Large-scale simulation of traffic flow using Markov model

Modeling and simulating movement of vehicles in established transportation infrastructures, especially large urban road networks is an important task. It helps with understanding handling traffic problems, optimizing regulations adapting the management real time for unexpected disaster events. A mathematically rigorous stochastic model that can be used analysis was proposed earlier by other researchers which based on interplay between graph Markov chain theories. This provides a transition probability matrix describes traffic's dynamic its unique stationary distribution network. In this paper, new parametrization presented introducing concept two-dimensional handle together vehicles' distribution. addition, weighted least squares estimation method applied estimating parameter using trajectory data. case study, we apply our Taxi Trajectory Prediction dataset network data from OpenStreetMap project, both available publicly. To test approach, have implemented software. We run simulations medium scales procedure, artificial datasets, been proved satisfactory. application, unfolded map Porto, dataset. The approach described here combines techniques whose use to analyze has not previously reported.