Assessing Pedestrian Safety on Roads Through Machine Learning Approaches for State Highways in Washington State - Final Project Report

Aug 5, 2026

Upload by Hazel

The report evaluates multiple machine learning approaches to understand pedestrian fatal collisions on Washington State’s highways. Four classification methods are applied to quantify how roadway features relate to fatal crash outcomes: Logistic Regression, Nearest Neighbor Classification, Decision Tree, and Random Forest Classifier. Project data come from the Highway Safety Information System (HSIS), combining statewide collision records with roadway characteristics for all state highways. Models are built with K-fold cross-validation and assessed using accuracy scores and confusion matrices. Results show the Decision Tree delivered the most consistent performance and strongest overall results.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages40

Reading Time1 h 41 min

AccessFree to download

Executive Summary

Chapter 1. Introduction

Research Background

Problem Statement

Research Objectives

Chapter 2. Literature Review

Traditional Statistical Pedestrian Safety Analysis

Machine Learning based Pedestrian Safety Analysis

Chapter 3. Roadway Pedestrian Safety Analysis

Assessing Pedestrian Safety on Roads Through Machine Learning Approaches for State Highways in Washington State FINAL PROJECT REPORT by Yinhai Wang, Wei Sun, Sam Ricord, Cesar Maia de Souza University of Washington for Center for Safety Equity in Transportation (CSET) USDOT Tier 1 University Transportation Center University of Alaska Fairbanks ELIF Suite 240, 1764 Tanana Drive Fairbanks, AK 99775-5910 In cooperation with U.S. Department of Transportation, Research and Innovative Technology Administration (RITA) 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 herein. This document is disseminated under the sponsorship of the U.S. Department of Transportation’s University Transportation Centers Program, in the interest of information exchange. The Center for Safety Equity in Transportation, the U.S. Government and matching sponsor assume no liability for the contents or use thereof. | TECHNICAL REPORT DOCUMENTATION PAGE | | | | | | --- | --- | --- | --- | --- | | 1. Report No. | 2. Government Accession No. | 3. Recipient’s Catalog No. | | | | 4. Title and Subtitle
Assessing Pedestrian Safety on Roads Through Machine Learning Approaches for State Highways in Washington State | | 5. Report Date
July 1, 2024 | | | | | | 6. Performing Organization Code | | | | 7. Author(s) and Affiliations
Yinhai Wang, Wei Sun, Sam Ricord, Cesar Maia de Souza | | 8. Performing Organization Report No.
INE/CSET 24.09 | | | | 9. Performing Organization Name and Address
Center for Safety Equity in Transportation
ELIF Building Room 240, 1760 Tanana Drive
Fairbanks, AK 99775-5910 | | 10. Work Unit No. (TRAIS) | | | | | | 11. Contract or Grant No.
69A34520501020620 | | | | 12. Sponsoring Organization Name and Address
United States Department of Transportation
Research and Innovative Technology Administration
1200 New Jersey Avenue, SE
Washington, DC 20590 | | 13. Type of Report and Period Covered
Final report, Sep 2019 – Sep 2022 | | | | | | 14. Sponsoring Agency Code | | | | 15. Supplementary Notes
Report uploaded to: | | | | | | 16. Abstract
The report presents a unique contrast for several Machine Learning approaches aiming at understanding pedestrian fatal collisions. Four classification techniques are applied to assess how roadway features mainly correlate to pedestrian fatal crashes: Logistic Regression, Nearest Neighbor Classification, Decision Tree, and Random Forest Classifier. The data used in this project was collected from the Highway Safety Information System (HSIS) database, which provides both collision data for the entire state of Washington and roadway characteristics for all state highways. Each of the four modeling approaches was implemented using K-fold crossvalidation, a process that allows choosing the best parameters for the model. Their results were evaluated and then compared in terms of accuracy score and confusion matrices for the testing data set. It was found that the Decision tree had consistent results and the best performance among all models, showing how the distinct predictors relate to each other to predict fatal pedestrian collisions. | | | | | | 17. Key Words
Safety Data Tool; Roadway Safety Assessment; Pedestrian safety, collisions, severity, HSIS database, Machine Learning, Statistical and Machine Learning Modeling, Classification methods; | | | 18. Distribution Statement | | | 19. Security Classification (of this report)
Unclassified. | 20. Security Classification (of this page)
Unclassified. | | 21. No. of Pages
40 | 22. Price
N/A | Form DOT F 1700.7 (8-72) Reproduction of completed page authorized. SI* (MODERN METRIC) CONVERSION FACTORS TABLE OF CONTENTS Disclaimer............................................................................................................................

FAQs

Which machine learning classification techniques are used to assess pedestrian fatal collisions?

The report applies Logistic Regression, Nearest Neighbor Classification, Decision Tree, and Random Forest Classifier.

What data source supports the modeling in this project?

Modeling uses the Highway Safety Information System (HSIS) database, which provides statewide collision data and roadway characteristics for Washington’s state highways.

How are the models trained and evaluated, and which approach performed best?

Each approach is implemented using K-fold cross-validation and evaluated with accuracy scores and confusion matrices. The Decision Tree shows the most consistent results and the best overall performance among the models.

Aug 5, 2026

Upload by Hazel