Linking behavioural indicators to safety: what is safe driving and what is not?

Safety is often quantified by the number of traffic accidents and the severity of those accidents. A significant part of the causes of traffic accidents can be traced back to the driver. In recent years, there were several efforts to describe driver behaviour and its relation with risk, by evaluating the output of driving behaviour (i.e. behavioural indicators) using measures like speed, time headway, time to collision and lateral position. But what determines a situation to be critical or unsafe? This is the central research question in a new TNO project. Although there are established functions that describe the relationship between some of these indicators and the level of risk, they are still widely based on historical accident data, are usually not linked to their surroundings or to other behavioural indicators, seldom include temporal factors, and researchers have almost never reached consensus concerning critical values.