Eli Schragenheim usually present great new ways of thinking or new approaches in the world of Theory of Constraints. He's been trying to do more on his blog as well, which usually makes me think. At TOCICO this year, he presented on Managing Under Uncertainty in the TOC Way.
Eli has presented previously about a "mystery analysis" approach in thinking through "failures" or other mysteries where the results of an action or decision were substantially different from expectations (my notes). I still like this general idea - a couple people have mentioned it at the conference already. In this discussion, he reflected back to the TOC principle of "Never Say, I know". Uncertainty has a connection here. You can never know everything about a situation - if so, there is no decision to make. Importantly, you also never know nothing. Our understanding of the cause-and-effect is limited. Professor Herbert Simon said as much, "It is impossible to have perfect and complete information at any given time to make a decision"
What is uncertainty? In the context of this conversation it is the combination of our incomplete understanding of a system AND the natural variation of processes within the system. The result is that we can't know for sure what will be the outcome of an action or decision. And for many people, particularly within organizations, this deepens a conflict: Act now or Wait. Act now because we need to respond or move. Wait because we don't want to take an action that might get us in trouble later (limit risk). And in the context of organizational decisions, people are strongly inclined to limit their personal risk.*
How can we help overcome this conflict? Can we help people and organizations make decisions while still keeping the risk of those decisions relatively low? Eli was adamant that direction of the solution does NOT include teaching people how to understand probabilities - our intuitive understanding of very low probabilities is not helpful in this arena.
The direction that Eli proposed is to combine intuition and analysis. (Intuition has appeared in a number of discussions so far.) We use intuition to come up with a reasonable range of potential outcomes. Since this is an area of the business that we care about, our intuitions should be fairly good. Then use analysis to consider the net benefit/damage at the extremes of the potential outcomes. He suggests not worrying about the "black swan" type events - we often have no intuition about these. Rely on intuition to clarify the possibilities and the options. And then use analysis to assess the impacts. What action(s) would generally take you to the better outcomes? The definition of "better" is still objective, as loss and gain are not counter-balanced (loss aversion). But at least we acknowledge a reasonable range of possibilities and then investigate reasonable actions to take.
And this all loops back to the idea of learning organization and continuous improvement. Stop allowing one-point estimates to dominate a discussion. Bring in reasonable ranges, so that people can see the likely effect and its possible boundaries. Take advantage of any opportunity to hone the knowledge of the group, team, individual: any time a result falls outside the range, investigate and update the flawed understandings. Refine the intuition!
This way of thinking can be applied anywhere. But within organizations that have implemented the TOC solutions, there is a natural link to uncertainty in the buffers. Buffers essentially provide this reasonable range of expected results. And they also help organizations stop over-reacting when there is variation. The buffer absorbs most of the regular variation, and enables the organization to focus when the system moves outside this range.
* This point in the discussion reminded me of a recent HBR Podcast on Making the Toughest Calls with Joseph Badaracco. He has described five questions to ask or think through when making those tough decisions, and many of Eli's comments resonate with those in the podcast.