🏥 CareFlow — Intelligent Caregiver Workload Optimization 🚀

CareFlow is a state-of-the-art, AI-driven dispatch and workload management engine designed to revolutionize elderly care facilities. By leveraging real-time telemetry and advanced machine learning scoring, CareFlow ensures the right caregiver reaches the right resident at the right time, while actively preventing staff burnout.

---

✨ Key Features

• 🎯 Smart Dispatch Engine: Automatically scores caregivers across 6 critical signals (workload, proximity, skill, availability, ETA, and fairness) to find the perfect match for every incident.
• 🧠 Neural Fatigue Analysis: Real-time tracking of staff "neural strain" based on shift duration, task intensity, and cumulative assignments to prevent caregiver burnout.
• 🚨 Emergency Protocol: Intelligent episode grouping that consolidates multiple alarms into single manageable incidents, reducing alert fatigue.
• 🤖 Explainable AI: Every recommendation comes with a plain-English explanation generated by Groq AI (Llama 3), building trust and clarity for coordinators.
• 📊 Admin & Caretaker Dashboards: Specialized views for facility managers to oversee the floor and for caregivers to manage their active tasks seamlessly.
• 🌓 Premium UI/UX: A sleek, dark-mode-first interface with glassmorphism, smooth animations, and a neural-mesh backdrop for a modern clinical feel.

---

🛠️ Tech Stack

Frontend

• React 18 + Vite (Lightning fast development)
• Tailwind CSS (Premium utility-first styling)
• Framer Motion-style Animations (Custom CSS/JS micro-animations)
• Modular Architecture: Clean separation of components, pages, and constants.

Backend

• Python (Fast and reliable)
• FastAPI (High-performance API framework)
• Groq AI Integration (Real-time Llama 3 inference for dispatch explanations)
• Weighted Scoring Engine: Custom algorithm for intelligent responder selection.

---

📂 Project Structure

careflow/
├── backend/
│   ├── main.py             # FastAPI Server & Scoring Logic
│   ├── fatigue_model.py    # ML Fatigue Prediction Engine
│   └── requirements.txt    # Python Dependencies
├── frontend/
│   ├── src/
│   │   ├── components/     # Reusable UI (Avatars, Charts, Cards)
│   │   ├── constants/      # Design Tokens & API Config
│   │   ├── pages/          # Major App Views (Admin, Caretaker, Fatigue)
│   │   ├── App.jsx         # Global State & Navigation Brain
│   │   └── index.css       # Core Design System

---

🚀 Getting Started

1️⃣ Clone the Repository

git clone https://github.com/YOURNAME/careflow-hackathon.git
cd careflow-hackathon

2️⃣ Backend Setup (Python)

cd backend
python -m venv venv
venv\Scripts\activate  # Windows
source venv/bin/activate  # Mac/Linux
pip install -r requirements.txt
# Add your GROQ_API_KEY to .env
uvicorn main:app --reload

3️⃣ Frontend Setup (Node.js)

cd frontend
npm install
npm run dev

Accessible at: http://localhost:3000

---

🧠 The Dispatch Algorithm

The "Brain" of CareFlow evaluates every responder based on a weighted matrix:

• Workload (25%): Prioritizes staff with fewer active tasks.
• Status (20%): Immediate points for "Available" caregivers.
• Proximity (20%): Prioritizes staff on the same floor as the resident.
• Skill Match (15%): Ensures specialized training matches the incident type.
• ETA (12%): Calculated arrival time based on current task status.
• Fairness (8%): Prevents task-stacking on the same individuals.

---

🏆 Hackathon Novelty

1. Dynamic Weight Shifting: Scoring weights automatically adjust based on severity (Critical incidents prioritize Proximity/Status).
2. Burnout Risk Flags: Staff with 7+ assignments are automatically deprioritized by 45% of their total score.
3. Episode Context: Consolidates related alarms to prevent coordinator overwhelm.

---

🤝 Team

Developed with ❤️ by the CareFlow Team for the HackSus Hackathon.

---

Note: This project was built using 100% free and open-source tools.