Due to computer resource and run time limitations the Ohio Statewide Travel Demand Model (OSTDM) was developed with spatial resolution sufficient to conduct corridor level analyses. However, it is employed in contexts that require model applications at varying spatial levels. For example, traffic forecasts for design projects need the model to be applied at a greater spatial resolution than is currently feasible in the full model. On the other hand, sketch planning uses and visioning exercises may not need the full resolution found in the model. In the context of an advanced activity based model, such varying spatial resolutions also have a bearing on both quality of results and model run times. In light of these issues, the current study proposes to examine the results of the OSTDM at three spatial levels: 1) The full Ohio statewide model with the original 5116 zone system, 2) a quick run version of the model with PUMA and county boundary based zone system with around 250 TAZs and 3) the focusing model tool based on a 14,000 TAZ system, only applied in the subarea of interest for project analyses.

For the base year, the full model run of Ohio’s advanced activity based model takes around 8 hours (sampling rate of 1 in 3 households) to run whereas the quick run version of the model runs in 1.5 hours (sampling rate of 1 in 100). The Quick run model is built as an automated application so that all the networks and input files are aggregated on the fly, thereby ensuring constancy with the full model. This study also uses ODOT’s unique focusing model tool that takes in the trip tables from an OSTDM run and disaggregates them to a much smaller zone system in the user supplied focus area using population and employment data to allocate trips. OSTDM networks are extracted from a master network with all public roads which is used in focus model runs ensuring consistency between the focus model and the full model. By comparing results for the subarea from these three versions, the study notes the differences in results and summarizes the overall model performance in terms of spatial and transportation outputs between these three versions.