Employees driving for work have a proportionately high risk of involvement in a fatal or serious road traffic collision. For example, it has been estimated that 25% of UK road accidents involve someone driving for work. Across the EU, 34% of work fatalities involve road traffic or transport accidents. These collisions are not confined to large trucks but include many smaller vehicles where driving is secondary to the employees’ main tasks. This gives relevance to the study of improving fleet safety. Increasingly fleet risk management has focused on safety and the behaviours of drivers rather than asset management and cost controls. Studies have shown that changing behaviour in both drivers and the rest of the organisation can make large improvements in safety. This study reports on the analysis of detailed insurance claims data from a large UK motor fleet of approximately 40,000 vehicles, since 2001 . A one size fits all approach cannot be expected to work with so many drivers in such a variety of roles. Although each worker is an individual, he or she share patterns of behaviour with those of fellow workers. Rather than treating employees as an amorphous mass they are clustered into homogeneous groups using multivariate statistical techniques based on driver claim histories. The influence of interventions on the behaviour of these groups will contribute to successful safety improvement strategies for each group. The performance of these groups has been studied and interventions successfully applied. Some of these are reported in the paper. The company has had a good track record in applying a range of strategies to produce significant performance improvements. Latterly the rate of improvement has shown signs of slowing so a new view of the drivers and the system in which they operate i s needed. One such approach could be a reconsideration of the segmentation scheme based on the clustering approach. (Author/publisher)
Abstract