The State of Ohio is considering a range of options for capturing additional financial value from the Ohio Turnpike, including maintaining the status quo, issuing bonds against future revenue, or implementing a public-private partnership (P3). In evaluating these options, it is critical for the state to have a strong understanding of the value of the asset to ensure the best possible outcome to the state. In support of this assessment, traffic and revenue forecasts were prepared using a combination of the Ohio Statewide Model and a spreadsheet-based Rapid Response Model. The presentation will detail the development of those forecasts and how they differ from the development of many public-sector traffic forecasts.

The Ohio Statewide Model is a sophisticated integrated economic-land use-transport model. After calibration to historical traffic and revenue data, the model was able to provide forecasted traffic for autos and six classes of trucks through the year 2040 under various economic and toll scenarios. By using a model along with neighboring roadway traffic count data, the likely traffic diversion onto parallel routes was identified, highlighting impact to other state facilities.

Sensitivity tests using the model enabled the development of diversion curves that allow the user to pivot from the baseline forecasts to quickly assess the impact from a range of toll schemes.

To limit the need to run the full model for each test, a spreadsheet-based rapid response forecast model was build from the statewide model data and diversion curve assumptions. This model was responsible for converting the results into an annual revenue stream and accounting for certain details such as the EZ-Pass penetration rate and the effect of inflation on the real value of future tolls. A number of scenarios were evaluated with the rapid response spreadsheet, including those with various assumptions on the region’s economic growth, future toll rate schemes, and local exclusion (no toll for short trips).

The combined models were used to perform a backcast to year 1997 conditions, which allowed the model to be validated against data from past toll increases and decreases.

Fintan Geraghty, Brad Ship, Howard Wood, Alistair Sawers and Youssef Dehghani of Parsons Brinckerhoff also contributed to this work.