Contemporary Business Studies - Academy of Knowledge Process ...

International Journal of Contemporary Business Studies
Vol: 3, No: 1. January, 2012 ISSN 2156-7506
Available online at http://www.akpinsight.webs.com
uncertainty. When Bayesian networks were first introduced, assigning the full probability distributions
manually was time intensive. Solving a Bayesian network with a considerable number of nodes is known
to be a nondeterministic polynomial time hard [NP hard] problem (Dagum and Luby, 1993). However,
significant advancements in computational capability along with the development of heuristic search
techniques to find events with the highest probability have enhanced the development and understanding
of Bayesian networks. Correspondingly, the Bayesian computational concept has become an emergent
tool for a wide range of risk management applications (Cowell et al, 2007). The methodology has been
shown to be especially useful when information about past and/or current situations is vague, incomplete,
conflicting, and uncertain.
Pai et al (2003) were among the first researchers to analyze supply chain risks using Bayesian networks.
Their study examined the risk profile associated with a US Department of Defense (DoD) supply chain
for trinitrotoluene (TNT). The supply chain was comprised of TNT recovery plants, storage facilities, and
ammunition depots. Using Bayesian networks, the researchers were able to establish risk factors and
acceptable risk limits for all assets contained in the DoD supply chain. Bayesian networks have also been
used to conduct diagnostics (Kauffmann et al, 2002; Kao et al, 2005), cost optimization studies
(Narayanan et al, 2005), and flexibility analysis (Wu, 2005; Milner and Kouvelis, 2005) in supply chains.
Since the work of Pai et al (2003), researchers have continued to explore the use of Bayesian networks to
analyze and manage supply chain risks. For example, there have been a number of studies which
examine the use of Bayesian networks as part of a decision support system to manage such risks (Li and
Chandra, 2007; Meixell et al, 2008; Shevtshenko and Wang, 2009; Makris et al, 2001; Taskin and Lodree,
2011). Studies by Tomlin (2009) and Chen et al (2010) demonstrate how Bayesian networks can be used
to manage supply chain uncertainty. The integration of Bayesian networks into supply chain forecasting
methodologies to mitigate risks has also been examined by several researchers (Yelland, 2010; Yelland et
al, 2010; Rahman et al, 2011). Lockamy and McCormack (2009) conducted a study which uses Bayesian
networks to examine operational risks in supply chains. The authors have also used these networks to
analyze outsourcing risks in supply chains (Lockamy and McCormack, 2010). Finally, Lockamy (2011)
has developed a methodology for benchmarking supplier risks using Bayesian networks.
This research study contributes to the current body of supply chain management literature by introducing
a methodology for evaluating network risks in supply chains. The methodology also includes an
assessment of the potential revenue impact a supplier can have on an organization as measured by valueat-risk
(VAR). The methodology in this study is offered as a tool to assist supply chain managers in the
formulation of strategies and tactics designed to mitigate supply chain risks.
4.2Assessment Model
An assessment model developed by Handfield and McCormack (2007) was used to evaluate the risk of
each supplier. This model incorporates data from several sources to provide a 360-degree view of a
supplier's risk profile. The risk assessment model is presented in Figure 1. A potential challenge regarding
the use of the assessment model is the need for a detailed review of data furnished by suppliers. Such
reviews can be costly and time consuming, and therefore are usually limited to the firm's strategic
suppliers. However, well designed, qualitative self- reporting can be very cost effective. As in the case of
this study, each supplier was required to answer an online multiple choice survey that takes less than 30
minutes to complete. In addition, given the large number of suppliers in many supply chains, the buyers
or category managers often do not know every detail about each supplier, thus making self-reporting by
suppliers a necessity. With these qualitative indicators, it is also feasible and affordable to assess several
tiers in the supply network, consequently making the network risk profile broad and deep. The risk
assessment model identifies and quantifies the risk of a supply disruption using a framework that
describes the attributes of suppliers, their relationships, and their interactions with the organization
performing the assessment.
10
Copyright © 2012. Academy of Knowledge Process