This study deals with 17,063 traffic accidents that have taken place on the main segments of expressways in Korea during the past seven years. The objectives of the study are to uncover the various heterogeneous factors linked to traffic accidents through developing accident severity models by crash...
This study deals with 17,063 traffic accidents that have taken place on the main segments of expressways in Korea during the past seven years. The objectives of the study are to uncover the various heterogeneous factors linked to traffic accidents through developing accident severity models by crash type (single vehicle, multi-vehicle and overall) based on the above data, and to propose the safety policies to reduce traffic accident severity using the analysis results.
In pursuing the above, this study gives particular attentions to applying a multi-level ordered logit model (MLOLM) that takes into account the hierarchical structure of traffic accident data and includes the spatial heterogeneity of the group level (highway segment unit). The existing models, however, consider just the heterogeneity of the single and individual level.
Main results of the study can be summarized as follows.
First, through developing the traffic accident severity models by crash type (single-vehicle, multi-vehicle and overall), 19, 27 and 31 significant accident severity factors are relatively derived and spatial heterogeneity is tested in accident severity with regard to multi-vehicle and overall accidents. On the other hand, eight common variables are adopted and other different (specific) variables are derived with regard to each crash type.
Second, the severity of multi-vehicle and overall accidents are analyzed to be more influenced by such the fixed factors as road factors (geometric) and accident factors (crash type) than the changeable road conditions like weather conditions. In contrast, the severity of single-vehicle accidents is evaluated to be significantly affected by the changeable factors including weather and season.
Third, the most relevant factors to the severity derived through the odds ratio (OR) are evaluated to be the speeding as a human factor, vans, special cars and trucks as vehicle factors, and downhill sections as a road factor.
Finally, the different accident severity factors by crash type shows that differentiated safety strategies are necessary in order to reduce traffic accident severity.
This study deals with 17,063 traffic accidents that have taken place on the main segments of expressways in Korea during the past seven years. The objectives of the study are to uncover the various heterogeneous factors linked to traffic accidents through developing accident severity models by crash type (single vehicle, multi-vehicle and overall) based on the above data, and to propose the safety policies to reduce traffic accident severity using the analysis results.
In pursuing the above, this study gives particular attentions to applying a multi-level ordered logit model (MLOLM) that takes into account the hierarchical structure of traffic accident data and includes the spatial heterogeneity of the group level (highway segment unit). The existing models, however, consider just the heterogeneity of the single and individual level.
Main results of the study can be summarized as follows.
First, through developing the traffic accident severity models by crash type (single-vehicle, multi-vehicle and overall), 19, 27 and 31 significant accident severity factors are relatively derived and spatial heterogeneity is tested in accident severity with regard to multi-vehicle and overall accidents. On the other hand, eight common variables are adopted and other different (specific) variables are derived with regard to each crash type.
Second, the severity of multi-vehicle and overall accidents are analyzed to be more influenced by such the fixed factors as road factors (geometric) and accident factors (crash type) than the changeable road conditions like weather conditions. In contrast, the severity of single-vehicle accidents is evaluated to be significantly affected by the changeable factors including weather and season.
Third, the most relevant factors to the severity derived through the odds ratio (OR) are evaluated to be the speeding as a human factor, vans, special cars and trucks as vehicle factors, and downhill sections as a road factor.
Finally, the different accident severity factors by crash type shows that differentiated safety strategies are necessary in order to reduce traffic accident severity.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.