A family of macroscopic node models. (April 2015)

• Record Type: Journal Article
• Authors: Smits, Erik-Sander; Bliemer, Michiel C.J.; Pel, Adam J.; van Arem, Bart
• Is Part Of: Transportation research. Volume 74(2015)
• Page Start: 20
• Page End: 39
• Publication Date: 2015-04
• Subjects: Node model -- Network loading -- Turn delay -- Driving behaviour -- Dynamic traffic assignment -- Generic Class of first-order Node Models; Transportation -- Research -- Periodicals; Transportation -- Mathematical models -- Periodicals
• DOI: 10.1016/j.trb.2015.01.002
• Languages: English
• ISSNs: 0191-2615

Abstract

Highlights: The family captures all models that comply to a set of basic requirements. Our formulation in terms of turn delays allows behavioural interpretation. We analyze four family members regarding solution methods and underlying behaviour. Pareto optimality and uniqueness of the family are analyzed. Properties are illustrated on a three-legged node example. Abstract: The family of macroscopic node models which comply to a set of basic requirements is presented and analysed. Such models are required in macro-, mesoscopic traffic flow models, including dynamic network loading models for dynamic traffic assignment. Based on the behaviour of drivers approaching and passing through intersections, the model family is presented. The headway and the turn delay of vehicles are key variables. Having demand and supply as input creates a natural connection to macroscopic link models. Properties like the invariance principle and the conservation of turning fractions are satisfied. The inherent non-uniqueness is analysed by providing the complete set of feasible solutions. The node models proposed by Tampère et al. (2011), Flötteröd and Rohde (2011) and Gibb (2011) are members of the family. Furthermore, two new models are added to the family. Solution methods for all family members are presented, as well as a qualitative and quantitative comparison. Finally, an outlook for the future development of empirically verified models is given.