Statistically-based Validation of Computer Simulation Models in Traffic Operations and Management

The process of model validation is crucial for the use of computer simulation models in transportation policy, planning and operations. The obstacles that must be overcome and the issues that must be treated in performing a validation are laid out here. We describe a general process that emphasizes five essential ingredients for validation: context, data, uncertainty, feedback, and prediction. We use a test-bed to generate specific (and general) questions and to give concrete form to answers and the methods used in providing them. The test-bed is the traffic simulation model, CORSIM; we apply it to assess signal-timing plans on a street network of Chicago. The validation process applied in the test-bed demonstrates how well CORSIM can reproduce field conditions, identifies flaws in the model that need to be overcome, and how well CORSIM predicts performance under new (untried) signal conditions. One specific conclusion: CORSIM, though imperfect, is effective in evaluating signal plans on urban networks, at least under some restrictions.