Beruflich Dokumente
Kultur Dokumente
Research Article
Research on a Risk Assessment Method considering
Risk Association
Copyright © 2016 Zhan Zhang et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Regarding risk assessment problems with multiple associated risks, a risk assessment method (RAM) is proposed in this paper.
According to the risk-associated assessment information offered by expert panel, a comprehensive associated matrix is constructed
to identify the influence relationship among risks so as to determine the hierarchical structure of risks. Then, based on the
determined divided or undivided risk hierarchical structure as well as the possibility and loss of risks provided by expert panel,
each value at risk (VAR) is calculated through knowledge related to probability theory. Finally, the feasibility and efficiency of the
proposed method are demonstrated through a calculating case.
paper and also the limitations and further research work to is set as situation B, the expert panel are required to offer
be carried out. the occurrence probability of bottom risk and conditional
probability according to the risk hierarchical structure. Fur-
thermore, the comprehensive probability of the occurrence of
2. Theories and Methods
each risk would be calculated through conditional probability
2.1. Description of Problems. In risk assessment problems, the formula and total probability formula specifically so as to
occurrence of a certain risk may lead to another risk, that is, calculate each value at risk in consideration of loss of each
the association among risks. The following symbols are used risk.
to express the set and quantity of a risk assessment problem
in consideration of risk-associated situations: 2.2. Risk Relationship Identification. According to the descrip-
tion of problems above, identification methods of relation-
(i) 𝑅 = {𝑅1 , 𝑅2 , . . . , 𝑅𝑚 }: the set of risks, where 𝑅𝑖 means ship among several risks (or risk hierarchical structures) are
the risk number 𝑖, 𝑖 = 1, 2, . . . , 𝑚 listed below.
(ii) 𝐸 = {𝐸1 , 𝐸2 , . . . , 𝐸𝑛 }: the set of the expert panel, where Firstly, 𝑛 risk-associated matrixes (𝐴 1 , 𝐴 2 , . . . , 𝐴 𝑛 ) are
𝐸𝑘 means the expert number 𝑘, 𝑘 = 1, 2, . . . , 𝑛 combined as a group-associated matrix 𝐴 𝐺 = [𝑎𝑖𝑗𝐺]𝑚×𝑚 ,
(iii) 𝑤 = (𝑤1 , 𝑤2 , . . . , 𝑤𝑛 ): the weight vector of experts, among which the calculating formula of 𝑎𝑖𝑗𝐺 is
where 𝑤𝑘 means the importance or weight of the
expert 𝐸𝑘 ; it satisfies 0 ≤ 𝑤𝑘 ≤ 1, ∑𝑛𝑘=1 𝑤𝑘 = 1, 𝑘 = { 𝑘
𝑛
𝑛
{
{ ∑ 𝑎𝑖𝑗 𝑤𝑘 , 𝑡𝑖𝑗 ≤
1, 2, . . . , 𝑛 2
𝑎𝑖𝑗𝐺 = {𝑘=1 𝑖, 𝑗 = 1, 2, . . . , 𝑚. (1)
(iv) 𝐴 𝑘 = [𝑎𝑖𝑗𝑘 ]𝑚×𝑚 : risk-associated matrix offered by {
{0, 𝑛
𝑡𝑖𝑗 >
the expert 𝐸𝑘 , where 𝑎𝑖𝑗𝑘 means the evaluation value { 2
offered by the expert 𝐸𝑘 for the direct influence degree In formula (1), 𝑡𝑖𝑗 means the number of experts who score
of the risk 𝑅𝑖 on the risk 𝑅𝑗 , in which five-point the direct influence degree of the risk 𝑅𝑖 on the risk 𝑅𝑗 as 0.
