Calibration of the aggregate transit assignment model of EMME/2 using genetic algorithms

The study objective is to calibrate the aggregate transit assignment model such that the assignment output volumes match ridership volumes obtained by the on board counts. Specifically, calibration involves the determination of the optimal values of the following parameters: Boarding time, Wait time factor, Wait time weight, Auxiliary time weight and Boarding time weight. Conventionally, we use default values of EMME/2 or use a trial and error method for calibration. But such conventional methods do not ensure the optimal set of values. A more efficient method to determine the optimal values of the combinatorial parameters is to use Genetic Algorithms. Each chromosome in GA represents a specific set of parameter values that has a specific goodness of fit value. Finally, the best set of values was obtained through GA optimizer. This automatic method will help to get rid of tedious conventional methods for calibration of transit assignment models.