TOD-Simulation: Transit-Oriented Development Land-Use Scenario Modeling

This repository contains spatial simulation scripts and supporting files developed for the master's thesis:

"Modeling Transit-Oriented Development: A Land-Use Approach for Urban Decarbonization"
Paulo Dimas, 2025 Thesis available at: https://purl.utwente.nl/essays/108255 Master’s Thesis, University of Twente

The simulation framework adapts and extends the SIMLANDER land-use change model using a set of original R scripts to evaluate decarbonization scenarios based on Transit-Oriented Development (TOD) principles. The main case study is the metropolitan area of Rome.

---

🗂️ Repository Structure

/
├── LICENSE              ← MIT license for all original code
├── README.md            ← Project overview (this file)
├── /scripts/            ← All R scripts used for modeling and scenario simulation
├── /analysis/           ← All Excel Sheets used for the analysis
├── /simlander_mod/      ← Modified script from SIMLANDER.

Online Files

Due to the large size of the examples, you can find it online through the link

├── /inputs/             ← Land-use maps, accessibility layers, zoning constraints
├── /outputs/            ← Simulation results, emissions tables, final rasters
---

📜 Description of Scripts

This repository includes eight R scripts (three of them are scenarios) organized to perform land-use simulation modeling, accessibility evaluation, and TOD scenario execution.

1. Trim_Raster.R

Trims all input rasters to a consistent spatial extent and resolution.
Ensures slope, road, transit, and land-use rasters are spatially aligned.

2. Stats_Raster.R

Generates a .csv file with per-pixel spatial statistics. As input it should be the latest land-use map. The features are:

• Transit network
• Road network
• Urban center
• Slope

3. TB_Vector.R (Traceback Algorithm)

Converts binary urban/non-urban maps into multi-class land-use rasters. Uses output from Stats_Raster.R.

4. Scenario Scripts

• Scenario_IB.R: Applies TOD densification logic (Intense Building)
• Scenario_RETRO.R: Implements green retrofitting strategy
• Scenario_BAL.R: Balanced strategy combining TOD and retrofitting

Each scenario applies rule-based transformations based on spatial suitability metrics and zoning constraints.

5. PixelCounter.R

Counts the number of pixels per land-use class.
Useful both before and after scenario simulation to assess land-use shifts.

---

🔁 Modeling Workflow

1. Prepare raster data (Trim_Raster.R)
   1. Land-use maps with 2 classes (1 = urban and 0 = non-urban)
   2. Road network (euclidean distance)
   3. Transit network (euclidean distance)
   4. Urban distance (euclidean distance)
   5. Slope terrain (degrees)
   6. Zoning/Protected areas with 2 classes (0 = protected and 1 = developable)
2. Compute spatial metrics (Stats_Raster.R)
3. Apply traceback to reclassify urban areas (TB_Vector.R)
4. Run scenario of choice (IB, RETRO, BAL)
5. Count results (PixelCounter.R)
6. Analyse through the different Excel Sheets and GIS software.

---

📍 Data Requirements

All input rasters must be in the same projection, resolution and aligned.

---

📄 License

This repository is licensed under the MIT License.

This work includes adapted logic inspired by the SIMLANDER land-use simulation model developed by Richard Hewitt, which is described by the author as Free and Open Source Software (FOSS).
The SIMLANDER source code itself is not included in this repository. Only original or adapted components are shared here under MIT terms.

---

📚 Citation

If you use this code in academic research, please cite:

Paulo Dimas. (2025). TOD Simulation Scripts and Data (v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.16635783

BibTeX

@mastersthesis{dimas2025tod,
  title     = {Modeling Transit-Oriented Development: A Land-Use Approach for Urban Decarbonization},
  author    = {Paulo Dimas},
  year      = {2025},
  school    = {University of Twente},
  url       = {https://github.com/Raverino/TOD_modelling}
}

And optionally cite the SIMLANDER model:

Hewitt, R. (2022). SIMLANDER: A land-use simulation model. International Journal of Geographical Information Science. https://doi.org/10.1080/13658816.2022.2098299

---

📬 Contact

📧 For questions, feedback, or collaboration: p.r.almeidadimas@student.utwente.nl