As we plan for new land use policies and investments in the transportation system over the next decade, we will face a new set of challenges tied to the changing demographic and economic conditions, in addition to the rising costs of energy and related policies aimed at reducing the carbon footprint of our economy. The first step in discovering the possible implications of these changes is a deeper understanding of the current relationships between land use and travel behavior, and how these might be impacted by future land use, transportation and energy policies.

This paper examines the links between land use and transportation patterns focusing on how land use patterns, built environment characteristics and socio-economics affect the resulting travel patterns in Ohio. This study forms the basis of a regional model that uses regional population and employment growth forecasts to allocate future land uses to appropriate sites, subject to environmental, zoning, utility, and other constraints that reflect alternative future policies. Based on these land allocation patterns, we estimate models to examine the resulting travel patterns. The estimated models are then used to test several scenarios to identify the impacts of Ohio’s changing economy and demographics, impacts of different land use policies, and changes in the transportation system characteristics.

The datasets used for this study come from several sources. Using the household travel surveys conducted by the Ohio Department of Transportation (ODOT) and Mid-Ohio Regional Planning Commission (MORPC), together with data on population, employment and land use patterns from ODOT, Census and CTPP (Census Transportation Planning Package), we develop models to examine the development patterns, determinants of trip making and trip lengths. The household travel surveys provide detailed information on the travel patterns of Ohioans, including data from around 28,800 households and over 260,000 trips. Although this study focuses on Ohio, the methodology and techniques can be adopted to assemble data and develop similar models elsewhere.