Supply Chain Management: An International Journal

Risk management models for supply chain: a scenario analysis of outsourcing to China
David L. Olson Desheng Wu
Article information:
To cite this document:
David L. Olson Desheng Wu, (2011),"Risk management models for supply chain: a scenario analysis of outsourcing to China",
Supply Chain Management: An International Journal, Vol. 16 Iss 6 pp. 401 - 408
Permanent link to this document:
http://dx.doi.org/10.1108/13598541111171110
Downloaded on: 29 November 2014, At: 17:07 (PT)
References: this document contains references to 45 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 4878 times since 2011*
Users who downloaded this article also downloaded:
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

Charlene Xie, Chimay J. Anumba, Tzong#Ru Lee, Rao Tummala, Tobias Schoenherr, (2011),"Assessing and managing risks
using the Supply Chain Risk Management Process (SCRMP)", Supply Chain Management: An International Journal, Vol. 16 Iss
6 pp. 474-483 http://dx.doi.org/10.1108/13598541111171165
Peter Finch, (2004),"Supply chain risk management", Supply Chain Management: An International Journal, Vol. 9 Iss 2 pp.
183-196 http://dx.doi.org/10.1108/13598540410527079
Richard Wilding, Beverly Wagner, Claudia Colicchia, Fernanda Strozzi, (2012),"Supply chain risk management: a new
methodology for a systematic literature review", Supply Chain Management: An International Journal, Vol. 17 Iss 4 pp. 403-418
http://dx.doi.org/10.1108/13598541211246558

Access to this document was granted through an Emerald subscription provided by 101358 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service
information about how to choose which publication to write for and submission guidelines are available for all. Please visit
www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of
more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online
products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics
(COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

Research paper

Risk management models for supply chain:

a scenario analysis of outsourcing to China
David L. Olson
Department of Management, University of Nebraska, Lincoln, Nebraska, USA, and
Desheng Wu
RiskLab, University of Toronto, Toronto, Canada

Abstract
Purpose – A key process involved in supply chains is a priori evaluation of potential partners, not only in terms of expected cost (which includes
exchange rate risk), but also in terms of other risks. These risks can include product failure, producing company failure (such as bankruptcy), and even
political risk. This paper aims to compare tools to aid supply chain organizations in measuring, evaluating, and assessing risk in this environment.
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

Design/methodology/approach – The authors demonstrate the use of DEA, followed by a DEA simulation model and also a Monte Carlo simulation
using a risk-adjusted cost concept. Once non-dominated partners are identified by DEA, simulation analysis is applied to compare expected
performance of vendors, and the range of expected outcomes can be identified, aiding supply chain core organizations to better select producing
partners.
Findings – The authors consider strategies of outsourcing to China, as well as other nations under various forms of risk. A scenario analysis using risk
management models indicates outsourcing to Great China is a good strategy.
Originality/value – The authors conducted a thorough review of supply chain risk management and identified criteria and various risk performance
measures for outsourcing under risk and uncertainty in a supply chain. The benefit of outsourcing to China is discussed. The authors have designed an
international outsourcing problem, where foreign exchange risk, product failure, organizational failure, and political risks are considered.

Keywords Supply chain outsourcing, Recall risk, Data envelopment analysis, Monte Carlo simulation, Supply chain management, Product recall

Paper type Research paper

1. Introduction 10 percent of total trade in the world and has become the
largest recipient of foreign direct investment (FDI) from
Supply chains have proven instrumental in improving OECD nations. Recently, the risks of recall have become a
efficiency within many industries, to include retail (witness major concern in supply chains. Drysdale (2008) estimated
Wal-Mart) and computers (witness Dell). But supply chain direct losses of over $100 billion due to recalls, increased
organizations involve many risks (Lucas et al., 2007). These warranties, and rework, along with additional indirect loss to
can be disruption (Kull and Closs, 2008), product failure security, image, and customer confidence. There have been
(Wade, 2007), or other forms of risk. While these risks can be widely publicized examples including recall of Chinese
daunting, doing business requires acceptance of some level of produced toys by Mattel Corporation (Enderwick, 2008),
risk within the organization’s area of expertise. Sustainable heparin contamination leading to recalls of drugs in over ten
development within supply chains has been proposed (Matos European countries (Payne, 2008), and melanine
and Hall, 2007). This paper demonstrates two tools (data contamination in Unilever supply chain partners in tea-
envelopment analysis – DEA and simulation) that can aid related products (Madden, 2008).
supply chain organizations in measuring, evaluating, and Responding to this developing problem, the US Food and
assessing risk involved with supply chain outsourcing. Drug Administration is working on ways to improve
Because of a large pool of low wage and efficient labor, traceability and ensure supply chain safety (Lewis, 2008).
China has become very attractive to companies seeking to Roth et al. (2008) proposed a six-sigma based framework of
outsource products and services. China’s trade is now around food supply chain control, relying on traceability,
transparency, testability, time, trust, and training. These
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1359-8546.htm
The authors would like to thank the anonymous reviewers for their
constructive and insightful comments that improved the quality and
presentation of this paper. They also appreciate the efforts and time from
the Guest Editor, Dr Charlene Xie, for managing the review process. In
Supply Chain Management: An International Journal
16/6 (2011) 401– 408 addition, this paper is particularly supported by “Aim for the Top
q Emerald Group Publishing Limited [ISSN 1359-8546] University Plan” of the National Sun Yat-Sen University and Ministry of
[DOI 10.1108/13598541111171110] Education, Taiwan (NSC 99-2410-H-110-053-MY3).

