# Slide 1: Title Slide
**Title:** BMS AI Assistant
**Subtitle:** Enterprise Supply Chain Optimization System
**Footer:** Technical Overview | Version 1.0

---

# Slide 2: Executive Summary
**Key Points:**
*   **Objective:** Streamline access to supply chain data via natural language.
*   **Solution:** Local AI Chatbot with integrated forecasting capabilities.
*   **Value Proposition:** Zero external API costs, 100% data privacy, instant response time.

---

# Slide 3: System Architecture
**Layers:**
1.  **Presentation:** Web-based UI (HTML5/JS).
2.  **API:** FastAPI Backend (Python).
3.  **Logic:** Intent Parser & Business Rules.
4.  **Data:** In-memory high-speed data store.

---

# Slide 4: Core Capabilities
**Features:**
*   **Demand Forecasting:** 30-day predictive analytics using ARIMA.
*   **Inventory Tracking:** Real-time multi-warehouse stock visibility.
*   **Supplier Intelligence:** Vendor lead times and contact details.
*   **Automated Reporting:** On-demand PDF generation.

---

# Slide 5: Technical Stack
**Technologies:**
*   **Language:** Python 3.10
*   **Framework:** FastAPI + Uvicorn
*   **AI Model:** TinyLlama 1.1B (Local Inference)
*   **Deployment:** Docker Container

---

# Slide 6: Process Flow
**Steps:**
1.  **User Input:** Natural language query.
2.  **Intent Recognition:** Pattern matching algorithm.
3.  **Data Retrieval:** SQL/CSV lookup or Model Inference.
4.  **Response Generation:** JSON formatting and UI rendering.

---

# Slide 7: Future Roadmap
**Next Steps:**
*   Integration with live ERP database (SQL).
*   Advanced multi-variate forecasting models.
*   User authentication and role-based access control.