Saturday, 15 December 2007

Problem Solving:: Heuristic Analysis vs. Formal Methods?

Brand Killer Robots reveal::
You went to school! Some of you went for a very long time. The longer you went, the better you became at applying formal methods. The better you became at interpreting a problem then solving the problem by formally breaking it into logical structures, sequences and algorithms. For you, solving a problem is a simple process of start, middle and end. For you, given sufficient data and direction, solving a problem is a simple process of start, middle and end. About starting your day at 9.00am, solving problems til lunchtime and working from after lunch to five o'clock solving more problems. Given sufficient data and direction, it is all a simple process of start, middle and end.

As the years go by your databank of logical structures, sequences and algorithms become larger and so burnt into your memory, that the speed of your RAM increases, so that you can solve known problem patterns faster and faster.

Your mind becomes stronger and stronger at solving known problems and for most businesses -- this is just the type of mind that is required.

But what happens when the "formal methods brain" is confronted by a problem pattern that it has never seen before? What happens when the formal methods brain is confronted with an "unknown problem". Where there is not necessarily a start, a middle and an end?

What happens is the formal method brain switches to "trial n error" mode, in other words - "switches in to heuristic thinking mode". Problem is, the formal method brain is not used to operating in the uncertain world of the "heurstic thinker". The formal thinker has ingrained patterns and biases which constantly try to override the heuristic process, making it much more difficult for formal methods thinkers to solve problems in a world of uncertainty.

But the heuristic thinker is trained to think in terms of 'trial n error' and is no stranger to solving problems within a domain of uncertaintly. In fact, the heuristic thinker has the reverse problem of the formal methods thinker, because whilst they are strong in a world of uncertainty - they are generally pretty weak at solving problems, in a world of certainty.

Conclusion is "that there is room for both types of thinker - just make sure you assess which of your staff are which type of thinker - and employ each to work on the problems that are most relevant to the way they think".

No comments: