Multi-commodity flow estimation in transportation networks with partial counts on selected links
Conference Paper
Overview
Additional Document Info
View All
Overview
abstract
An important activity for state transportation departments is to perform travel demand modeling for urban areas. Typically about half of traffic links have counts and the remainder of the links require the development of count estimates. The proposed research represents a network flow methodology for estimating traffic volumes for multiple vehicle classes on the basis of partial data on a given set of highway links. Given an origin-destination demand matrix and link cost functions that represent traffic congestion in highway links, a multi-commodity traffic assignment problem is solved to provide the input to a generalized least squares optimization problem which gives final traffic flow estimates.