It is possible that these three ‘residual’ peaks
may reflect time on task effects since it is clear that
the start time of journeys will tend to cluster around
certain times of day. Although few studies have
examined time on task effects in transport safety,
those that have tend to find a similar trend to that
observed in occupational injuries, for example,
Folkard (1996). Surprisingly, risk does not show a
monotonic increase over time on task. Rather, there
is a transient 2-4 hour peak in risk superimposed on
an otherwise exponential increase in risk over time
on task. The nature of this trend is such that only
when individuals have been working for ca 12 hours
or more will their risk rise above that found during
the 2-4 hour peak. This implies that there may be an
optimal duration of task to minimize risk and that,
with the exception of duties of up to 2.5 hours, the
safest duration may be ca 8-10 hours. While there
appear to be no published studies on the underlying
cause(s) of the transient 2-4 hour peak in risk, it may
reflect on the need to ‘reautomatize’ even highly
learned complex skills during the first few hours of
any period of duty. Clearly there is a strong need for
further research aimed at both elucidating the underlying
cause(s) of this 2-4 hour transient peak and
determining whether it is indeed responsible for the
‘residual’ time of day peaks in accident risk.