As supply chains grow in complexity and reach around the world, it becomes more challenging to understand the impact on operations from an unexpected disruption at one supplier’s site. To address this issue, David Simchi-Levi ( a professor of civil and environmental engineering and engineering systems at the Massachusetts Institute of Technology and the founder of LogicTools, a provider of software for optimizing supply chains that is now part of IBM) writes in Harvard Business Review that he and his colleagues William Schmidt of Cornell and Yehua Wei of Duke developed a method to help prioritize the financial or operational impact of risk so companies may focus their mitigation efforts on key suppliers and risk.


The model, a mathematical description of the supply chain that can be computerized, focuses on the impact of potential failures at points along the supply chain, rather than the cause of the disruption. Using the model, companies can quantify what the financial and operational impact would be if a critical supplier’s facility were out of commission for, say, two weeks—regardless of the cause, Simchi-Levi explains. The method was implemented successfully at Ford Motor Company.


A central feature of the original model was time to recovery (TTR), or the time it would take for a particular node—a supplier facility, a distribution center or transportation hub—to be restored to full functionality after a disruption. By combining suppliers’ TTR information with the details of Ford’s supply chain, product bill-of-material, volume, and profit margins by product line and pipeline inventory, the method identifies the risk exposure associated with a disruption in each site of Ford’s network, Simchi-Levi writes in the article. This is done by simulating the firm’s response to a disruption at a specific site for the duration of TTR.


The TTR values are determined by combining historical experience and the results of surveying the firm’s buyers or suppliers. The problem, however, is suppliers tend to be overly optimistic about their TTR because they realize a long TTR won’t be accepted by the manufacturer. That prompted Simchi-Levi and his colleagues to realize the need for a way to identify bottleneck suppliers for which it’s critical to obtain accurate TTR information and distinguish them from other suppliers where even plus or minus 30 percent error in TTR information will have very little impact on the supply chain.


Simchi-Levi and his colleagues then developed a metric they call “time to survive” (TTS), which is the maximum duration the supply chain can match supply with demand after a node disruption. To determine TTS associated with a specific node, the node is removed from the supply chain and a calculation is used to determine how long—using inventory in the pipeline and other available supply sources—the supply chain can continue to meet demand without that node. If the TTS of a specific site is greater than its TTR, this site doesn’t expose the firm to any risk since during the time the site is recovering from a disruption, the firm can still match supply with demand, Simchi-Levi writes in the article. On the other hand, if the TTS of a specific facility is smaller than its TTR, its disruption will expose the firm to financial and operational problems.


The new metric then drove the development of a model to assess the level of strategic inventory: inventory used to respond to a disruption anywhere in the supply chain. That is, TTS and TTR metrics can be combined to determine how much strategic inventory the organization needs, as well as where to position this inventory so each site’s TTS is greater than its TTR, Simchi-Levi writes. This leads to a robust supply chain because a disrupted node will always recover before it exceeds its ability to apply the mitigation strategies the firm has in place.


This also presents an opportunity to cut costs since a long TTS is often achieved by adding a great deal of strategic inventory. Cutting inventory for these suppliers by 50 percent, for example, will have very little impact on the ability to respond to a disruption, Simchi-Levi writes.


What are your thoughts on time to recover and time to survive? Do you know how long your supply chain can continue to meet demand if various nodes experience a disruption of two weeks or more?