scale is adopted. 0 means “no influence,” 1 means Furthermore, the group-associated matrix 𝐴 𝐺 is transformed
“weak influence,” 2 means “rather weak influence,” 3 into regulated group-associated matrix 𝐵𝐺 = [𝑏𝑖𝑗𝐺]𝑚×𝑚 , among
means “medium influence,” 4 means “rather strong which the calculating formula of 𝑏𝑖𝑗𝐺 is
influence,” and 5 means “strong influence”
(v) 𝜆 𝑘 : the thresholds offered by the expert 𝐸𝑘 for com- 𝑎𝑖𝑗𝐺
prehensive influence degree. Setting up thresholds is 𝑏𝑖𝑗𝐺 = 𝑖, 𝑗 = 1, 2, . . . , 𝑚. (2)
to reject the influence relationship that is insignificant max𝑖 {∑𝑛ℎ=1 𝑎𝑖ℎ
𝐺
}
and has little comprehensive influence degree accord-
Secondly, comprehensive influence matrix with indirect
ing to expert panel. 0 ≤ 𝜆 𝑘 ≤ 1, 𝑘 = 1, 2, . . . , 𝑛
association 𝐶 = [𝑐𝑖𝑗 ]𝑚×𝑚 is set up and the calculating formula
(vi) 𝑃𝑖𝑘 : the probability of occurrence of the risk 𝑅𝑖 offered is
by the expert 𝐸𝑘 , 𝑘 = 1, 2, . . . , 𝑛, 𝑖 = 1, 2, . . . , 𝑚 −1
𝐶 = 𝐵𝐺 (𝐼 − 𝐵𝐺) . (3)
(vii) 𝑙𝑖𝑘 :
the loss after the occurrence of the risk 𝑅𝑖 offered
by the expert 𝐸𝑘 , 𝑘 = 1, 2, . . . , 𝑛, 𝑖 = 1, 2, . . . , 𝑚 Meanwhile, 𝑛 thresholds (𝜆 1 , 𝜆 2 , . . . , 𝜆 𝑛 ) of comprehen-
(viii) 𝑃𝑖𝐺: the comprehensive probability of the occurrence sive influence degree are combined as a group threshold 𝜆 𝐺,
of the risk 𝑅𝑖 . 𝑖 = 1, 2, . . . , 𝑚 and the calculating formula is
1, 2, . . . , 𝑚 𝜆 𝐺 = ∑ 𝑤𝑘 𝜆 𝑘 . (4)
𝑘=1
(x) 𝑧𝑖 : the VAR of the risk 𝑅𝑖 , 𝑖 = 1, 2, . . . , 𝑚
According to the group threshold 𝜆 𝐺, the comprehensive
We propose the detailed procedure of RAM in consid- influence matrix 𝐶 = [𝑐𝑖𝑗 ]𝑚×𝑚 is transformed into 𝜆 𝐺-cut
eration of risk association. Firstly, several risk-associated matrix 𝐶𝜆 = [𝛼𝑖𝑗 ]𝑚×𝑚 , where
matrixes offered by expert panel are combined as group risk-
associated matrix, while several thresholds of comprehensive {1, 𝑐𝑖𝑗 ≥ 𝜆 𝐺
influence degree offered by expert panel are combined as 𝛼𝑖𝑗 = { 𝑖, 𝑗 = 1, 2, . . . , 𝑚. (5)
group threshold; secondly, risk relationship identification 0, 𝑐𝑖𝑗 < 𝜆 𝐺
{
is performed through methods like matrix transform and
there are usually two situations, divided and undivided risk Assume 𝐷 = [𝑑𝑖𝑗 ]𝑚×𝑚 as 0-1 comprehensive associated
hierarchical structures in the result of risk relationship iden- matrix, in which
tification. If the undivided risk hierarchical structure is set as
situation A, each value at risk would be calculated according {𝛼𝑖𝑗 + 1, 𝑖=𝑗
to the occurrence probability and loss of each risk offered 𝑑𝑖𝑗 = { 𝑖, 𝑗 = 1, 2, . . . , 𝑚. (6)
𝛼 , 𝑖 ≠ 𝑗
by the expert panel; if the divided risk hierarchical structure { 𝑖𝑗
Mathematical Problems in Engineering 3
Matrix 𝐷 reflects the preference of the expert panel for the risk 𝑅𝑗 offered by the expert 𝐸𝑘 ; 𝑑𝑓 is the number of
influence degree among each risk with indirect influence risks on the layer number 𝑓. 𝑅𝑗,𝑓−1 is assumed as the risk
relationship. The influence of the risk 𝑅𝑖 on itself is 1. number 𝑗 on the layer number 𝑓 − 1 related to the risk
According to matrix 𝐷, hierarchical structure of each risk 𝑖𝑓 𝑖𝑓
𝑅𝑖𝑓 , 𝑗 = 1, 2, . . . , 𝑑𝑓−1 ; 𝑑𝑓−1 is the number of risks on the layer
is divided. Assume number 𝑓 − 1 related to the risk 𝑅𝑖𝑓 . 𝑅𝑖𝑓 = 𝑇 is assumed as
the occurrence of the risk 𝑅𝑖𝑓 , and 𝑅𝑖𝑓 = 𝐹 means that the
𝐻𝑖 = 𝐻𝑖 ∩ 𝐽𝑖 𝑖 = 1, 2, . . . , 𝑚, (7)
risk 𝑅𝑖𝑓 does not happen. In consideration of two conditions,
𝑖𝑓
where 𝐻𝑖 = {𝑅𝑗 | 𝑑𝑗𝑖 = 1} and 𝐽𝑖 = {𝑅𝑗 | 𝑑𝑖𝑗 = 1}. 𝐻𝑖 means 𝑅𝑗,𝑓−1 = 𝑇 and 𝑅𝑗,𝑓−1 = 𝐹, there will be 2𝑑𝑓−1 conditional
the risk set corresponding to the element valued 1 in the list probabilities related to 𝑅𝑖𝑓 = 𝑇.
