API Documentation
Decision Intelligence Platform API
Base URL: http://localhost:8000
Source Label
All API responses include metadata indicating the data source:
{"metadata": {"source": "simulation/historical_replay"}}
This platform uses historical data and simulation. No real-time NYC taxi data is connected.
Endpoints
GET /health
Health check.
curl http://localhost:8000/health
Response:
{
"status": "ok",
"version": "3.0.0",
"models_loaded": ["hot_zone", "single_step", "two_step"]
}
GET /ready
Readiness check — verifies models can be loaded.
GET /version
Version and build info.
GET /v1/models
List available models.
POST /v1/recommendations
Generate zone recommendation.
Request:
{
"vehicle_id": "vehicle_001",
"latitude": 40.758,
"longitude": -73.985,
"zone_id": 161,
"model_name": "two_step",
"deterministic": true
}
Response:
{
"recommendation": {
"vehicle_id": "vehicle_001",
"timestamp": "2026-08-07T18:30:00Z",
"current_zone": 161,
"recommended_zone": 132,
"ranked_zones": [
{
"zone_id": 132,
"score": 0.91,
"expected_demand": 41.7,
"expected_supply": null,
"expected_revenue": null,
"travel_time_minutes": null
}
],
"confidence": 0.87,
"model_name": "two_step",
"model_version": "two-step-v1",
"explanations": ["Deterministic recommendation"]
},
"alternatives": [],
"metadata": {
"model_version": "two-step-v1",
"source": "simulation/historical_replay"
}
}
POST /v1/demand/forecast
Get demand forecast for a zone.
POST /v1/fleet/optimize
Optimize recommendations for an entire fleet.
Error Handling
Errors return:
{
"error": "description",
"detail": "optional detail message",
"timestamp": "2026-08-07T18:30:00Z"
}
Model Names
Name |
Description |
|---|---|
|
Ranks zones by historical pickup demand |
|
Greedy single-step utility maximization |
|
Truncated two-step horizon planning (default) |
Interactive Docs
Start the API and visit:
Swagger: http://localhost:8000/docs
ReDoc: http://localhost:8000/redoc