Laboratory experiments to assess the reliability of traffic assignment map

The assignment models, representing the relationship between travel demand and flows on the network, play a key role in almost all applications of both planning and transport network management. Although there are countless specifications for the assignment models, their reliability and dispersion structure are rarely discussed in the literature. Then, this paper addresses the errors that the usual assignment models introduce, both in terms of network/route choice modelling and in terms of OD matrix estimation based on the traffic measurements. This is done through laboratory experiments in a controlled environment. Specifically, assignment models are compared with the results of direct estimates, obtained by randomly sampling route choices observations within the population, with variable sampling rates.