number 𝑖 in the matrix 𝐷; 𝐽𝑖 means the risk set corresponding Furthermore, assuming that 𝑃𝑘 (𝑅𝑖𝑓 = 𝑇 | 𝑅1,𝑓−1 =
to the element valued 1 in the row number 𝑖 in the matrix 𝐷. 𝑇, 𝑅2,𝑓−1 = 𝑇, . . . , 𝑅𝑑𝑖𝑓 ,𝑓−1 = 𝑇) is the probability of occur-
If formula (7) works for a certain 𝑖, 𝑅𝑖 should be regarded 𝑓−1
as the bottom element in 𝐷 and the list number 𝑖 and the rence of the risk 𝑅𝑖𝑓 offered by the expert 𝐸𝑘 with the risks
row number 𝑖 in 𝐷 should be deleted to form a new matrix. 𝑅1,𝑓−1 , 𝑅2,𝑓−1 , . . . , 𝑅𝑑𝑖𝑓 ,𝑓−1 happening simultaneously, 𝑘 =
𝑓−1
During the process, if the bottom element cannot be found, 1, 2, . . . , 𝑛; 𝑃𝑘 (𝑅𝑖𝑓 = 𝑇 | 𝑅1,𝑓−1 = 𝑇, 𝑅2,𝑓−1 = 𝐹, . . .,
risk assessment should be performed regarding situation A. 𝑅𝑑𝑖𝑓 ,𝑓−1 = 𝑇) is the probability of occurrence of the risk
If there is bottom element, searching bottom element in the 𝑓−1
new matrix should be performed until all the elements in 𝑅𝑖𝑓 offered by the expert 𝐸𝑘 with the risks 𝑅1,𝑓−1 , 𝑅3,𝑓−1 ,
the matrix are deleted. Then the hierarchical structure of 𝑅4,𝑓−1 , . . . , 𝑅𝑑𝑖𝑓 ,𝑓−1 happening simultaneously but no risk
𝑓−1
matrix 𝐷 should be built up according to the order of deleting. 𝑅2,𝑓−1 , 𝑘 = 1, 2, . . . , 𝑛; . . . , 𝑃𝑘 (𝑅𝑖𝑓 = 𝑇 | 𝑅1,𝑓−1 = 𝐹, 𝑅2,𝑓−1 =
Finally, risk assessment should be performed for situation B. 𝐹, . . . , 𝑅𝑑𝑖𝑓 ,𝑓−1 = 𝐹) is the probability of occurrence of the
𝑓−1
risk 𝑅𝑖𝑓 without 𝑅1,𝑓−1 , 𝑅2,𝑓−1 , . . . , 𝑅𝑑𝑖𝑓 ,𝑓−1
offered by the
2.3. Risk Assessment. RAMs for situations A and B are listed 𝑓−1
ture cannot be divided, each risk is regarded as elements of the 𝑅1,𝑓−1 = 𝑇, 𝑅2,𝑓−1 = 𝐹, . . . , 𝑅𝑑𝑖𝑓 ,𝑓−1
= 𝑇), . . . , 𝑃𝑘 (𝑅𝑖𝑓 =
𝑓−1
same layer to process. Firstly, the probability of occurrence of 𝑇 | 𝑅1,𝑓−1 = 𝐹, 𝑅2,𝑓−1 = 𝐹, . . . , 𝑅𝑑𝑖𝑓 = 𝐹) are
,𝑓−1
risk offered by the expert panel (𝑃𝑖1 , 𝑃𝑖2 , . . . , 𝑃𝑖𝑛 ) is combined 𝑘
𝑓−1
𝑘
as comprehensive probability of occurrence of risk 𝑃𝑖𝐺, and recorded as 𝑃 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝑇, . . . , 𝑇), 𝑃 (𝑅𝑖𝑓 = 𝑇 |
the calculating formula is 𝑇, 𝐹, . . . , 𝑇), . . . , 𝑃𝑘 (𝑅𝑖𝑓 = 𝑇 | 𝐹, 𝐹, . . . , 𝐹) for abbreviation.
