The aim of this study is to see to what extent accidents are preceded by evasive manoeuvres. North Carolina accident data has been used for the analysis. All reported fatal accidents having occurred in the state during three months of 1988, and a small number of non- fatal accidents for comparison reasons are analyzed. The conclusion must be, however, that a rather high percentage of accidents are not preceded by evasive manoeuvers. An optimal conflict definition should therefore include hazardous situations of types other than evasive actions.
Abstract