MPC 17-338 | J. Wood and S. Zhang Evaluating Relationships Between Perception- Reaction Times, Emergency Deceleration Rates, and Crash Outcomes Using Naturalistic Driving Data A University Transportation Center sponsored by the U.S. Department of Transportation serving the Mountain-Plains Region. Consortium members: Colorado State University University of Colorado Denver Utah State University North Dakota State University University of Denver University of Wyoming South Dakota State University University of Utah Evaluating Relationships Between Perception-Reaction Times, Emergency Deceleration Rates, and Crash Outcomes Using Naturalistic Driving Data Jonathan Wood, PhD South Dakota State University Phone: (435) 760-2781 Email: [email protected] Shaohu Zhang South Dakota State University Email: [email protected] December 2017 Acknowledgments The Mountain-Plains Consortium (MPC) and South Dakota State University (SDSU) funded this research. Virginia Tech University (VTU) provided the data used in the analysis. Disclaimer The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the information presented. This document is disseminated under the sponsorship of the Department of Transportation, University Transportation Centers Program, in the interest of information exchange. The U.S. Government assumes no liability for the contents or use thereof. NDSU does not discriminate in its programs and activities on the basis of age, color, gender expression/identity, genetic information, marital status, national origin, participation in lawful off-campus activity, physical or mental disability, pregnancy, public assistance status, race, religion, sex, sexual orientation, spousal relationship to current employee, or veteran status, as applicable. Direct inquiries to: Vice Provost, Title IX/ADA Coordinator, Old Main 201, 701-231-7708, [email protected]. ABSTRACT Perception-reaction times (PRT) and deceleration rates are critical components in the design of highways and streets. This research has several objectives, including 1) evaluate differences in PRT and deceleration rates between crash and near-crash events, 2) assess the correlation between PRT and deceleration rate, 3) determine if there is a causal relationship between PRT and deceleration rate (and what it is), and 4) develop predictive models for PRT and deceleration rate that can be used for roadway design and crash reconstruction. These objectives were met by applying multiple statistical analysis techniques to the SHRP2 naturalistic driving data. The analysis results indicated that crash events were associated with longer PRT values and lower deceleration rates. The Pearson correlation between PRT and deceleration rate was low. However, PRT was a causal factor of deceleration rate in both crash and near-crash events. In crash events, longer PRT values were associated with lower deceleration rates. In near-crash events, longer PRT values were associated with higher deceleration rates. Regression models for crash reconstruction were estimated using panel and quantile regression methods. Applications of these models for both purposes are illustrated and discussed. The results for design applications are compared with existing AASHTO design guidance. TABLE OF CONTENTS 1. Introduction ............................................................................................................................................... 1 1.1 Problem Statement ........................................................................................................................ 1 1.2 Objectives ..................................................................................................................................... 2 1.3 Scope ............................................................................................................................................. 3 1.4 Outline of Report .......................................................................................................................... 3 2. Background 4 2.1 Overview ....................................................................................................................................... 4 2.2 Literature Review .......................................................................................................................... 4 2.2.1 Deceleration Rate .................................................................................................................. 4 2.2.2 Crash Reconstruction ............................................................................................................ 6 2.2.3 Perception-reaction Time ...................................................................................................... 7 3. Research Methods ..................................................................................................................................... 9 3.1 Overview ....................................................................................................................................... 9 3.2 Counterfactual Framework and Statistical Matching .................................................................... 9 3.2.1 Matching Method .................................................................................................................. 9 3.2.2 Covariate Balance ............................................................................................................... 10 3.2.3 Estimating the Treatment Effect ......................................................................................... 11 3.2.4 Hidden Bias Sensitivity Analysis ........................................................................................ 11 3.3 Correlation .................................................................................................................................. 12 3.4 Graphical Causal Models ............................................................................................................ 12 3.5 Panel Regression ......................................................................................................................... 14 3.6 Quantile Regression .................................................................................................................... 16 4. Data Collection and Processing .............................................................................................................. 17 4.1 Overview ..................................................................................................................................... 17 4.2 Data ............................................................................................................................................. 17 4.3 Data Exploration ......................................................................................................................... 20 4.3.1 Event Severity ..................................................................................................................... 20 4.3.2 Crash Severity ..................................................................................................................... 20 4.3.3 Gender ................................................................................................................................. 20 4.3.4 Age ...................................................................................................................................... 21 4.3.5 Road Alignment .................................................................................................................. 22 4.3.6 Pavement Surface Condition ............................................................................................... 22 4.3.7 Lighting ............................................................................................................................... 23 4.3.8 Speed ................................................................................................................................... 24 5. Analysis and Results ............................................................................................................................... 25 5.1 Overview ..................................................................................................................................... 25 5.2 Differences Between Crash and Near-crash Events .................................................................... 25 5.3 Correlation and Causation Between Perception-reaction Times and Deceleration Rates .......... 26 5.3.1 DAG for Crash Events ........................................................................................................ 26 5.3.2 DAG for Near-crash Events ................................................................................................ 28 5.3.3 DAG for Combined Events ................................................................................................. 29 5.4 Predictive Models ....................................................................................................................... 30 5.5 Design Recommendations........................................................................................................... 33 5.6 Crash Reconstruction .................................................................................................................. 37 6. Conclusions 39 6.1 Summary ....................................................................................................................................