Live demo: https://kopi-marihack26.netlify.app/
A React dashboard built for the 2026 Maritime Hackathon. It presents the results of a multi-objective fleet selection optimisation model for the route Port Hedland → Singapore (55M tonnes annual demand), with interactive charts covering cost, emissions, safety, and scenario analysis.
- KPI summary — total cost, total DWT, average safety score, CO2 emissions, fleet size
- Pareto efficiency frontier — cost vs safety trade-off across 21 optimisation runs
- Fuel composition breakdown — vessel count and emissions by fuel type
- Shapley value analysis — game-theoretic ranking of each vessel's contribution to fleet value
- Carbon sensitivity — total cost impact at varying carbon prices ($40–$160/tonne)
- Sensitivity heatmap — joint effect of carbon price and safety threshold on fleet cost
- Scenario comparison table — baseline (safety ≥ 3.0) vs sensitivity (safety ≥ 4.0) metrics
- MCMC robustness score — proportion of vessels classified as essential across posterior samples
- Fleet table — full vessel list with type, DWT, fuel, safety score, cost, and CO2
| Layer | Library |
|---|---|
| Framework | React 19 + TypeScript |
| Build | Vite |
| UI | Material UI v7 |
| Charts | Recharts |
| Deploy | Netlify |
Prerequisites: Node 20+, Bun
# Install dependencies
bun install
# Start dev server
bun run dev
# Build for production
bun run buildThe dev server runs at http://localhost:5173 by default.
All data is mock-generated for demonstration. To regenerate:
python3 generate_mock_data.pyThis writes 8 JSON files to src/data/, covering fleet results, Pareto frontier, Shapley values, MCMC robustness, fuel summaries, carbon sensitivity, and the sensitivity heatmap.
src/
components/ # UI components (charts, tables, KPI cards, layout)
data/ # Generated JSON data files
App.tsx # Root layout and data wiring
generate_mock_data.py # Mock data generator
The project deploys to Netlify automatically on push. Configuration is in netlify.toml. The build command is bun run build and the publish directory is dist.