Friday, April 11, 2008, 2 PM
Olsson Hall 120

How much is too much? A Quest to Identify Normal Traffic Patterns

Ramkumar Venkatanarayana, Ph.D. Candidate

Abstract

Knowledge of the "normal" traffic flow pattern is required for a number of transportation applications. An example of such a pattern is the time-series of traffic volumes experienced on freeways during a.m. and p.m. peak commuter travel. Traditionally, the simple historic average (a simple average of all the data in the history) has been considered as the best way to derive the traffic pattern. However, this method may often be significantly biased by the presence of incidents. One solution to avoid this bias is through visual inspection of the data by experts. The experts could potentially identify anomalies caused by incidents, 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 the potential of information technology, new data analysis tools are essential. In this research, a new tool, the Quantum-Frequency algorithm, was developed, based on density-based clustering. A methodology to evaluate such tools was developed and is being applied to several promising algorithms. These include the Quantum-Frequency algorithm, the traditional k-means clustering, wavelet denoising, and the median. All these algorithms are being applied to several real-world datasets.

During preliminary evaluations, compared to the historic average, the pattern identified by the Quantum-Frequency algorithm resulted in 39 % lower cumulative deviation from the pattern identified manually by experts. A concrete application of the patterns from the Quantum-Frequency algorithm will also be demonstrated, using the newly developed Pattern-Based Model for traffic data archiving.

Biography

Ram Venkatanarayana is a Transportation Systems Engineer with the Smart Travel Laboratory, University of Virginia. He is a Eno Fellow and recipient of the 'Outstanding Student of the Year 2001' Award for the George Mason Consortium University Transportation Center. Ram has a Bachelors degree in Civil Engineering from IIT Madras, and a MS in Civil Engineering from the University of Virginia. He is currently pursuing his PhD in Civil Engineering at UVA, expecting to be a double Hoo by this August.

The Civil Engineering seminar series is open to the University community.
Civil Engineering undergraduate students are especially invited to attend.