Prediction of arrival times of freight traffic on us railroads using support vector regression

Machine learning;Railroad;Prediction;Support vector machine;Arrival time;Data mining;Feature engineering

Barbour, William Walker ; Work ; Daniel B.

关键词: Machine learning;    Railroad;    Prediction;    Support vector machine;    Arrival time;    Data mining;    Feature engineering;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/97803/BARBOUR-THESIS-2017.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Variability of the travel times on the United States freight rail network is high due to large network demand relative to infrastructure capacity especially when traffic is heterogeneous. Variable runtimes pose significant operational challenges if the nature of runtime variability is not predictable. To address this issue, this article proposes a data-driven approach to predict estimated times of arrival (ETAs) of individual freight trains, based on the properties of the train, the properties of the network, and the properties of potentially conflicting traffic on the network. The ETA problem is posed as a machine learning regression problem and solved using a support vector regression machine trained and cross validated on over two years of historical data for a 140 mile stretch of track located primarily in Tennessee, USA. The article presents the data used in this problem and details on feature engineering and construction for predictions made across the full route. It also highlights findings on the dominant sources of runtime variability and the most predictive factors for ETA, identified by applying the data framework. ETA improvement results exceeded 20% over baseline methods for predictions made at some locations and averaged over 15% across the study area. Ideas for further ETA improvement using the prediction algorithms are also discussed.