Resilience and the Ability to Anticipate

Resilience and the Ability to Anticipate

This is a chapter of Resilience Engineering in Practice written by David Woods. In this chapter Woods goes over the patterns of anticipation that resilience systems go through.

Similar to the patterns of how adaptive systems fail, knowing these patterns and being able to recognize them (or their absence) in your own systems can help highlight sources of resilience.

There are 5 patterns, we’ll talk a little bit about each one.

Pattern 1: Resilient systems are able to recognize the signs that adaptive capacity is falling or inadequate to the contingencies and squeezes or bottlenecks ahead.

If our system (of which people are a part), cannot notice that resources and sources of resilience are declining, then how will it be able to adapt to the future? Similarly, if a system is aware of it’s adaptive capacity, but cannot look ahead and see if they may be sufficient, it is also unable to adapt.

Woods gives us some questions we can ask of our environment to try and determine if our systems are able to fit this pattern:

  • “Are there new kinds or sizes of disrupting events beginning to occur or beginning to appear on the horizon?”
  • “Is the ability to respond eroding or declining?”
  • “Is one perspective over confident and miscalibrated about the adaptive capacity of the system and how it achieves that capacity?”

I think that last one is especially important. If the answer is “yes,” then that means that the system works differently than those people or that person imagines. This can be more difficult if it is a “stakeholder” or management. It also means that there are potentially hidden (to that view) or unappreciated sources of resilience somewhere in the system that are helping to keep the system functioning.

Having those sources of resilience is good of course, the danger here lies in not being able to identify them or having them be hidden. It opens up the opportunity for them to be compromised or removed.

Pattern 2: Resilient systems are able to recognize the threat of exhausting buffers and reserves

This is somewhat similar to the previous, but specifically focusing on reserve resources (which could be sources of adaptive capacity).

Woods gives the example of incident commanders, who “maintain reserves which could be deployed to fill a gap or handle a new turn of events,” as opposed to always deploying all resources to a single incident.

He was referring to urban firefighting, but the same principle applies to our own incident response.

Pattern 3: Resilient systems are able to recognize when to shift priorities across goal trade-offs.

I like the way he puts this, shifting across goal trade-offs. It makes it clear that “trade-offs are fundamental and inescapable,” whether its a trade off between optimality and brittleness or efficiency and thoroughness, or some other trade-off, there is always one occurring.

The question here for us and our systems are, where are we positioned along these trade offs? Is that positioning the right place for us to be given the environment we are operating in? If not, how can we move to a better positioning?

As Woods reminds us, “how organizations manage situations where goals conflict is a critical indicator of resilience.”

Further, “resilient systems are able to know when to sacrifice acute production goals and prioritize chronic safety goals.”

Pattern 4: Resilience systems are able to make perspective shifts and contrast diverse perspectives that go beyond their nominal system position.

Woods doesn’t discuss this one in too much detail here, but it is a key part of what makes systems resilient. For a bit more on perspective shifts, take a look at the earlier cognitive systems engineering work

If we ask and answer questions about a system from a single perspective, we open ourselves up to the possibility of missing other signals. These could be signals about coming disturbances or signals about where reserves are being built, or signals about how adaptive capacity is being preserved.

Pattern 5: Resilient systems are able to navigate interdependencies across roles, activities, levels.

This is similar to pattern 4, but also includes operating across different scales. Woods returns to the example of urban firefighting, where various operations will take place across scales and there is a need to keep multiple groups in sync.

Without the ability to carry out this form of anticipation, systems are at risk of the adaptive breakdown pattern of working at cross-purposes or being locally adaptive but globally maladaptive.

Both pattern 4 and 5 are increasingly being researched under guise of “polycentric control architectures” including polycentric governance. That is, ways of working that dynamically adapt and consist of interdependent roles and activities.

Takeaways

  • Recognizing the role of anticipation and the patterns in which it manifests in resilience systems can help when evaluating your own systems.
  • These patterns are a look at the other side of the coin from how adaptive systems fail, you can think of these as how adaptive systems succeed.
  • Be aware of perspectives that are miscalibrated about the adaptive capacity in a system and how it is created.
  • “How organizations manage situations where goals conflict is a critical indicator of resilience.”
  • “Resilient systems are able to know when to sacrifice acute production goals and prioritize chronic safety goals.”

Don't miss out on the next issue!