Building trip generation models from national database for medium sized- communities.

摘要

Freight data is propreitery and it involves high cost to acquire relevant freight data. Due to this reason freight was often excluded in travel demand modeling or included implicitly. The changing preference in the mode of transportation of freight to trucks, puts more truck traffic on the roadways. The implicit inclusion of freight in travel demant models does not do justice to the freight trips accounted for truck. The knowledge of freight data has led to exclusive freight demand modeling.;The national freight database like TRANSEARCH is propreitery but is available at both aggregated and disaggregated level. On the other hand, Freight Analysis Framework version 3(FAF3) data provides a national view of freight volumes for major metropolitan areas throughout the nation. FAF3 dataavailable is specific to freight analysis zones(FAZ). The developing communities with reasonable freight activity that are not included as freight analysis zone have to invest excessive funds to model freight. These counties are termed as medium-sized.;Present study is an effort to build trip generatin models from the 74 metro regions from 123 FAZs. The goal was to provide the models that can be used in national context. This document presents a methodology to build the trip generation statistical regression models.;The employment data is the explanatory variable in the models. North American Industry Classification System(NAICS) employment data is aggregated to one-digit level and two-digit level and two sets of models were developed at one-digit aggregate level and two-digit aggregate level.The trip generation models developed can be incorporated in commodity-specific modeling approach.;The models developed were further tested against statistical parameters. The adequacy of the models is examined. A comparision between models at one-digit level and two-digit level is done. Models developed are more adequate compared to the previous research studies.