Hi, Terminal 5 has certainly struck a cord with you. I've been interested in your experience and your suggestion the other day about whether any modelling has been done.
I am reliably informed (a friend of mine worked on T5 building) that the luggage systems were modelled at X2 capacity and things worked well.

I work in healthcare modelling to mainly to understand disease and model treatment effects. However one of the concepts in modelling that is beautifully illustrated here is the inclusion to the model of complex human interaction. I'm not sure but my guess is that the model for T5 didn't include any facility to include level of skill of the operatives and probably treated them as automatons.
So lets make sure that in the field of OR people are explicit in what they've modelled and how complex it is.
Also here's a piece in the Guardian which illustrates how training was an issue here.
http://www.guardian.co.uk/business/2008/may/08/britishairwaysbusiness.theairlineindustry?gusrc=rss&feed=networkfront

Happy flying
Mike

I think it illustrates the limit of modelling - as Azelrod or someone said the only valid model of a complex system is the system itself.

It doesn't take much to stress test a system with real people - just takes money. I feel sorry for the guys fired, I bet they were under cost targets.