🚀 Appian Predictive Operations

Welcome to the Appian Predictive Operations repository!
This project is a Full-Stack Intelligent Operations Dashboard built to optimize workforce allocation using Generative AI (Groq / Llama-3) and Predictive Analytics.

---

📌 Overview

Operations teams often face:

• SLA breaches
• Inefficient staff allocation
• Poor workload visibility
• Manual and delayed decision-making

Appian Predictive Operations AI addresses these challenges by combining:

• AI-driven workload forecasting
• Intelligent staff optimization
• Conversational AI for real-time insights

All within a single, interactive dashboard.

---

🌟 Key Features

🧠 HONEY(Holistic Operations Navigation & Efficiency Yielder) AI Chatbot

A context-aware AI assistant acting as an Operations Expert.

• Answers queries related to SLAs, staffing, and workload
• Powered by Groq Llama-3
• Embedded directly into the dashboard UI

---

📊 Predictive Forecasting

• AI Confidence Intervals (Upper & Lower bounds)
• Time Range Toggle (24-Hour / 7-Day forecasts)
• Staffing Heatmap to highlight high-stress periods

---

👥 Intelligent Resource Planning

• Persistent Database
  • Uses staff_db.json
  • Retains employee data across server restarts
• Real-Time AI Audit
  • Detects staffing imbalances automatically
• One-Click Optimization
  • Instantly apply AI-recommended staff reallocations

---

⚙️ Admin Control Center

• Simulation Engine
  • What-if analysis for volume spikes and staff reductions
• Configurable SLA Thresholds
• Audit Log Export
  • Download system logs as CSV

---

🛠️ Tech Stack

Frontend

• React
• Vite
• Tailwind CSS
• Recharts

Backend

• Python
• FastAPI
• Pandas
• Pydantic

AI & Analytics

• Groq API
• Llama-3
• Predictive modeling with confidence intervals

---

🛠️ Prerequisites

Ensure the following are installed on your system:

1. Node.js (v16 or higher)
   https://nodejs.org/

2. Python (v3.9 or higher)
   https://www.python.org/

3. Git
   https://git-scm.com/

---

🚀 Setup Guide (Step-by-Step)

1️⃣ Clone the Repository

git clone https://github.com/HemangDubey/appian-predictive-project.git
cd appian-predictive-project

2️⃣ Backend Setup (Python/FastAPI)

Open a terminal in the root folder:

1. Navigate to backend:

cd backend

2. Install Dependencies:

pip install fastapi uvicorn pandas openai pydantic

3. Configure API Key:

• Open backend/main.py.
• Locate the client setup (approx Line 26).
• Replace gsk_... with your valid Groq API Key.
• (Free keys available at console.groq.com).

4. Start Server:

uvicorn main:app --reload

*Terminal should show: Uvicorn running on http://127.0.0.1:8000*

3️⃣ Frontend Setup (React/Vite)

Open a NEW terminal (keep the backend running) and run:

1. Navigate to frontend:

cd frontend

2. Install Dependencies:

npm install

3. Start Application:

npm run dev

4. Launch:

• Click the Local Link displayed (usually http://localhost:5173).

---

📂 Project Structure

appian-predictive-project/
│
├── backend/                # Python FastAPI Server
│   ├── main.py             # Core Logic, API Endpoints & AI Integration
│   ├── staff_db.json       # Persistent Database (Auto-generated on run)
│   └── appian_historical_data.csv # Dataset for analysis
│
├── frontend/               # React Application
│   ├── src/
│   │   ├── App.jsx         # Main Dashboard & Logic
│   │   ├── ChatAssistant.jsx # Honey AI Chatbot Component
│   │   └── index.css       # Tailwind CSS Styling
│   └── package.json
│
└── README.md               # Project Documentation

---

⚠️ Troubleshooting

• **Error: pip is not recognized**

• Ensure Python is added to your system PATH variables.

• Error: Connection Refused on Frontend

• Verify that the Backend Terminal is running and shows "Uvicorn running".

• Error: 401 Unauthorized in Chatbot

• Check backend/main.py and ensure your API Key is correct and inside quotes " ".

---

🤝 Contributions

This project was designed, developed, and maintained by:

• Hemang Dubey
  🔗 LinkedIn: https://www.linkedin.com/in/hemang-dubey-7b801628b/

• Akshat Agrawal
  🔗 LinkedIn: https://www.linkedin.com/in/akshat-agrawal-3b0b13255/

• Megha Gautam
  🔗 LinkedIn: https://www.linkedin.com/in/megha-gautam-8533b2292/

• Jayesh Pani
  🔗 LinkedIn: https://www.linkedin.com/in/jayesh-pani-a9ab7628a/

Contributions, suggestions, and improvements are always welcome.
Feel free to fork the repository, raise issues, or submit pull requests.