A Disaggregate Analysis of Urban Travel Behavior

Abstract

This research was directed toward an evaluation of tra- vel behavior at a disaggregate level of analysis.The ob- jectives of the research were to develop trip generation models at the household level of analysis, to evaluate the stability of the relationships over time, to examine the causal structure of the variables found to be most signifi- cant in explaining trip production, and to examine individu- al's attitudes and perceptions of factors important in travel choice.A quasi-laboratory experimental design was established to evaluate the stability of household trip generation models over a seven year period.A sample of 357 single family households was selected from the 1964 Indianapolis, Indiana transportation study.The identical families were reinter- viewed in 1971.Disaggregate trip generation models from the two data sets were equally successful for estimating the zonal trips reported by all single family units from which the survey sample was drawn.Also, disaggregate models based on data from all dwelling unit types were comparable to aggregate models for estimating total zonal travel.The dis- aggregate models provide the additional benefit of reduced sample size requirements for estimation of model parameters.Examination of the causal structure of the variables indicated that family size and auto ownership are the pri- mary explanatory variables in household trip generation equa- tions. Income, labor force and relative accessibilityXll provide an indirect influence on trip production due to their effect on auto ownership levels.Attitudinal responses indicated only minor differences among subgroups of the sample in regard to factors important in mode and route choice and in selection of a trip destina- tion.Travel time and travel costs were not generally perceived as the important factors in the respondent's decision. CHAPTER 1.Committee on Urban Transportation (NCUT) .This committee drew upon the experience of many of the leading authorities in traffic engineering, highway planning, transit operations and city planning.Their experiences and procedural recommendations were combined and published in a transportation planning guide and a series of seventeen procedural manuals (NCUT, 1958).These manuals, along with continuous advisory aid of the Bureau of Public Roads, helped cities to develop transportation plans for various study areas.The need for uniform, accepted procedures for developing transportation planning studies was emphasized by the 1962 Federal Highway Act.This act required every urban area of over 50,000 population to have a continuing, comprehensive planning process on which to base their requests for Federal Aid monies.Because of the shortage of persons trained in transportation planning, personnel with limited backgrounds were utilized to perform the technical analyses in many of the smaller areas.To meet the basic requirements of the 1962 Act, the cities relied heavily on the model formulations and forecasting experience developed in other cities; andthe guidelines and directives provided by the Bureau of Public Roads.It is fair to say that the development of the urban transportation planning procedures has been formed through a lack of understanding of the complexities of urban travel, rather than in light of such knowledge.There has been, and still is, a lack of understanding of the entire decision process which controls urban travel behavior.The inter- relationships among economic, social, and psychological factors which influence travel choice have not been integrated into a single theory of travel.Consequently, rather than attempting to model the complex behavior as a continuous decision process, the transportation planner has developed a four stage, sequential modeling process known as the Urban Transportation Planning (UTP) package.These models

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 3 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.