Theoretical and Empirical Analysis of the Effects of Transit System Operations on Urban Networks

Georgia Tech Research Corporation, Atlanta GA

Investigators

Abstract

This award explores the premise that the recently developed theory of the Network Fundamental Diagram (NFD) can be extended to account for the main effects that transit systems' operational characteristics (such as bus frequency, number of stops, number of routes) have on macroscopic network performance measures (such as, network capacity, mean travel time and probability of gridlock) and for the effects of network performance on transit operations. To accomplish this, the research approach will be to combine NFD theory with traffic flow theory of moving bottlenecks to produce the proposed NFD-transit theory and evaluate its results using both empirical data and a microscopic simulation model especially developed for this project. The empirical data will come from the metro Atlanta region and includes traffic data from loop detectors at signalized intersections and transit data from buses' GPS devices and ridership surveys. If successful, this award will improve our understanding of the cause-and-effect mechanisms between transit systems and transportation network performance. The research could be potentially transformative since traditional travel forecasting models cannot address this problem satisfactorily because of their reliance on travelers' origins and destinations. The proposed method is macroscopic and therefore does not require microscopic travel patterns. Although it is expected that the real-world data will deviate from the assumptions of the theory, valuable insights will be obtained by carefully studying how these deviations affect predictions and how this theory can be improved. The project will develop urban congestion reduction strategies that consider both public and private transportation alternatives. The results will be beneficial to the population of large urban areas, especially to underrepresented groups who often live in transit-dependent areas. The main findings will be disseminated through graduate-level courses, simulation and analysis tools made available online and through two workshops for planning agencies.