This is quite interesting. In the US, in order to decide how fatigue plays
a part in accidents, they've used biomathematical fatigue models.

Their Summary Report which has just come out about their fatigue assessment
tool makes fascinating reading. For the short press release see
http://www.fra.dot.gov/us/press-releases/121

For the actual report, you can read the 42-page document he

http://www.fra.dot.gov/downloads/saf...a_ord_0621.pdf

In a nutshell, it demonstrates how a biomathematical fatigue model can be
used to assess how much work schedule factors can contribute to increased
fatigue, which in turn makes accidents and errors more likely. The validity
of the models is assessed using FAST (Fatigue Avoidance Scheduling Tool)
software. The report is quite mathematical in nature (as one would expect),
correlating accident risk with effectiveness and coming up with complex
equations such as x2 = 7.201, p 0.01. In other words, this report may
not be everyone's cup of tea. But I'm sure some of you will find it
interesting, nonetheless.

The main conclusion of the study is that you can indeed utilise
biomathematics when looking at preventing accidents. Whilst this is the
case with the railway industry, I would have thought it could be relevant
in many other sectors. To demonstrate this, the SAFTE (Sleep, Activity,
Fatigue and Task Effectiveness) model that was used for the study was
originally developed for the US armed forces.

I also would have thought that the findings would be equally applicable to
the UK. It's certainly a clever way of assessing risk.

I guess some of you will think "D'oh - why spend so much money in coming to
a conclusion that tiredness makes people make mistakes". And you'd be
right, of course. But it's the methodology in which they have come to the
conclusion that is particularly interesting.