On April 14 2016 a 6.5 magnitude earthquake struck the island of Kyushu in southern Japan. Two days later the island was hit by a second, even more powerful, quake. The widespread damage caused by these natural disasters fractured supply chains, causing a number of companies to halt production lines in various locations.
The quakes delivered a powerful reminder that supply chains remain vulnerable to low-probability, high-impact disruptions. But they also highlight how far we have come in dealing with this type of risk.
After the quakes hit in April, car manufacturer Toyota shut down 26 car assembly lines across Japan because the flow of parts from a supplier that suffered damage was severely disrupted. Honda and Nissan implemented shutdowns for similar reasons. Quake-related damage forced General Motors to temporarily stop production lines in four plants in North America. The after effects were felt outside the auto industry too. For example, Sony Corp halted production at a plant in Kumamto that makes parts for digital cameras.
Although these supply chain disruptions are damaging, on the Richter scale of disaster outcomes they are relatively low. For example, the havoc wreaked on the US Gulf Coast by Hurricane Katrina in 2005, and the earthquake and tsunami that devastated parts of Japan in 2011, were much more calamitous.
Also, terrible events like these have taught us much about preparing supply chains for unexpected disasters. There is more support for risk managers, and in general, enterprises now take disaster response more seriously that they did a decade or so ago.
For example, over the last several years we have witnessed the development of “fast assessment” software applications. Such systems follow product bills of material, and once a disruption takes place anywhere in the supply chain, an OEM can find quickly which parts are affected, the products these parts feed, and the customers using these products. Coupled with inventory and alternate supplier information, these systems can be used to prioritize the response and communicate with customers.
On the corporate side, enterprises have created emergency management centers that can be activated when crises hit. There is also more disaster preparedness training for employees. A key lesson many enterprises have taken aboard, is that quick reactions are vitally important, especially when evaluating the consequences of a major disruption. The ability to rapidly assess which suppliers, parts, and customers are impacted, enables companies to prioritize supply and put in place the resources they need for a quick recovery. Past disasters have shown that the most agile companies tend to switch to alternative sources of supply ahead of the competition, and are more effective as minimizing the disruption – both short- and long-term – to their businesses.
An example of a company that has this level of agility is Cisco. As I describe in my book The Power of Resilience: How the Best Companies Manage the Unexpected (MIT Press, October 2015), during the 2011 Japan earthquake that occurred at 9:47 pm Cisco headquarters’ local time, the company detected and understood the significance of the event within 40 minutes and had escalated it to senior management 17 minutes later.
Even if few organizations are able to match such reaction times today, many are better prepared than they were in 2011 and have developed quicker responses to large-scale disruptions that hit with almost no warning.
Toyota is an example of such an organization. For instance, when the manufacturer was struck by the Kyushu earthquakes in April, the Wall Street Journal reported that the company had reassessed its disaster response mechanisms as a result of the damage inflicted by the 2011 quake. Toyota had looked at the challenge of balancing efficiency and risk, and re-examined contingency scenarios.
It was suggested that Toyota’s just-in-time (JIT) production method, which relies on parts supplies that are carefully synchronized with production requirements, is vulnerable to sudden disruptions. This might be the case, but that is no reason to change the company’s world-famous production model. JIT delivers huge efficiency gains, one being that it obviates the need to maintain a lot of costly buffer inventory. Also, the fast production flows that are part of JIT help companies to detect how disruptions impact lower tier suppliers; another key lesson gained from past experience.
“There are always risks, and they wouldn’t necessarily be resolved even if we have parts inventory worth two weeks. Rather, we have been focused on how to understand and identify issues in case of a problem, and how quickly we could recover,” a Toyota official said in the Wall Street Journal article.
These comments reflect a more thoughtful approach to supply chain risk management. We still have a long way to go – particularly in managing the risks associated with low-tier suppliers. But in these doom-laden times, it’s encouraging to see that companies are making progress in preparing for disasters that seem to be increasing in both frequency and severity.
This post was written by Yossi Sheffi, Elisha Gray II Professor of Engineering, MIT, Director, MIT Center for Transportation & Logistics.