401
Risk management models for supply chain Supply Chain Management: An International Journal
David L. Olson and Desheng Wu Volume 16 · Number 6 · 2011 · 401 –408

governmental actions, especially in the US and in Europe, quality of products, and avoidance of highly publicized events
threaten to make supply chain operations more complex and such as lead paint on toys.
costly (Hoffman, 2008). There have been reports of US
manufacturer investment in China declining due to recalls 2.2 Supply chain risk categories
such as those cited in the prior paragraph (Sowinski, 2008). Internal sources of supply chain uncertainty include capacity
Models aid in the analysis of outsourcing opportunities. We availability, information delays, and regulatory compliance
demonstrate the use of DEA, followed by a DEA simulation (Cucchiella and Gastaldi, 2006). External sources include
model and also a Monte Carlo Simulation using a risk- competitor actions, political environment, market price
adjusted cost concept. DEA method aids the buyer in fluctuations, uncertain costs, and supplier quality.
classifying the suppliers (or their initial bids) into two Outsourcing increases some external uncertainties for core
categories: the efficient suppliers and the inefficient suppliers. supply chain entities (for instance reliability of supply,
Weber has primarily discussed the application of DEA in compliance with quality) and reduces others (outsourcing
supplier selection in several publications; see Weber and will be expected to yield lower costs, which reduces the
Ellram (1993), and Weber and Desai (1996). Apart from probabilities of losing customers in all likelihood).
simply categorizing suppliers, Weber demonstrated how DEA Table I is based on supply chain risks given by Cucchiella
can be used as a tool for negotiating with inefficient suppliers. and Gastaldi (2006). We focus on those risks that are
However, classical DEA often fails to work effectively since expected to have impact in supply chain outsourcing.
classical DEA is very sensitive to statistical noise (Wu, 2009a),
which motivates the utilization of a DEA simulation model. 2.3 Responses to risk
We do this with hypothetical data, with the intent of showing Supply chain organizations have a number of responses
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

how alternative vendors in supply chains can be evaluated. available to manage and mitigate risks. Insurance is risk
Using hypothetical data is somewhat lack of empirical mitigation by definition. But other means can be used, to
evidence, but can keep approaches and insights in a general include information sharing schemes (Olson and Wu, 2007;
framework. Real data would be specific to each organization Olson and Xie, 2010). With respect to outsourcing, different
making the selection decision. The rest of the paper is levels of coordination can still be applied, although the term
organized as follows. Section 2 of the paper presents outsourcing implies a looser degree of control. Still,
outsourcing risk categories. Section 3 discusses outsourcing organizations such as Wal-Mart require their supply chain
to China. Section 4 presents simulation of DEA Model results participants to have compatible information systems
and analysis, Section 5 presents Monte Carlo Simulation for (Fishman, 2006). Wal-Mart also has worked toward greater
Analysis, and section 6 concludes the paper. use of technology such as radio-frequency identification.
Other responses can include development of closer
relationships with outsourcing agencies, imposition of
2. Outsourcing risk categories performance standards, joint training, joint strategy
development, and marketing initiatives (Ritchie and
Outsourcing refers to subcontracting a process such as
Brindley, 2007). Kleindorfer and Saad (2005) found trends
product design or manufacturing, to a third-party company.
from more independent responses, such as insurance and
It is very important in cost-effective supply chains, and
contractual standards, to more cooperative efforts (such as
involves a number of inherent risks. Ritchie and Brindley
shared information or development of closer relationships).
(2007) gave a framework for supply chain risk management.
Outsourcing reduces many risks to core organizations.
This framework included a process including identification of
Cucchiella and Gastaldi (2006) classified benefits of
risk sources, consequences, evaluation of management
outsourcing in terms of reducing financial exposure and
responses, and measurement of performance outcomes.
overcoming limits of productive capacity on the part of core
organizations. Outsourcing can allow for easier compliance
2.1 Risk sources and consequences with local regulations. Outsourcing also makes it possible to
Risks in supply chains can come from a number of sources. react to market timing, as the core organization could be more
The first step of the process is to identify risks associated with agile in terms of response to market demand than they would
a specific operation. These can arise from the environment, be if they had to construct all facilities needed throughout the
and can be specific to particular industries. Supply chain supply chain. Outsourcing organizations could also avoid
features involving risk include supply chain configuration. many political problems.
Outsourcing involves greater opportunities to take advantage
of low cost solutions, but may well involve riskier supply chain
participants. Organizations can adopt different strategies,
3. Outsourcing to China
some of which involve more risk than others. Each problem Outsourcing can be domestic when the company providing
will have specific risks associated with it. goods or services is located inside the same country; or
Ritchie and Brindley (2007) labeled those risks having international when the company providing goods or services is
significant impact on likelihood of negative consequences to outside the country of the purchaser. Recent research indicates
be risk drivers. Guinipero and Eltantawy (2004) cited product that important factors that attract various companies to seek
technology, security needs, and experience of both the core outsourcing activities in China involve improved infrastructure,
supply chain entity and outsourcing organizations as creators growing reputation for product quality and on time delivery, the
of risk. Risk drivers are those sources of risk with higher reliability of middlemen to organize transactions and improved
probability and/or impact. Risk performance of course legal structure (Whalley, 2007).
includes financial performance, but also includes The recent growth in Chinese trade can be taken as an
nonfinancial factors, to include avoidance of disruption, evidence of growth of outsourcing to China from outsiders

