Multilevel model for ramp crash frequency that reflects heterogeneity among ramp types

Freeway ramps on a trumpet interchange are classified into six types by their configuration (loop-, direct-, and semi-direct-) and function (on-/off-). Since the effects of crash factors vary for different ramp types, each ramp type has different characteristics for crash occurrence, and this yields heterogeneity among ramp types. Therefore, this study compares the effects of crash factors that vary for different ramp types by developing a multilevel model that reflects the heterogeneity of crashes among ramp types. Based on 1,155 ramp crashes data over a four-year period, three multilevel models with two hierarchies were estimated; an unconditional model, a random coefficients regression model, and a full random coefficients model. The fixed effects of models indicated that offramps are more vulnerable to crash than on-ramps, and crash frequency increases as AADT increases and/or ramp length shortens. The results also implied that the ramp length has different effects on crashes according to ramp types. The random effects showed that the intra-class correlation was 0.185, indicating that 18.5% of the total variance is contributed to the heterogeneity among different ramp types. The findings from the estimated models provide an enhanced understanding about ramp crashes and contribute to the safe design and maintenance of freeway ramps.

[1]  Anthony S. Bryk,et al.  Hierarchical Linear Models: Applications and Data Analysis Methods , 1992 .

[2]  S. Miaou.  The relationship between truck accidents and geometric design of road sections: Poisson versus negative binomial regressions. , 1994, Accident; analysis and prevention.

[3]  Ezra Hauer,et al.  On exposure and accident rate , 1995 .

[4]  Karin M Bauer,et al.  STATISTICAL MODELS OF ACCIDENTS ON INTERCHANGE RAMPS AND SPEED-CHANGE LANES , 1998 .

[5]  R. B. Albin,et al.  Evaluating Median Crossover Likelihoods with Clustered Accident Counts: An Empirical Inquiry Using the Random Effects Negative Binomial Model , 1998 .

[6]  Andrew P Jones,et al.  The use of multilevel models for the prediction of road accident outcomes. , 2003, Accident; analysis and prevention.

[7]  James A Bonneson,et al.  Calibration of Predictive Models for Estimating Safety of Ramp Design Configurations , 2005 .

[8]  Andrew Gelman,et al.  Data Analysis Using Regression and Multilevel/Hierarchical Models , 2006 .

[9]  Vikash V. Gayah,et al.  Crash Risk Assessment Using Intelligent Transportation Systems Data and Real-Time Intervention Strategies to Improve Safety on Freeways , 2007, J. Intell. Transp. Syst..

[10]  S. Emerson,et al.  AASHTO (American Association of State Highway and Transportation Officials). 2001. A Policy on Geometric Design of Highways and Streets. Fourth Edition. Washington, D.C. , 2007 .

[11]  Byoung-Jo Yoon.  Development of Traffic Accident Forecasting Models for Trumpet IC Ramp Types of Freeway , 2007 .

[12]  Darren J Torbic,et al.  SafetyAnalyst: Software Tools for Safety Management of Specific Highway Sites , 2007 .

[13]  Mohamed Abdel-Aty,et al.  Multilevel data and bayesian analysis in traffic safety. , 2010, Accident; analysis and prevention.

[14]  Paul P Jovanis,et al.  Spatial Correlation in Multilevel Crash Frequency Models , 2010 .

[15]  Priyanka Alluri,et al.  Effects of State-Specific SPFs, AADT Estimations, and Overdispersion Parameters on Crash Predictions Using SafetyAnalyst , 2012 .

[16]  Chungwon Lee,et al.  Safety Performance Functions Reflecting Categorical Impact of Exposure Variables for Freeways , 2013 .

[17]  Kaan Ozbay,et al.  Spatial Analysis of County Level Crash Risk in New Jersey Using Severity-Based Hierarchical Bayesian Models , 2014 .

[18]  Bhagwant Persaud,et al.  Methodology to develop crash modification functions for road safety treatments with fully specified and hierarchical models. , 2014, Accident; analysis and prevention.

[19]  H. Teng,et al.  Safety Analysis of Freeway Segments with Random Parameters , 2015 .

[20]  Mohammed A. Quddus,et al.  Effects of Geodemographic Profiles of Drivers on Their Injury Severity from Traffic Crashes Using Multilevel Mixed-Effects Ordered Logit Model , 2015 .