𝑛
Firstly, the probability vectors 𝑃11 = (𝑝111 , 𝑝211 , . . . , 𝑝𝑑11 ),
𝑃𝑖𝐺 = ∑ 𝑤𝑘 𝑃𝑖𝑘 𝑖 = 1, 2, . . . , 𝑚. (8) 𝑃21 = (𝑝121 , 𝑝221 , . . . , 𝑝𝑑21 ), . . . , 𝑃𝑛1 = (𝑝1𝑛1 , 𝑝2𝑛1 , . . . , 𝑝𝑑𝑛1 ) are
𝑘=1 combined as the comprehensive probability vector 𝑃1 =
(𝑝11 , 𝑝21 , . . . , 𝑝𝑑1 ) and the calculating formula is
Secondly, the loss of risks (𝑙𝑖1 , 𝑙𝑖2 , . . . , 𝑙𝑖𝑛 ) offered by the
𝑛
expert panel is combined as comprehensive loss of risks 𝐿𝐺𝑖,
𝑝𝑗1 = ∑ 𝑤𝑘 𝑝𝑗𝑘1 𝑗 = 1, 2, . . . , 𝑑1 . (11)
and the calculating formula is 𝑘=1
𝑛
𝐿𝐺𝑖 = ∑ 𝑤𝑘 𝑙𝑖𝑘 𝑖 = 1, 2, . . . , 𝑚. (9) Secondly, the conditional probabilities 𝑃1 (𝑅𝑖𝑓 = 𝑇 |
𝑘=1 𝑇, 𝑇, . . . , 𝑇), 𝑃2 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝑇, . . . , 𝑇), . . . , 𝑃𝑛 (𝑅𝑖𝑓 = 𝑇 |
𝑇, 𝑇, . . . , 𝑇) are combined as the comprehensive conditional
Finally, on the basis of the comprehensive probability of probability 𝑃𝐺(𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝑇, . . . , 𝑇); the conditional prob-
occurrence of risk 𝑃𝑖𝐺 and comprehensive loss of risks 𝐿𝐺𝑖, the abilities 𝑃1 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝐹, . . . , 𝑇), 𝑃2 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝐹,
value at risk is calculated, and the calculating formula is
. . . , 𝑇), . . . , 𝑃𝑛 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝐹, . . . , 𝑇) are combined as
the comprehensive conditional probability 𝑃𝐺(𝑅𝑖𝑓 = 𝑇 |
𝑧𝑖 = 𝑃𝑖𝐺𝐿𝐺𝑖 𝑖 = 1, 2, . . . , 𝑚. (10)
𝑇, 𝐹, . . . , 𝑇); . . .; the conditional probabilities 𝑃1 (𝑅𝑖𝑓 = 𝑇 |
𝐹, 𝐹, . . . , 𝐹), 𝑃2 (𝑅𝑖𝑓 = 𝑇 | 𝐹, 𝐹, . . . , 𝐹), . . . , 𝑃𝑛 (𝑅𝑖𝑓 = 𝑇 |
𝐹, 𝐹, . . . , 𝐹) are combined as the comprehensive conditional
Situation B. As for the situation that risk hierarchical struc- probability 𝑃𝑛 (𝑅𝑖𝑓 = 𝑇 | 𝐹, 𝐹, . . . , 𝐹), and the calculating
ture can be divided, 𝑅𝑓 = {𝑅𝑖1 , 𝑅𝑖2 , . . . , 𝑅𝑖𝑚 } is assumed as formulas are as follows:
the set of risk on the layer number 𝑓, and 𝑅𝑖𝑓 is set as the
risk number 𝑖 on the layer number 𝑓, 𝑓 = 1, 2, . . . , 𝑚; 𝑃𝑘1 = 𝑃𝐺 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝑇, . . . , 𝑇)
(𝑝1𝑘1 , 𝑝2𝑘1 , . . . , 𝑝𝑑𝑘1 ) is the probability vector of the occurrence
𝑛 (12a)
of the risk on the first layer (the bottom layer) offered by
= ∑ 𝑤𝑘 𝑃𝑘 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝑇, . . . , 𝑇) ,
the expert 𝐸𝑘 , where 𝑝𝑗𝑘1 is the probability of occurrence of 𝑘=1
4 Mathematical Problems in Engineering
𝑃𝐺 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝐹, . . . , 𝑇) ..
