01622577

Component

This paper first introduces the concept of traffic safety region to real-time crash risk evaluation. A hybrid intelligent algorithm, combining sequential forward selection (SFS), principal components analysis (PCA) and least squares support vector machines (LSSVM), is presented to estimate traffic safety region and classify the traffic safety states. Based on the estimated traffic safety region, safety margin is calculated to measure the traffic crash risk in real time. To demonstrate the advantage of the proposed method, this paper develops two crash risk evaluation models, namely SFS-LSSVM model and PCA-LSSVM model, based on crash data and non-crash data collected on freeway I-880N in Alameda. Validation results show that the method is of reasonably high accuracy for identifying traffic safety states, and then the safety margin is a meaningful indicator for real-time crash risk evaluation.

This paper was sponsored by TRB committee ABJ70 Standing Committee on Artificial Intelligence and Advanced Computing Applications.

01618707

17-02829

English

17p

2017

Transportation Research Board 96th Annual Meeting

Digital/other

Figures; References; Tables

Classification; Crash risk forecasting; Reliability; Traffic safety

Interstate 880

Principal component analysis

Data and Information Technology; Highways; Safety and Human Factors

Transportation Research Board Annual Meeting 2017 Paper #17-02829

TRIS, TRB, ATRI

Dec 8 2016 11:03AM