Project Details
Description
Most conventional methods for estimating trip matrices from traffic counts assumes either that the counts are error-free determin-istic variables or they use a simplified traffic assignment model. Without these quite rough assumptions, the methods often demand prohibitive calculation times. In the project a new matrix-estimation method, 'Multiple Path Matrix Estimation' (MPME) has been developed which do not have these properties. Regarding route choices it corresponds to the models developed in 2a.
For practitioners, MPME is most promising in cases where an old trip-matrix needs to be updated for use in sketch-plan models or as pivot-matrix in larger traffic models. MPME has been tested and used successfully in several full-size cases (from small cities with 25,000 inhabitants to metropolitan areas with 5 million inhabitants). In all cases, the method gave lower deviations between traffic counts and estimated traffic than other tested methods. It converged smoothly within acceptable calculation times. It is now being implemented in the US GIS-T, 'TransCAD', which is produced by the Boston-company, 'Caliper Corporation'.
For practitioners, MPME is most promising in cases where an old trip-matrix needs to be updated for use in sketch-plan models or as pivot-matrix in larger traffic models. MPME has been tested and used successfully in several full-size cases (from small cities with 25,000 inhabitants to metropolitan areas with 5 million inhabitants). In all cases, the method gave lower deviations between traffic counts and estimated traffic than other tested methods. It converged smoothly within acceptable calculation times. It is now being implemented in the US GIS-T, 'TransCAD', which is produced by the Boston-company, 'Caliper Corporation'.
| Status | Finished |
|---|---|
| Effective start/end date | 01/09/1994 → 07/12/1997 |
Collaborative partners
- Technical University of Denmark (lead)
- Tetraplan A/S (Project partner)
Funding
- Unknown
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.