DisSModel 🌍

"Science should not need to be rewritten to go into production."
(A ciência não deve ser reescrita para ir para a produção.)

📖 Research Trajectory

DisSModel did not emerge from a blank slate. It is the current expression of a research agenda that began in 2001 with an undergraduate thesis on geographic data interoperability using XML and open standards — a time when the central question was already forming:

How can geospatial models be built so that others can understand, reuse, and trust them?

Three principles unite this trajectory:

1. 🔓 Openness as method — open source and open data as conditions for scientific validation.
2. 🧩 Interoperability as architecture — systems designed to communicate, avoiding silos.
3. ♻️ Reproducibility as requirement — publishing conditions for re-execution, not just results.

DisSModel is the synthesis: a Python-native, FAIR-aligned, cloud-ready simulation framework where the same scientific code runs unchanged from a Jupyter notebook to a distributed production cluster.

🎯 About

DisSModel is a modular Python framework for spatially explicit dynamic simulation models. Developed by the LambdaGeo group at the Federal University of Maranhão (UFMA), it provides the simulation layer that connects domain models (LUCC, coastal dynamics) to a reproducible execution environment.

🌟 Key Features

• Dual substrate — same model logic runs on vector (GeoDataFrame) and raster (RasterBackend/NumPy).
• Lightweight scheduler — pure-Python time-stepped engine; models auto-register at instantiation and receive clock ticks via setup / pre_execute / execute / post_execute lifecycle hooks.
• Executor pattern — strict separation between science (models) and infrastructure (I/O, CLI, reproducible execution).
• Experiment tracking — every run generates an immutable ExperimentRecord with SHA-256 checksums, TOML snapshot, and full provenance.
• Storage-agnostic I/O — dissmodel.io handles local paths and s3:// URIs transparently.
• Platform-ready contract — the ModelExecutor interface is designed so the same model code can later run on the DisSModel Platform, a separate project under development.

🏗 Architecture

┌──────────────────────────────────────────────────────────┐
│  Science Layer  (Model)                                  │
│  FloodModel, AllocationClueLike, MangroveModel, ...      │
│  → only knows math, geometry and time                    │
├──────────────────────────────────────────────────────────┤
│  Infrastructure Layer  (ModelExecutor)                   │
│  CoastalRasterExecutor, LUCCVectorExecutor, ...          │
│  → only knows URIs, local/S3, column_map, parameters     │
├──────────────────────────────────────────────────────────┤
│  Core modules                                            │
│  dissmodel.core      — Environment, Model, SpatialModel  │
│  dissmodel.geo       — RasterBackend, neighborhoods      │
│  dissmodel.executor  — ModelExecutor ABC, ExperimentRecord│
│  dissmodel.io        — load_dataset / save_dataset       │
│  dissmodel.visualization — Map, RasterMap, Chart         │
└──────────────────────────────────────────────────────────┘

🚀 Quick Start

1. Install

pip install dissmodel

# Or latest development version
pip install "git+https://github.com/DisSModel/dissmodel.git@main"

2. Write a Model

# forest_fire_model.py
from dissmodel.core import Environment
from dissmodel.geo import SpatialModel, vector_grid

# 10x10 grid where every cell starts as forest
gdf = vector_grid(dimension=(10, 10), resolution=1.0, attrs={"state": "forest"})

class ForestFireModel(SpatialModel):
    def setup(self, prob_spread=0.3):
        self.prob_spread = prob_spread

    def execute(self):
        # Called every step — only math here, no I/O
        burning = self.gdf["state"] == "burning"
        # ... apply spread logic ...

env = Environment(end_time=50)
ForestFireModel(gdf=gdf, prob_spread=0.4)
env.run()

