3 degrees of complexity

I found out yesterday that there is a difference between complicated and complex, terms that I had been using interchangeably. The book I am currently reading, The Checklist Manifesto, tries to address the issue of whether checklists will work for any level of complexity in situations we deal with in our professional lives. The author refers to a paper by Sholom Glouberman and Brenda Zimmerman which categorizes problems as simple, complicated, and complex.

Simple situations are those that have been addressed multiple times and sometimes tools and processes are available to reduce errors further. An example would be baking a cake; if you have the recipe and follow the instructions faithfully, it is easy to imagine that you would end up with an acceptable cake. The process is replicable and the availability of moulds and ready-mixes makes the process easier and the outcome more predictable.

Complicated problems contain subsets of simple problems but are not merely reducible to them; there are issues of coordination and specialized expertise. Sending a spacecraft to the moon is a complicated problem. However, once the problem has been solved, the solution can be replicated if we develop rigourous processes; these are likely to assure a high degree of success if followed faithfully.

Complex problems, the highest order in their classification, can encompass both simple and complicated sub-problems but may not be reducible to either. There is the added complexity brought in by uniqueness of situations, subjects and objects that makes replication of success difficult to assure. Consider the problem of raising a child. One can apply oneself and put in a lot of effort and have tremendous success with one child. But all the expertise, inherent and learned during the process cannot assure that success can be replicated in raising another child with the same process.

Expertise can contribute to success in complex situations, but it is neither a necessary nor sufficient condition for success.

Outcomes will always remain uncertain with complex problems.