Understanding and Predicting Travel Time with Spatio-Temporal Features of Network Traffic Flow, Weather and Incidents

Abstract

Travel time on a route varies substantially by time of day and from day to day. It is critical to understand to what extent this variation is correlated with various factors, such as weather, incidents, events or travel demand level in the context of dynamic networks. This helps a better decision making for infrastructure planning and realtime traffic operation. We propose a data-driven approach to understand and predict highway travel time using spatio-temporal features of those factors, all of which are acquired from multiple data sources. The prediction model holistically selects the most related features from a high-dimensional feature space by correlation analysis, principle component analysis and LASSO. We test and compare the performance of several regression models in predicting travel time 30 min in advance via two case studies: (1) a 6-mile highway corridor of I-270N in D.C. region, and (2) a 2.3-mile corridor of I-376E in Pittsburgh region. We found that some bottlenecks scattered in the network can imply congestion on those corridors at least 30 minutes in advance, including those on the alternative route to the corridors of study. In addition, real-time travel time is statistically related to incidents on some specific locations, morning/afternoon travel demand, visibility, precipitation, wind speed/gust and the weather type. All those spatio-temporal information together help improve prediction accuracy, comparing to using only speed data. In both case studies, random forest shows the most promise, reaching a root-mean-squared error of 16.6% and 17.0% respectively in afternoon peak hours for the entire year of 2014.

Seasonal Disorder in Urban Traffic Patterns: A Low Rank Analysis

“…Matrix factorization can be viewed as an approach to reduce the dimensionality or compress Asif et al (2013) traffic data. Traffic data dimensionality techniques span more classical techniques building on principal component analysis Li et al (2007), Yang and Qian (2019), Li et al (2015), to more recent developments using variational autoencoders Boquet et al (2020). Exploiting spatio-temporal patterns can both support data reduction and also prediction of future traffic states on the network, Yang and Qian (2019).…”

Learning to Recommend Signal Plans under Incidents with Real-Time Traffic Prediction

“…In these cases, past traffic dynamics on the target road segments may have little useful information implying their future traffic states. A widely used solution found in literature is to take into account spatiotemporal correlations between target road segments and nearby segments ( 19, 20 , 42 ). As it takes time for traffic to propagate, abnormal traffic nearby can work as longer-term predictors.…”

“…Linear regression models with L 1 regularization, that is, LASSO, which use the same feature set as the neural network model, are built for each segment i and prediction horizon h independently ( 19, 43 ). The model learns the weights w h ( i ) such that the loss in Equation 8 is minimized.…”

“…Recent models are built with spatiotemporal traffic flows, traffic events and incidents, and weather data to trigger traffic volume or speed prediction 5 to 30 min ahead (16)(17)(18)(19)(20). Historical average, linear models such as autoregression or LASSO, local regression and nearest neighborhood methods, graphical models, and deep neural networks are common modeling choices (16,(18)(19)(20)(21)(22)(23)(24)(25)(26). Encoder-decoder recurrent neural network is a popular deep learning architecture initially proposed for machine translation (27).…”

Towards an End-to-End Framework of CCTV-Based Urban Traffic Volume Detection and Prediction

“…As reported in [ 46 ], RF can handle overfitting with sufficient speed when large data series are used, compared to other similar regression tree algorithms. A more recent study in [ 47 ] reported that RF achieved the lowest average errors compared to other regression models for predicting travel times across an urban road network. An empirical analysis in [ 48 ] acknowledged the limitation that fine-tuning is a time-consuming procedure.…”