-A technique is proposed for estimating the probability distribution of total network travel time, in the light of normal day-to-day variations in the travel demand matrix over a road traffic network. A solution method is proposed, based on a single run of a standard traffic assignment model, which operates in two stages. In stage one, moments of the total travel time distribution are computed by an analytic method, based on the multivariate moments of the link flow vector. In stage two, a flexible family of density functions is fitted to these moments. It is discussed how the resulting distribution may in practice be used to characterise unreliability. Illustrative numerical tests are reported on a simple network, where the method is seen to provide a means for identifying sensitive or vulnerable links, and for examining the impact on network reliability of changes to link capacities. Computational considerations for large networks, and directions for further research, are discussed.1 Corresponding author: dwatling@its.leeds.ac.uk .

ContributorsGIW wrote and revised the manuscript in response to co-author comments. He finalized all the figures and tables, performed the literature search, and assisted with data interpretation. HJK critically reviewed the manuscript and made important suggestions to improve it. He assisted with data interpretation. IBA performed the data analysis, constructed the figures and tables, and made important suggestions to improve the manuscript. H-CK assisted with the data analysis and also reviewed the manuscript. GRC critically reviewed the manuscript and made important suggestions to improve it. He assisted with data interpretation. All other authors were given the opportunity to review the manuscript and make suggestions which GIW received, either revising the paper or providing explanations. All who are not deceased were involved with approval of the manuscript.

Article

It is well known that coordinated, area-wide traffic signal control provides great potential for improvements in delays, safety, and environmental measures. However, an aspect of this problem that is commonly neglected in practice is the potentially confounding effect of drivers re-routing in response to changes in travel times on competing routes, brought about by the changes to the signal timings. This article considers the problem of optimizing signal green and cycle timings over an urban network, in such a way that the optimization anticipates the impact on traffic routing patterns. This is achieved by including a network equilibrium model as a constraint to the optimization. A Genetic Algorithm (GA) is devised for solving the resulting problem, using total travel time across the network as an illustrative fitness function, and with a widely used traffic simulation-assignment model providing the equilibrium flows. The procedure is applied to a case study of the city of Chester in the UK, and the performance of the algorithms is analyzed with respect to the parameters of the GA method. The results show a better performance of the signal timing as optimized by † The work was carried out while the author was at the University of Leeds.

No abstract