Skip to content

Chandrikajoshi123/Predictive-Maintenance-System

Repository files navigation

Rolls-Royce Digital Twin – Predictive Maintenance & Agentic AI

Overview

This is more than just a data science project — it’s a story of innovation, inspired by Rolls-Royce’s IntelligentEngine vision. A few months ago, I became fascinated with how Rolls-Royce builds intelligent engines that "talk back" — digital twins that predict and communicate when something will break, long before it actually does.

So I built one myself.

This platform simulates an intelligent aircraft engine environment:

  • Predicts Remaining Useful Life (RUL) with machine learning
  • Uses a LangChain-powered AI agent to make autonomous decisions
  • Generates GPT-based natural language maintenance reports
  • Offers real-time dashboards for engineers and managers

All deployed in a production-grade stack (FastAPI + Streamlit + Docker + Cloud).

Key Features

Feature Description
RUL Prediction Predicts Remaining Useful Life using NASA CMAPSS sensor data
FastAPI Backend RESTful API for serving RUL predictions
Streamlit Dashboard Real-time dashboard with sensor visualizations + decision outcomes
LangChain Agent Makes autonomous decisions based on RUL and sensor health
GPT-4 Maintenance Reports Generates human-readable insights from raw predictions
Dockerized & Cloud-ready Easy to deploy on Render, GCP, AWS, or Docker Desktop

Workflow Diagram

image

Contact & Credits

Built with ❤️ to demonstrate intelligent aircraft monitoring systems. Inspired by Rolls-Royce's "IntelligentEngine" vision.

  • Project by: Chandrika Joshi
  • GitHub:

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published