Chapter 4 - Data-Driven Approaches for Estimating Travel Time Reliability

Abstract

Due to the extensive availability of various traffic sensors on roadways, traffic data collection has become easier and cheaper. Travel time reliability (TTR), which benefits from data availability, has attracted increasing attention in recent years, and is often listed as one of the major roadway performance and service quality measures for both traffic engineers and travelers. Measuring TTR is the first step toward improving TTR, ensuring on-time arrivals and reducing travel costs. Four basic components are usually considered to measure TTR, including travel time estimation/collection, quantity of travel time selection, probability distribution selection, and TTR index selection. This chapter empirically proves the concept that “the same data tells you the same story” and TTR measures are insensitive to the probability distribution selection. This chapter also covers an additional component beyond the basic components, which is the estimation of the accuracy of TTR measures. The bootstrap technique, which is a data-driven technique and is based on resampling with replacement, plays an important role in estimating the accuracy. The accuracy estimations provide more general pictures for estimated TTR rather than a single TTR value. In addition, the concept of segment-based TTR on roadways is extended to origin-destination (OD)-based TTR over roadway networks. The characters of OD-based TTR are briefly discussed in this chapter. This chapter summarizes our continuous efforts to improving the accuracy of TTR estimation and concept extensions, resulting in contributing to data-driven traffic studies and applications.