402
Risk management models for supply chain Supply Chain Management: An International Journal
David L. Olson and Desheng Wu Volume 16 · Number 6 · 2011 · 401 –408

Table I Supply chain risks assessed for outsourcing impact

Supply chain risk Elaboration Outsourcing impact
Accounting risk Risk of ruin High
Asset impairment risk Utilization of assets Increased risk to core entity
Characteristics of supplier Can select most innovative outsourcing organization
Country Risk propensity can vary by country Outsourcing opportunities with low cost tend to be in riskier locations
Competitive risk Need to differentiate products Outsourced products available to competitors
Customer risk Customer likelihood of placing orders; product Low quality products may drive away customers;
obsolescence Outsourcing may reduce risk of product obsolescence
Downside risk Risk of negative outcome (failure) Outsourcing vendors can be replaced
Financial risk Loss due to financial markets Core entity less likely to be brought down by outsourcing vendor failure than
by subordinate failure
Interaction Communication coordination Outsourcing vendors are more independent; can impose requirements for
shared systems
Legal risk Litigation exposure Risk lower for core entity, as burden shifted to outsourcing vendor
Product risk Product technical complexity Core entities should make sure outsourcing organization is competent
Regulatory risk Core entity risk lower, as outsourcing vendors assume local risk
Reputation risk Customer confidence Higher to core entities, as customers will hold core entity responsible
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

Shared risk Outsourcing allows access of all potential vendors in the market
Supplier size Smaller organizations have greater risk of performance
Supply disruption If outsourcing supplier fails, can be replaced
Source: Developed from Cucchiella and Gastaldi (2006)

because the trade growth reflects low wage production, which as well. Shareholders care about return on capital. Lenders care
also drives outsourcing. China became the number one about lending security assurance. Suppliers care about
exporter of technology products in the world, with the total assurance of being paid, as well as about future opportunities.
amount over US$ 400 billion in 2004. The trend of Employees care about future employment. An effective system
outsourcing to China is also obvious from the evidence of needs to take care of risk issues in all of these areas, as well as other
fast FDI growth in China: the FDI growth rate being aspects of supply chains that can become problems.
15 percent, 13 percent and 14 percent in three successive Many different types of models have been proposed in the
years after China joined the WTO in 2001. literature. Because of the uncertainty involved, statistical
China has become the global manufacturing center and the analysis and simulation is very appropriate to consider supply
winner of most IT outsourcing contracts from developed chain risk. We will only report a few of the many studies,
Asian countries such as Japan and South Korea. Fierce relying on more recent articles. Bayesian analysis has been
competition exists between China and the traditional proposed to model information and knowledge integration
outsourcing destination countries such as India, Ireland and within complex networks (Li and Chandra, 2007). Simulation
Israel, for the much bigger and more profitable North was proposed in a number of studies, to include discrete-event
American and European market. Outsourcing to China can simulation to estimate survival over long-range periods given
take advantage of low labor cost, while suffering from other assumed probabilities of supply chain linkage failure (Klimov
business risks. As to IT outsourcing problems, two types of and Merkuryev, 2008). Wu and Olson (2008b) used Monte
risks are perceived in international business operations in Carlo simulation to evaluate risks associated with vendor
China (Olson and Wu, 2007). First, because China is not a selection, following up on similar modeling from many
full market economy based on a democratic political system, sources. System dynamics models have been widely used,
there is some political risk in the government’s interfering especially with respect to the bullwhip-effect (Towill and
with free enterprises. Such risks are deemed negligible based Disney, 2008) and to model environmental, organizational,
on the open and reform policies of the central government in and network related risk issues (Kara and Kayis, 2008).
the past two decades, and the economic boom derived from a Other modeling approaches have been applied to supply
more transparent political environment. Second, whereas chain risk as well. Tang et al. (2008) applied a fuzzy genetic
China’s lack of protection of intellectual properties is widely algorithm approach to evaluate logistics strategies to lower
reported, there have been very few cases where business supply chain risks. As mentioned in the cases section,
software was pirated. Using unreliable pirated software can Hewlett-Packard applied regression models as the basis of
result in organizational failure and product failure. forecasting, and optimization models to select how much to
Performance measures have been proposed to evaluate purchase from each available source (Nagali et al., 2008).
outsourcing strategies accommodating risk factors. Bogataj and Bogataj (2007) used parametric linear
programming based on net present value to estimate supply
3.1 Risk performance measures chain vulnerability. Goh et al. (2007) applied a stochastic bi-
The effectiveness of risk management measures can be measured criterion algorithm to analyze a multi-stage global network
at least subjectively in the form of expert opinion. These problem with objectives of profit maximization and risk
measures include not only financial aspects, but other risk factors minimization. Many studies applied analytic hierarchy

