Modelling Animal-Vehicle Collision Counts across Large Networks Using a Bayesian Hierarchical Model with Time-Varying Parameters

Animal-vehicle collisions (AVCs) are common around the world and result in considerable loss of animal and human life, as well as significant property damage and regular insurance claims. Understanding their occurrence in relation to various contributing factors and being able to identify locations of high risk are valuable to AVC prevention, yielding economic, social and environmental cost savings. However, many challenges exist in the study of AVC datasets. These include seasonality of animal activity, unknown exposure (i.e., the number of animal crossings), very low AVC counts across most sections of extensive roadway networks, and computational burdens that come with discrete response analysis using large datasets. To overcome these challenges, a Bayesian hierarchical model is proposed where the exposure is modeled with nonparametric Dirichlet process, and the number of segmentlevel AVCs is assumed to follow a Binomial distribution. A Pólya-Gamma augmented Gibbs sampler is derived to estimate the proposed model. By using the AVC data of multiple years across about 100,000 segments of state-controlled highways in Texas, U.S., it is demonstrated that the model is scalable to large datasets, with a preponderance of zeros and

[1]  Nathan P. Snow,et al.  A landscape-based approach for delineating hotspots of wildlife-vehicle collisions , 2014, Landscape Ecology.

[2]  W. F. Porter,et al.  Temporal, spatial, and landscape habitat characteristics of moose-vehicle collisions in Western Maine. , 2010 .

[3]  Guohui Zhang,et al.  A latent class approach for driver injury severity analysis in highway single vehicle crash considering unobserved heterogeneity and temporal influence , 2019 .

[4]  Anuj Sharma,et al.  Exploring spatio-temporal effects in traffic crash trend analysis , 2017 .

[5]  V. Radeloff,et al.  Difference in spatiotemporal patterns of wildlife road-crossings and wildlife-vehicle collisions , 2012 .

[6]  Jie He,et al.  Temporal analysis of crash severities involving male and female drivers: A random parameters approach with heterogeneity in means and variances , 2021 .

[7]  R. Jensen,et al.  Landscape factors that contribute to animal–vehicle collisions in two northern Utah canyons , 2014 .

[8]  Liping Fu,et al.  Using a flexible multivariate latent class approach to model correlated outcomes: A joint analysis of pedestrian and cyclist injuries , 2017 .

[9]  D. Croft,et al.  Frequency and causes of kangaroo–vehicle collisions on an Australian outback highway , 2006 .

[10]  J. Bissonette,et al.  Spatial-temporal patterns in Mediterranean carnivore road casualties: Consequences for mitigation , 2009 .

[11]  María J. Alonso,et al.  Testing pole barriers as feasible mitigation measure to avoid bird vehicle collisions (BVC) , 2015 .

[12]  Konstantina Gkritza,et al.  Deer-vehicle collisions, deer density, and land use in Iowa's urban deer herd management zones. , 2010, Accident; analysis and prevention.

[13]  Emilio Díaz-Varela,et al.  Assessing methods of mitigating wildlife-vehicle collisions by accident characterization and spatial analysis , 2011 .

[14]  F. Mannering,et al.  Unobserved heterogeneity and temporal instability in the analysis of work-zone crash-injury severities , 2020, Analytic Methods in Accident Research.

[15]  Giorgos Mountrakis,et al.  Multi‐scale spatiotemporal analyses of moose–vehicle collisions: a case study in northern Vermont , 2009, Int. J. Geogr. Inf. Sci..

[16]  C. Dormann,et al.  Effectiveness of light-reflecting devices: A systematic reanalysis of animal-vehicle collision data. , 2016, Accident; analysis and prevention.

[17]  Antonio Canale,et al.  Bayesian Kernel Mixtures for Counts , 2011, Journal of the American Statistical Association.

[18]  Li Song,et al.  Day-of-the-week variations and temporal instability of factors influencing pedestrian injury severity in pedestrian-vehicle crashes: A random parameters logit approach with heterogeneity in means and variances , 2021 .

[19]  Anuj Sharma,et al.  Using the multivariate spatio-temporal Bayesian model to analyze traffic crashes by severity , 2018 .

[20]  F. Mannering.  Temporal instability and the analysis of highway accident data , 2018 .

[21]  Prateek Bansal,et al.  A new spatial count data model with time-varying parameters , 2020, 2008.03760.

