Abstract

This research proposes a Least Squares method and a Maximum Likelihood method for estimating the “Markovian Traffic Equilibrium” model. This model performs a traffic assignment based on absorbing Markov Chains, adding stochasticity in route choices of individuals. The parameters to be estimated are those present in flow-time performance functions and utility functions on the arcs composing the routes. The estimation procedure is complex as the model considers traffic congestion, then the service variables depend on the equilibrium. Tests built on synthetic data are conducted; we analyze the capability of each approach to approximate the population parameters under different scenarios.