Predicting accident frequency at their severity levels and its application in site ranking using a two-stage mixed multivariate model.

Accident prediction models (APMs) have been extensively used in site ranking with the objective of identifying accident hotspots. Previously this has been achieved by using a univariate count data or a multivariate count data model (e.g. multivariate Poisson-lognormal) for modelling the number of accidents at different severity levels simultaneously. This paper proposes an alternative method to estimate accident frequency at different severity levels, namely the two-stage mixed multivariate model which combines both accident frequency and severity models. The accident, traffic and road characteristics data from the M25 motorway and surrounding major roads in England have been collected to demonstrate the use of the two-stage model. A Bayesian spatial model and a mixed logit model have been employed at each stage for accident frequency and severity analysis respectively, and the results combined to produce estimation of the number of accidents at different severity levels. Based on the results from the two-stage model, the accident hotspots on the M25 and surround have been identified. The ranking result using the two-stage model has also been compared with other ranking methods, such as the naïve ranking method, multivariate Poisson-lognormal and fixed proportion method. Compared to the traditional frequency based analysis, the two-stage model has the advantage in that it utilises more detailed individual accident level data and is able to predict low frequency accidents (such as fatal accidents). Therefore, the two-stage mixed multivariate model is a promising tool in predicting accident frequency according to their severity levels and site ranking.

[1]  Gary A Davis,et al.  Possible aggregation biases in road safety research and a mechanism approach to accident modeling. , 2004, Accident; analysis and prevention.

[2]  K. Kockelman,et al.  Bayesian Multivariate Poisson Regression for Models of Injury Count, by Severity , 2006 .

[3]  Gudmundur F. Ulfarsson,et al.  Bicyclist injury severities in bicycle-motor vehicle accidents. , 2007, Accident; analysis and prevention.

[4]  Bhagwant Persaud,et al.  Empirical Bayes before-after safety studies: lessons learned from two decades of experience and future directions. , 2007, Accident; analysis and prevention.

[5]  Joon Jin Song,et al.  Bayesian multivariate spatial models and their applications , 2004 .

[6]  J. T. Wulu,et al.  Regression analysis of count data , 2002 .

[7]  Md. Mazharul Haque,et al.  Right-Angle Crash Vulnerability of Motorcycles at Signalized Intersections , 2010 .

[8]  J. Guzmán.  Regression Models for Categorical Dependent Variables Using Stata , 2013 .

[9]  Li-Yen Chang,et al.  Analysis of traffic injury severity: an application of non-parametric classification tree techniques. , 2006, Accident; analysis and prevention.

[10]  Fred L. Mannering,et al.  An exploratory multinomial logit analysis of single-vehicle motorcycle accident severity , 1996 .

[11]  Bryan K Allery,et al.  Relationships between Safety and Both Congestion and Number of Lanes on Urban Freeways , 2008 .

[12]  Robert B. Noland,et al.  Congestion and Safety: A Spatial Analysis of London , 2003 .

[13]  Chao Wang,et al.  Impact of traffic congestion on road accidents: a spatial analysis of the M25 motorway in England. , 2009, Accident; analysis and prevention.

[14]  Toshiyuki Yamamoto,et al.  Underreporting in traffic accident data, bias in parameters and the structure of injury severity models. , 2008, Accident; analysis and prevention.

[15]  M. Quddus.  Modelling area-wide count outcomes with spatial correlation and heterogeneity: an analysis of London crash data. , 2008, Accident; analysis and prevention.

[16]  J. Ivan,et al.  Explaining two-lane highway crash rates using land use and hourly exposure. , 2000, Accident; analysis and prevention.

[17]  Bani K. Mallick,et al.  ROADWAY TRAFFIC CRASH MAPPING: A SPACE-TIME MODELING APPROACH , 2003 .

[18]  Ezra Hauer,et al.  How Best to Rank Sites with Promise , 2004 .

[19]  P Bartlett,et al.  Road casualties Great Britain: 2007 - annual report , 2008 .

[20]  Chao Wang,et al.  Road Traffic Congestion and Crash Severity: Econometric Analysis Using Ordered Response Models , 2010 .

[21]  P. Jovanis,et al.  Spatial analysis of fatal and injury crashes in Pennsylvania. , 2006, Accident; analysis and prevention.

[22]  Shaw-Pin Miaou,et al.  Bayesian ranking of sites for engineering safety improvements: decision parameter, treatability concept, statistical criterion, and spatial dependence. , 2005, Accident; analysis and prevention.

[23]  A. Hole.  Fitting Mixed Logit Models by Using Maximum Simulated Likelihood , 2007 .

[24]  D. Graham,et al.  Spatial Variation in Road Pedestrian Casualties: The Role of Urban Scale, Density and Land-use Mix , 2003 .

