Graduation Semester and Year
2023
Language
English
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Civil Engineering
Department
Civil Engineering
First Advisor
Mohsen Shahandashti
Abstract
Monitoring real-time information on road conditions, especially during winter storms, is crucial to establishing winter maintenance strategies by the State Departments of Transportation (State DOTs) in the United States. States and local highway agencies allocate substantial resources each year for winter operations to improve road safety during winter storms. However, most weather-related vehicle accidents happen on snowy, slushy, or icy surfaces, leading to a significant number of fatalities and injuries annually. Traditionally, transportation agencies rely on the information provided by Road Weather Information Systems for monitoring road conditions along roadways. However, these systems are costly and only provide estimates at specific locations, resulting in distant areas being underrepresented. Additionally, the data acquisition systems integrated into snowplows, which offer real-time road condition images, comprise several components and are not extensively employed in states that encounter infrequent snowstorms. This is primarily due to the high costs associated with installation and maintenance. The main objective of this study is to develop a cost-effective system that enables the monitoring of road conditions by providing real-time road condition images and estimating road surface temperatures. The approach developed to collect and transfer real-time road conditions images proved advantageous as it eliminated the need for complex and expensive multi-component systems, while also reducing the training requirements for users. Furthermore, a methodology was developed to leverage publicly accessible weather forecasts provided by the National Weather Service for estimating road surface temperatures on roadways (excluding bridges). This innovative approach enabled the estimation of road surface temperature without depending on expensive road weather information systems, which may not be accessible in many locations. In particular, statistical models were developed to establish relationships between road surface temperature and the weather forecasts that were publicly available in high resolution. The findings of the study indicated that linear statistical models, like multiple linear regression, could achieve an acceptable level of accuracy for estimating road surface temperature. However, it was observed that nonlinear models, such as Random Forest, could enhance accuracy by capturing the intricate and nonlinear interactions among the explanatory variables. Lastly, a data visualization platform (i.e., digital twin system) was created to display the real-time road conditions by combining the functionalities of mobile devices and capabilities of the ArcGIS application programming interface The findings of this study emphasize the practicality of utilizing gridded weather forecasts, supplied by the National Weather Services, to estimate the temperature of road surface as well as utilizing the functionalities of mobile devices to communicate road conditions information. The proposed methodology can be integrated into a winter operation decision-making system to visualize the road conditions images and map the estimated road surface temperature on roadways without the need for road weather information systems. The estimated road surface temperatures on roadways assists highway agencies to plan winter maintenance strategies more efficiently by taking proactive measures in areas where low surface temperatures are estimated.
Keywords
Winter operations, Road conditions Information, Publicly available data
Disciplines
Civil and Environmental Engineering | Civil Engineering | Engineering
License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.
Recommended Citation
Darghiasi, Pooya, "Digital Twin Enabled Winter Operations Management Through the Integration of Artificial Intelligence, Sensory Level Data, and Publicly Available Data" (2023). Civil Engineering Dissertations. 436.
https://mavmatrix.uta.edu/civilengineering_dissertations/436
Comments
Degree granted by The University of Texas at Arlington