.
𝑛
= ∑ 𝑤𝑘 𝑃𝑘 (𝑅𝑖𝑓 = 𝑇 | 𝑇, 𝐹, . . . , 𝑇) , (12b) 𝑃𝐺 (𝑅1,𝑓−1 = 𝐹, 𝑅2,𝑓−1 = 𝐹, . . . , 𝑅𝑑𝑖𝑓
𝑓−1
,𝑓−1
𝑘=1
Table 1 𝑅1 𝑅2 𝑅3 𝑅4 𝑅5
𝐸1 𝐸2 𝐸3 𝐸4 𝐸5 𝑅1 0 2 3 4 0
𝑃(𝑅1 = 𝑇) [ ]
0.36 0.24 0.28 0.32 0.24 𝑅2 [0 0 0 0 5]
𝑃(𝑅2 = 𝑇) 𝐴3 = [ ],
0.32 0.34 0.25 0.25 0.35
𝑅3 [0 0 0 0 0]
[ ]
[ ]
𝑅4 [0 3 1 0 0]
[ ]
𝑅5 [0 0 1 4 0]
Table 2
𝑙𝑖𝑘 𝐸1 𝐸2 𝐸3 𝐸4 𝐸5 𝑅1 𝑅2 𝑅3 𝑅4 𝑅5
𝑅1 20 22 25 22 22 𝑅1 0 0 5 3 0
𝑅2 28 30 28 29 28 [ ]
𝑅2 [0 0 0 0 5]
𝐴4 = [ ],
𝑅3 [0 0 0 0 0]
[ ]
[ ]
𝑅4 [0 3 4 0 1]
[ ]
𝑛 𝑅5 [0 0 0 1 0]
𝐿𝐺2 = ∑ 𝑤𝑘 𝑙2𝑘
𝑘=1 𝑅1 𝑅2 𝑅3 𝑅4 𝑅5
= 0.25 ∗ 28 + 0.2 ∗ 30 + 0.3 ∗ 28 + 0.1 ∗ 29 𝑅1 0 0 5 2 0
[ ]
+ 0.15 ∗ 28 = 28.5, 𝑅2 [0 0 1 0 5]
𝐴5 = [ ].
