The objective of project AVV AIUTO - Dynamic Assignments are: (a) to test whether conclusions that have been drawn on the future development of congestion and the effectiveness of policy measures depend on the assumptions within the National Model System (NMS) regarding assignment techniques; and (b) to investigate inhowfar these conclusions change under varying assignment assumptions. This is achieved through a theoretical and practical comparison of assignments with two dynamic assignment techniques (3DAS and CONTRAM), compared with a standard application of the current assignment model within the NMS, QBLOK. In this report a comparison between the three assignment techniques involved is carried out, containing: (1) a description of the background and key features of the dynamic assignment techniques; (2) data requirements of these assignment techniques, related to networks, matrices, etc.; and (3) a comparison between current assignment technique and the two dynamic assignment techniques, paying attention to theoretical concepts, constraints for application and their applicability for forecasting, policy analysis and traffic management purposes. From the comparison may be concluded that the envisaged practical comparison should be feasible. On theoretical ground there is no clear advantage of a single method over the others. All three have been applied to a range of traffic management measures and are most applicable in the long term. Any immediate constraints may be resolved through minor software adjustments, but practical constraints with respect to run time and memory requirements can only be determined through application. For none of the three techniques convergence to the sought user equilibrium can be proved - this would appear to become a more commonly accepted feature now more and more realism is introduced into assignment techniques. Some doubts exist about the solution method applied for 3DAS. Key issues of importance for a comparative application of the three techniques are addressed. This includes the transformation of speed-flow-density relationships, the representation of demand, and the equivalence of key link characteristics. The report concludes with a proposal Evaluation Framework. (A)
Abstract