Intrazonal and external trips are generally omitted from the modeling process because they are difficult to measure. The exclusion of these trips can affect the accuracy of a calibrated distribution model. This research is aimed at overcoming the limitation of estimating the travel cost of intrazonal and external trips. The increase or decrease in accuracy resulting from the omission of various trip types is also predicted. Further, the effects of different shares of intrazonal trips pertaining to different zoning systems on the accuracy of other trip types is examined. These objectives were met by analyzing a database pertaining to Texas commodity flows. A set of doubly constrained gravity models was calibrated for two zoning systems, 254-county and 24-district. The findings indicate that the accuracy of estimates of different trip types increases if the intrazonal trips are omitted. A two-step method to predict complete, interzonal, intrazonal, and external trip matrices is recommended as follows: 1) calibrate the complete matrix to estimate the intrazonal and external trips while ignoring the predictions of interzonal trips and 2) exclude intrazonal and external trips and calibrate the remaining origin-destination matrix to predict the interzonal trips, resulting in a new distribution model or a new cost parameter.


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00716495
  • Record Type: Publication
  • Files: TRIS, ATRI
  • Created Date: Feb 22 1996 12:00AM