𝑅3 [0 0 0 0 0]
[ ]
𝑧2 = 𝑃2𝐺𝐿𝐺2 = 0.29 ∗ 22.3 = 28.8005. [ ]
𝑅4 [0 3 4 0 1]
(15) [ ]
𝑅5 [0 0 0 1 0]
(16)
The risk value of contract modification is much bigger
than the value of bad booking. The company should consider
this result to design the outsourcing plan. Firstly, according to formula (1), the 5 risk-associated
matrixes (𝐴 1 , 𝐴 2 , . . . , 𝐴 5 ) are combined as the group-
Case 2 for Situation B. On the same background, through associated matrix 𝐴 𝐺 = [𝑎𝑖𝑗𝐺]5×5 :
relevant analysis and review of feedback suggestions from
questionnaires, the panel of experts determines 5 outsourcing
risks: bad book (𝑅1 ), contract modification (𝑅2 ), decreased 𝑅1 𝑅2 𝑅3 𝑅4 𝑅5
service quality (𝑅3 ), hidden costs (𝑅4 ), and damaged com-
pany image (𝑅5 ). The 5 risk-associated matrixes offered by the 𝑅1 0 0 3.7 2.9 0
[ ]
experts are as follows: 𝑅2 [0 0 0 0 5]
𝐴𝐺 = [ ]. (17)
𝑅3 [0 0 0 0 0]
[ ]
[ ]
𝑅4 [0 3 1.75 0 0]
[ ]
𝑅1 𝑅2 𝑅3 𝑅4 𝑅5
𝑅5 [0 0 0 0 0]
𝑅1 0 2 3 2 0
[ ]
𝑅2 [ 0 0 0 0 5]
𝐴1 = [ ], Secondly, according to formula (2), the group-associated
𝑅3 [
[ 0 0 0 0 0]
] matrix 𝐴 𝐺 is transformed into regulated group-associated
[ ]
𝑅4 [
[ 0 3 1 0 0]
] matrix 𝐵𝐺 = [𝑏𝑖𝑗𝐺]5×5 ; namely,
𝑅5 [ 0 0 0 1 0]
𝑅1 𝑅2 𝑅3 𝑅4 𝑅5 𝑅1 𝑅2 𝑅3 𝑅4 𝑅5
𝑅1 0 0 4 3 1 𝑅1 0 0 0.56 0.44 0
[ ] [ ]
𝑅2 [ 0 0 2 0 5] 𝑅2 [0 0 0 0 0.76 ]
[ ], 𝐵𝐺 = [ ]. (18)
𝐴2 = [0 0 ]
𝑅3 [
[0 0 0 0 0]
] 𝑅3 [
[
0 0 0 ]
]
[ ] [0
𝑅4 [ 0] 𝑅4 [ 0.45 0.27 0 0 ] ]
[0 3 1 0 ]
𝑅5 [ 0 0 0 1 0] 𝑅5 [0 0 0 0 0 ]
6 Mathematical Problems in Engineering
𝐸1 𝐸2 𝐸3 𝐸4 𝐸5
𝑃(𝑅1 = 𝑇) 0.36 0.24 0.28 0.32 0.24
R4 R2
𝑃(𝑅2 = 𝑇 | 𝑅1 = 𝑇) 0.42 0.44 0.45 0.45 0.45
𝑃(𝑅2 = 𝑇 | 𝑅1 = 𝐹) 0.22 0.20 0.23 0.23 0.22
𝑃(𝑅4 = 𝑇 | 𝑅1 = 𝑇) 0.33 0.35 0.33 0.32 0.36
R1
𝑃(𝑅4 = 𝑇 | 𝑅1 = 𝐹) 0.15 0.12 0.11 0.15 0.14
𝑃(𝑅3 = 𝑇 | 𝑅4 = 𝑇) 0.40 0.38 0.42 0.40 0.40
Figure 1: Hierarchical structure of risks. 𝑃(𝑅3 = 𝑇 | 𝑅4 = 𝐹) 0.20 0.22 0.20 0.23 0.21
𝑃(𝑅5 = 𝑇 | 𝑅2 = 𝑇, 𝑅4 = 𝑇) 0.44 0.42 0.44 0.43 0.44
𝑃(𝑅5 = 𝑇 | 𝑅2 = 𝑇, 𝑅4 = 𝐹) 0.40 0.38 0.37 0.38 0.37
According to formula (3), group comprehensive influence 𝑃(𝑅5 = 𝑇 | 𝑅2 = 𝐹, 𝑅4 = 𝑇) 0.33 0.32 0.28 0.26 0.28
matrix 𝐶 = [𝑐𝑖𝑗 ]5×5 is constructed; that is,
𝑃(𝑅5 = 𝑇 | 𝑅2 = 𝐹, 𝑅4 = 𝐹) 0.18 0.15 0.20 0.16 0.16
𝑅1 𝑅2 𝑅3 𝑅4 𝑅5
Table 4: Loss of risks offered by experts.
