Mode choise : implications for planning.

To appraise land-use and transportation plans for the future, us census journey-to-work survey data were used to develop distribution and modal split models for work trips in the New York region. Gravity distribution models were developed for each of three income groups. Using multiple regression techniques, an equation was derived that related walk-to-work trips to measures of a zone's self-containment. An equation to allocate work trips to transit and automobile modes was developed. It was found that the percent of transit trips for an origin-destination pair depends on employment density at the work end of the trip, residential density at the home end of the trip, availability of adequate rail service between origin and destination, relative times of the automobile and transit modes, the cost of tolls, and the cost of parking. Analyses of the root-mean -square error, geographical and transportation biases, and sensitivity to changes in the variables showed the equation to be a reasonable forecasting tool. For each of three income levels, modal split equations were derived, including all of the same variables. Comparisons were made of the parameters of the three stratified equations and it was found that as the income level rose, the significance of transportation-related variables increased the significance of land-use variables decreased. Analyses of the stratified equations showed them to be reasonable forecasting tools. Examination was made of sensitivities for the assumption of a radically different distribution of income levels, and it was found that the sensitivities as well as predictive values would change appreciably. It was concluded that only if income level distribution were not changing radically, would unstratified equation be adequate for modal split forecasts. The model calibration process required no trip data collection beyond that provided by the census survey. Future refinements have been suggested, particularly the use of a second point in time. Also, a recognition of the dynamic nature of the social structure is necessary to insure proper urban planning. /author/.