Supply Chain Intelligence Dashboard

An AI-powered responsive dashboard for comprehensive supply chain management with multi-agent worflows where each node is an agent performing individual task.

🚀 Overview

This system provides intelligent supply chain insights through three specialized analysis modules coordinated by an graph-based orchestration framework. This dashboard provides and data-driven demand simulation.

🏗️Architecture

Core Components:

• Multi-Agent Workflow: LangGraph-coordinated analysis pipeline
• AI Analysis Engine: TinyLLaMA LLM insights across domains
• Interactive Dashboard: Python library Streamlit-based interface

Analysis Modules:

1. Procurement Analysis: Contract document processing and risk assessment from text summarization.
2. SKU Rationalization: Product portfolio optimization and classification
3. Scenario Planning: Demand change simulation and revenue impact

🔄 Workflow

1. User Input - Parameters provided via dashboard interface

2. LangGraph Orchestration - Intelligent routing and coordination

3. Multi-Agent Processing - Parallel analysis by specialized modules

4. Dashboard Visualization - Interactive results presentation

5. Shared State Management - Centralized data persistence across all steps

Running the System

Start Agent Pipeline (Backend) python agent.py

Launch Dashboard (Frontend) streamlit run dashboard.py

📊 Dashboard Features

Unified Intelligence Display: Correlated insights across all modules

Interactive Scenario Controls: Real-time parameter adjustment

Multi-format Output: Text summaries, metrics, and visual charts

🔧 Technology Stack

Language Model: TinyLLaMA (via Ollama)

Dashboard: Streamlit

Visualization: Plotly