|
|
|
Title: Modeling Commercial Vehicle Travel
Accession Number: 01024670
Record Type: Component
Availability: Transportation Research Board Business Office 500 Fifth Street, NW Abstract: A new Commercial vehicle trip forecasting model has been developed for the Baltimore Metropolitan Council (BMC). BMC has included trucks in its regional modelling for many years. However, it has not specifically included a fairly important category of nonpersonal travel that uses passenger cars, light trucks, and other vehicles not included in the “Truck” model. These trips include light-duty delivery vehicles, taxis, government vehicles, service personnel, craftsmen, tradesmen, and similar vehicles. BMC recently revised its Medium and Heavy Truck trip models. In conjunction with this effort, a completely new Commercial vehicle trip model was also developed. The principal challenge in estimating Commercial vehicle trips is to define them and obtain any kind of data on observed trip patterns. Traditional surveys were judged unlikely to be useful, so the project researchers devised an innovative way to estimate Commercial vehicle counts at those locations where Maryland DOT classification count data already existed. BMC staff conducted new manual counts at a representative sample of locations throughout the region. Commercial vehicles were defined as those bearing text, a logo, or carrying equipment of an obvious commercial nature. A database was created with various link characteristics. The consultant used this database to calibrate a “count model”, which was then applied to estimate daily Commercial counts at the 550+ classification count locations for the year 2000. These counts indicated that Commercial traffic (excluding Medium and Heavy Trucks) is almost 8% of the total daily volume. BMC retained a consultant who has refined the practical application of a methodology to synthesize a trip table from count data. Working “backwards” from the count data, the consultant not only created such a trip table, but then used it to develop a Commercial vehicle trip forecasting model that would produce link-level volumes with reasonable accuracy. A Commercial vehicle trip model from another region was adapted as a starting point. The consultant then applied a procedure called “adaptable assignment” to systematically adjust this interim model so as to better match the counts. This process resulted in a number of changes to the interim model and also produced a calibration adjustment table. This adjustment table is then multiplied by the nominal output of the model, producing a new table whose assignment comes much closer to matching the count data. The resulting assignment error is -1.9% and the regional percent root-mean-square error is 13%, both of which are excellent results. The result is a process that both exhibits reasonable sensitivities to the key input variables (employment by type and households) and has been shown to match the synthesized counts to a fairly high degree of accuracy. A 2025 forecast was made with this new process and the results found to be reasonable. The entire procedure has been implemented within the framework of BMC’s regional model set, using the TP+/Viper software package.
Monograph Title: Monograph Accession #: 01024684
Language: English
Corporate Authors: Transportation Research Board ADB50: Committee on Transportation Planning Applications Authors: Allen Jr, William GAgnello, PaulEditors: Bennett, GeorgiaDonnelly, RickPagination: 12p
Publication Date: 2004-3
Conference:
Ninth TRB Conference on the Application of Transportation Planning Methods
Location:
Baton Rouge Louisiana, United States Media Type: CD-ROM
Features: Figures; Tables
TRT Terms: Identifier Terms: Uncontrolled Terms: Geographic Terms: Subject Areas: Geotechnology; Highways; Planning and Forecasting; I10: Economics and Administration; I72: Traffic and Transport Planning
Files: TRIS, TRB
Created Date: May 9 2006 11:19AM
More Records from this Conference:
|
