CoolFace
Apppublic

AK-21/Graphite-Industrial-Intelligence

sourceHugging Faceupdated 3mo agoView on Hugging Face
0likes
App README

Graphite Industrial Intelligence

Graphite Industrial Intelligence is a state-of-the-art predictive maintenance and root-cause analysis platform designed for critical industrial infrastructure. It bridges the gap between raw hardware telemetry and unstructured technical documentation.

Instead of operators having to cross-reference sensor alerts with 500-page OEM manuals, the platform ingests everything into a unified ecosystem. It uses AI to semantically search documents and visually map relationships in real-time.

Key Features

  1. 1.Asset Registry: Manage the digital twins of your industrial equipment. View live health statuses and track maintenance requirements.
  2. 2.Document Ingestion: Upload technical manuals and standard operating procedures (SOPs). Files are instantly parsed, chunked, and vectorized for semantic search by the AI engine.
  3. 3.Interactive Knowledge Graph: Visually explore complex relationships between physical assets, detected anomalies, and technical documentation. Run AI Root Cause Analysis directly from the visual interface.
  4. 4.AI Knowledge Q&A: Chat directly with the Retrieval-Augmented Generation (RAG) pipeline to ask complex maintenance questions and get instant answers sourced and cited from your uploaded manuals.

Technology Stack

  • —Frontend: React, Vite, Tailwind CSS, Framer Motion, React Flow (@xyflow/react)
  • —Backend: Node.js, Express, tRPC (Type-safe APIs)
  • —Database: SQLite (better-sqlite3) for local fast storage
  • —AI & Embeddings: Google Generative AI (Gemini)

Getting Started

Prerequisites

  • —Node.js (v18 or higher recommended)
  • —An active Google Gemini API Key. You must set this in your environment for the AI features (Q&A, Root Cause Analysis) to work.

Installation

  1. 1.Clone the repository:
bash
   git clone https://github.com/AK-2110/Graphite-Industrial-Intelligence.git
   cd Graphite-Industrial-Intelligence
  1. 1.Set up Environment Variables:
  2. 2.Navigate to the server directory and create a .env file:
bash
     cd server
     # Create .env and add your Gemini API Key
     # GEMINI_API_KEY=your_api_key_here
  1. 1.Install Dependencies:
  2. 2.Open two terminal windows.
  3. 3.Terminal 1 (Backend):
bash
     cd server
     npm install
  • —Terminal 2 (Frontend):
bash
     cd client
     npm install

Running the Application

  1. 1.Start the Backend Server: In Terminal 1 (inside the server directory), run:
bash
   npm run dev

(Alternatively, if `npm run dev` is not configured, run `npx tsx src/index.ts`)

  1. 1.Start the Frontend Client: In Terminal 2 (inside the client directory), run:
bash
   npm run dev
  1. 1.Open the Application: Open your browser and navigate to http://localhost:5173 to access the Graphite Industrial Intelligence dashboard.

Usage Workflow

  1. 1.Register Assets: Start by adding your physical equipment in the Asset Registry.
  2. 2.Ingest Docs: Upload technical manuals via the Document Ingestion module.
  3. 3.Detect Anomalies: When an asset goes offline, open the Knowledge Graph to trace the failure.
  4. 4.Query AI: Click on failing nodes or use the Q&A bot to extract exact repair procedures from your documentation.