Route Set Generation for Quick Scan Applications of Dynamic Traffic Assignment

Abstract

Recent years have shown an interest in developing and using quick-scan
analysis tools for the evaluation of policy options, such as planning issues,
investments in infrastructure or public transport or other measures to cope with the increasing mobility problem. These tools can give a quick and easy insight into the impacts of all kinds of measures, without the large amount of work associated with the use of transport models. The assignment procedure of these tools can be improved and made more consistent with transport models. Therefore, it is needed that the calculation time of standard assignment algorithms is decreased. One possibility is to decrease the size of the route sets used. In this paper, this possibility was investigated for a number of small and medium-sized networks, using a dynamic traffic assignment framework. It was found that a route set in which each OD-pair has a maximum of 4-6 routes is sufficient to get comparable results with the situation with larger route sets. This rule of thumb for the maximum number of routes seems stable if demand increases and is not influenced by the overlap factor, which is an important parameter in the generation of route sets. Further research should focus on the scale factor in the route set generation algorithm and also larger networks need to be studied to be able to come a better founded conclusion about the size of the route set, which can be used in quick-scan tools.