403
Risk management models for supply chain Supply Chain Management: An International Journal
David L. Olson and Desheng Wu Volume 16 · Number 6 · 2011 · 401 –408

process, to include recent studies such as assessment of an

offshoring decision (Schoenherr et al., 2008), the similar max
m ;n
mT Y 0
decision to select suppliers (Kull and Talluri, 2008), overall
s:t: vT X 0 ¼ 1
supply chain risk evaluation (Gaudenzi and Borghesi, 2006), ð2Þ
and inbound supply risk evaluation (Wu et al., 2006). mT Y 2 vT X # 0
Blackhurst et al. (2008) presented a study considering ~ 2 vT # 21 · 1~
2 mT # 21 · 1
multiple objectives for supplier risk assessment utilizing a
generic multiple criteria analysis similar to the simple multi- where m and v are weight vectors attached to the output and
attribute rating theory (SMART) method. input respectively. Performing a DEA analysis requires the
We will demonstrate two analytic approaches useful in solution of n linear programming problems of the above form
supporting risk management using a scenario involving (1) or (2), one for each DMU. The optimal value of the
selection of outsourcing supply vendors. We will consider objective function indicates the proportional reduction of all
ten available alternatives for delivery of a small product inputs for DMU0 that will move it onto the frontier which is
component. These alternatives represent hypothetical the envelopment surface defined by the efficient DMUs in the
suppliers from China, Indonesia, Taiwan, Vietnam, sample. A DMU is termed efficient if and only if the optimal
Germany, and the United States. Each of these outsourcing value u * is equal to 1 and all the slack variables are zero.
options have expected characteristics over a number of Each DEA model seeks to determine which of the n
features. A model is proposed to identify non-dominated candidate vendors define an envelopment surface that
alternatives (Edwards, 1977; Keeney and Raiffa, 1976; Olson, represents best practice, referred to as the efficient frontier.
1996; Wu, 2009b). We will demonstrate how simulation can Units that lie on the non-dominated surface are deemed
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

be used to better understand uncertainties involved in price efficient in DEA while those units that do not, are termed
(and vendor survival relative frequencies). We will inefficient. DEA provides a comprehensive analysis of relative
demonstrate how DEA and simulation models can be used efficiencies for multiple input-multiple output situations by
together to focus on multiple aspects of decision problems, evaluating each vendor and measuring its performance
and trade-off analysis to rationally select preferred options. relative to an envelopment surface composed of other
Note that practical data on outsourcing for China are very vendors. DEA “scans” the available suppliers and identifies
difficult to obtain, and we thus use artificial data for the efficient subset of suppliers. Those vendors known as the
demonstration purpose. The data used in this paper is efficient reference set are the peer group for the inefficient
purely hypothetical, and does not imply relative performance units. Thus, a DEA model can serve as a “filter” before the
for specific organizations in any country. determination of final selected suppliers. Various DEA
models are available and many model extensions can be
used to provide a more comprehensive review of candidate
4. Simulation of DEA Model vendor performance. For example, we can use categorical
variables such as the competitive environment and vendor
4.1 DEA Model
location to enhance the model and add additional variables to
DEA can be used for a comparison of alternatives over multiple
either model if more vendors become available. We can also
criteria and provides a useful complement to the available
add weight restrictions to the models to refine the results and
decision models for outsourcing decision (Talluri et al., 2006;
lead to more realistic recommended improvements. We will
Wu, 2007). DEA is used to establish a best practice group
develop DEA simulation models to implement some
amongst a set of observed units and to identify the units that are
extensions in Section 4.2.
inefficient when compared to the best practice group. DEA also
indicates the magnitude of the inefficiencies and improvements
possible for the inefficient units. Consider n DMUs to be 4.2 Analytic procedure
evaluated, DMUj (j ¼ 1; 2. . .n) that consumes the amounts Xj ¼ Step 1: Define the probability distribution of exchange rate,
{xij } of m different of inputs (i ¼ 1; 2; . . . ; m) and produces the product failure, organizational failure, and political failure;
amounts Yj ¼ {yrj } of r outputs (r ¼ 1; . . . ; s). Denote by Calculate price after exchange rate, relative frequency of
X ¼ {Xj ; j ¼ 1; 2. . .n}, Y ¼ {Yj ; j ¼ 1; 2. . .n}, the input oriented product survival, organizational survival, and political
survival. Set the number of simulation runs. These are the
efficiency of a particular DMU0 under the assumption of
inputs to the DEA model.
constant returns to scale can be obtained from the following
Step 2: Generate experimental data according to defined
primal-dual linear programs (input-oriented CCR model
probability distributions.
(Charnes et al., 1978)).
Step 3: Denote by i the indicator of the candidate supplier,
set the initial value of i to unity.
~ þ 2 1 · 1s
min u 2 1 · 1s ~2 Step 4: Solve the following linear program:
u;l;sþ ;s2

s:t: Yl 2 sþ ¼ Y 0 ~ þ 2 1 · 1s
min zi ¼ ui 2 1 · 1s ~2
ð1Þ ui ;l;sþ ;s2
uX 0 2 X l 2 s2 ¼ 0
s:t:Yl 2 sþ ¼ Y i
l; sþ ; s2 $ 0
ui X i 2 X l 2 s2 ¼ 0
l; sþ ; s2 $ 0
where s þ and s- are the slacks in the system, l is a multiplier
*
vector with n elements, 1 is the non-Archimedean Denote by ui the optimal objective value to above linear
infinitesimal. program.

