The growing demand for better coordination of metropolitan land use and transportation planning has led to the need for new analytical tools to examine the potential impacts of land-use and transportation policies and investments. In this paper I report on the residential and market-clearing components of the recently developed UrbanSim land-use model, which is currently being implemented in Hawaii, Oregon, and Utah. The model is based on parcel-level land-use data and simulates the interaction between demand and supply of real estate by using a dynamic behavioral approach that operates on an annual time schedule. The model system is now operational, and in this paper the design and calibration results for the Eugene-Springfield, Oregon metropolitan area are described.