The German Network Control Method Motion -- New Methods and Evaluations

Adaptive network control systems are more and more recognized in Germany and its neighbor countries, as they make an increase in transport quality possible. A major player contributing to this progress is motion, which up to now has been implemented by Siemens in more than ten cities in Germany and internationally. Its development is based on a two-level architecture in order to preserve the existing concept of "intelligent" and autonomous controllers. The paper is split into two separate parts. The first part deals with the aspect of optimizing the coordination in signalized street networks. This paper will describe the experiences when applying Genetic Algorithms (a modern optimization method) in order to optimize traffic signal coordination and will discuss some restrictions which arise implicitly with this class of optimization algorithms when used for calculating the coordination. Due to the described problems when using Genetic Algorithms we decided to resign from using this class of algorithms and to re-use and update a deterministic method we had in our portfolio. This deterministic algorithm was under operation when an extended evaluation study had been performed by the Ruhr-University of Bochum, Germany, analyzing the impact of Motion on car traffic, pedestrians and public transport in Muenster, Germany. In the second part of the paper this algorithm will be summarized by sketching its basic features, followed by a presentation of the most important results of the evaluation study. With this second algorithm savings of 21% to 35% in waiting times and 12% to 35% in number of stops could be achieved.

Language

  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; Maps; References; Tables;
  • Pagination: 13p
  • Monograph Title: ITS in Daily Life

Subject/Index Terms

Filing Info

  • Accession Number: 01146224
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Dec 13 2009 12:05AM