University of Houston civil and environmental engineering professor Lu Gao published research in Accident Analysis and Prevention applying large language models to roadway crash analysis, in a study funded by the Texas Department of Transportation. The LLM component converts unstructured police crash narratives into structured, mechanism-specific labels such as hydroplaning or curve-related loss of control, details typically absent from conventional structured crash fields. A case study linked more than 24,000 police crash narratives to a pavement management dataset of roughly 180,000 records and found strong associations between friction and texture measures and wet-pavement crash mechanisms. Gao states the results can help transportation agencies screen road segments and prioritize pavement-safety projects where crash risk is elevated.