𝑅1 0 0.198 0.679 0.44 0.15
[ ] 𝑙𝑖𝑘 𝐸1 𝐸2 𝐸3 𝐸4 𝐸5
𝑅2 [ 0 0 0 0 0.76 ]
𝐶= [ ] . (19) 𝑅1
𝑅3 [ 0 ]
20 22 25 22 22
[ 0 0 0 0 ] 𝑅2
[ ] 28 30 28 29 28
𝑅4 [
[ 0 0.45 0.27 0 0.342 ]
] 𝑅3 56 56 54 55 54
𝑅5 [ 0 0 0 0 0 ] 𝑅4 64 65 66 66 64
𝑅5 80 78 78 80 82
Furthermore, according to the situation of the company,
the thresholds about comprehensive influence matrix pro- Table 5: Probability of occurrence of risks, loss of risks, and value
vided by 5 experts separately are 𝜆 1 = 0.1, 𝜆 2 = 0.1, 𝜆 3 = at risk.
0.15, 𝜆 4 = 0.1, 𝜆 5 = 0.1. According to formula (4), the
group threshold is 𝜆 𝐺 = 0.115. According to formula (5), 𝑃𝑖𝐺 𝑙𝑖𝐺 𝑧𝑖
the comprehensive influence matrix 𝐶 is transformed into 0-1 𝑅1 0.29 22.4 6.5
comprehensive influence matrix 𝐷 = [𝑑𝑖𝑗 ]5×5 ; that is, 𝑅2 0.284 28.5 8
𝑅3 0.245 55 13.5
𝑅1 𝑅2 𝑅3 𝑅4 𝑅5 𝑅4 0.19 65 12.4
𝑅1 1 1 1 1 1 𝑅5 0.253 79.3 20
[ ]
𝑅2 [0 1 0 0 1]
𝐷= [ ]. (20)
𝑅3 [0 0 1 0 0] (𝑙𝑖1 , 𝑙𝑖2 , . . . , 𝑙𝑖5 ) is combined as the comprehensive loss of risks
[ ]
[ ]
𝑅4 [0 1 1 1 1] 𝐿𝐺𝑖, as shown in the third list of Table 3. Finally, according
[ ]
to formula (10), each value at risk 𝑧𝑖 is calculated as shown
𝑅5 [0 0 0 0 1] in fourth list of Table 3. Thus, the calculating results of risk
assessment provide decision support for risk management of
According to matrix 𝐷 and formula (7), the hierarchical the biopharmaceutical company.
structure of 5 kinds of risks is divided as shown in Figure 1.
According to the hierarchical structure of risks shown
in Figure 1, combined with the historical data and reality of 4. Conclusion
the market, the panel of experts offered the probability of
occurrence of the bottom risk 𝑅1 , the conditional probability A risk assessment method in consideration of risk-associated
of occurrence of the risks on the second layer 𝑅2 and 𝑅4 , situations is provided in this paper. Based on various kinds
and the conditional probability of occurrence of the risks on of evaluating information about risks offered by the group
the third layer 𝑅3 and 𝑅5 , as shown in Table 3. The loss of of experts, this method calculates each value at risk through
risks offered by experts (unit: ten thousand Yuan) is listed in identification of hierarchical structure of risks by adopting
𝑖𝑓 knowledge related to probability theory. According to calcu-
Table 4. Furthermore, according to formulas (11)–(14.2𝑑𝑓−1 ), lation analysis, the proposed method is feasible and proved to
the comprehensive probability of occurrence of risks 𝑃𝑖𝐺 is have certain application value. For structured risk assessment
calculated, shown in the second list of Table 5. For example, problems, the proposed method is general. However, the
the probability of occurrence of the risk 𝑅2 is 𝑃2𝐺 = 𝑃𝐺(𝑅1 = requirements of different industries and different types of
𝑇)𝑃𝐺(𝑅2 = 𝑇 | 𝑅1 = 𝑇) + 𝑃𝐺(𝑅1 = 𝐹)𝑃𝐺(𝑅2 = 𝑇 | business for service outsourcing are not the same. So the
𝑅1 = 𝐹) = 0.284. According to formula (9), the loss of risks method we give in this paper may not be directly applicable
Mathematical Problems in Engineering 7
to all industries or all types of enterprises, especially in some [14] E. G. Baranoff and T. W. Sager, “The relations among asset risk,
special industries. As service outsourcing progresses, some product risk, and capital in the life insurance industry,” Journal
of the expected outsourcing risks will change, and on the of Banking & Finance, vol. 26, no. 6, pp. 1181–1197, 2002.