404
Risk management models for supply chain Supply Chain Management: An International Journal
David L. Olson and Desheng Wu Volume 16 · Number 6 · 2011 · 401 –408

Step 5: Set i ¼ i þ 1. If i . total number of vendors, go to the times of simulations change, the above analysis does not
Step 6; Otherwise go to Step. change. This is consistent with that resulted from a Monte
Step 6: Find the value of i that achieves the maximum for all Carlo simulation in Section 5.
u*i , if the number of simulation runs is achieved, go to Step 7;
otherwise, go to Step 2. 5. Monte Carlo simulation for analysis
Step 7: Generate statistical results after all simulation runs
are done. Simulation models are sets of assumptions concerning the
To investigate and illustrate the performance of the relationship among model components, usually described in
candidate vendors and compare the results to those from terms of probability distributions. Simulations can be time-
other models, we have carried out three simulation scenarios: oriented (for instance, involving the number of events such as
100 simulation runs, 500 simulation runs and 1,500 demands in a day, or system dynamics models involving
simulation runs. Input and output data are randomly feedback) or process-oriented (for instance, involving queuing
generated using the assumed distributions in each of these systems of arrivals and services). Uncertainty can be included
simulation runs. by using probabilistic inputs for elements such as demands,
DEA was run using: inter-arrival times, or service times. These probabilistic inputs
.
price after exchange rate; need to be described by probability distributions with
.
probability of product survival; specified parameters. Probability distributions can include
.
probability of organizational survival; and normal distributions (with parameters for mean and
. probability of political survival. variance), exponential distributions (with parameter for
mean), lognormal (parameters mean and variance), or any
Table II gives these parameter values, which will also be used
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

of a number of other distributions. A simulation run is a

in a Crystal Ball model later. In Table II, expected prices are sample from an infinite population of possible results for a
generated by taking quoted price, inflating by mean exchange given model. After a simulation model is built, the number of
rate and dividing by the probability of avoiding failure from trials is established. Statistical methods are used to validate
the three failure categories of product, organizational, and simulation models and design simulation experiments.
political failure. This is the standard treatment for probability Many financial simulation models can be accomplished on
and statistical analysis. While actual data may follow other spreadsheets, such as Excel. There are a number of
distributions better than the normal, our purpose is to commercial add-on products that can be added to Excel,
demonstrate, so we use the more widely known normal such as @Risk or Crystal Ball, that vastly extend the
distribution. simulation power of spreadsheet models (Evans and Olson,
We estimated the DEA efficiency score given to each vendor 2002). These add-ons make it very easy to replicate
each time we run DEA. We finally calculated the average DEA simulation runs, and include the ability to correlate
efficiency scores and standard deviation in the total runs. Table variables, expeditiously select from standard distributions,
III presents the average efficiency score, standard derivation, the aggregate and display output, and other useful functions.
95 percent of confidence interval for the mean and the number of In supply chain outsourcing decisions, a number of factors
efficient DMUs in the DEA simulation models. can involve uncertainty, and simulation can be useful in
From Table III, it is evident that based on the DEA gaining better understanding of systems (Wu and Olson,
simulation, the vendor with the worst performance was from 2008a). We begin by looking at expected probability
Germany because its average efficiency score is less than 0.24 distributions of prices for the component to be outsourced
in all three scenarios of DEA simulation. Table III gives five from each location. China C in this case has the lowest
dominated solutions: Taiwan, Germany, Alabama, Ohio, and estimated price, but it has a wide expected distribution of
Arizona because none of them achieved efficiency among even exchange rate fluctuation. These distributions will affect the
1,500 simulation runs. All other five suppliers have some actual realized price for the outsourced component. The
chance of being a winner during the simulations. The best two Chinese C vendor is also rated as having relatively high
performers are from Chinese B and C with mean scores being probabilities of failure in product compliance with contractual
around 0.8 (the probability of being an efficient solution). As standards, in vendor financial survival, and in political

Table II Simulation distributions and probabilities

Quoted Product Organizational Political Expected
Outsourcing vendor price Exchange distribution failure failure failure price
A China A 0.82 Normal(1.3,0.2) 0.10 0.15 0.05 1.47
B China B 0.70 Normal(1.3,0.2) 0.08 0.11 0.05 1.17
C China C 0.60 Normal(1.3,0.2) 0.13 0.17 0.05 1.14
D Indonesia 0.80 Normal(1.2,0.12) 0.16 0.10 0.07 1.37
E Taiwan 1.36 Normal(1.03,0.02) 0.01 0.01 0.10 1.59
F Vietnam 0.85 Normal(1.1,0.1) 0.15 0.25 0.05 1.54
G Germany 3.20 Normal(1.05,0.02) 0.01 0.02 0.01 3.50
H Alabama 2.05 1 0.03 0.20 0.03 2.72
I Arizona 1.80 1 0.04 0.15 0.03 2.27
J Ohio 2.50 1 0.02 0.25 0.03 3.51

405
Risk management models for supply chain Supply Chain Management: An International Journal
David L. Olson and Desheng Wu Volume 16 · Number 6 · 2011 · 401 –408

Table III Simulation of DEA Model using probability of survival

Supplier ChinaA ChinaB ChinaC Indonesia Taiwan Vietnam Germany Alabama Arizona Ohio
100 simulation runs
Mean 0.679 0.81 0.865 0.757 0.571 0.734 0.236 0.363 0.41 0.301
STD 0.221 0.186 0.175 0.198 0.149 0.195 0.061 0.092 0.103 0.076
No. efficient 14 28 44 15 0 12 0 0 0 0
Max 1 1 1 1 0.942 1 0.378 0.58 0.654 0.481
Confidence level (95 percent) 0.044 0.037 0.035 0.039 0.029 0.039 0.012 0.018 0.021 0.015
500 simulation runs
Mean 0.663 0.795 0.861 0.734 0.541 0.704 0.224 0.346 0.391 0.287
STD 0.218 0.206 0.179 0.206 0.143 0.203 0.058 0.09 0.102 0.075
No. efficient 78 176 232 80 0 75 0 0 0 0
Max 1 1 1 1 0.887 1 0.398 0.568 0.642 0.471
Confidence level (95 percent) 0.019 0.018 0.016 0.018 0.013 0.018 0.005 0.008 0.009 0.007
1,500 simulation runs
Mean 0.667 0.79 0.866 0.733 0.543 0.707 0.224 0.347 0.392 0.288
STD 0.22 0.206 0.182 0.206 0.146 0.201 0.06 0.092 0.104 0.076
No. efficient 223 528 696 246 0 225 0 0 0 0
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