[22]  Jessica Tressou,et al.  Bayesian nonparametric model for clustering individual co-exposure to pesticides found in the French diet. , 2011 .

[23]  James G. Scott,et al.  Modeling unobserved heterogeneity using finite mixture random parameters for spatially correlated discrete count data , 2016 .

[24]  Prateek Bansal,et al.  A Dynamic Choice Model with Heterogeneous Decision Rules: Application in Estimating the User Cost of Rail Crowding , 2020 .

[25]  Panagiotis Ch. Anastasopoulos.  Random parameters multivariate tobit and zero-inflated count data models: addressing unobserved and zero-state heterogeneity in accident injury-severity rate and frequency analysis , 2016 .

[26]  J. Kolowski,et al.  Using Penrose distance to identify potential risk of wildlife–vehicle collisions , 2008 .

[27]  Guohui Zhang,et al.  Modeling animal-vehicle collisions considering animal-vehicle interactions. , 2011, Accident; analysis and prevention.

[28]  Giorgos Mountrakis,et al.  Spatial wildlife-vehicle collision models: a review of current work and its application to transportation mitigation projects. , 2011, Journal of environmental management.

[29]  Srinivas R. Geedipally,et al.  A semiparametric negative binomial generalized linear model for modeling over-dispersed count data with a heavy tail: Characteristics and applications to crash data. , 2016, Accident; analysis and prevention.

[30]  Helai Huang,et al.  Modeling unobserved heterogeneity for zonal crash frequencies: A Bayesian multivariate random-parameters model with mixture components for spatially correlated data , 2019 .

[31]  Yao-Jan Wu,et al.  Modeling animal-vehicle collisions using diagonal inflated bivariate Poisson regression. , 2011, Accident; analysis and prevention.

[32]  Torsten Hothorn,et al.  Temporal patterns of deer-vehicle collisions consistent with deer activity pattern and density increase but not general accident risk. , 2015, Accident; analysis and prevention.

[33]  Dominique Lord,et al.  Multilevel Dirichlet process mixture analysis of railway grade crossing crash data , 2016 .

[34]  Fred L. Mannering,et al.  The heterogeneous effects of guardian supervision on adolescent driver-injury severities: A finite-mixture random-parameters approach , 2013 .

[35]  Michel Bierlaire,et al.  Variational Bayesian Inference for Mixed Logit Models with Unobserved Inter- and Intra-Individual Heterogeneity , 2019, 1905.00419.

[36]  E. Solberg,et al.  Temporal patterns of moose-vehicle collisions with and without personal injuries. , 2017, Accident; analysis and prevention.

[37]  Fred L. Mannering,et al.  A temporal analysis of driver-injury severities in crashes involving aggressive and non-aggressive driving , 2020 .

[38]  Richard Andrášik,et al.  The KDE+ software: a tool for effective identification and ranking of animal-vehicle collision hotspots along networks , 2015, Landscape Ecology.

[39]  Chandra R. Bhat,et al.  Big data, traditional data and the tradeoffs between prediction and causality in highway-safety analysis , 2020, Analytic Methods in Accident Research.

[40]  Ali S Al-Ghamdi,et al.  Warning signs as countermeasures to camel-vehicle collisions in Saudi Arabia. , 2004, Accident; analysis and prevention.

[41]  Rico Krueger,et al.  A Dirichlet Process Mixture Model of Discrete Choice , 2018, 1801.06296.

[42]  Xiaoyan Huo,et al.  Comparative analysis of alternative random parameters count data models in highway safety , 2021 .

[43]  Lancelot F. James,et al.  Gibbs Sampling Methods for Stick-Breaking Priors , 2001 .

[44]  George A. Conway,et al.  Characteristics of Moose-vehicle Collisions in Anchorage, Alaska, 1991–1995 , 1999 .

[45]  Andrew P Tarko,et al.  Markov switching negative binomial models: an application to vehicle accident frequencies. , 2008, Accident; analysis and prevention.

[46]  F. Mannering,et al.  Time-of-day variations and temporal instability of factors affecting injury severities in large-truck crashes , 2019, Analytic Methods in Accident Research.

[47]  Nataliya V Malyshkina,et al.  Zero-state Markov switching count-data models: an empirical assessment. , 2008, Accident; analysis and prevention.

[48]  Sylvia Richardson,et al.  Sampling from Dirichlet process mixture models with unknown concentration parameter: mixing issues in large data implementations , 2013, Statistics and Computing.

