Abstract

The AI-Powered Ticket Management System leverages cutting-edge technologies, including LangChain, Ollama's Llama 3.2, SQLite, and SQL Agent, to streamline customer support interactions. This system enables automated customer authentication, ticket tracking, and administrative oversight while ensuring secure and efficient query handling. A user-friendly Gradio-based interface enhances accessibility, allowing customers and administrators to interact seamlessly. The integration of Large Language Models (LLMs) with SQL Agents ensures intelligent query resolution, improving response accuracy and efficiency. This project provides a scalable and intelligent approach to managing customer support tickets while maintaining security and ease of use.

Methodology

1. Database Design & Setup:

   • A relational database (SQLite) is created to store customer details and support tickets.
   • Two primary tables: Customers (CustomerID, Name, Email) and SupportTickets (TicketID, CustomerID,
     Issue, Status).
   • Sample data is populated for demonstration purposes.

2. LLM, SQL Agent, and Database Integration:

   • LangChain and Ollama’s Llama 3.2 are used to process user queries and generate intelligent responses.
   • SQL Agent dynamically interprets and executes SQL queries based on user input, ensuring accurate data
     retrieval.
   • A role-based authentication system differentiates between customer and admin functionalities.

3. Customer Query Processing:

   • Customers authenticate via email.
   • Queries such as "What is the status of my ticket?" or "List all my tickets" are processed using the SQL
     Agent.
   • The system retrieves relevant ticket details and returns structured responses.

4. Admin Query Processing:

   • Admins can execute queries like "How many tickets are open?" or "View all customer tickets."
   • SQL Agent ensures efficient execution of complex queries, providing comprehensive insights.

5. Gradio-Based Interactive Interface:

   • A Gradio web-based UI enables easy customer and admin interactions.
   • Users enter their email and queries, receiving instant AI-generated responses.
   • The interface is designed for simplicity and ease of navigation.

Results

• Enhanced Customer Support Efficiency:

  • AI-powered automation significantly reduces response times for customer inquiries.
  • Customers receive instant updates on ticket status without human intervention.

• Scalability & Accuracy:

  • The system efficiently handles a growing number of tickets and customer interactions.
  • LLM-powered responses and SQL Agent execution ensure high accuracy in ticket-related queries.

• Seamless Admin Insights:

  • Administrators can monitor open, closed, and total tickets in real-time.
  • Structured SQL queries executed through the SQL Agent provide detailed analytics for better decision-
    making.

• User-Friendly Interface:

  • The Gradio interface enables smooth interactions for both customers and admins.
  • Authentication mechanisms ensure secure access to customer data.

• Secure & Robust System:

  • Role-based authentication protects sensitive data.
  • SQLite ensures data integrity and quick retrieval times.

The AI-Powered Ticket Management System presents an innovative approach to customer support, blending AI intelligence with SQL Agent capabilities to create a seamless, automated, and scalable ticketing solution.