Max 1 1 1 1 0.968 1 0.383 0.594 0.67 0.492

Confidence level (95 percent) 0.011 0.01 0.009 0.01 0.007 0.01 0.003 0.005 0.005 0.004

stability of host country. The simulation is modeled to Figure 2 Distribution of results for Taiwanese vendor costs
generate 10,000 samples of actual realized price after
exchange rate variance, to include having to rely upon an
expensive ($5 per unit) price in case of outsourcing vendor
failure. The $5 is arbitrary, and effective as long as it exceeds
quoted prices as it does here.
Monte Carlo simulation output is exemplified in Figure 1,
which shows the distribution of prices for the hypothetical
Chinese outsourcing vendor C, which was the low price
vendor very nearly half of the time. Figure 2 shows the same
for the Taiwanese vendor, and Figure 3 for the safer but
expensive German vendor.
The Chinese vendor C has a higher relative frequency of
failure (over 0.31 from all sources combined, compared to
0.30 for Indonesia). This raises its mean cost, because in case
of failure, the $5 per unit default price is used. There is a
cluster around the contracted cost of $0.60, with a minimum
dropping slightly below 0 due to exchange rate variance, a
Figure 3 Distribution of results for Germany vendor costs
mean of $0.78 and a maximum of $1.58 given survival in all
three aspects of risk modeled. There is a spike showing a
default price of $5.00 per unit in 0.3134 of the cases. Thus

Figure 1 Distribution of results for Chinese vendor C costs

while the contractual price is lowest for this alternative, the

average price after consideration of failure is $2.10.
Table IV shows comparative output. Simulation provides a
more complete picture of the uncertainties involved.

406
Risk management models for supply chain Supply Chain Management: An International Journal
David L. Olson and Desheng Wu Volume 16 · Number 6 · 2011 · 401 –408

Table IV Simulation output

Relative frequency AvgCost if did
Vendor Mean cost Min cost Max cost of failure Probability low not fail Average overall
China B 0.70 20.01 1.84 0.2220 0.1370 0.91 1.82
Taiwan 1.36 1.22 1.60 0.1180 0.0033 1.41 1.83
China C 0.60 0.05 1.58 0.3134 0.4939 0.78 2.10
China A 0.82 20.01 2.16 0.2731 0.0188 1.07 2.14
Indonesia 0.80 0.22 1.61 0.2971 0.1781 0.96 2.16
Arizona 1.80 1.80 1.80 0.2083 0.0001 2.71 2.47
Vietnam 0.85 0.40 1.49 0.3943 0.1687 0.94 2.54
Alabama 2.05 2.05 2.05 0.2472 0 2.78
Ohio 2.50 2.50 2.50 0.2867 0 3.22
Germany 3.20 2.90 3.81 0.0389 0 3.42
Note: Average overall assumes cost of $5 to supply chain should vendor fail

Relative frequencies of being the low-cost alternative are also References

shown. The greatest relative frequency was for China C at
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

0.4939, with Indonesia next at 0.1781. The expensive (but Blackhurst, J.V., Scheibe, K.P. and Johnson, D.J. (2008),
safer) alternatives of Germany and Alabama both were never “Supplier risk assessment and monitoring for the
low (and thus were dominated in the DEA model). But automotive industry”, International Journal of Physical
Germany had a very high relative frequency of survival, and in Distribution & Logistics Management, Vol. 38 No. 2,
the simulation could appear as the best choice (rarely). Should pp. 143-65.
vendor fails, the average overall costs (risk-adjusted) from Bogataj, D. and Bogataj, M. (2007), “Measuring the supply
Germany, Ohio and Alabama are the highest, which verifies chain risk and vulnerability in frequency space”,
results from Table III where efficiency scores given to these International Journal of Production Economics, Vol. 108
three are far less than 0.39 and thus are dominated. Vendors Nos 1/2, pp. 291-301.
from the Great China yields lower risk-adjusted costs, which is Charnes, A., Cooper, W.W. and Rhodes, E. (1978),
to some extent consistent to DEA results in Table III. “Measuring the efficiency of decision-making units”,
European Journal of Operational Research, Vol. 3 No. 2,
6. Conclusions pp. 429-44.
Cucchiella, F. and Gastaldi, M. (2006), “Risk management in
Outsourcing within supply chains has grown in importance, as supply chain: a real option approach”, Journal of
the Internet opens communications around the globe, and as Manufacturing Technology Management, Vol. 17 No. 6,
international business rapidly develops. Outsourcing offers pp. 700-20.
many cost advantages as low-cost producers anywhere can Drysdale, C. (2008), “Counterfeiting a ‘real’ issue”, Circuits
compete. There is greater risk in outsourcing, but this risk is Assembly, Vol. 19 No. 5, p. 12.
reduced with the ability to communicate in real time, and as Edwards, W. (1977), “How to use multiattribute utility
international production standards are expressed and measurement for social decisionmaking”, IEEE Transactions
outsourcing vendors comply with them. We used hypothetical on Systems, Man, and Cybernetics, Vol. 7 No. 5, pp. 326-40.
data to demonstrate risk aspects of the outsourcing decision. Enderwick, P. (2008), “Quality assurance and upgrading in
This paper presented DEA as a way to consider cost along
global supply chains: implications for management in a
with various aspects of risk in a priori selection of supply chain
transitory economy”, Thunderbird International Business
producers under conditions of product recall risk. DEA
Review, Vol. 50 No. 4, pp. 217-30.
provides a very effective method to analyze data involving
Evans, J.R. and Olson, D.L. (2002), Introduction to Simulation
multiple factors or criteria, as are present in this case. Monte
Carlo simulation provides a tool to consider uncertainties, and and Risk Analysis, 2nd ed., Prentice Hall, Upper Saddle
is very effective when probability is present, as in this case. River, NJ.
Here we use simulation to see specifics about the performance Fishman, C. (2006), The Wal-Mart Effect: How the World’s
of alternatives identified as both non-dominated and Most Powerful Company Really Works – And How It’s
dominated by DEA models. When vendor fails, our Transforming the American Economy, Penguin, New York,
simulation generates the average overall costs, i.e. risk- NY.
adjusted costs, based on which preferred outsourcing sites Gaudenzi, B. and Borghesi, A. (2006), “Managing risks in the
could be determined. In that sense, simulation gives a more supply chain using the AHP method”, The International
complete picture of the probabilistic aspects of decisions. Journal of Logistics Management, Vol. 17 No. 1, pp. 114-36.
Overall, our Monte Carlo simulation generates similar decision Goh, M., Lim, J.Y.S. and Meng, F. (2007), “A stochastic
results to DEA and DEA simulation models: Vendors from the model for risk management in global supply chain
Great China are preferred to those from western nations due to networks”, European Journal of Operational Research,
lower risk-adjusted costs and higher efficiencies. Further work Vol. 182 No. 1, pp. 164-73.
has to be done by showing how insights in this paper work Guinipero, L.C. and Eltantawy, R.A. (2004), “Securing the
using real data from specific organization. upstream supply chain: a risk management approach”,

