Road Traffic Accident Prediction using an SCGM (1,1)c-Markov Model

Date

2018

Authors

Zou, X.
Yue, W.L.
Li, C.

Editors

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Journal article

Citation

Asian Transport Studies, 2018; 5(1):191-205

Statement of Responsibility

Conference Name

Abstract

The prediction of road traffic accidents is of importance in locating the most hazardous sites in order to improve the safety management level. Through the comparative analysis of traditional road traffic accident prediction methods, an SCGM (1,1)c?Markov prediction model using the grey system theory and Markov chain theory is constructed. By taking the West, North, and South Terraces of Adelaide in South Australia as a case, training and verifying the prediction model is conducted, and the mean absolute percentage error of the combined prediction model is 2.55% which means a high prediction accuracy being achieved. Then, the predicted crash rates of West, North, and South Terraces, which are 99.50, 59.50, and 18.97 crashes/km, respectively, are resulted by the combined models respectively, with West Terrace shown to be the most hazardous road section. The results indicate that the prediction model can be well applied in road traffic accident prediction with strong engineering practicability.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

Copyright 2018 Eastern Asia Society for Transportation Studies

License

Grant ID

Call number

Persistent link to this record