A methodology for determining optimum cell size for a grid based traffic flow model for
heterogeneous traffic is proposed in this paper. The cell size is an important factor to determine as it
affects the computational efficiency and model accuracy. The objective function minimizes three aspects
namely the difference of distance headway in case of cellular automata and grid based traffic flow model,
the total number of cells to represent different types of vehicles and multiple of cell width that gives
closer representation of the different road widths. The presented method is found better then the previous
attempt which tries to find the cell size by trial and error.
