TwinEdge AI

Edge-Based Digital Twin for Predictive Maintenance and Smart Manufacturing

Project Overview

TwinEdge AI is an AI-powered predictive maintenance solution designed for smart manufacturing environments. The system uses machine sensor data and machine learning algorithms to predict equipment failures before they occur, reducing downtime and maintenance costs.

This project was developed as a Proof of Concept (POC) for Tata Technologies InnoVent 2027 under the category:

AI at the Edge Solutions for Industrial Heavy Machinery

---

Problem Statement

Industrial machines often experience unexpected failures that lead to:

• