 2
Nine-point pair-wise comparison scale.
Numerical value Verbal meaning for risk factor evaluation

Verbal meaning for alternative evaluation

1
2

Equally preferred
Equally to moderately
preferred
Moderately preferred
Moderately to strongly
preferred
Strongly preferred
Strongly to very strongly
preferred
Very strongly preferred

Equally important
Equally to moderately more
important
Moderately more important
Moderately to strongly more
important
Strongly more important
Strongly to very strongly
more important
Very strongly more
important
Very strongly to extremely
more important
Extremely more important

3
4
5
6
7
8
9

d /d
dli [0, d1]
li 1
f1(dli ) = 1
dli [d1, )

0 dli ( , 0)

(2) A medium measurement whitened function, grey number

[0, d1, 2d1], its whitened function is expressed as follows:

d /d
dli [0, d1]
li 1

f2 (dli ) = 2 dli /d1 dli [d1, 2d1]

dli (0, 2d1)

0

Very strongly to extremely

preferred
Extremely preferred

(4) Calculate weight of index Wi . Wi = n j = 1 (aij /k = 1 akj ), aij is the

n
element of judgment matrix, and i = 1 Wi = 1, 0 < Wi < 1
(5) Perform all necessary calculations to obtain the maximum
eigenvalue, consistency ratio CR and consistency index CI :

CI
max =

n
,
max is the
= max
n1
n
n
1
i = 1 ( j = 1 aijwj/wi ); n
n

maximal eigenvalue of matrix,

is the number of factors in the

judgment matrix.
Satty dened the consistency ratio CR as , when , consistency of
judgment matrix can be acceptable, when , when consistency fails
to meet the requirement, the judgment matrix need to be revised.
RI represents the average random consistency index and is shown
in Table 3.
Once all comparisons have been made, and the relative weights
of each criterion have been evaluated, the numerical probability of
each alternative can be calculated.
Step 3: Build the evaluation sample matrix.
Let number of experts equal r : E = {E1, E2, , Er}. Evaluation
sample of the ith index Oi is represented as dli from the lth expert.
Evaluation matrix D was obtained in accordance with data
provided by the expert panel, as follows:

d d
11 12
d d
D = (dli )r n 21 22

dl1 dl2

d1n

d2n
(1 l r )

dln

3
0.58

4
0.90

5
1.12

dli [0, d1]

1

d2 dli
dli [d1, d2]
f3 (dli ) =
d2 d1

0
dli (0, d2)

(4)

The turning point value of whitened function called thresholds d1

and d2. There are two methods used to get the value of turning
point: one is an objective threshold according to the criterion or
experience analogies; another is a relatively threshold according to
maximum value found, medium value found and minimum value
found from the evaluation matrix.
Step 6: Calculate grey statistics.
Calculate the weight f j (dli )(1 j m) according to the whitened function. Then, calculate the grey statistics (nij ) of evaluation
matrix and total grey statistics ( ni ).
r

nij =

f j (dli)

(5)

(1)

Table 3
Average random index.
2
0.00

(3) A lower limit measurement whitened function, grey number

[0, d1, d2], its whitened function is expressed as follows:

ni =

(1) An upper limit measurement whitened function, grey number

1
0.00

(3)

i=1

Step 4: Determine evaluation grade.

The comprehensive evaluation set is determined based on
measurement theory, V = [V1, V2, V3, , Vm], m indicates grade
number. For instance, the grades include: good, slightly good,
medium, slightly bad, bad.
Step 5: Determine grey cluster.
This step includes: determining the grade number and the grey
number of the grey cluster and determines whitened function of
grey number. The grey cluster is determined by the qualitative
analysis, in accordance with the evaluation grades. A commonly
used whitened function has three types as follows (Fig. 1):

n
RI

(2)

6
1.24

7
1.32

8
1.41

9
1.45

nij

(6)

j=1

Step 7: Calculate the grey evaluation weight value and weight

matrix.
With regard to the i th evaluation factor, the grey weight evaluation value is rij = nij /ni . The weight matrix can be obtained according to rij .

r11
r
R = 21

rn1

r12
r22

rn2

r1m
r2m

rnm

(7)

Step 8: Calculate the grey comprehensive evaluation vector.

Grey comprehensive evaluation vector A can be calculated
based on the weight set W and the grey evaluation weight matrix
R:

r11
r
A = WR = W1, W2, , Wn 21

rn1

r12
r22

rn2

r1m
r2m

rnm

(8)

C. Tan et al. / Ocean Engineering 117 (2016) 238245

241

Fig. 1 Three types whitened function.

Step 9: Calculate the grey comprehensive evaluation value.

Each grey grade was assigned according to grey level, the
hierarchical vector V = [V1, V2, V3, , Vm] of the grey cluster was
obtained, and the comprehensive evaluation value was obtained
as follows:

= A VT

(9)

4. Application
This paper took an ageing shallow sea platform in Bohai Bay, as
a research project. The platform was completed in 1994, and was
designed for the last 15 years. The design life of this ageing platform was exceeded in 2009. In order to meet the production demands, this platform must continue to function safely and
correctly.
Since 2003, leadership was forced to extend the life of this
platform, using all of the available information. Leaders were able
to apply the new repair decision-making model to this platform.
For leadership, the decision-making model can improve the science, objectivity, pertinence and validity of life extension and
repair.
Management must understand and master not only platform
design and reconstruction, but also the material performance and
load as they change over time change in order to prolong the life of
this platform. In addition, failure risk and risk tolerance level must
factor into life extension decisions. Risk factors inherent to the
ageing platform must be taken into account, and scientic and
effective evaluation system must be employed in order to make
informed decisions.
In view of the actual situation of the platform, set up life extension steering group (LESG), its members include top executives,
experienced staffs, ocean engineering experts, and safety experts.

Life extension and repair of offshore platform involves many

knowledge areas, LESG should include appropriate subject matter
experts. Ensure enough expertize available for LESG when the
need of any specic discipline arises. LESG emphasized the importance of the team working together and put a high value on
interpersonal, communication and coordination with each other.
Through regular and irregular meeting continuously to increase
the frequency of information exchange, and meet the various
needs for communication.
Experts of LESG determined the effecting factors of life extension, and determined criterion layer and index layer of index
system based on Delphi method.
This study built a hierarchical structure of inuential decisionmaking factors for ageing platform status based on historical records, current conditions, and future service risks (Fig. 2).
Offshore platforms are a complex system, each comprised of
multiple levels, variables, and factors. Safety problems in the
platforms cover a wide range, including dynamic factors and static
factors, thus the hierarchy of the index system must closely represent actual conditions.
As shown in Fig. 2, each criterion layer contains four secondary
indexes: project, risk, load and structure. The index layer contains
18 indexes (O1 O18). The Analytic Hierarchy Process (AHP)
method dened the weight of the criterion layers relative to the
target layer.
According to the steps of AHP method, the index of each criterion layer was compared to the target; the results of this comparison are shown in Table 4.
The analysis was made by using Excel software according to the step

n
and of AHP method. max = 4.0062, RI = 0.90, CI = max
=
n1
4.0062 4
41

= 0.0021 , CR = CI /RI = 0.0021/0.90 = 0.0023 < 0.10, meet

requirement of consistency test.

W = AT = (0.0929, 0.1966, 0.2948, 0.4157)

Fig. 2. Hierarchical structure of the inuencing factors for ageing platform status assessment.

242

C. Tan et al. / Ocean Engineering 117 (2016) 238245

Table 4
Pair wise comparison of criterion layer safety evaluation index.
Risk factors

Project

Risk

Load

Structure

Weight

Project

1
2
3
5

1/2
1

1/3

1/5

2/3
1

1/2

0.0929
0.1966

Risk
Load
Structure

3/2
2

4/3

0.2948

3/4
1

0.4517

In the same manner, the weight set Wi of the index layer can be
obtained in accordance with the criterion layer.

W1 = B1T = (0.2928, 0.7072)

W2 = B2T = (0.1896, 0.1218, 0.1747, 0.4221, 0.0918)

W3 = B3T = (0.1013, 0.1618, 0.0712, 0.3251, 0.3406)

W4 = B4T = (0.1015.0.3124, 0.3821, 0.0589, 0.0468, 0.0983)
Though accurate risk assessments are necessary to extend the
life and to make informed repair decisions, research in this area is
relatively scant at present and is lacking complete theoretical
analysis, as mentioned above. The present study attempts to remedy this. In accordance with the latest research results, as well as
information provided by the expert panel, risk grading was further
veried by calculations and fuzzy evaluation.
The index range in Fig.2 was determined based on the historical
information for the platforms, as well as expert studies (Stacey
et al., 2008; Puskar et al., 2006; Manzocchi, 2008; Yang, 2011). An
evaluation criterion table of risk factors is provided below.
= max [s1(x ) , s2(x ) , , sn(x )]
s(x ) = x/r
where n is the number of platform dents; x is the maximum
range of radial dents, r is the jacket radius, and is the dent
structure importance.
= max [g1(a) , g2(a) , , gm(a)]
g (a) = va/t
where m is the number of platform aws, a is the maximum
crack depth; t is the wall thickness of the jacket, and v is the crack
structure importance.

= max [1(b) , 2(b) , , k(b)]

(b ) = b / c
where k is the amount of platform corrosion, b is equivalent
corrosion depth; c is the wall thickness of the corrosive structure,
and is corrosion structure importance.
This study invited 10 eld experts Ei in ageing platform status
evaluation to make judgments according to Table 5. To facilitate
the evaluation, the experts rate the indices on a scale of 15. The
evaluation matrix is shown as below Table 6.
Determine the evaluation grade is 5 according to measurement
theory (Table 7).
The four relative thresholds were 5, 4, 3, and 2 respectively.
Corresponding grey numbers and whitened functions can be obtained according to the evaluation matrix.
Let the rst grey cluster Good equal 1, e = 1, and grey number
1 [5, ], its whitened function is expressed as follows:
d /5 d [0, 5]
li
li
f1(dli ) = 1 dli [5, )

0 dli [0, )
Let the second grey cluster Slightly good equal 2, e = 2, and
grey number 2 [0, 4, 8], so its whitened function is expressed
as follows:
d /4
dli [0, 4]
li

f2 (dli ) = 2 dli /4 dli [4, 8]

dli [0, 8]
0

Let the third grey cluster Medium equal 3, e = 3, and grey

number 3 [0, 3, 6], its whitened function is expressed as
follows:
d /3
dli [0, 3]
li

f3 (dli ) = 2 dli /3 dli [3, 6]

0
dli [0, 6]

Let the fourth grey cluster Slightly bad equal 4, e = 4 , and grey
number 4 [0, 2, 4], so its whitened function is expressed as
follows:
d /2
dli [0, 2]
li

f4 (dli ) = 2 dli /2 dli [2, 4]

0
dli [0, 4]

The grey statistics of evaluation matrix was then calculated

using Eq. (5). For index C11, the grey statistics of evaluation matrix

Table 5
Scoring criteria for ageing platform life extension and repair decisions.
Index

Renovation
Service time/year
Ship collision
Fire or explosion
Submarine landslide
Environment pollution
Falling objects
Max wind speed ( ms1)
Highest wave (m)
Max ow velocity ( ms1)
Maximum ice load
Extreme load history
Collision damage
Fatigue cracks
Corrosion
Material degradation
Pile foundation scour/m
Marine fouling/cm

Score

Not degraded
<12
No risk
No risk
No risk
No risk
No risk
<18

Slightly degraded
12 18
Low risk
Low risk
Low risk
Low risk
Low risk
18 25

Obviously degraded
18 22
Signicant risk
Signicant risk
Signicant risk
Signicant risk
Signicant risk
>25

Seriously degraded
22 25
High risk
High risk
High risk
High risk
High risk

Extremely
>25
Serious risk
Serious risk
Serious risk
Serious risk
Serious risk

<3
<80

36
80 150

>6
>150

150
0
0.002
No aw

150 < 600

1
0.002 < 0.005
0 < 0.001
0.005 < 0.03
Slight
0.06 0.6
19

> 600
2
0.005 < 0.02
0.001 < 0.03
0.03 < 0.08
Obvious
0.6 1.8
10 20

3
0.02 < 0.1
0.03 < 0.15
0.08 < 0.25
Serious
1.8 4.0
20 35

>3
> 0.1
> 0.15
> 0.25
Extreme
1.8 4.0
>35

0.005
No
<0.06
<1

is the dent factor, is the crack factor, is the corrosion factor; is the ice load factor.

C. Tan et al. / Ocean Engineering 117 (2016) 238245

243

Table 6
Evaluation score of experts.

dli

O1

O2

O3

O4

O5

O6

O7

O8

O9

O10

O11

O12

O13

O14

O15

O16

O17

O18

E1
E2

5
4
4
3
4
3
4
4
3
3

4
4
3
3
4
3
4
4
4
3

4
4
5
5
4
4
4
3
3
5

3
3
2
3
4
3
3
3

3
4
3

4
3
3

3
4
5

4
3

4
3

4
3
4
5
4

4
4
5
3
3

4
4
5
3
4
4
3

4
3
2
3
3
4
2
3
3
2

4
3
4
3

5
4
3
5
4
5
3
5
5
4

4
4
4
5
4
4
3
4
3
3

5
5
5
5
5
5
5
5
5
5

4
4
3
4
4
3
5
4
3
5

3
2
3
4
3
3
2
3
4
3

3
2
4
3
2
3
3
3
3
3

4
5
3
5
4
5
4
5
4
5

3
2
4
3
3
2
3
2
3
2

4
2
3
2
4
3
3
2
4
3

E3
E4
E5
E6
E7
E8
E9
E10

4
3
4
3

Table 7
Evaluation grade.
Evaluation grade

Good

Relatively good

Medium

Relatively bad

Bad

Grade scale

According to the weight set W built based on the AHP method,

the above-mentioned results were plugged into Eq. (8) in order to
calculate grey comprehensive evaluation vector A.

A = WR = (0.2818, 0.3224, 0.2859, 0.1092, 0)

The hierarchical vector V = (5, 4, 3, 2, 1) of the grey cluster was
then substituted in Eq. (9) to acquire the comprehensive evaluation value of the platform:

was obtained as follows:

If e = 1,

= A V T = 3.77

10

n11 =

f1(dli) = f1(5) + f1(4) + f1(4)

k=1

+ f1(3) + f1(4) + f1(3) + f1(4) + f1(4)

+ f1(3) + f1(3) = 1 + 4/5 + 4/5 + 3/5
+ 4/5 + 3/5 + 4/5 + 4/5 + 3/5 + 3/5 = 7.4
Other necessary results were obtained in the same manner. If
e = 2, n12 = 8.75. If e = 3, n13 = 7.67. If e = 4 , n14 = 2.
The above results were then plugged into Eq. (6) to calculate
the total grey statistics.
4

n1 =

n1e = 7.4 + 8.75 + 7.67 + 2 = 25.82

e= 1

The weight matrix of grey evaluation can also be obtained according to rij = nij /ni
0.2866

0.2748
0.3377
0.2259

0.2748
0.2284

0.2989
0.2158
0.2866
R=
0.3656
0.3114

0.4801
0.3312
0.2209

0.2105
0.3724

0.2001
0.2264

0.3389
0.3435
0.3604
0.2823
0.3244
0.3796
0.3343
0.2697
0.3389
0.3508
0.3693
0.3601
0.3715
0.2760
0.2632
0.3597
0.2502
0.2830

0.2971
0.3053
0.2607
0.3279
0.3053
0.2912
0.2882
0.3099
0.2971
0.2411
0.2794
0.1598
0.2793
0.3191
0.3267
0.2256
0.3087
0.3019

0.0775
0.0763
0.0412
0.1639
0.0954
0.1008
0.0786
0.2046
0.0775
0.0425
0.0399
0
0
0.1840
0.1996
0.0423
0.2409
0.1887

0
0
0
0
0

0
0
0
0
0

0
0
0
0
0

0
0

The number is rounded to 4, demonstrating that the safety

condition of platform is relatively good.
For repair and maintenance of platforms, many experts have
been concerned about its structural integrity. In this paper, the
authors offer a new thought for the research of repair and life
extension decisions in ageing offshore platforms. According to the
repair records of China National Petroleum Corporation, the
company owns a lot of offshore platforms, made three hundred
times repairs from 2012 to 2014, and assessed repair grade according to repair costs and each repair time. Assess the safety
condition of platforms when platforms are being repaired. A range
of status values can be obtained by statistical data. The statistics is
shown as follows.
According to Table 8, the safety grade and life extension time of
an ageing offshore platform can be divided into 10 grades, and
repair grade can be divided into 8 grades, for ease of decisionmaking. Repair decisions can then be made based on these grades.
Life extension time for platforms is decreased each year, and repair
grade will increase annually.
According to the standards shown in Table 9, the safety grade of
this platform is 5, with the repair decision being a Major repair.
According to the scores provided by experts, the platform's most
pressing issues are fatigue cracks, corrosion, and marine fouling
these factors must be addressed rst, and then adequate measures
taken to prevent them from recurring in future. In accordance with
the evaluation results, the life extension of this platform was determined to be 6 years, with immediate repairs recommended. In
view of this, it should adopt the following measures.
(1) Schedules and executes major repair, repair defective outer
members of offshore platform on time. Marine fooling should
clean up in time, and ensure the length of marine fooling of less
than 1 cm. (2) Fatigue crack growth is of concern when service life
extension of offshore installations is undertaken. Fatigue life

Table 8
The statistics of repair records.
Repair grade

Current repair

Minor repair

Medium repair

Major repair

Overhaul

Fully repair

Repair times
Range of status value

183
[4.38, 4.84]

103
[3.91, 4.52]

59
[3.78, 4.19]

23
[3.46, 3.82]

11
[3.13, 3.56]

3
[2.81, 3.32]

244

C. Tan et al. / Ocean Engineering 117 (2016) 238245

Table 9
Platform life extension and repair decision-making reference table.
Items

Score

5.0 4.7

4.7 4.4

4.4 4.1

4.1 3.8

3.8 3.5

3.5 3.2

3.2 2.9

2.9 2.6

2.6 2.3

<2.3

Safety grade
Repair grade

1
a

3
c

6
f
5

7
g

8
g

10

4
d
7

5
e

Life extension time/year

2
b
9

9
h
2

10
h
1

a. No repair; b. Current repair; c. Minor repair; d. Medium repair; e. Major repair; f . Overhaul; g. Fully repair; h. Ready to obsolete.

Table 10
Comprehensive score of platform service state.
Service time/year

10

11

12

13

14

15

16

17

18

19

20

Score (before repair)

Score (after repair)
Life extension time/year(before repair)
Life extension time/year(after repair)
Repair grade

4.88

4.62
4.72
9
10
b

4.54
4.66
9
9
b

4.34
4.53
8
9
c

4.26
4.47
8
9
c

4.12
4.35
8
8
c

3.96
4.43
7
9
d

4.14
4.32
8
8
c

3.85
4.37
7
8
d

3.82
4.31
7
8
d

3.77
4.71
6
10
e

10

Fig. 3. Score and life extension time of platform before and after repairs.

extension techniques as ultrasonic peening could prevent fatigue

cracks to reappear and thus to avoid costly recurrent weld repair.
(3) For corrosive status, the associate anti-corrosion measures of
combining cathodic protection with coating is considered to be
suitable.
To maintain the structural integrity of offshore installations the
following actions could be taken: crack repair and subsequent
ultrasonic peening treatment, underwater welding, plate and shell
reinforcement, clamps and grouted clamping. An offshore structure needs to maintain its structural integrity during its service
life. As the life extension project is undertaken some structural
components, which shows the need of some of the above actions
to maintain the structural integrity during the extended service
life.
Considering possible multi failure modes of ageing platform
members, including static strength failure, fatigue failure and
corrosion failure. So the routine repair and maintenance is espically needed for the aging platform. In this model, test period of
risk-based inspection (RBI) is one year.
Experts graded each index based on the calendar year data for
the platform. After careful calculation and evaluation, a comprehensive score of the platform's service state was completed as

shown in Table 10 and Fig. 3.

