Traffic collision is a complex process that involves the interactions between roadways, vehicles, and drivers. There is a relationship between collision and traffic congestion, however, the nature of such relationship is not well understood due to lack of congestion data at time of collision. Accordingly, the objective of this study is to leverage big data analytics techniques to quantify the relationship between safety and traffic level of service (LOS). The 2011–2014 state wide collision data and continuous count station data in Virginia are used for such purpose. An innovative 3-D basic spline filtering technique is adopted to link time, flow, and density for estimating the flow and density values at the time of collisions. The results show the validity of such filtering technique for predicting the flow/density values. In addition, the pattern of collision severity associated with LOS suggests that severe collisions are more likely to occur under good service levels (LOS A and B) while non-severe collisions are more likely to happen at relatively worse service levels (LOS C and D); LOS C has the largest likelihood of collision occurrence in general.
Investigating the Relationship between Collision Characteristics and Traffic Level of Service through Big Data Analytics: A Case Study in the State of Virginia
International Conference on Transportation and Development 2018 ; 2018 ; Pittsburgh, Pennsylvania
2018-07-12
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Framework for Mid-Air Traffic Collision Detection Using Data Analytics
Springer Verlag | 2021
|Trans Tech Publications | 2013
|Investigating the relationship between jobs-housing balance and traffic safety
Online Contents | 2017
|