skybed there's some interesting work being done in relation to full factor simulation - granted these are essentially multi layer computer models but they have the advantage of taking physical data and overlaying factor that simply cannot be tested without serious risk to the participants.
I a number of these research papers (pdfs) which I'm happy to fwd.
The big trap is that CC ratios cannot be legislated on a type specific basis - and importantly determining a CC ratio is a 'whole of duty' consideration - this includes other essential functions such as fire, security and medical.
Remember Cranflield's work is evac focused and reliant largely on physical tests which are at best surreal and not representative of real life ops on an a/c. - Try repeating some of their tests with tired crew and 'pax' that don't know what they are attending the event for...... In fact watch the A380 evac demo carefully - the 78 seconds of well rehearsed actions even by the pax.