Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 128


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 127
116 I N N O VAT I O N S I N T R AV E L D E M A N D M O D E L I N G , V O L U M E 2 sis. For the DTA approximator, this period was broken and the strictly capacity-constrained CTM. The shift into eighteen 10-min intervals, with three additional 10- from freeways to arterials and collectors is felt to be min intervals provided for network clearance. A total of more consistent with CTM's more detailed traffic flow 2.56 million vehicle trips were assigned. Because Trans- model. However, actual traffic counts and speed checks CAD's approximator does not recognize tolls, delay-based would be needed to determine which model's predictions tolls were added to the free-flow travel time for each link, are more accurate. using an assumed travel time value of $10 per vehicle hour. Tables 1 and 2 summarize the results for comparing the approximator and VISTA with STA. CONCLUSIONS The most notable differences between the approxima- tor and STA are in links that are predicted to be con- While congestion-reduction policies and DTA each have gested under the static analysis. For these, DTA predicts attracted considerable interest in recent years, efforts at an even higher level of congestion, often significantly so. using the latter to evaluate the former on large-scale net- While static analysis predicts a TSTT of 1.27 million works are relatively few. Several issues arise when trying vehicle hours, the DTA approximator predicts 2.53 mil- to do this. A comparison of static and dynamic traffic lion vehicle hours. Unfortunately, the approximator's assignment is nontrivial due to fundamental differences use of Bureau of Public Roadstype functions, which between the models; however, the increase in capacity allow arbitrarily high volumes, precludes taking advan- induced by clearance intervals in the DTA approximator tage of the queue spillback features available in other can be accounted for by an appropriate increase in the DTA implementations (such as VISTA) that can provide capacities used in static assignment. With models such as added realism. Additionally, it should be noted that the CTM, which are vastly different from static assign- much of this increased congestion can be found on free- ment, it is much more difficult to compare the results on ways, which carry far more traffic than other functional a link-by-link basis, and in this work only global mea- classes. sures of system performance were compared. As with the DTA approximator, TSTT is much higher The CTM may produce results that are considerably under VISTA than under static assignment: VISTA pre- different than traditional assignment, because it models dicts a TSTT of 3.09 million vehicle hours for the same traffic flow at a more detailed level. This is particularly 3-h application. Another, perhaps more significant, apparent in congested networks because many assump- result is that static assignment tends to designate consid- tions in static traffic assignment about steady-state con- erably more vehicles to freeways, whereas VISTA's ditions and link performance functions are less realistic. assignment relies more on arterials and collectors. This is In this investigation, the effects of these assumptions in contrast to the DTA approximator, in which the dis- were amplified with the DTA approximator, which also tribution of traffic among the roadway classes was more uses link performance functions. comparable. This arises from fundamental distinctions However, the additional computation time required between the link performance function-based approach to find a DTA solution, particularly with the CTM, can- TABLE 1 Comparison Between Static Assignment and DTA Approximator Vehicle-miles Traveled (millions) Average v/c (VMT-weighted) Category N Static Dynamic Static Dynamic Freeway 6,292 25.88 29.86 1.02 1.64 Principal Arterial 4,936 6.46 5.39 0.72 0.86 Minor Arterial 10,434 7.65 6.25 0.43 0.52 Collector 14,596 2.97 2.74 0.77 1.39 Frontage Road 2,783 1.56 1.44 0.60 0.96 Congested 2,995 18.35 22.61 1.26 2.08 Uncongested 25,997 28.61 25.11 0.62 0.79 TABLE 2 Comparison Between Static Assignment and VISTA Total travel time (h x 103) Proportion (%) Functional class Static Dynamic Static Dynamic Freeway 505 325 40.6 10.5 Principal Arterial 174 543 14.0 17.6 Minor Arterial 237 715 19.1 23.2 Collector 227 738 18.2 23.9 Frontage Road 45 390 3.7 12.6 Total System 1,266 3,086 100 100