WHEN S. Travis Waller, a native of tiny Ironton, OH, first experienced the traffic snarls of Chicago, he was shocked. Clearing congestion is now his work. The fundamental research tool in his field is software that models traffic flow, but until recently, the models were capable of representing only static traffic conditions. Waller wrote algorithms that allow for dynamic modeling of changing conditions. As a postdoc at Northwestern University, he led the development of modeling tools based on the algorithms and made them available on the Web. Waller and others are using these tools to analyze traffic-congested areas and specify the most effective long-term investments in infrastructure improvements. Now an assistant professor of civil engineering at the University of Texas at Austin, Waller is also devising an online routing system that uses onboard computers to tell drivers which routes offer the quickest paths given current traffic patterns. Hoping to identify the conditions that slow traffic, he is working with the city of Chicago to test the system by tracking transponder-equipped buses. Eventually, the system could feed data to a central controller that would change red lights to green to improve traffic flow.