407
Risk management models for supply chain Supply Chain Management: An International Journal
David L. Olson and Desheng Wu Volume 16 · Number 6 · 2011 · 401 –408

International Journal of Physical Distribution & Logistics Roth, A., Tsay, A.A., Pullman, M.E. and Gray, J.V. (2008),
Management, Vol. 34 No. 9, pp. 698-713. “Unraveling the food supply chain: strategic insights from
Hoffman, W. (2008), “Regulation revisits recalls”, Traffic China and the 2007 recalls”, Journal of Supply Chain
World, Vol. 272 No. 22, p. 15. Management: A Global Review of Purchasing & Supply,
Kara, S. and Kayis, B. (2008), “Proactive logistics risk Vol. 44 No. 1, pp. 22-39.
management – a system dynamics modeling approach”, Schoenherr, T., Rao Tummala, V.M. and Harrison, T.P.
International Journal of Risk Assessment & Management, (2008), “Assessing supply chain risks with the analytic
Vol. 10 No. 3, pp. 224-37. hierarchy process: providing decision support for the
Keeney, R.L. and Raiffa, H. (1976), Decisions with Multiple offshoring decision by a US manufacturing company”,
Objectives: Preferences and Value Tradeoffs, John Wiley & Sons, Journal of Purchasing & Supply Chain Management, Vol. 14
New York, NY. No. 2, pp. 100-11.
Kleindorfer, P.R. and Saad, G.H. (2005), “Managing Sowinski, L.L. (2008), “A shift in supply chain strategy”,
disruption risks in supply chains”, Production and World Trade, Vol. 21 No. 9, pp. 16-18.
Operations Management, Vol. 14 No. 1, pp. 53-68. Talluri, S., Narasimhan, R. and Nair, A. (2006), “Vendor
Klimov, R. and Merkuryev, Y. (2008), “Simulation model for performance with supply risk: a chance-constrained DEA
supply chain reliability evaluation”, Technological & Economic approach”, International Journal of Production Economics,
Development of Economy, Vol. 14 No. 3, pp. 300-11. Vol. 100 No. 2, pp. 212-22.
Kull, T. and Closs, D. (2008), “The risk of second-tier Tang, C.X.H., Lau, H.C.W. and Ho, G.T.S. (2008),
supplier failures in serial supply chains: implications for “A conceptual fuzzy-genetic algorithm framework for
order policies and distributor autonomy”, European Journal assessing the potential risks in supply chain
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