other hand, new and unpredictable outsourcing risks will [15] M. Abdellaoui, P. Klibanoff, and L. Placido, “Experiments on
emerge. In order to ensure the smooth realization of service compound risk in relation to simple risk and to ambiguity,”
outsourcing, the enterprise risk control during the whole Management Science, vol. 61, no. 6, pp. 1306–1322, 2015.
service outsourcing process is worth studying. Also further [16] X.-W. Liao, W. Tao, and L. Yuan, “A Bayesian network model
research is required for large calculation amount of the risk under group decision making for evaluating IT outsourcing
occurrence. risk,” in Proceedings of the International Conference on Risk
Management & Engineering Management (ICRMEM ’08), pp.
559–564, Beijing, China, November 2008.
Competing Interests [17] G. Büyüközkan and D. Ruan, “Choquet integral based aggre-
gation approach to software development risk assessment,”
The authors declare that there is no conflict of interests Information Sciences, vol. 180, no. 3, pp. 441–451, 2010.
regarding the publication of this paper.
References
[1] M. Doumpos and C. Zopounidis, “Assessing financial risks
using a multicriteria sorting procedure: the case of country risk
assessment,” Omega, vol. 29, no. 1, pp. 97–109, 2001.
[2] T. M. Williams, “Risk-management infrastructures,” Interna-
tional Journal of Project Management, vol. 11, no. 1, pp. 5–10,
1993.
[3] S. Piramuthu, “On preprocessing data for financial credit risk
evaluation,” Expert Systems with Applications, vol. 30, no. 3, pp.
489–497, 2006.
[4] M. Farrell and R. Gallagher, “The valuation implications of
enterprise risk management maturity,” Journal of Risk and
Insurance, vol. 82, no. 3, pp. 625–657, 2015.
[5] J. C. Wei, Z. J. Li, L. Q. Shi et al., “Comprehensive evaluation
of water-inrush risk from coal floors,” Mining Science and
Technology (China), vol. 20, no. 1, pp. 121–125, 2010.
[6] M. Nazam, J. Xu, Z. Tao, J. Ahmad, and M. Hashim, “A fuzzy
AHP-TOPSIS framework for the risk assessment of green sup-
ply chain implementation in the textile industry,” International
Journal of Supply and Operations Management, vol. 2, no. 1, pp.
548–568, 2015.
[7] B. Bahli and S. Rivard, “Validating measures of information
technology outsourcing risk factors,” Omega, vol. 33, no. 2, pp.
175–187, 2005.
[8] C. Oksel, C. Y. Ma, and X. Z. Wang, “Structure-activity rela-
tionship models for hazard assessment and risk management of
engineered nanomaterials,” Procedia Engineering, vol. 102, no. 1,
pp. 1500–1510, 2015.
[9] K. L. Astles, M. G. Holloway, A. Steffe, M. Green, C. Ganassin,
and P. J. Gibbs, “An ecological method for qualitative risk
assessment and its use in the management of fisheries in New
South Wales, Australia,” Fisheries Research, vol. 82, no. 1–3, pp.
290–303, 2006.
[10] A. S. Markowski and M. Sam Mannan, “Fuzzy risk matrix,”
Journal of Hazardous Materials, vol. 159, no. 1, pp. 152–157, 2008.
[11] D. Donoho and M. Gavish, “Minimax risk of matrix denoising
by singular value thresholding,” The Annals of Statistics, vol. 42,
no. 6, pp. 2413–2440, 2014.
[12] A. Khashman, “Neural networks for credit risk evaluation:
investigation of different neural models and learning schemes,”
Expert Systems with Applications, vol. 37, no. 9, pp. 6233–6239,
2010.
[13] E. Angelini, G. Tollo, and A. Roli, “A neural network approach
for credit risk evaluation,” The Quarterly Review of Economics
and Finance, vol. 48, no. 4, pp. 733–755, 2008.
Advances in Advances in Journal of Journal of
Operations Research
Hindawi Publishing Corporation
Decision Sciences
Hindawi Publishing Corporation
Applied Mathematics
Hindawi Publishing Corporation
Algebra
Hindawi Publishing Corporation
Probability and Statistics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014
International
Journal of Journal of
Mathematics and
Mathematical
Discrete Mathematics
Sciences