Chaos, Construction sites, and Bayes

21 Jan

The last article was a case study about construction site decision making. To summarise briefly, a project manager was in charge of six construction sites, each ran by a site manager. Three of the site managers were “trusted” by the project manager and had excellent performance records. The other three site managers were “not trusted”, took up large amounts of senior management time and required significant performance improvement. We conducted research to uncover the difference between “trusted” and “untrusted”, and then turn this information into a resource which could be used to improve performance.

Our research discovered that “trust” in this specific context meant adopting an adaptive mind-set when carrying out plans. When a “trusted” site manager arrived at a job with a plan, they would mentally simulate what could potentially go wrong with executing the plan as it was written down. In other words, they would see the plan as a hypothesis, and revise it in relation to the situation they found at the site on the day of arrival. By contrast, the “untrusted” managers would see the plan as rigid, and attempt to execute it regardless of the current site situation. The result was initiative and adaption based on new information for the “trusted”, and blindly following the original plan into trouble for the “untrusted”. The high performing site managers applied a sort of cognitive Bayesian reasoning.

Bayes rule and Bayes theorum can be incredibly sophisticated and complex statistics**, but I’ll be using an extremely basic and crude example to illustrate a point. The rule works very well in situations where there is a lot of incomplete and missing data (McGrayne, 2011, Lee, 2012). It allows a hypothesis to be generated, and then updated based on the emergence\discovery of new data. The majority of human situations are chaotic, meaning, past information has a bearing on the future, but any error\deviance in calculating that information can lead to dramatically different outcomes to those predicted and\or seen before. Therefore, regular updating of beliefs based on new information, and maintaining the capacity to seek and absorb new information, is important when operating in chaotic domains.

Generating an initial belief (in the form of a plan), actively seeking new information, and then updating the initial belief (plan), manifested itself as “trust”  and high performance in the construction site case study. I would suggest that a form of Bayesian reasoning is essential to high performance in chaotic domains.


McGrayne, S. B. (2011). The Theory That Would Not Die: How Bayes’ Rule Cracked the Enigma Code, Hunted Down Russian Submarines & Emerged Triumphant from Two Centuries of Controversy. Yale University Press

Lee, P. M. (2012). Bayesian Statistics. Wiley.

**Want to know more about Bayes? Excellent introduction at the link below

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: