Skip to main content
ICTD 2024 Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Abstract for presentation (Poster or Podium)

Street & Highway Operations

PTV-VISSIM Car Following Model Calibration via Headway Distributions

Sunday, June 16, 2024
1:15 PM - 2:30 PM ET
Room: Galleria
Abstract:

The calibration and validation of microscopic traffic simulation models are an essential, yet often overlooked, step in traffic modeling. In PTV-VISSIM, calibration could potentially involve over 100 parameters and is often highly iterative. Many current guidelines and research efforts either do not offer a methodology for model calibration, or use black-box algorithms and machine learning to find the optimal parameter set, running the risk of overfitting or determining counter-intuitive parameter settings. This study introduces a methodology for car following model (W74 and W99) calibration using headway distributions. A visual calibration is obtained by overlaying the cumulative distribution function (CDF) of field-collected headways with simulated CDF plots generated using parameter combinations that cover the entire parameter value surface. The use of graph-based calibration provides a highly operational and accessible procedure, as they visually highlight the effects of parameter changes on simulated outputs, providing a more transparent calibration methodology.

Learning Objectives:

Attendees can expect to learn the following from this session:
  • Upon completion, participants will be able to appreciate the need for better calibration procedures and guidelines for VISSIM modeling.
  • Upon completion, participants will be able to understand the relationship between model parameters and real-world, measurable quantities (such as headways or travel time).
  • Upon completion, participants will be able to explore how different models need drastically different calibrations, depending on the use case and the project objectives.