- Accept that you're operating in an uncertain world - - this should be a natural for most engineers. The really historical difference is the degree and breadth of uncertainty we face. From earthquakes to climate change driven hurricanes to energy prices to political stability - - you can't be realistic about assessing the chances of a given event occurring unless you first confront all the other possibilities that might come true instead.
- Assess the level of uncertainty you face - - look at your statistical models. Have you considered the high level of uncertainty we face in our uber-connected world? Read widely and deeply - - are your assumptions valid and reasonable. Historically, a model with a 95% confidence interval would have been appropriate - - we may be looking at a world were 90% or 80% become the givens for confidence intervals.
- Augment the range of uncertainty - - chances are you were not very good at #1 or #2. Our brains are not wired for estimating the range of uncertainty we face. Most people consistently underestimate uncertainty - - their powers of imagination are unusually worse than their power of mathematics. Come up with your own rules of thumb when trying to make sense of small data sets - - double the difference between the largest and smallest observation. If you have a huge data set - - look at multiplying by 1.5.
Monday, October 11, 2010
The Three A's of Engineering
Engineers need to learn from history what we can and cannot predict. Increasing our comfort level with uncertainly will require us to combine engineering with psychology with statistics. It also involves our utilization of the three A's - - Accept, Assess, and Augment. The following explores and explains the three A's in greater depth.
Labels:
Problem Solving
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.