Monday, February 22, 2010

The Complexities of Certainty

At the most basic level, engineering is about "things" and how others define and measure the performance certainty of these things. Certainty as defined by many end users is perfect knowledge of the performance attributes of their "things" in any and all situations. Fundamentally the desire is to predict any and all outcomes without any doubt. The bridge will not fall and the car will come to a complete stop - - the public has an expectation of absolute certainty. The overreaching goal of engineering should be to give the public some level of certainty - - and over time the advancement of any civilization can be measured by the increasing knowledge and increasing certainty in matters relating to technology. The key question facing engineering is how do we discuss and debate certainty - - or in our case, uncertainty.

Engineers, for example civil engineers, understand that many of the "things" the public utilizes and takes for granted are inherently spatial and complex - - where our understanding of these systems is less than complete. They may never be complete. We are put in positions where we must explain that many of our systems are either fully stochastic, or part stochastic and part deterministic - - to a general public that has expectations that that their 401(k) balance will automatically increase by 8.5% every year (the last two or three years in the financial markets may have helped the general public understand the differences between stochastic versus deterministic). In some respects, the claims and demands of certainty have become a universal philosophy in a world marked by increasing complexity and risk.

Jana Eggers, the CEO of Spreadshirt, the maker of personalized clothing, has an interesting thought on the complexities of certainty:

I have another one that stirs up quite a bit of controversy, which is a Madeleine Albright saying - - “Be confident, not certain.” And it’s funny to me how many people don’t like that - - “Well, what do you mean? If you’re not certain about it, you’re supposed to be confident about it?”

The public face of engineering that is “Confident, not Certain” may be our biggest hurdle and challenge. The homeowner on the other side of the levee is probably going to have much the same reaction - - “If you’re not certain about it, you’re supposed to be confident about it?” Engineering does an excellent job in explaining uncertainty in terms of risk, probabilities, distributions, models, etc. The problem is that our culture and language is one of certainty - - and in many, many cases, absolute certainty. People want “Confident and Certain" - - and in some areas and endeavors, we will never be able to make delivery on "Confident and Certain."

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.