This study examined the relationship between working time and the incidence of safety-related incidents for British Rail train drivers. It took in data about booking-on and booking-off times for all shifts of all 16,000 BR drivers over two years and almost 7,500 safety-related incidents occurring to them. This volume of data enabled statistically significant conclusions to be drawn and tested in relatively fine detail. The study found that, within the shift, incident rates peaked in the early hours (2nd to 4th hour) and remained below the average to the limit of 12 hours. This conclusion held when allowance was made for the proportion of each hour spent driving. It was not explained by traffic variation within the day, age, type of work, variation in start time or type of incident. The BR study also found, over the range observed, no worsening of safety performance with long weekly hours, sustained runs of consecutive shifts, or short rest intervals. It did find some evidence of increased risk on the first shift after a break. Neither the duration nor the cause of the break had a significant effect on the risk. The study was later extended to other groups of staff in safety critical jobs with similar findings. (A)
Samenvatting