Skip to content

Repository files navigation

QVDFE: queue-consistent link emissions for traffic assignment

Pages Reproducibility Code license: MIT

This repository is the review and reproduction archive for:

A queue-consistent link emission function for traffic assignment: Bridging QVDF delay and fundamental-diagram states Mohammad Abbasi and Xuesong (Simon) Zhou

The project connects a queue volume-delay function (QVDF), a triangular fundamental diagram, and an age-weighted MOVES operating-mode benchmark. It contains the independent manuscript, the self-contained QVDFE-CBI package, all three corridor inputs, and the complete output used by the latest numerical experiment.

Current evidence

The experiment is a transparent in-sample diagnostic, not an external validation:

Stage/evidence set Count
Input sensor/TMC series 317
Candidate congestion episodes 4,270
Clean episodes after robust screening 2,186
Reconstructed episodes, all directions 1,827
Focus-direction episode comparisons 1,006
Physically admissible focus episodes 403
Strict measured-volume/rate-domain episodes 402

On the strict subset, exact two-regime emission MAPE ranges from 11.5% to 40.5%. The cubic approximation differs from exact episode totals by 4.5% at the median. These results support a reproducible regime test, but not a global monotonicity or convexity claim.

Repository map

QVDFE_paper/current/                 independent main_v8 manuscript and PDF
experiments/codes/qvdfe_cbi/         only experiment code package
experiments/datasets/                I-10, I-405, and I-17 inputs
experiments/emission_input/          vehicle-age and operating-mode rates
experiments/output/qvdfe_cbi/        complete staged outputs, logs, and report
docs/                                code and reproduction documentation site
scripts/                             input and publication verification

The retired experiments/output/qvdfe_paper/ workflow is intentionally ignored and is not part of the published repository.

Fast orientation

Clone and install

The I-17 readings are stored with Git LFS. Install Git LFS before cloning or pull the LFS object after an existing clone:

git lfs install
git clone https://github.com/Mmdabb/QVDFE.git
cd QVDFE
git lfs pull

python -m venv .venv
# Windows: .venv\Scripts\activate
# macOS/Linux: source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -r requirements-lock.txt
python -m pip install -e experiments/codes/qvdfe_cbi

Verify and rerun

python scripts/check_inputs.py
python scripts/verify_publication.py
python -m pytest experiments/codes/qvdfe_cbi/tests
python experiments/codes/qvdfe_cbi/run_pipeline.py

To regenerate only downstream figures and emissions from completed traffic outputs:

python experiments/codes/qvdfe_cbi/generate_stage_figures_and_emissions.py
python experiments/codes/qvdfe_cbi/run_two_regime_emissions.py

Interpretation boundaries

  • I-10 and I-405 contain measured flow, speed, and density.
  • I-17 is speed-only; its flow and density are synthesized after FD estimation, so it cannot independently validate discharge or emission totals.
  • The five-minute operating-mode benchmark does not resolve acceleration.
  • The reconstructed-speed emission benchmark measures propagation of traffic reconstruction error; it is not the proposed two-regime emission model.
  • Assignment use requires an explicit check of the complete generalized-cost derivative over the calibrated domain.

Licensing and citation

Software is provided under the repository MIT license. The manuscript and third-party traffic/emission inputs retain their own licensing boundaries; see LICENSES.md. Citation metadata are available in CITATION.cff.

About

Reproduction and review archive for the QVDFE manuscript

Topics

Resources

Contributing

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages