Managing unpredictable supply-chain disruptions
19 January 2015
Human Capital Alliance Senior advisor Somsak Jaitrong looks at a new disaster risk-mitigation model.
The Fukishima earthquakes and the unprecedented Thai floods in 2011 created unforeseen havoc for many supply-chain dependent global manufacturers. Not long after these “low-probability, high-impact” events, many global manufacturers suddenly realized that their highly-disbursed efficient low-cost supply-chains were seriously disrupted for extended periods.
The Fukushima earthquakes disrupted Toyota’s highly-efficient Japan-based supply-chain, a critical part of its global manufacturing operations while the prolonged Thai floods made everyone aware that Nikon’s production of popular cameras was highly dependent on its Thai lens manufacturers.
Earlier this year, David Simchi-Levy, MIT professor of engineering systems, William Schmidtt, Cornell University Graduate School of Management assistant professor and Yehua Wei, Duke University Fuqua School of Business assistant professor wrote in the Harvard Business Review that major disaster such as Katrina and Fukushima clearly showed that traditional methods for managing supply-chain risks were seriously inadequate.
Traditional risk mitigation methods
The authors said that traditional methods for managing supply- chain risks primarily relied on knowing the likelihood of occurrence and the magnitude of impact, for every potential event that could materially disrupt a firm’s operations.
These methods that used historical data to quantify risks may be adequate for analyzing common supply-chain disruptions such as poor supplier performance, forecast errors and transportation breakdowns.
“However, because historical data on rare events are limited or non-existent, their risk is hard to quantify using traditional models.”
Developed new computer-driven risk-mitigation model
To address the challenge, the authors developed a computerized risk management model that targets specific points along supply-chains. Their model that was tested as a case study with Ford Motor Co also delivered several most interesting findings.
Obviates need to determine any specific risk
The authors’ new model’s basic premise relies on the continuous quantifying of the financial and operational impacts if critical suppliers facilities are out of commission. By focusing on the shuttering’s operational impacts, companies don’t have to determine when low-probability, high-impact disruptions will occur.
Greatest surprise – little correlation to size of spend
Using their model to analyze Ford’s exposure to supply-chain disruptions, the authors’ greatest surprise was the low-correlation between procurement spend from a particular site and the disruptive impact on company performance.
The study discovered that the supplier sites cause the greatest damage were those which Fords annual purchases were relatively small. These findings even surprised Ford managers
“Many of these suppliers had not previously been identified by the company’s risk mangers as high-exposure suppliers.”
The Ford example
In March 2013, Ford suddenly experienced a shortage of specialty resin Nylon 12, used for manufacturing fuel tanks, brake components and seat fabrics because a key supplier, Evonik’s plant exploded in Marl, Germany. Evonik took more than six months to restart production, during which time downstream production facilities of Ford and other major automakers severely disrupted.
The authors said that if Ford had used the framework model prior to the disruption, it would have detected the risk exposure and associated production bottlenecks. Moreover, it would have encouraged Ford to proactively work with Evonik to fast track its new Singapore production plant.
High value parts may not be as critical
As supply-chains become increasingly globalized, complex and extended, the authors said that highest disruptive impact points of failure in many cases may not be from high-value parts’ suppliers.
Most importantly, the authors said their new model allows supply-chain managers to avoid guessing the likelihood of infrequent high-impact events and instead concentrate on evaluating their organization’s vulnerability to disruptions, regardless of their cause and where they strike.
At the same, using quantitative measures to determine risk, it provides an easy to understand segmented process that results in more resilient supply networks.