A 3D digital twin platform that allows city planners to test housing proposals inside an existing neighborhood and instantly see projected impacts on traffic congestion, transit load, infrastructure strain, and financial feasibility.

GrowthSync re-engineers traditional planning review by transforming static development reports into an interactive, scenario-based simulation tool. Instead of reviewing consultant PDFs after the fact, planners can model impacts immediately.

Place a proposed subdivision into the 3D environment, define unit count and housing mix, and watch real-time projections update across transportation, transit, infrastructure, and cost metrics.

• 3D Digital Twin - Real-time rendering of 4,600+ buildings, roads, and transit infrastructure in the Toronto area
• Voice-Driven Planning - Describe developments naturally: "Add 800 mixed units near transit with 20% townhouses"
• Transportation Impact - Peak hour trip generation, congestion scoring per road segment, color-coded road load visualization
• Transit Load Analysis - Population-based rider projections against route capacity with overload indicators
• Infrastructure Index - Composite scoring across water, sewer, school capacity, and parkland ratio
• Financial Projections - Developer build cost, municipal servicing cost, and development charge recovery estimates
• AI Constraint Explanation - Ask "Why did congestion spike?" and get causal analysis
• AI Optimization - Request "Keep units at 1,000 but reduce congestion" for density redistribution suggestions
• Randomized Building Generation - Each placed building gets unique dimensions, orientation, roof style, and facade
• Environmental Reports - Carbon footprint, noise propagation, habitat, and community impact analysis

• Node.js 18+
• npm or pnpm

git clone https://github.com/EVAnunit1307/CHD_city_pathing.git
cd CHD_city_pathing
npm install

Create a .env.local file:

GEMINI_API_KEY=your_gemini_api_key
ELEVENLABS_API_KEY=your_elevenlabs_api_key

npm run dev

Open http://localhost:3002 in your browser.

1. View baseline neighborhood in 3D
2. Add 1,200 detached homes — watch congestion spike, transit underused
3. Switch to transit-oriented mixed density
4. Congestion drops, transit usage increases, infrastructure index stabilizes
5. Generate environmental impact report
6. Ask AI for optimization suggestions

Voice Input --> Web Speech API --> /api/design --> Gemini 2.5 Flash
                                                       |
                                               3D Building Editor (Three.js)
                                               + AI Sound Effects (ElevenLabs)
                                                       |
                                               3D City Map (traffic, transit, zoning)
                                                       |
                               /api/environmental-report --> Impact Analysis (Gemini)
                               /api/tree-advisor --> Species Recommendations (Gemini)

├── app/                    # Next.js app router
│   └── api/                # API routes (design, environmental reports, map data)
├── components/             # React components
│   ├── editor/             # Building editor (viewport, controls)
│   └── ThreeMap.tsx        # Main 3D map component
├── lib/                    # Core logic
│   ├── editor/             # Building spec types, geometry builders, randomizer
│   ├── traffic/            # Vehicle behavior, collision, signal coordination
│   ├── buildingRenderer.ts # Building rendering with varied facades
│   ├── projection.ts       # Web Mercator coordinate projection
│   ├── roadNetwork.ts      # Road graph and A* pathfinding
│   └── simulationConfig.ts # Traffic simulation parameters
├── public/
│   ├── map-data/           # Pre-processed building, road, and signal data
│   └── sounds/             # AI-generated sound effects
└── scripts/                # Data processing utilities

• Evan Liem
• Herman Isayenka
• Ishaan Sharma

MIT