Heterogeneous Traffic Flow Modelling Using Cellular Automata

Developing countries have heterogeneous traffic pattern which gives considerable difficulty in modeling traffic flow phenomenon. Due to the complexity in develop mathematical models for heterogeneous traffic, simulation based approach is adopted by many researchers. Cellular automata (CA) based modeling has emerged as an efficient tool for homogeneous traffic flow modeling. In this study, an attempt has been made to develop a heterogeneous traffic flow model using CA. In this approach the road stretch is divided into cells of uniform size and vehicles are represented as cells, including clearance required for each vehicle type. Vehicles are updated at each time step based on simple rules, namely acceleration, deceleration, and randomization. The arrival pattern, arrival speed distribution, classified volume, and driver behaviour probability are the required inputs for the model. With these inputs, the model generates vehicles and follows the CA rules for updating the vehicle position at each time step. A small road stretch is taken for validating the model. The observed and simulated speeds for each type of vehicle are compared and found satisfactory. A simulation experiment is also done to get the speedflow- density relationship for a given vehicle composition.