Unmet demand is measured in vehicles and is typically estimated by dividing the queue length by the average vehicle occupancy length. The unmet demand queue length is equal to the total queue length minus the served queue length. Average vehicle occupancy lengths typically range from 25 to 50 feet and vary based on local conditions.

Field measurement is a common method to estimate queue length. Real-time online congestion maps (e.g., Google maps, Bing maps, INRIX, etc.) may be used to understand average queue lengths. For intersections, queue length is collected at the start of red and averaged over the entire analysis period. For freeway facilities, queue length is collected and averaged over the entire analysis period. Unmet demand data may also be collected from Traffic Management Centers (TMCs). In circumstances where vehicles are diverting from their normal route due to congestion, a diversion analysis using a TDM or other methods may be used to estimate the number of diverting vehicles.

After unmet demand is estimated, the additional volumes are added to the microsimulation model. Appropriate measures are taken to confirm that all vehicles are allowed entry into the model (e.g., review error logs to see how many vehicles were denied entry). It is important to confirm all vehicles enter the model, because vehicles that are denied entry to the model are not tracked which may lead to inaccurate model results.