Cost-effectiveness of cordon studies for trip matrix estimation from traffic counts

Abstract

Cordon studies can substantially improve estimation of origin-destination (O/D) trip matrices from link traffic counts. This paper proposes a method for integrating cordon survey results in O/D matrix estimation. This method is applied to 5 inter-urban networks in the Valencia Region, focusing on the specific formulation of the O/D matrix estimation technique, the assignment model to be implemented, the selection of traffic count locations in the cordoned area and, especially, the selection of cordon survey stations and data collection. Cost-effectiveness of cordon survey selection is addressed. The results show that cost-effectiveness ratios tend to decrease with the number of surveyed cordons in the 5 networks analyzed. This is due to non-linear reduction of origin-destination estimation errors and economies of scale on conducting cordon surveys. The results of this study can be useful to assess decision-making when conducting cordon surveys in interurban networks. Once such decisions have been taken, the method presented in this paper can be applied to the whole O/D matrix estimation procedure.