United States Department of Transportation
i
Segment-Level Crash Risk Analysis for New Jersey Highways Using Advanced Data Modeling
-
2020-06-01
-
[PDF-883.82 KB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:Rutgers University. Center for Advanced Infrastructure and Transportation ; United States. Department of Transportation. Federal Highway Administration ; United States. Department of Transportation. Office of the Secretary for Research and Technology ; United States. Department of Transportation. University Transportation Centers (UTC) Program
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report 01/01/2020 – 05/31/2020
-
Corporate Publisher:
-
Abstract:Highway crashes are the most significant challenge to the goal of providing a safe and efficient highway transportation system. They result in significant societal toll reflected in numerous fatalities, personal injuries, property damage, and traffic congestion. To that end, much attention has been given to developing models to study and predict crash occurrence and severity. Most of these models are reactive: they aim to identify the significant crash factors, crash hot-spots and crash-prone roadway locations, analyze and select the most effective countermeasures for reducing the number and severity of crashes. More recently advancements have been made in developing proactive crash risk models, aiming to assess crash risks in a short term, and inform traffic management strategies to prevent and mitigate the negative effects of crashes. This study developed and tested several models for segment-level crash risk and severity assessment considering the data available to most transportation agencies in real time on a regional network scale. The data included roadway geometry characteristics, traffic flow characteristics, and weather condition data. The models included Bayesian Logistics Regression, Decision Tree, Random Forest, Gradient Boosting Machine, K-Nearest Neighbor, and Gaussian Naïve Bayes (GNB). The models were trained and tested using a dataset containing records of 10,155 crashes that occurred on two interstate highways in New Jersey over a period of two years. It was found that for the given dataset the models provided limited predictive value.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
