Author Type

Graduate Student

Date of Award

Fall 11-5-2025

Document Type

Dissertation

Publication Status

Version of Record

Submission Date

December 2025

Department

Computer and Electrical Engineering and Computer Science

Degree Name

Doctor of Philosophy (PhD)

Thesis/Dissertation Advisor [Chair]

Mehrdad Nojoumian

Abstract

This dissertation advances transportation safety through three interconnected contributions that address fundamental challenges in crash data analysis and application. First, we present a comprehensive survey of road crash analysis methodologies, tracing the evolution from traditional statistical approaches to modern machine learning techniques while identifying critical gaps between research sophistication and practical implementation. This survey reveals that while methodological advances have been substantial, ongoing data quality issues, particularly in location accuracy and completeness, limit the effectiveness of even the most sophisticated analytical approaches.

Building on these findings, we develop a novel system to validate and correct crash location data using multi-modal large language models (LLMs) integrated with geospatial analysis. Our credibility-based scoring system evaluates location accuracy by comparing multiple data sources, analyzing crash narratives and diagrams, and applying spatial validation techniques. Empirical testing on 5,000 Ohio crash reports demonstrates that approximately 20% require location corrections, with our method successfully identifying and correcting these errors through automated post-processing.

The third contribution introduces a risk-aware navigation solution applicable to both human-driven and autonomous vehicles. By integrating historical crash patterns with predicted traffic volumes, we create standardized risk metrics for individual road segments that can be incorporated into routing algorithms. Validation across 56 high-volume commute routes in Ohio demonstrates that safety-prioritized routing reduces crash risk exposure by an average of 22% while increasing travel time by only 9% on average.

Together, these contributions form a cohesive approach to transportation safety from data quality assessment through practical application, demonstrating how advances in data science and analytical methods translate into tangible safety benefits for current transportation systems and emerging navigation technologies.

Download

Share

COinS