Ceri wrote:

>The reason I'm getting aereated about all this is that there seems
>to be an American, mangement viewpoint that uses computer simulation
>in real life situations to simply improve efficiency. Complexity is
>more than efficiency. It may even be the opposite - necessary
>inefficiency to establish an emergent phenomenon. Life,
>consciousness and everything.........

Ah- ha

Of course complexity is more than efficiency. Pseudo-biological simulations may just develop non-mechanical algorithms, but - and this is where we differ I think - those algorithms can generate something which has "many of the features of a complex system". I don't think we can say more than this because there is no First Law of Complexity, against which to compare it.

What's more such "complex" (small c) simulations will, in some cases be so abstracted from reality as to have all determinants included, and therefore be explicable in a A does B while C is doing D model, but still they do Complex things like displaying emergence (ie not foreseeable from components at component level), scale-independence and clustering around attractors.

There is a valid argument that such complex features may actually not just be the sole preserve of self-organisation but will actually arise from evolution, and even (shock horror) good design. This stems not from management gurus but from statistical physicists. Incidentally such systems do have high inherent efficiency in the long run: although they look inefficient in the short term, by having flexibility and residual capacity most of the time, they are more resilient and thus productive for much more of the time. Whether it needs a full-scale buy in of complexity theory, this is a message health managers could do with getting their heads round - as you will know more than most.

And at the end of the day, if the computer sim uses enough richness of data to model what really happens then surely use it - but the key is that the sim has to be continually compared against the thing it is modelling. And when they differ, it's the simulation which is wrong.

Clarification or just more snow in your forecast blizzard?

--
Chris Burton, [log in to unmask] on 28/11/2002

>
>Ceri