What has been investigated
CMC's accident analysis is based on real, reconstructed accidents from the German In-Depth Accident Study (GIDAS) and its pre-crash matrix simulation database (GIDAS-PCM).
The reason to choose this database, which includes data from all kinds of traffic accidents with personal damage, is its extensive accident information collection such as accident site, vehicle, participant, and injury; including the accident reconstruction and possibility for proper weighting. Furthermore, the accident information includes items like road conditions, speed, visibility, trajectories, actions of the participants etc.
CMC has started its accidentology research with some of the most frequent types of motorcycle accidents that were found in the database: 'Crossing Traffic' and 'Left Turn' scenarios.
In CMC's analysis, a total of 23 potential influencing factors were investigated and reported, including the ones that eventually did not appear to have an important contribution to the accident. The simulation database allowed an additional 5 investigations to understand the accident situation more clearly, including the 'Time-To-Collision'. All this is gathered in the reports "Accident Analysis - Crossing Traffic and Left Turn" and 'Accident Analysis - Additional Crossing Traffic'.
In addition, 'Single accidents' (driving accidents where no other party was involved) were investigated. These include left curve / right curve / straight scenarios, as presented in the report "Accident Analysis - Single Accidents'.
‘Longitudinal Traffic’ analyses the situation where a road user collides with a road user who is driving directly in front.
The ‘U-turn’ report analyses the situation where a road user is turning around and collides with upcoming or following traffic.
Finally, 'Lane Change' is about situations where a road user changes lanes or enters lanes or changes direction and then collides with road users who were going in the same direction.
The reason to choose this database, which includes data from all kinds of traffic accidents with personal damage, is its extensive accident information collection such as accident site, vehicle, participant, and injury; including the accident reconstruction and possibility for proper weighting. Furthermore, the accident information includes items like road conditions, speed, visibility, trajectories, actions of the participants etc.
CMC has started its accidentology research with some of the most frequent types of motorcycle accidents that were found in the database: 'Crossing Traffic' and 'Left Turn' scenarios.
In CMC's analysis, a total of 23 potential influencing factors were investigated and reported, including the ones that eventually did not appear to have an important contribution to the accident. The simulation database allowed an additional 5 investigations to understand the accident situation more clearly, including the 'Time-To-Collision'. All this is gathered in the reports "Accident Analysis - Crossing Traffic and Left Turn" and 'Accident Analysis - Additional Crossing Traffic'.
In addition, 'Single accidents' (driving accidents where no other party was involved) were investigated. These include left curve / right curve / straight scenarios, as presented in the report "Accident Analysis - Single Accidents'.
‘Longitudinal Traffic’ analyses the situation where a road user collides with a road user who is driving directly in front.
The ‘U-turn’ report analyses the situation where a road user is turning around and collides with upcoming or following traffic.
Finally, 'Lane Change' is about situations where a road user changes lanes or enters lanes or changes direction and then collides with road users who were going in the same direction.
Important outcomes
From the Crossing Traffic and Left Turn analysis, an important outcome is that no explicit view obstruction existed in the majority of cases, but still participant A (mainly cars/ trucks) overlooked participant B (mainly PTWs) or misjudged the situation. This implies that there is a need for technology support to inform participant A of the existence of oncoming participant B.
In addition, the study of TTC shows that the earliest possibility to notify a vehicle driver/ rider is 4.5 seconds (median) before the collision in case of crossing traffic accident type 302, while it is 1.5 seconds (median) in case of left turn accident type 211.
In addition, the study of TTC shows that the earliest possibility to notify a vehicle driver/ rider is 4.5 seconds (median) before the collision in case of crossing traffic accident type 302, while it is 1.5 seconds (median) in case of left turn accident type 211.
Visual overview
Scroll through the slides to see a short introduction to this topic
Documentation
Nr. 1 - Crossing Traffic and Left Turn - Scroll through the complete document here below, or download it as PDF:
Nr. 2 - Additional Crossing Traffic scenarios - Scroll through the complete document here below, or download it as PDF:
Nr. 4 - Longitudinal Traffic scenarios - Scroll through the complete document here below, or download it as PDF:
Nr. 5 - U-turn scenarios - Scroll through the complete document here below, or download it as PDF:
Nr. 6 - Lane change scenarios - Scroll through the complete document here below, or download it as PDF: