A BILEVEL PROGRAMMING APPROACH TO ESTIMATING O-D MATRIX BY TRAFFIC COUNTS

Abstract

In this paper the authors present a bilevel programming formulation for the trip demand estimation (TDE) problem in a congested network where it is assumed that the flows satisfy the user optimal principle. It is possible both to provide an a priori estimate of the O-D matrix or to use only the information available from traffic counts. The most important contribution of this paper is an exact algorithm to solve the bilevel TDE problem. This algorithm uses the penalty function approach and the trust region strategy; it converges to a local minimum of the equivalent nonlinear programming problem. By using the relaxation and decomposition principles, as well as the underlying network structure, the proposed algorithm may be used to solve medium size TDE problems. Moreover, the algorithm may be applied to other convex bi-level programming problems in which the lower level problem has a network structure. (A)

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 11 citations exist upstream.