[25]  Srinivas Reddy Geedipally,et al.  Examining Methods for Estimating Crash Counts According to Their Collision Type , 2010 .

[26]  Gregory K. Leonard,et al.  A utility-consistent, combined discrete choice and count data model Assessing recreational use losses due to natural resource damage , 1995 .

[27]  Andrew P. Jones,et al.  District Variations in Road Curvature in England and Wales and their Association with Road-Traffic Crashes , 2007 .

[28]  Paul Damien,et al.  A multivariate Poisson-lognormal regression model for prediction of crash counts by severity, using Bayesian methods. , 2008, Accident; analysis and prevention.

[29]  C J O'Donnell,et al.  Predicting the severity of motor vehicle accident injuries using models of ordered multiple choice. , 1996, Accident; analysis and prevention.

[30]  Eun Sug Park,et al.  Multivariate Poisson-Lognormal Models for Jointly Modeling Crash Frequency by Severity , 2007 .

[31]  Bhagwant Persaud,et al.  Comparison of empirical Bayes and full Bayes approaches for before-after road safety evaluations. , 2010, Accident; analysis and prevention.

[32]  Fred L. Mannering,et al.  The relationship among highway geometrics, traffic-related elements and motor-vehicle accident frequencies , 1998 .

[33]  R Elvik,et al.  Black Spot Management and Safety Analysis of Road Networks , 2007 .

[34]  C. Bhat.  Simulation estimation of mixed discrete choice models using randomized and scrambled Halton sequences , 2003 .

[35]  L J Mountain,et al.  The identification of accident blackspots: a comparison of current methods. , 1988, Accident; analysis and prevention.

[36]  Simon Washington,et al.  Validation of FHWA Crash Models for Rural Intersections: Lessons Learned , 2003 .

[37]  P. Haan,et al.  Estimation of Multinomial Logit Models with Unobserved Heterogeneity using Maximum Simulated Likelihood , 2006 .

[38]  David J. Spiegelhalter,et al.  WinBUGS user manual version 1.4 , 2003 .

[39]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .

[40]  Paul P Jovanis,et al.  Bayesian Multivariate Poisson Lognormal Models for Crash Severity Modeling and Site Ranking , 2009 .

[41]  Yuanchang Xie,et al.  Predicting motor vehicle collisions using Bayesian neural network models: an empirical analysis. , 2007, Accident; analysis and prevention.

[42]  Andrew Gelman,et al.  General methods for monitoring convergence of iterative simulations , 1998 .

[43]  D. Hensher,et al.  A mixed generalized ordered response model for examining pedestrian and bicyclist injury severity level in traffic crashes. , 2008, Accident; analysis and prevention.

[44]  Fred L. Mannering,et al.  The statistical analysis of crash-frequency data: A review and assessment of methodological alternatives , 2010 .

[45]  Fred L Mannering,et al.  Highway accident severities and the mixed logit model: an exploratory empirical analysis. , 2008, Accident; analysis and prevention.

[46]  Chao Wang,et al.  A spatio-temporal analysis of the impact of congestion on traffic safety on major roads in the UK , 2013 .

[47]  F Mannering,et al.  Effect of roadway geometrics and environmental factors on rural freeway accident frequencies. , 1995, Accident; analysis and prevention.

[48]  Dominique Lord,et al.  The prediction of accidents on digital networks, characteristics and issues related to the application of accident prediction models , 2000 .

[49]  Wen Cheng,et al.  Experimental evaluation of hotspot identification methods. , 2005, Accident; analysis and prevention.

[50]  K. El-Basyouny,et al.  Collision prediction models using multivariate Poisson-lognormal regression. , 2009, Accident; analysis and prevention.

[51]  Fred Mannering,et al.  Probabilistic models of motorcyclists' injury severities in single- and multi-vehicle crashes. , 2007, Accident; analysis and prevention.

[52]  Richard Williams,et al.  Review of Regression Models for Categorical Dependent Variables Using Stata, Second Edition, by Long and Freese , 2006 .

[53]  F Mannering,et al.  Analysis of injury severity and vehicle occupancy in truck- and non-truck-involved accidents. , 1999, Accident; analysis and prevention.

[54]  J. Wielinski,et al.  Annual Meeting of the Transportation Research Board , 2010 .

[55]  Md. Mazharul Haque,et al.  Empirical Evaluation of Alternative Approaches in Identifying Crash Hot Spots , 2009 .

[56]  F Mannering,et al.  Statistical analysis of accident severity on rural freeways. , 1996, Accident; analysis and prevention.

[57]  B. Mallick,et al.  Bayesian multivariate spatial models for roadway traffic crash mapping , 2006 .