Abstract:
Calibrating and validating a traffic simulation model for use on a transportation network is a process that depends on field data that is often limited, but essential for determining inputs to the model and for assessing its reliability. A quantification and systemization of the calibration/validation process exposes statistical issues inherent in the use of such data. Our purpose is to elucidate these issues and describe a methodology to address them.
The formalization of the calibration/validation process leads naturally to the use of Bayesian methodology for assessing uncertainties in model predictions arising from a multiplicity of sources (randomness in the simulator, statistical variability in estimating and calibrating input parameters, inaccurate data and model discrepancy). We exhibit the methods and the approach on an urban street network, using the micro-simulator.
CORSIM, while calibrating the demand and turning movement parameters. We also indicate how the process can be extended to deal with other model parameters as well as with the possible misspecification of the model. While the methods are described in a specific context they can be used generally, inhibited at times by computational burdens that must be overcome, often by developing approximations to the simulator.
Keywords:
Bayesian Analysis; Posterior Distribution; Stochastic Model Approximation; Traffic Simulation; Model Validation; Calibration; CORSIM
