Estimation of origin-destination matrices from traffic counts: theoretical and operational development

This thesis deals with the o-d estimation problem from indirect measures, addressing two main aspects of the problem: the identification of the set of indirect measures that provide the maximum information with a resulting reduction of the uncertainty on the estimate; once defined the set of measures, the choice of an estimator to identify univocally and as much reliable as possible the estimate. As regards the former aspect, an innovative and theoretically founded methodology is illustrated, explicitly accounting for the reliability of the o-d matrix estimate. The proposed approach is based on a specific measure, named Synthetic Dispersion Measure (SDM), related to the trace of the dispersion matrix of the posterior demand estimate conditioned to a given set of sensors locations. Under the mild assumption of multivariate normal distribution for the prior demand estimate, the proposed SDM does not depend on the specific values of the counted flows †" unknown in the planning stage †" but just on the locations of such sensors. The proposed approach is applied to real contexts, leading to results outperforming the other methods currently available in the literature. In addition, the proposed methodology allows setting a formal budget allocation problem between surveys and counts in the planning stage, in order to maximize the overall quality of the demand estimation process. As regard the latter aspect, a “quasi-dynamic” framework is proposed, under the assumption that o-d shares are constant across a reference period, whilst total flows leaving each origin vary for each sub-period within the reference period. The advantage of this approach over conventional within-day dynamic estimators is that of reducing drastically the number of unknowns given the same set of observed time-varying traffic counts. The quasi-dynamic assumption is checked by means of empirical and statistical tests and the performances of the quasi-dynamic estimator - whose formulation is also given †" are compared with other dynamic estimators.