Automated identification of traffic patterns : final report.

Author(s)
Smith, B.L. & Venkatanarayana, R.
Year
Abstract

Knowledge of the 'normal traffic patterns' on the roadways is essential for a number of transportation applications such as signal system retiming and performance measurement. The simple historic average – the average of all the traffic data in a dataset, by the time of day – has traditionally been used widely to derive these traffic patterns. However, this method is significantly biased by the presence of traffic abnormalities (such as crashes, and inclement weather). One solution to avoid this bias is through visual inspection of the data by experts. The experts could potentially identify and eliminate the traffic abnormalities, and thereby identify the underlying 'normal traffic patterns'.Three main challenges of this approach are: (1) the bias introduced due to subjectivity, (2) the additional time required to analyze the data manually, and (3) the increasing sizes of the available traffic data sets. To address the above challenges, and also exploit new opportunities in the form of traffic data archives, new data analysis tools are essential. In this research study, a new method, the Quantum-Frequency algorithm, was developed, based on density-based clustering. Three other algorithms – K-Means Clustering, Wavelet-based Clustering and Median – were identified as promising algorithms, and developed further. A 10-step methodology was developed to evaluate these promising algorithms, along with the traditional historic average algorithm. Under this methodology, the above algorithms were applied to several real-world datasets. Based on their regular convergence and high accuracy, the Quantum-Frequency Algorithm (with optimized parameter values) and the Median algorithms are recommended for practical applications. Key contributions of this study include (1) a new definition for 'normal traffic pattern', (2) development, application and optimization of the Quantum-Frequency Algorithm, (3) development and application of the 10-step evaluation methodology, and (4) first documented quantification of the bias in the widely-used historic average algorithm. (Author/publisher)

Request publication

2 + 0 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.

Publication

Library number
20081005 ST [electronic version only]
Source

Charlottesville, VA, University of Virginia, Center for Transportation Studies, ITS Implementation Research Center, 2008, VII + 83., 83 ref.

Our collection

This publication is one of our other publications, and part of our extensive collection of road safety literature, that also includes the SWOV publications.