Identifier

etd-07062016-154527

Degree

Master of Science (MS)

Department

Geography and Anthropology

Document Type

Thesis

Abstract

Severe weather events can have a significant impact on transportation networks. Many previous studies tried to analyze and explore the tremendous impact of extreme weather events on traffic behavior, speed, travel time and capacity. The purpose of this research was to analyze and discuss the impact of precipitation, temperature, visibility and wind speed on hourly weekday traffic flow volume in Atlanta, Georgia. This study focused on investigating which weather variables affect traffic volume, developing a machine learning based predictive technique to derive weather-traffic volume decision rules, and building a decision support tool. The correlation between extreme weather events and traffic volume was investigated by comparing traffic volume between a base case scenario and an extreme weather scenario. This research used 2 main data sources: hourly traffic data as recorded by 50 Automatic Traffic Recorder (ATR) sites around Atlanta and hourly precipitation data from 4 climate stations retrieved from the Integrated Surface Hourly (ISH) weather data archives of the National Centers of Environmental Information (NCEI). The statistical analysis and spatiotemporal relationships between traffic volume and weather variables were analyzed individually and evaluated using statistical tests. A machine learning technique was also used to simultaneously examines weather variables and hours of days to predict leading variables that greatly contributed towards a reduction in traffic volume. According to the results of this analysis, there were significant impacts of visibility, precipitation and temperature on traffic volumes especially during certain hours of the day. Extensive statistical analysis proved that during certain hours of the day, individual weather elements such as precipitation, minimum temperature, visibility and wind can have statistically significant individual impact in reducing traffic volume. Machine learning techniques helped derive models that can be used to predict conditions resulting in a decrease in traffic volume. A decision support tool was also developed to visualize traffic volume and weather interactions.

Date

2016

Document Availability at the Time of Submission

Student has submitted appropriate documentation to restrict access to LSU for 365 days after which the document will be released for worldwide access.

Committee Chair

Sathiaraj, David

DOI

10.31390/gradschool_theses.3477

Download

Share

COinS