The repair cycle is a period of three or ve year period, during a
conventional evaluation. The model proposed in this study uses a
one-year cycle, conversely, which provides more accurate results
and better reects cost-saving benets.
The comprehensive score and repair decision-making of the
platform every time can be used as a reference. Check whether the
last repair reached its intended effect. In offshore platform life
cycle, the process of life extension and repair is essentially dynamic with feedback owing both ways and its decision-making is
a continuous process. This process is distributed into four stages,
namely evaluation, decision making, repair, and feedback. The four
stages are a system of closed loop. They are mutual connection,
mutual restrict, and linked together, recycling, and forming a
continuous process.
For some structural components, project team should adopt
various strategies and try to avoid repair several times for the
same structural component. In the case of fatigue cracks the application of fatigue life improvement techniques will extend the
fatigue life of the concerned weld joint hopeful beyond the targeted max service life. So welded joins of structural integrity
concern will then be reassessed with the intention to avoid any

C. Tan et al. / Ocean Engineering 117 (2016) 238245

further weld repair during the remaining service life.

The possible failure mechanisms must be identied by the
LESG. LESG can make adjustment to the new conditions. Continuously conrm every repair predictions of the model with inspections, and it will be helpful to improve the life extension and
repair decision-making model.
According to the analyses described above, this study reached
the following conclusions:
For this platform, the lifespan was 10 years in the 11th year, but
was then reduced to 8 years in the 13th year. A total of two current
repairs, four minor repairs, three medium repairs, and one major
repair will be necessary by the 20th year. The comprehensive score
for this platform stayed nearly above a 4.0 throughout evaluation,
suggesting that the platform already possesses high quality safety
standards as well as economic advantages. Life extension and repair should focus on the structural integrity, and the platform's
rst task is to strengthen these weak points: fatigue cracks, corrosion, and marine foulingthese factors must be addressed rst,
and then adequate measures taken to prevent them from recurring in future.
The life cycle of the ageing platforms is a process of dynamic
change, under the inuence of signicant uncertainties. Uncertainty factors are mostly materialized on the risk aspect. For
example, there are typical emergencies like re and explosion,
falling objects, and ship collision. On the basis of risk analysis, risk
management should be strengthened to avoid these accidents. The
denition of the platform's design life must include more than just
its service life in order to ensure it functions safely and effectively,
as well economic viability.

5. Conclusion
The case analysis presented in this study found that the design
life of an ageing offshore platform is not only just dependent on its
service life. Moreover, the design life alone does not reect the
dynamic effects of the highly uncertain environment surrounding
a platform. A fuzzy life extension and repair decision-making
model was introduced, which provides an exact description of the
life process for offshore platforms, as they vary with time. The
proposed model is particularly applicable to many ageing platforms that are currently functioning safely and cost-effectively.
This paper focuses on life extension and repair decision-making
technology, in addition to an adaptive adjustment mechanism for
inuential factors according to the service conditions of ageing
platforms. The process of life extension and repair decision-making is a continuous closed loop. This cycle includes the stages of
evaluation, decision making, repair, and feedback.
Analysis showed that the safety grade of this platform is 5, with
the repair decision being a Major repair. Fatigue cracks, corrosion,
and marine fouling need to be strengthened as well.
There are currently no uniform standards for grading offshore
oil platform life extension and a general dearth of relevant theories. Comprehensive analysis of risk is a difcult endeavor in
ageing platforms, as they are highly complex systems under the
inuence of a wide variety of factors. The specic processes of
choosing indexes and determining clusters still require further
study.

