A travel demand model for rural areas
Berger, Aaron Damien.
MetadataShow full item record
Gallatin County has experienced dramatic population growth over the past decade. The Sonoran Institute has been developing land use planning tools to help address the issues arising from this high growth. One of the most serious impacts of high population growth is increased vehicle traffic, typically measured through vehicle miles traveled. The Sonoran Institute's land use planning tools were used to develop two future growth scenarios, one representing growth following current trends and the other representing a higher population density and higher employment/residential mix of development. The project task was to predict the total daily vehicle miles traveled for the two future scenarios using a travel demand model. The project goals were to develop a travel demand model that was simple to run, accurate, sensitive to changes in urban form, and could output daily vehicle miles traveled for future land development scenarios. The project used a literature review to determine expected effects of changes in urban form on travel behavior, and the sensitivity of current travel modeling methods to changes in urban form. For simplicity, the travel modeling method known as the fourstep process was selected for the project with modifications to add sensitivity to urban form. Current literature almost entirely states that the basic four-step model is not sensitive to urban form. One source did indicate that the four-step process might have some sensitivity to changes in urban form, but this sensitivity was not quantified. A basic four-step process model was run for a base year and the two future scenarios. The results indicated that the high density/high mix scenario reduced daily vehicle miles traveled by about 13%. This shows that the basic four-step model is sensitive to urban form. This sensitivity was analyzed through a series of idealized land development scenarios, and the sensitivity of the process to changes in urban form was quantified. The results were then incorporated into a now modified four-step model that produced a 16% reduction in daily vehicle miles traveled between the future scenarios while also providing better comparative accuracy between the two scenarios. This method met the initial objectives of the project.