Improving Interstate Freeway Travel Time Reliability Analysis by Clustering Travel Time Distributions

Abstract

Travel time reliability quantifies variability in travel times and has become a critical aspect for evaluating transportation network performance. The empirical travel time cumulative distribution function (CDF) has been used as a tool to preserve inherent information on the variability and distribution of travel times. With advances in data collection technology, probe vehicle data has been frequently used to measure highway system performance. One challenge with using CDFs when handling large amounts of probe vehicle data is deciding how many different CDFs are necessary to fully characterize experienced travel times. This paper explores statistical methods for clustering CDFs of travel times at segment level into an optimal number of homogeneous clusters that retain all relevant distributional information. Two clustering methods were tested, one based on classic hierarchical clustering and the other used model-based functional data clustering, to find out their performance on clustering distributions using travel time data from Interstate 64 in Virginia. Freeway segments and those within interchange areas were clustered separately. To find the proper data format as clustering input, both scaled and original travel times were considered. In addition, a non-data-driven method based on geometric features was included for comparison. The results showed that for freeway segments, clustering using travel times and the Anderson–Darling dissimilarity matrix and Ward’s linkage had the best performance. For interchange segments, model-based clustering provided the best clusters. By clustering segments into homogenous groups, the results of this study could improve the efficiency of further travel time reliability modeling.

Prediction of Travel Time Reliability on Interstates Using Linear Quantile Mixed Models

“…To improve efficiency, the analysis was based on clustered data such that prediction models were only necessary for each cluster rather than for hundreds of individual segments (5,6).…”

“…On the other hand, developing a universal CDF for all studied segments could likely risk the loss of substantial information. Zhang et al ( 5 ) showed that one solution to this challenge is to cluster segments with similar CDF shapes into the same group, then estimate CDFs at the group level to quantify impact factors and make predictions. This paper uses travel time data from interstate segments that has been clustered into approximately homogeneous groups using the methods proposed by Zhang et al ( 5 , 6 ).…”

“…Zhang et al ( 5 ) showed that one solution to this challenge is to cluster segments with similar CDF shapes into the same group, then estimate CDFs at the group level to quantify impact factors and make predictions. This paper uses travel time data from interstate segments that has been clustered into approximately homogeneous groups using the methods proposed by Zhang et al ( 5 , 6 ). A brief overview of the clustering process is provided; however, the focus is on developing LQMMs at a group level, which could be used to improve the target setting process required by MAP-21 as well as to assess potential impacts of future projects.…”

Methodology for Predicting MAP-21 Interstate Travel Time Reliability Measure Target in Virginia

“…In the transportation literature, CART have been used for many applications including safety ( 12 – 14 ), transit ( 15 ), and workzone ( 16 ) studies. A parallel study tested linear quantile mixed models and generalized random forests for predicting Interstate reliability and recommended the latter because they were found to be more sensitive to the impact of events ( 17 ).…”

An expressway traffic congestion measurement under the influence of service areas

“…Caceres, H further estimated route travel time distribution by the probabilistic model with considering the time of day, inclement weather, and traffic incidents; then, he found that route travel time variability increases when the mean travel time increases [24]. Some studies show that when the length of a segment is long enough and the impact of traffic accidents is small, the value of Travel Time Reliability (TTR) will be normal [29].…”