The partial trip data file format is similar to the dynamic screenline files with a couple of exceptions. First, it must include two nodes corresponding to paired detector locations similar to nodes A and B as in the screenline case but here the two nodes are not required to be directly connected on the network. We will refer to these detector nodes as DNode A and DNode B to avoid confusion with the connected screenline and turncount nodes. The other key difference is the inclusion of a scaling factor for each partial trip data set corresponding to a paired detector location, this value corresponds to S in Equation 11. Once the user has identified all detector pairs and their corresponding scaling factors, they may set up the dynamic partial trip data file using corresponding counts for each time interval. The general format for a partial trip data file using the equation notation above can thus be given as

where t=time segment, and N=total number of time segments.

The confidence values Vc are optional. Note that unlike the screenline and turncount case, confidence values for partial trip data counts must take into consideration the validity of the scaling factor in addition to any uncertainty in the count itself. An example partial trip data file is provided in Figure 18.