In a review of the literature it was found that the current yellow-timing in the Netherlands should be prolonged by about one second (given the present three seconds for a 50 km/h intersection and four seconds for a 70 or 80 km/h intersection) to adapt to drivers' needs in normal circumstances and to what under more extreme conditions can be expected. In literature no information was available on the long term influence of a prolongation of the yellow time on the decision behaviour of car drivers. Therefore a field experiment (before/after study, period one year) was conducted. Also the influence of different detection distances and different waiting times was investigated. At six locations (the total experimental area included 30 signalised intersections where the yellow time was prolonged) the behaviour of car drivers in a relevant area in front of the stop line was observed and analysed quantitatively by means of time-lapse video. Although a small adaptation of drivers' behaviour can be distinguished, (no difference in behaviour between six and twelve months after the prolongation of the yellow time was found) the net result of the longer yellow timing appears to be a halving of the number of run red offences. Therefore, it is recommended that a yellow phase of four seconds for 50 km/h intersections and one of five seconds for 80 km/h intersections should be used. Consistent effects of different detection distances and different waiting times on drivers' behaviour could not be found, either because of specific local circumstances or because the number of observations was too small.
Abstract