of Operational Research, Vol. 186 No. 3, pp. 1158-74. management”, International Journal of Risk Assessment
Kull, T.J. and Talluri, S. (2008), “A supply risk reduction & Management, Vol. 10 No. 3, pp. 263-71.
model using integrated multicriteria decision making”, Towill, D.R. and Disney, S.M. (2008), “Managing bullwhip-
IEEE Transactions on Engineering Management, Vol. 55 No. 3, induced risks in supply chains”, International Journal of Risk
pp. 409-19. Assessment & Management, Vol. 10 No. 3, pp. 238-62.
Lewis, L. (2008), “Healing the food supply chain”, Stores Wade, J. (2007), “The bittersweet poison in the prescription
Magazine, Vol. 90 No. 11, pp. 6-8. drug supply chain”, Risk Management, Vol. 54 No. 7, p. 12.
Li, X. and Chandra, C. (2007), “A knowledge integration Weber, C.A. and Desai, A. (1996), “Determination of paths
framework for complex network management”, Industrial to vendor market efficiency using parallel coordinates
Management & Data Systems, Vol. 107 No. 8, pp. 1089-109. representation: a negotiation tool for buyer”, European
Lucas, C., Mitra, G. and Poojari, C.A. (2007), “Risk based Journal of Operational Research, Vol. 90 No. 2, pp. 142-55.
methods for supply chain planning and management”,
Weber, C.A. and Ellram, L.M. (1993), “Supplier selection
Journal of the Operational Research Society, Vol. 58 No. 11,
using multi-objective programming: a decision support
p. 1397.
system approach”, International Journal of Physical
Madden, N. (2008), “Behind the scenes in a Chinese
Distribution & Logistics Management, Vol. 23 No. 2, pp. 3-14.
factory”, Advertising Age, Vol. 79 No. 40, p. 14.
Whalley, J. (2007), “China and outsourcing”, in Trefler, D.
Matos, S. and Hall, J. (2007), “Integrating sustainable
(Ed.), volume on outsourcing for Industry Canada
development in the supply chain: the case of life cycle
(forthcoming).
assessment in oil and gas and agricultural biotechnology”,
Wu, D. (2009), “Supplier selection: a hybrid model using
Journal of Operations Management, Vol. 25 No. 6,
DEA, decision tree and neural network”, Expert Systems
pp. 1083-102.
with Applications, Vol. 36 No. 5, pp. 9105-12.
Nagali, V., Hwang, J., Sanghera, D., Gaskins, M.,
Wu, D. (2009), “Performance evaluation: an integrated
Pridgen, M., Thurston, T., Mackenroth, P., Branvold, D.,
Scholler, P. and Shoemaker, G. (2008), “Procurement risk method using data envelopment analysis and fuzzy
management (PRM) at Hewlett-Packard Company”, preference relations”, European Journal of Operational
Interfaces, Vol. 38 No. 1, pp. 51-60. Research, Vol. 194 No. 1, pp. 227-35.
Olson, D.L. (1996), Decision Aids for Selection Problems, Wu, D. and Olson, D.L. (2008), “A comparison of stochastic
Springer, New York, NY. dominance and stochastic DEA for vendor evaluation”,
Olson, D.L. and Wu, D. (2007), Enterprise Risk Management, International Journal of Production Research, Vol. 46 No. 8,
World Scientific, Singapore. pp. 2313-27.
Olson, D.L. and Xie, M. (2010), “Comparison of Wu, D. and Olson, D.L. (2008), “Supply chain risk,
coordinated supply chain inventory management simulation, and vendor selection”, International Journal of
systems”, Journal of Service and Operations Management, Production Economics, Vol. 114 No. 2, pp. 646-55.
Vol. 6 No. 1, pp. 73-88. Wu, T., Blackhurst, J. and Chidambaram, V. (2006),
Payne, P. (2008), “Is pharma’s supply chain safe?”, “A model for inbound supply risk analysis”, Computers in
Pharmaceutical Technology Europe, Vol. 20 No. 7, pp. 15-16. Industry, Vol. 57 No. 4, pp. 350-65.
Ritchie, B. and Brindley, C. (2007), “An emergent framework
for supply chain risk management and performance
Corresponding author
measurement”, Journal of the Operational Research Society,
Vol. 58 No. 11, pp. 1398-411. Desheng Wu can be contacted at: DWu@Rotman. Utoronto.Ca

To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints

408
This article has been cited by:

1. Seyed Hamed MoosaviRad, Sami Kara, Suphunnika Ibbotson. 2014. Impact of international outsourcing on the value adding of
industries. The International Journal of Logistics Management 25:3, 463-486. [Abstract] [Full Text] [PDF]
2. Kristian Rotaru, Carla Wilkin, Andrzej Ceglowski. 2014. Analysis of SCOR’s approach to supply chain risk management.
International Journal of Operations & Production Management 34:10, 1246-1268. [Abstract] [Full Text] [PDF]
3. Pui-Sze Chow, Tsan-Ming Choi, Bin Shen, Jinhui Zheng. 2014. Supply Contracting with Risk-Sensitive Retailers under
Information Asymmetry: An Exploratory Behavioral Study. Systems Research and Behavioral Science 31:10.1002/sres.v31.4,
554-564. [CrossRef]
4. Chun-Hung Chiu, Tsan-Ming Choi, Yongjian Li, Lei Xu. 2014. Service Competition and Service War: A Game-Theoretic
Analysis. Service Science 6, 63-76. [CrossRef]
5. Tsan-Ming Choi. 2013. Multi-period risk minimization purchasing models for fashion products with interest rate, budget, and
profit target considerations. Annals of Operations Research . [CrossRef]
6. Anna Nagurney, Dong Li, Ladimer S. Nagurney. 2013. Pharmaceutical supply chain networks with outsourcing under price and
quality competition. International Transactions in Operational Research n/a-n/a. [CrossRef]
7. Simulation modeling 373-386. [CrossRef]
8. Risk mechanics 387-408. [CrossRef]
Downloaded by Carleton University At 17:07 29 November 2014 (PT)

9. Li Zhao, Baofeng Huo, Linyan Sun, Xiande Zhao. 2013. The impact of supply chain risk on supply chain integration and
company performance: a global investigation. Supply Chain Management: An International Journal 18:2, 115-131. [Abstract]
[Full Text] [PDF]
10. Theodosios Tsiakis, Panagiotis TsiakisManaging the Risks of Outsourcing IT Security in Supply Chain 477-495. [CrossRef]