Dynamic stochastic macroscopic traffic model considering vehicle type information based lane flow distributive ratio

T. Pan, Agachai Sumalee

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review


This paper develops a dynamic stochastic model considering multiple lane flow distribution based on the Cell Transmission Model (SCTM) which was originally proposed in combined single lane style with uniform parameters for simplification. However, based on the previous research and numerical analysis, it can be observed that each lane on expressway possesses a specific lane feature to be distinguished with other lanes, including the dominant vehicle size, the relatively stable distributive ratio of lane flow over the total amount, the specific flow-density relationship and so on. Such features are brought about by several reasons, e.g., transportation geography, traffic regulations, drivers' decisions surrounding the studied site, etc. That is to say, although these features might remain stable in time domain, longitudinal variations is inevitable. Thus variations of lane flow distributions along the expressway not only demonstrate the dynamics of traffic state, but also indicate the results induced by series of lane changing behaviors happened between detectors. Therefore, in order to reproduce reality, the new model is improved by introducing lane based elements such as specific lane features, vehicle type based lane flow distributive ratio and lane-changing process. Given the lane based discrete real-time measurements provided by detectors nowadays scattering along the expressways and weaving areas, the model, which has been determining to estimate the traffic conditions for non-monitored region with higher simulating frequency than systematic setting, is developed by following steps as below: firstly, as the most dominant variables or traffic state propagation in SCTM, the sending and receiving functions are adjusted according to the advanced complicated cross-lane matching network instead of the simple one-to-one correspondence. In addition, the capacity drop effect is considered in updating linear equation of traffic mode via speed reduction due lane changing maneuver. Finally, a vehicle type based lane changing ratio algorithm is adopted. The development of this algorithm is proposed by fulfilling lane changing requirement motivated by ramps and real-time measurement related to vehicle type distributions on main section.
Original languageEnglish
Title of host publicationProceedings of the 18th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2013 - Travel Behaviour and Society
PublisherHong Kong Society for Transportation Studies Limited
Number of pages8
ISBN (Print)9789881581426
Publication statusPublished - 1 Jan 2013
Event18th International Conference of Hong Kong Society for Transportation Studies on Travel Behaviour and Society, HKSTS 2013 - Hong Kong, Hong Kong
Duration: 14 Dec 201316 Dec 2013


Conference18th International Conference of Hong Kong Society for Transportation Studies on Travel Behaviour and Society, HKSTS 2013
Country/TerritoryHong Kong
CityHong Kong

ASJC Scopus subject areas

  • Transportation

Cite this