If a truck is equipped with GPS, one can programmatically track its movement within the container terminal. It is then possible to identify the yards that were visited, infer the types of containers dropped off or picked up, and measure the time spent stopped in each yard, as well as the time spent driving. This allows a much more granular measurement of terminal service times than the commonly reported aggregate statistic ''truck turn-time.'' Furthermore, the process requires no resources from the terminal. Our methodology is adapted from map-matching, which attempts to align a sequence of GPS readings with a road network. In this case, we attempt to align GPS readings with each of the typical sequences of terminal operations, so as to find the most likely sequence.