This is a variant of a post I recently made to the Lean-Agile user group. On the group there has been a lot of talk about the apparent disconnect or conflict between Complex Adaptive Systems (CAS) and Lean-thinking. Complex adaptive systems are diverse and made up of multiple interconnected elements (and so are a part of network science) and adaptive in that they have the capacity to change and learn from experience. Social networks and the stock market are examples. I believe all of the examples of where Lean-thinking is in apparent conflict with CAS are not examples of true Lean-thinking but are more misunderstandings of Lean.
Lean thinking does not imply any sort of determinism as much as it takes advantage of a scientific approach. Since respecting people is a one of the cornerstones of Lean being aware of how people work would be a necessity. Deming talks a lot about how you have to make people enjoy their jobs and bring their intellect to it.
Anyway, in another mini-project of mine (I am re-organizing many of my blogs so people can see my own personal journey to Lean-Agile) I came across a blog I wrote in May, 2007. It's called "Challenging Why (not if) Scrum works. http://www.netobjectives.com/blogs/challenging-why-not-if-scrum-works
One might say the most surprising thing about this blog is that it discusses how teams doing waterfall achieved a 3-to-1 improvement over other teams doing waterfall. Not by anything they did, so to speak, as much as how they were organized. I found this a particularly valuable insight since we should look for improvement everywhere we can get it. No, to me the most surprising thing about this blog was that no one in the Scrum community was interested in exploring why Scrum was truly working (at least at the time). However, again on reflection, this may not be very surprising, since Scrum is founded on a notion that our processes must be black box in nature and therefore can't really be explained.
The case I describe illustrates how one can look at issues that affect complex-adaptive-systems (organization, reporting structure, physical location, level of communication) and how these can be managed for improvement. In other words, even though we may have complex systems which we can't predict their exact outcomes, we can set up situations so that the outcomes will improve in their own emergent manner.
There is a big difference between causality and predictability. I may not be able to predict things even when I know the causes of them. A simple example is adding one pound of weight to a table top until it breaks. Chaos theory will tell me I can't predict exactly when it'll break - but I know the cause. OK, this is a chaos situation - what about CAS?
Traffic is an example of CAS. Last night it rained in Seattle (who'da thunk!?!). Traffic was a disaster. This typically happens, btw. Now, 'how' the disaster showed up is unpredictable (last night it was 5 jammed big time). But that it will be bad is very predictable.
The fact that we are working with Complex Adaptive Systems does not mean we cannot take a scientific approach to it. We just have to know we need to continuously refine/redefine our thinking.
CEO, Net Objectives
Achieving Enterprise and Team Agility