Crash frequency modeling for signalized intersections in a high-density urban road network

Kun Xie, Xuesong Wang, Kaan Ozbay, Hong Yang

Research output: Contribution to journalArticle

Abstract

Conventional crash frequency models rely on an assumption of independence among observed crashes. However, this assumption is frequently proved false by spatially related crash observations, particularly for intersection crashes observed in high-density road networks. Crash frequency models that ignore the hierarchy and spatial correlation of closely spaced intersections can lead to biased estimations. As a follow-up to our previous paper (Xie et al., 2013), this study aims to address this issue by introducing an improved crash frequency model. Data for 195 signalized intersections along 22 corridors in the urban areas of Shanghai was collected. Moran's I statistic of the crash data confirmed the spatial dependence of crash occurrence among the neighboring intersections. Moreover, Lagrange Multiplier test was performed and it suggested that the spatial dependence should be captured in the model error term. A hierarchical model incorporating a conditional autoregressive (CAR) effect term for the spatial correlation was developed in the Bayesian framework. A deviance information criterion (DIC) and cross-validation test were used for model selection and comparison. The results showed that the proposed model outperformed traditional models in terms of the overall goodness of fit and predictive performance. In addition, the significance of the corridor-specific random effect and CAR effect revealed strong evidence for the presence of heterogeneity across corridors and spatial correlation among intersections.

Original languageEnglish (US)
Pages (from-to)39-51
Number of pages13
JournalAnalytic Methods in Accident Research
Volume2
DOIs
StatePublished - 2014

Fingerprint

road network
Lagrange multipliers
multiplier
deviant behavior
urban area
statistics
Statistics
performance
evidence

Keywords

  • Crash frequency model
  • Hierarchical conditional autoregressive model
  • Hierarchy
  • High-density network
  • Signalized intersection
  • Spatial correlation

ASJC Scopus subject areas

  • Safety Research
  • Transportation

Cite this

Crash frequency modeling for signalized intersections in a high-density urban road network. / Xie, Kun; Wang, Xuesong; Ozbay, Kaan; Yang, Hong.

In: Analytic Methods in Accident Research, Vol. 2, 2014, p. 39-51.

Research output: Contribution to journalArticle

Xie, Kun ; Wang, Xuesong ; Ozbay, Kaan ; Yang, Hong. / Crash frequency modeling for signalized intersections in a high-density urban road network. In: Analytic Methods in Accident Research. 2014 ; Vol. 2. pp. 39-51.
@article{9380c5f5dd2f44d696d6357a2ee37d85,
title = "Crash frequency modeling for signalized intersections in a high-density urban road network",
abstract = "Conventional crash frequency models rely on an assumption of independence among observed crashes. However, this assumption is frequently proved false by spatially related crash observations, particularly for intersection crashes observed in high-density road networks. Crash frequency models that ignore the hierarchy and spatial correlation of closely spaced intersections can lead to biased estimations. As a follow-up to our previous paper (Xie et al., 2013), this study aims to address this issue by introducing an improved crash frequency model. Data for 195 signalized intersections along 22 corridors in the urban areas of Shanghai was collected. Moran's I statistic of the crash data confirmed the spatial dependence of crash occurrence among the neighboring intersections. Moreover, Lagrange Multiplier test was performed and it suggested that the spatial dependence should be captured in the model error term. A hierarchical model incorporating a conditional autoregressive (CAR) effect term for the spatial correlation was developed in the Bayesian framework. A deviance information criterion (DIC) and cross-validation test were used for model selection and comparison. The results showed that the proposed model outperformed traditional models in terms of the overall goodness of fit and predictive performance. In addition, the significance of the corridor-specific random effect and CAR effect revealed strong evidence for the presence of heterogeneity across corridors and spatial correlation among intersections.",
keywords = "Crash frequency model, Hierarchical conditional autoregressive model, Hierarchy, High-density network, Signalized intersection, Spatial correlation",
author = "Kun Xie and Xuesong Wang and Kaan Ozbay and Hong Yang",
year = "2014",
doi = "10.1016/j.amar.2014.06.001",
language = "English (US)",
volume = "2",
pages = "39--51",
journal = "Analytic Methods in Accident Research",
issn = "2213-6657",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Crash frequency modeling for signalized intersections in a high-density urban road network

AU - Xie, Kun

AU - Wang, Xuesong

AU - Ozbay, Kaan

AU - Yang, Hong

PY - 2014

Y1 - 2014

N2 - Conventional crash frequency models rely on an assumption of independence among observed crashes. However, this assumption is frequently proved false by spatially related crash observations, particularly for intersection crashes observed in high-density road networks. Crash frequency models that ignore the hierarchy and spatial correlation of closely spaced intersections can lead to biased estimations. As a follow-up to our previous paper (Xie et al., 2013), this study aims to address this issue by introducing an improved crash frequency model. Data for 195 signalized intersections along 22 corridors in the urban areas of Shanghai was collected. Moran's I statistic of the crash data confirmed the spatial dependence of crash occurrence among the neighboring intersections. Moreover, Lagrange Multiplier test was performed and it suggested that the spatial dependence should be captured in the model error term. A hierarchical model incorporating a conditional autoregressive (CAR) effect term for the spatial correlation was developed in the Bayesian framework. A deviance information criterion (DIC) and cross-validation test were used for model selection and comparison. The results showed that the proposed model outperformed traditional models in terms of the overall goodness of fit and predictive performance. In addition, the significance of the corridor-specific random effect and CAR effect revealed strong evidence for the presence of heterogeneity across corridors and spatial correlation among intersections.

AB - Conventional crash frequency models rely on an assumption of independence among observed crashes. However, this assumption is frequently proved false by spatially related crash observations, particularly for intersection crashes observed in high-density road networks. Crash frequency models that ignore the hierarchy and spatial correlation of closely spaced intersections can lead to biased estimations. As a follow-up to our previous paper (Xie et al., 2013), this study aims to address this issue by introducing an improved crash frequency model. Data for 195 signalized intersections along 22 corridors in the urban areas of Shanghai was collected. Moran's I statistic of the crash data confirmed the spatial dependence of crash occurrence among the neighboring intersections. Moreover, Lagrange Multiplier test was performed and it suggested that the spatial dependence should be captured in the model error term. A hierarchical model incorporating a conditional autoregressive (CAR) effect term for the spatial correlation was developed in the Bayesian framework. A deviance information criterion (DIC) and cross-validation test were used for model selection and comparison. The results showed that the proposed model outperformed traditional models in terms of the overall goodness of fit and predictive performance. In addition, the significance of the corridor-specific random effect and CAR effect revealed strong evidence for the presence of heterogeneity across corridors and spatial correlation among intersections.

KW - Crash frequency model

KW - Hierarchical conditional autoregressive model

KW - Hierarchy

KW - High-density network

KW - Signalized intersection

KW - Spatial correlation

UR - http://www.scopus.com/inward/record.url?scp=84904544866&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904544866&partnerID=8YFLogxK

U2 - 10.1016/j.amar.2014.06.001

DO - 10.1016/j.amar.2014.06.001

M3 - Article

VL - 2

SP - 39

EP - 51

JO - Analytic Methods in Accident Research

JF - Analytic Methods in Accident Research

SN - 2213-6657

ER -