Calibration – Making Predictions Match Performance
Current state-of-the art pavement design incorporates mechanistic principles for modeling pavement responses, including stresses, strains, and deflections, that are induced by specific material properties, traffic loading, and climatic events. These responses are then related directly or indirectly to pavement distresses and smoothness through prediction models or transfer functions. Relating mechanistic responses to empirical observations requires the act of calibration and validation.
Viewers of this webinar will learn about:
The meaning of calibration, validation, and verification for pavement design modeling
Basic steps required for calibration and validation
Setting up a test sampling matrix for a distress type
Assessing and minimizing the bias and standard error in the prediction model
The Calibration Assistance Tool (CAT)