PSA Strategic Insights AI

An AI-powered chatbot interface for analyzing Port of Singapore Authority (PSA) operational metrics with real-time Power BI integration and intelligent data retrieval using RAG (Retrieval-Augmented Generation).

🚀 Features

Core Capabilities

• 🤖 AI-Powered Analysis: Uses Google Gemini AI for intelligent insights and recommendations
• 📊 Power BI Integration: Connects to Microsoft Power BI dashboards for live operational data
• 🧠 RAG (Retrieval-Augmented Generation): Vector store retrieves historical context to enhance AI responses
• 🎯 Strategic Alignment: Insights mapped to PSA's four strategic pillars

Strategic Pillars

1. Operational Excellence - Port operations optimization, reduced berth times
2. Digital Integration - Real-time data, AI-driven decisions, seamless coordination
3. Sustainability - Carbon reduction, fuel optimization, environmental stewardship
4. Customer-Centric Innovation - Predictable service, proactive communication

Key Metrics Tracked

• Port Time Savings (%)
• Arrival Accuracy (%)
• Bunker Savings ($M)
• Carbon Abatement (K tonnes)

📋 Prerequisites

• Node.js (v14 or higher)
• npm or yarn
• Google Gemini API Key (Get one here)

🛠️ Installation

1. Clone the Repository

git clone https://github.com/pranavpro9/PSA-Code-Sprint.git
cd PSA-Code-Sprint

2. Install Dependencies

npm install

3. Configure Environment Variables

Copy the example environment file:

cp .env.example .env

Edit .env and add your Gemini API key:

# Required: Google Gemini API Key
REACT_APP_GEMINI_API_KEY=your-actual-gemini-api-key

4. Get Your Google Gemini API Key

1. Visit Google AI Studio
2. Sign in with your Google account
3. Click "Create API Key"
4. Copy the key and paste it into your .env file

🚀 Running the Application

Development Mode

npm start

The app will open at http://localhost:3000

Production Build

npm run build

Run Tests

npm test

📖 Usage

Quick Start

1. Launch the application
2. Wait for the vector store to initialize (shows "connected" status)
3. Try one of the quick prompts or ask your own question
4. View AI-generated insights with strategic pillar alignment

Example Queries

• "Analyze current port time savings performance"
• "What are the key trends in arrival accuracy?"
• "Compare bunker savings across recent periods"
• "Provide insights on carbon abatement progress"

How It Works

The app loads Power BI dashboard data and makes it available for conversational analysis:

1. Data Loading: Dashboard metrics are loaded on startup
2. AI Analysis: Gemini AI analyzes the data and detects patterns
3. Proactive Insights: Automatically identifies anomalies and opportunities
4. Chat Interface: Ask questions in natural language to get insights
5. Recommended Actions: Every insight includes specific action items

Current Data Includes:

• 5 months of performance metrics (May-September 2025)
• Port Time Savings, Arrival Accuracy, Bunker Savings, Carbon Abatement
• Automatic anomaly detection and trend analysis
• Strategic recommendations aligned with PSA pillars

🏗️ Architecture

┌─────────────────┐
│   React UI      │
│  (Frontend)     │
└────────┬────────┘
         │
    ┌────┴────┐
    │         │
┌───▼──┐  ┌──▼────────┐
│Gemini│  │ Power BI  │
│  AI  │  │    API    │
└───┬──┘  └──┬────────┘
    │        │
    └────┬───┘
         │
    ┌────▼────────┐
    │ Vector Store│
    │    (RAG)    │
    └─────────────┘

Key Components

• App.js: Main application component with chat interface
• VectorStore: In-memory RAG implementation for context retrieval
• Power BI Integration: Live data fetching and caching
• Gemini AI: Natural language processing and insight generation

🔧 Configuration

Switching AI Providers

The app is configured for Google Gemini. To use a different provider, modify src/App.js:

• Update API endpoint in callGeminiAI() function
• Adjust request/response format
• Update environment variables

Power BI Setup (Optional)

1. Create an Azure AD application
2. Grant Power BI API permissions
3. Get workspace and dataset IDs from Power BI Service
4. Add credentials to .env file

🐛 Troubleshooting

Common Issues

"Missing API Key" Error

• Ensure .env file exists in project root
• Verify REACT_APP_GEMINI_API_KEY is set
• Restart the development server after changing .env

Data Not Loading

• Check browser console for errors
• Ensure Gemini API key is configured correctly
• Verify the app has fully loaded (all status indicators green)
• Try refreshing the page

Dependencies Installation Failed

# Clear cache and reinstall
rm -rf node_modules package-lock.json
npm install

Security Vulnerabilities

# Fix non-breaking issues
npm audit fix

# Fix all issues (may cause breaking changes)
npm audit fix --force

Tailwind CSS CDN Warning

• This is a development-only warning
• The app uses Tailwind via CDN for simplicity
• For production, consider installing Tailwind as a PostCSS plugin
• Current setup works fine for demos and development

🔒 Security Notes

⚠️ Important Security Considerations:

• Never commit .env file to version control
• API keys in frontend are visible to users
• For production, implement a backend proxy to secure API calls
• Use environment-specific configurations
• Rotate API keys regularly

📦 Dependencies

Core

• React 19.2.0 - UI framework
• Lucide React - Icon library
• TailwindCSS - Styling (via CDN)

Testing

• @testing-library/react - Component testing
• @testing-library/jest-dom - DOM assertions

🚧 Known Limitations

1. Frontend API Calls: API keys exposed in browser (needs backend proxy)
2. In-Memory Vector Store: Data lost on refresh (needs persistent storage)
3. No Authentication: Open access (needs user auth)
4. Limited Error Handling: Basic error messages (needs improvement)
5. No Rate Limiting: Uncontrolled API usage (needs throttling)

🛣️ Roadmap

• Backend API proxy for secure key management
• Persistent vector store (PostgreSQL + pgvector)
• User authentication and session management
• Advanced error handling and retry logic
• Real-time Power BI streaming
• Export insights to PDF/Excel
• Multi-language support
• Mobile responsive improvements

📄 License

This project is private and proprietary.

👥 Contributors

• Team Tembusu Meal Enhancements (Min Wenn, Pranav, Han Sheng)

📞 Support

For issues or questions, please contact the development team.

---

Built with ❤️ for PSA International