Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Information Science

Content Description

1 online resource (ix, 118 pages) : color illustrations, color maps.

Dissertation/Thesis Chair

Catherine T Lawson

Committee Members

Sue R Faerman, Eliot Rich


automated vehicles, big data, early warning, policy, traffic, weather, Big data, Traffic engineering, Traffic congestion, Weather forecasting

Subject Categories

Library and Information Science | Public Policy | Urban Studies and Planning


The proliferation of big data has allowed researchers to delve deeper into data and gain better understandings within almost every field. In the fields of transportation planning and traffic management, past research has shown direct relationships among weather conditions and traffic speed, volume, and congestion. However, these studies have mostly relied on static data that were spatially and temporally sparse or collected on a specific roadway for a specific time period for research purposes. With the need to address the impacts of climate change, including an increasing number of extreme weather events as well as an increasing intensity of such events, big data provides researchers with an opportunity to understand the above relationships in more detail. The transportation sector had been one of the early fields to exploit the potentials of big data by developing applications that can detect the amount and speed of traffic on different roads and providing the information to users and facility managers. This research combines traffic and weather conditions data for three different types of road segments in the Capital Region of New York State to understand the link between these weather conditions and traffic speed that can ultimately be used to provide timely advance warning to travelers. The research demonstrates that big data indeed provide advantages over the simple datasets that are commonly available by allowing for more granular level assessments. In particular, it can provide insights into the most important times of day/year and types of roads for safety interventions. The findings are expected to improve traffic risk prediction during severe weather conditions, which would provide a new approach for early warning and impending deployment of automated vehicles. The research suggests a broad array of implications both for the scholarly literature and for practical applications of this type of analysis.