RESIDENTIAL TRIP GENERATION: GROUND COUNTS VERSUS SURVEYS

Residential trip generation rates, i.e., the total number of vehicle trips per household during a 24-hour period, are a fundamental component of transportation planning. When agencies have different estimates of these rates for the same metropolitan area, the cost of the planning process increases since agencies must collect additional field data. To investigate discrepancies in these rates, residential trip generation rates based on four sources were compared: (1) ground counts collected by the Virginia Transportation Research Council (VTRC) at nine suburban neighborhoods, (2) household surveys distributed to the same neighborhoods, (3) national trip generation rates published by the Institute of Transportation Engineers (ITE), and (4) rates derived from the trip generation component of VDOT regional urban travel demand models. For neighborhoods composed solely of single-family detached homes, the average residential trip generation rate was 10.8 based on VTRC ground counts and 9.2 based on VTRC household surveys. Although underreporting of trips on written questionnaires may have contributed to this disparity, these rates were not significantly different at the 95% confidence level. Further, ground counts collected by VTRC and ground count rates reported by ITE were not significantly different. However, rates based on VTRC household surveys and those derived from VDOT regional models were significantly different when the VDOT rates were based on person trips rather than vehicle trips. This disparity resulted even though the person trips predicted by the VDOT long-range model were converted to vehicle trips using average automobile occupancies. The implication, therefore, is that when a data source gives the number of "vehicle trips per household" it is important to know if vehicle trips were measured directly or were calculated from person trips. When a consistent method of determining trip generation rates is used, the differences in rates between neighborhoods are explained by the large and random variations that are fundamental to trip generation studies. Accordingly, when a precise trip generation rate is required to forecast travel from a single neighborhood, the rate should be determined from field data if possible. When a trip generation rate is required for a group of neighborhoods (as is often the case with subarea studies), the average rate should be presented as part of a confidence interval as has been done in this study. For example, ground count data collected in this study for a set of seven neighborhoods of single-family detached homes produced a mean trip generation rate of 10.81, with a range of 9.4 to 12.2 vehicle trips per dwelling unit at the 95% confidence level. The ITE mean rate based on 348 neighborhoods was 9.57 vehicle trips per dwelling unit. As the number of neighborhoods increases, the confidence interval for the mean rate will decrease.

Language

  • English

Media Info

  • Features: Appendices; Figures; Photos; References; Tables;
  • Pagination: 53 p.

Subject/Index Terms

Filing Info

  • Accession Number: 00960144
  • Record Type: Publication
  • Report/Paper Numbers: FHWA/VTRC 03-R18,, Final Report
  • Contract Numbers: UPC 60710 6940
  • Files: NTL, TRIS, USDOT, STATEDOT
  • Created Date: Jul 22 2003 12:00AM