The Ohio Statewide Travel Demand Model is a state-of-the-art economic, land use and transport model. Its person transport component is a tour-based, micro-simulation model of the entire population of the State of Ohio. Person transport consists of two broad categories of travel: short distance travel (within 50 miles of home) and long distance travel. This paper describes the design, estimation and calibration of the short distance travel (SDT) models.

The SDT models forecast the person movements arising from household (or population) production and consumption established as part of the determination of the spatial distributions of economic activity and population. The SDT models use the list of persons and households, whose attributes are determined by the economic & population models, to develop a set of tours and trips for each person in the list. Data for the model design and estimation were obtained from statewide and MPO-based surveys; they consist of one-day trip diaries for the members of more than 26,000 households. For calibration, additional data were obtained from the 2000 Census and other sources.

The SDT models are a series of (mostly) discrete choice models, applied in the following order: auto ownership, work place location, activity day pattern, primary non-work activity location, work-based activity location, tour schedule, tour mode, intermediate stop location, intermediate stop duration, work-based activity duration, and trip mode. The paper discusses the overall structure of the model, including its sequence, segmentation and feedback components, as well as the structure of each individual model component. Estimation and calibration results for the worker activity pattern model, the work tour scheduling model and the work tour intermediate stop duration model will be discussed.

The SDT models expand upon previous statewide tour-based micro-simulation models in several ways: the activity pattern model was segmented and generalized in order to simplify the model estimation and application, but minimizing loss of observed pattern variability; the tour and intermediate stop scheduling models have an one-hour resolution, the work place location model has direct linkages to the economic/land use model, and all models include traveler and transport-related explanatory variables with highly significant coefficients, which make the models responsive to changes in the population, in the spatial distribution of activities and in travel conditions over time.