Resumen
A vehicle traveling in an urban network exhibits a fluctuating speed profile, affected by signals, congestion and other factors. Therefore, it is very difficult to determine the volume, speed for multiple intersections without extensive instrumentation. Moreover, urban networks do not show relatively homogeneous sections like freeways where delay calculations can be more complex due to local bottlenecks and multiple queueing patterns. Along the same lines, isolating the impacts of various factors also become problematic. In this paper, an ad-hoc delay calculation approach is employed using the green and yellow taxi GPS dataset (augmented with weather information) in New York City (NYC), which has the aforementioned dense urban street network with controlled intersections ?signalized or stop-signs. Using this approach, delay patterns associated with both NYC green and yellow taxi trips are investigated using their fare structure rather than conventional data sets such as volume, speed and signal timings. Results indicate that the intersection delay patterns differ spatially between different boroughs of NYC. Major delays in the lower Manhattan during the day especially seem to be critical since these delays not only affect the overall traffic flow but also have explicit adverse effects such as elevated air pollution and noise, which is a critical problem in NYC.