Resilient Supply Chains

Stephen DeAngelis

April 15, 2010

In a recent blog post, Lora Cecere noted that too many companies have been pursuing supply chain efficiency instead of supply chain resilience [“We should have Listened to Mother Nature,” 26 March 2010]. She makes some excellent points. She writes:

“Yes, we got what we asked for…. Efficient supply chains that snap and break under stress. Have you ever asked yourself the question, in the design of supply chain networks, should supply chain efficiency be the goal? Based on seven years of supply chain research, I think not. I think that the goal should be resilience.”

Cecere indicates that a dinner conversation with “two wonderful thought leaders,” John Hagel and Valdis Krebs, started her pondering on this subject. Part of that conversation, she reports, was about “what nature teaches us about networks.” She continues:

“Networks with the greatest resilience — [i.e., that demonstrate an] ability to stand the test of time– are not the most efficient. In fact the least resilient networks in nature are the MOST efficient. The most resilient networks have two characteristics:

Redundancy: In nature, the networks that last over the generations are those that have the right amount of redundancy. This includes excess capacity, extra nodes, and bypass routes. The efficient network with minimal capacity, node-to-node connections, and no parallel routes fail. This is especially true in push environments.

Dynamic reconfiguration: Likewise, in nature, the networks that stand the test of time have the ability to reconfigure dynamically under stress. Networks with the most centrality have the greatest failure rate.

“Ring a bell? Yes, we have designed efficient supply chains that are primarily push-based, with little redundancy and a high dependency on centralized hubs. Most are in trouble. Most do not know it YET.”

Cecere, of course, in not the first analyst to note the importance of redundancy in resilient supply chains. In 2005, Professor Yossi Sheffi from the Massachusetts Institute of Technology wrote a seminal work entitled The Resilient Enterprise that examined disruptions in corporate supply chains. Sheffi, using a traditional definition of resilience, discussed how businesses could build in “capabilities for bouncing back quickly” when disruptions occur. He noted that redundancies in the supply were essential to minimize the impact of disruptions. Cecere, by adding the notion of dynamic reconfiguration, goes beyond the traditional definition of resilience. In fact, together her two characteristics more closely describe what could be called a consilient supply chain. According to Harvard biologist Edward O. Wilson, consilience is more than a mere bouncing back it is a bringing together of resources to both anticipate change and adapt to it. He invented the word “consilience” as a way to argue for a “fundamental unity of all knowledge” that underlies the organization of nature and which can be found in a few principles common to all branches of learning. Since consilience is a term that is little understood, describing supply chains that are both resilient and adaptable as simply “resilient” is probably best.

When Sheffi describes a resilient enterprise, he primarily focuses on preventing or mitigating disruptions to corporate supply chains. In biological terms, Sheffi’s resilient enterprises would better be described as robust enterprises. A robust system is stable. In the world where change is a constant and its pace is increasing, stability eventually means diminishing returns. Stuart A. Kauffman, a recipient of a MacArthur Foundation “genius award” and holder of professorships at the University of Calgary and the Santa Fe Institute, avers that a living system must first be able to strike an internal compromise between malleability and stability. To survive in a variable environment, it must be stable, to be sure, but not so stable that it remains forever static. Nor can it be so unstable that the slightest internal chemical fluctuation causes the whole teetering structure to collapse. Supply chains, although not biological, are nonetheless living systems. They, therefore, must demonstrate the same kind of dynamic stability that Kauffman discusses if they are to move beyond robustness to resilience. Kevin Kelly, who also believes information age networks are more biologic than industrial, sees this same tension but instead of calling it dynamic stability he calls it sustainable or persistent disequilibrium. In order to survive in this environment, you must constantly seek a proper balance or you are likely to fall. I believe that is what Cecere is saying as well. She continues:

“I reflected back to when I used to install supply chain planning software. When I asked a client to define an objective function for an optimization program [they] would struggle for an answer. They were not sure what good looks like. However, when I asked if the output needed to be efficiency, they would always nod their head. In fact, ask anyone if they want an efficient network–especially a financial controller–and you will get a rousing YES!!! But, ask if they want a network that will snap and break under stress and you will probably get an emphatic NO! Few know the trade-offs. Most companies do not realize that they have designed networks that are not resilient until it is too late. Yep, you got it. We designed supply chain networks with the wrong goal. We defined supply chain excellence as a set of efficient networks with a high level of centrality and little redundancy. It is time to consult mother nature, because she got it right. … We need to recognize and fix the problem. Enterprise Resource Planning (ERP) and Advanced Planning Systems (APS) were installed with efficiency as their goal. There is too little redundancy, and less than 1% have dynamic reconfiguration capabilities. It will make for great stories on the front page of the Wall Street Journal. I just hope that it is not your name in the article. I think that is time to listen to Mother Nature.”

A company that develops a resilient supply chain is fundamentally better equipped to respond, perform, compete and win than its non-resilient peers. Robert Frenay reminds us that Eugenius Warming, a Danish botanist writing over a century ago, noted that competition is strongest among members of the same species, because they compete for exactly the same resources. Such competition normally results in winners and losers. As Frenay writes, “one of them will die out unless it can adapt to a slightly different niche.” [Pulse: The Coming Age of Systems and Machines (New York: Farrar, Straus and Giroux, 2006)] Every company must discover its niche and develop the supply chain that fits it best; but as Cecere, points out, successful supply chains will be both redundant and adaptable.