The Institute of Electrical and Electronics Engineers (IEEE) published the Texas A&M Transportation Institute’s (TTI’s) paper “Using an Interpretable Machine Learning Framework to Understand the Relationship of Mobility and Reliability Indices on Truck Drivers’ Route Choices.” (Note: To access the full article, readers will need a subscription.) A member of the TTI Mobility Division, Graduate Research Assistant Xiaoqiang “Jack” Kong served as lead author on the paper; his co-authors included Bill Eisele, Mobility Division head and senior research engineer; Yunlong Zhang, professor and associate department head of graduate programs in Texas A&M University’s Zachry Department of Civil and Environmental Engineering; and Xiao Xiao, graduate student worker.
IEEE published TTI’s paper in IEEE Transactions on Intelligent Transportation Systems, a journal focused on information technology design, analysis and control as they relate to transportation. In this paper, the research team tracks relationships between travel-time index and planning-time index and offers insights into how truckers make daily decisions about routes. These indices are often used to measure traffic congestion and travel-time reliability, which impact freight costs, schedules and planning.
Kong shares, “Truck routing has always been a critical but overlooked issue in transportation. Through the lens of Big Data analytics, our research team provides a better understanding of truck routing behavior from a unique perspective: travel-time reliability. As lead author, I’ve enjoyed working with my co-authors — including prominent researchers in the artificial intelligence (Dr. Zhang) and freight research (Dr. Eisele) fields. I’ve realized the power of teamwork and collaboration, and glimpsed the promising future of seeking solutions for transportation issues through Big Data analytics.”
The TTI team developed a model based on an interpretable machine learning algorithm. Through the study, the team discovered that truck drivers are more likely to be influenced by real-time congestion information if drivers are traveling the majority of the total trip time. The study also revealed complex, nonlinear impacts of travel-time reliability and real-time congestion information on truck drivers’ route choices. Only a limited number of research studies have addressed how mobility and reliability indices play into truckers’ route decisions, but this paper opens the door for comprehensive studies on this topic in the future.
“Freight transportation is critical to transportation systems and the economy in general, and route choices for trucks substantially affect operational costs and overall traffic flow conditions,” notes Eisele. “A better understanding of route choices from mobility and reliability perspectives — based on both current and historical experiences — could help transportation agencies provide information to truck drivers at the most appropriate time and fashion for their individual route choice decision-making.”
You must be logged in to post a comment.