[49]  Targeting mitigation efforts: The role of speed limit and road edge clearance for deer-vehicle collisions , 2014 .

[50]  Chris J. Johnson,et al.  Utility of Expert-Based Knowledge for Predicting Wildlife–Vehicle Collisions , 2009 .

[51]  E. Hazebroek,et al.  Ungulate Traffic Collisions in Europe , 1996 .

[52]  Chandra R. Bhat,et al.  Unobserved heterogeneity and the statistical analysis of highway accident data , 2016 .

[53]  Frank T. van Manen,et al.  Effectiveness of Wildlife Underpasses and Fencing to Reduce Wildlife–Vehicle Collisions , 2010 .

[54]  J. Sullivan.  Trends and characteristics of animal-vehicle collisions in the United States. , 2011, Journal of safety research.

[55]  E. Díaz-Varela,et al.  Spatiotemporal analysis of vehicle collisions involving wild boar and roe deer in NW Spain. , 2013, Accident; analysis and prevention.

[56]  Prateek Bansal,et al.  Fast Bayesian Estimation of Spatial Count Data Models , 2020, Comput. Stat. Data Anal..

[57]  Aksel Bo Madsen,et al.  Effectiveness of Wildlife Warning Reflectors in Reducing Deer-Vehicle Collisions: A Behavioral Study , 1998 .

[58]  Changxi Ma,et al.  Temporal stability of driver injury severity in single-vehicle roadway departure crashes: A random thresholds random parameters hierarchical ordered probit approach , 2020 .

[59]  Panagiotis Ch. Anastasopoulos,et al.  Analysis of accident injury-severity outcomes: The zero-inflated hierarchical ordered probit model with correlated disturbances , 2018, Analytic Methods in Accident Research.

[60]  James G. Scott,et al.  Bayesian Inference for Logistic Models Using Pólya–Gamma Latent Variables , 2012, 1205.0310.

[61]  Panagiotis Ch. Anastasopoulos,et al.  A correlated random parameters with heterogeneity in means approach of deer-vehicle collisions and resulting injury-severities , 2021 .

[62]  Fred L. Mannering,et al.  The analysis of vehicle crash injury-severity data: A Markov switching approach with road-segment heterogeneity , 2014 .

[63]  D. Croft,et al.  Assessing the impacts of roads in peri-urban reserves: Road-based fatalities and road usage by wildlife in the Royal National Park, New South Wales, Australia , 2006 .

[64]  A. Caughey,et al.  Cost-benefit analysis of state- and hospital-funded postpartum intrauterine contraception at a university hospital for recent immigrants to the United States. , 2010, Contraception.

[65]  J. Tash,et al.  An Approach Toward Understanding Wildlife-Vehicle Collisions , 2008, Environmental management.

[66]  M. Boyce,et al.  Predicting deer-vehicle collisions in an urban area. , 2011, Journal of environmental management.

[67]  Changxi Ma,et al.  Analysis of injury severity of rear-end crashes in work zones: A random parameters approach with heterogeneity in means and variances , 2020 .

[68]  Wei Li,et al.  Multivariate random parameters zero-inflated negative binomial regression for analyzing urban midblock crashes , 2018 .

[69]  F. Mannering,et al.  The role of gender and temporal instability in driver-injury severities in crashes caused by speeds too fast for conditions. , 2021, Accident; analysis and prevention.

[70]  Nang-Ngai Sze,et al.  Temporal instability of truck volume composition on non-truck-involved crash severity using uncorrelated and correlated grouped random parameters binary logit models with space-time variations , 2021 .

[71]  Francisco Suárez,et al.  Can we mitigate animal–vehicle accidents using predictive models? , 2004 .

[72]  Peter J. Rowden,et al.  Road crashes involving animals in Australia. , 2008, Accident; analysis and prevention.

[73]  R. Courtois,et al.  Electric Fencing as a Measure to Reduce Moose–Vehicle Collisions , 2007 .

[74]  Salvador Hernandez,et al.  Temporal stability of driver injury severities in animal-vehicle collisions: A random parameters with heterogeneity in means (and variances) approach , 2020, Analytic Methods in Accident Research.

[75]  M. Borkovcová,et al.  Estimated mortality of mammals and the costs associated with animal–vehicle collisions on the roads in the Czech Republic , 2013 .

[76]  A. Seiler.  Predicting locations of moose–vehicle collisions in Sweden , 2005 .