Specifying, Estimating and Validating a New Trip Generation Model: Case Study in Montgomery County, Maryland

Abstract

This paper discusses the development of an afternoon peak period trip generation model for both work and non-work trips. Three data sources are used in model development, a Household Travel Survey, a Census-Update Survey, and a Trip Generation Study. Seven one-direction trip purposes are defined, specifically accounting for stops made on the return trip from work to home. Trips are classified by origin and destination activities rather than by production and attraction, so reframing the conventional schema of home-based and non-home-based trips. Prior to estimating the model, the Household Travel Survey was demographically calibrated against the Census-Update to minimize demographic bias. A model of home-end trip generation is estimated using the Household Travel Survey as a cross-classification of the demographic factors of age and household size in addition to dwelling type. Non-home-end generation uses employment by type and population. The model was validated by comparison with a site based Trip Generation Study, which revealed an under-reporting of the relatively short and less regular shopping trips. Normalization procedures are developed to ensure that all ends of a chained trip were properly accounted for.