DynusT applies the widely accepted dynamic user equilibrium (DUE) definition as discussed in the US Transportation Research Board (TRB) published DTA Primer. This primer was authored by the TRB ADB30 Transportation Network Modeling Committee task group led by Professor Yi-Chang Chiu.
The Dynamic User Equilibrium (DUE) seeks to establish equal experienced travel times for vehicles departing at the same time between the same O-D pair taking different routes.
The major innovation in DynusT algorithmic implementation is the Gap Function Vehicle (GFV) Based Traffic Assignment. As shown in the flowchart below, this GFV method explicitly searches for vehicles experiencing less desirable travel time and assign them to better routes within the path set built for the same origin, destination, departure time triplet. As a result, this method would guarantee the gap function -the function that measures the travel time deviation to the shortest path for the same origin, destination, and departure time triplet - value to continue to decrease over iterations.
A typical DynusT DUE run will produce steadily decreasing gap function values over iterations as shown below. A user will typically get below a 10% gap function value around the 10th iteration.
DynuStudio also provides a more detailed view of the gap values for each departure time by iteration. As shown below, the chart on the left shows the progressive gap value reduction for each of the departure times. The chart on the right shows the 24-hr gap value profile over iterations.