A Framework for the Calibration of Microscopic Traffic Flow Models

In view of the widespread use of microscopic traffic simulation models in common practice an ever more attention on their calibration related issues is bearing. These models, indeed, depend on many parameters, and the value chosen per each of them can sensibly affect model output realism. The only way to undertake is, therefore, their calibration on real traffic data. Hence the work proposes a framework for calibrating microscopic traffic flow models that aims, in particular, to address the issues of MoP choices and calibration parameters. The whole procedure focuses on the execution of one or more tests with laboratory data. This is the only way to overcome the effect of the errors that arise when using empirical data and that regards mainly o/d flow estimation, the data measuring process and mathematical modeling of real dynamics. A crucial element of the proposed procedure is model sensitivity analysis as well, performed by using an Analysis of Variance (ANOVA). Preliminary results, obtained using the microscopic traffic simulation model AIMSUN in freeway conditions, show the capability of the framework of successfully facing the calibration problem.

Language

  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; References; Tables;
  • Pagination: 14p
  • Monograph Title: TRB 86th Annual Meeting Compendium of Papers CD-ROM

Subject/Index Terms

Filing Info

  • Accession Number: 01049337
  • Record Type: Publication
  • Report/Paper Numbers: 07-0106
  • Files: TRIS, TRB
  • Created Date: Feb 8 2007 4:40PM