Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
time period was divided into 18 10-minute intervals. Three additional 10-minute intervals were used for net- work clearance. A total of 2.56 million vehicle trips were assigned to the network. The TransCAD approximator cannot directly recognize tolls, so the delay- based tolls were added to the free- flow travel time for each link using an assumed value of travel time of $10 per vehicle hour. The computation time for the three models varied, ranging from approximately 30 minutes for STA to 3 weeks for VISTA. ⢠The most noticeable difference in the results from the STA and the DTA approximator is in projection of links to be congested under static conditions. The DTA approximator predicts a higher level of congestion. The STA predicted total system travel time of 1.27 million vehicle hours. The DTA approximator prediction of 2.53 million vehicle hours was nearly double the STA estimate. VISTA predicted a total system travel time of 3.09 million vehicle hours. These results indicate that STA models may underpredict congestion due to changes in demand over the peak period. The three applications also resulted in different traffic routing. STA assigns more vehicles to freeways, VISTA assigns more vehicles to arterials and collectors, and the DTA approximator distributed traffic more comparably among the roadway types. These dif- ferences result from the fundamental distinctions between the link performance function- based approach and the capacity- constrained cell transmission model. ⢠The assessment shows that it is possible to com- pare the results from the three different models based on global measures of system performance. It is more diffi- cult to compare the results on a link- by- link basis because of the fundamental differences in the models. The results from both dynamic models showed much higher congestion levels than the results from the static model. Additionally, the cell transmission model and link performance functions provided very different results. The suitability of each model depends on the particular application. S. Travis Waller moderated this session. 41ASSIGNMENT ADVANCES