3. Wrap an Executor (for CLI + Provenance)

# my_executor.py
from dissmodel.executor import ExperimentRecord, ModelExecutor
from dissmodel.executor.cli import run_cli
from dissmodel.io import load_dataset, save_dataset

class ForestFireExecutor(ModelExecutor):
    name = "forest_fire"

    def load(self, record: ExperimentRecord):
        gdf, checksum = load_dataset(record.source.uri)
        record.source.checksum = checksum
        return gdf

    def run(self, data, record: ExperimentRecord):
        from dissmodel.core import Environment
        env = Environment(end_time=record.parameters.get("end_time", 50))
        ForestFireModel(gdf=data, **record.parameters)
        env.run()
        return data

    def save(self, result, record: ExperimentRecord) -> ExperimentRecord:
        uri = record.output_path or "output.gpkg"
        checksum = save_dataset(result, uri)
        record.output_path = uri
        record.output_sha256 = checksum
        record.status = "completed"
        return record

if __name__ == "__main__":
    run_cli(ForestFireExecutor)

4. Run via CLI

# Execute a simulation
python my_executor.py run \
  --input data/forest.gpkg \
  --output data/result.gpkg \
  --param end_time=50 \
  --toml model.toml

# Validate data contract without running
python my_executor.py validate --input data/forest.gpkg

# Show resolved parameters
python my_executor.py show --toml model.toml

📦 ExperimentRecord: Reproducibility by Design

Every run produces an immutable provenance record:

{
  "experiment_id": "abc123",
  "model_commit": "a3f9c12",
  "code_version": "0.5.0",
  "resolved_spec": { "...TOML snapshot..." },
  "source": { "uri": "s3://...", "checksum": "e3b0c44..." },
  "artifacts": { "output": "sha256...", "profiling": "sha256..." },
  "metrics": { "time_run_sec": 2.15, "time_total_sec": 2.34 },
  "status": "completed"
}

The record.json written next to every output contains everything needed to re-run the experiment: the input URI and its SHA-256 checksum, the resolved parameters (TOML + CLI overrides), per-phase timings, and the output checksum.

📊 Performance Telemetry

Every run via the executor lifecycle generates a profiling_{id}.md alongside the output:

🧩 Ecosystem: Models & Examples

DisSModel is a core framework. To maintain a clean and specialized environment, all simulation models and implementation examples are hosted in separate repositories within the DisSModel ecosystem.

🔬 Specialized Model Libraries

🛠 Implementation Templates

Each repository demonstrates how to:

1. Define a Model: Using SpatialModel and Environment.
2. Wrap an Executor: Using ModelExecutor for I/O and provenance.
3. Deploy: Running via CLI or API.

🔭 Roadmap: DisSModel Platform

The DisSModel Platform is a distributed execution environment (FastAPI, Redis, Docker, MinIO/S3) currently under development as a separate project. It consumes the same ModelExecutor contract documented above — executors run through its job queue without any change to their scientific code, including remote experiment reproduction from a stored ExperimentRecord.

The platform is not part of this package. Everything described in this README works locally with pip install dissmodel alone.

📚 Documentation

• 📘 User Guide: https://dissmodel.github.io/dissmodel/
• 🧪 API Reference: https://dissmodel.github.io/dissmodel/api/core/
• 🎓 Examples: https://dissmodel.github.io/dissmodel/examples/

🤝 Contributing

Contributions are welcome! Please read our Contributing Guidelines and Code of Conduct before submitting a pull request.

• 🐛 Report bugs → GitHub Issues
• 💡 Request features → GitHub Issues
• 📝 Improve docs → Fork, edit, and submit a PR

🎓 Citation

@software{dissmodel2026,
  author = {Costa, Sérgio Souza and Santos Junior, Nerval de Jesus and Sousa, Felipe Martins and Alves, José Magno Pinheiro and Bezerra, Denilson da Silva},
  title = {DisSModel: A Python Framework for Spatially Explicit Dynamic Modeling},
  year = {2026},
  publisher = {LambdaGeo, Federal University of Maranhão (UFMA)},
  url = {https://github.com/DisSModel/dissmodel},
  version = {0.6.0}
}

⚖️ License

MIT © DisSModel — UFMA
See LICENSE for details.