245

References
Bea, Robert, Perez, Faustino, Ortega, Roberto, 2000. Requalication of platforms
offshore Tampico-Tuxpan, Mexico: Arenque platform[A]. In: Proceedings of
International Offshore and Polar Engineering Conference. Seattle, pp. 300306.
Chang, C.M., Chen, I.K., Chen, H., 1994. X-Ray inspection reliability for welded joints
[A]. In: Proceedings of International Conference on Structural Safety and Reliability. Innsbruck, Austria, pp. 991996.
Chen, Guo-ming, 2001. Reliability based defect assessment for offshore structure
[A]. In: Proceedings of Pacic Structural Steel Conference. Beijing, pp. 278283.
Galbraith, D.N., Sharp, J.V., Terry, E., 2005. Managing life extension in aging offshore
installations [A]. In: Proceedings of International Petroleum Exhibition and
Conference. Aberdeen, Paper SPE 96702.
Faber Michael, Havbro, Straub, Daniel, Chakrabarti, Partha, et al., 2005. Fatigue
analysis and risk based inspection planning for life extension of xed offshore
platform [A]. In: Proceedings of International Conference on Ocean, Offshore
and Artic Engineering, pp. 511519.
Feng, You-ling, Yang, Yu-hong, 2011. Quantitative evaluation of the international
inuence of the world expos on DHGF model. J. Jiamusi Univ. (Nat. Sci. Ed.) 7
(4), 626627.
Gu, Jifa, Fei, G.A.O., 1998. To see WuliShiliRenli systems approach from the view
of management science. Syst. Eng. Theory Pract. 8, 15.
Gu, Jifa, 2011. Practice on WuliShiliRenli system approach. Chin. J. Manag. 8 (3),
317322.
Jannie Jessen, Nielsen, Sorensena, John Dalsgaard, 2011. On risk-based operation
and maintenance of offshore wind turbine components. Reliab. Eng. Syst. Saf.
96 (1), 218229.
Kaisa, Simoda, Pulkkinen, Urho, 1998. Models for nondestructive inspection data.
Reliab. Eng. Syst. Saf. 60 (1), 112.
Liu, Kai, Jiong, Sun, 2009. Supportability assessment of torpedo electro-dynamic
system based on DHGF algorithm. Ship Electron. Eng. 9, 168171.
Manzocchi, M., Shetty, N., Stacey, A., 2008. Reliability-based assessment for fatigue
integrity management. In: Proceedings of the ASME 27th International Conference on Offshore Mechanics and Arctic Engineering. Estoril.
Ma, Zhi-qiang, Bo, Deng, 2011. Performance evaluation of university teachers based
on DHGF method. Stat. Decis. 4, 6971.
Moan, Torgeir, 2000. Recent research and development relating to platform requalication. J. Offshore Mech. Arct. Eng. 122 (1), 2032.
Moan, Torgeir, Johannes, M., Ole, Tom, 1999. Probabilistic inspection planning of
jacket structures [A]. In: Proceedings of Offshore Technology Conference.
Houston, Paper OTC 10848.
Puskar, F.J., Westlake, H.S., OConnor, P.E., 2006. The development of a recommended practice for structural integrity management (SIM) of xed offshore platforms. In: Proceedings of the 2006 Offshore Technology Conference,
Houston.
Qian, Xuesen, 1954. Engineering Cybernetics. McGraw Hill, New York.
Satty, T.L., 1994. How to make a decision: the analytic hierarchy process. Interfaces
24 (6), 1943.
Satty, T.L., 2005. Theory and Applications of the Analytic Network Process: Decision
Making with Benets, Opportunities, Costs, and, Risks. RWS Publications,
Pittsburgh.
Stacey, A,. Birkinshaw, M., Sharp, J.V., 2008. Life Extension Issues for Ageing Offshore Installations. In: Proceedings of the ASME 27th International Conference
on Offshore Mechanics and Arctic Engineering Estoril.
Tang, Qiu-sheng, Ma, Xian-ting, Ye, Jian, 2010. Risk management of logistic park
based on DHGF evaluation model. J. ChongQing JiaoTong Univ. (Nat. Sci.) 5 (10),
824827.
Wang, Wen, Wu, Qing-liang, 2009. Effectiveness evaluation of command and
control sub-system for digitized armored regiment based on DHGF method.
Command Control Simul. 10 (5), 5155.
Xu, Wei-xiang, Zhang, Quan-shou, 2000. A meta-synthesis of DHGF for evaluating
information. Comput. Eng. Appl. 5, 6062.
Xu, Wei-xiang, Zhang, Quan-shou, 2001. Study on evaluation method for information system projects from qualitative analysis to quantitative analysis.
Syst. Eng. Theory Pract. 3 (3), 124127.
Yang, Dong-ping, Niu, Geng-qi, Zhi, Jing-bo, Liu, Ling-ling, Peng, Xing-lai, 2011. Life
extension decision-making model for ageing offshore platforms. China Saf. Sci.
J. 21 (5), 97103.
Zhang, Jian, Li, Tie-li, 2010. Integrated evaluation of ship maneuverability based on
DHGF algorithm. Chin. J. Ship Res. 12 (6), 5660.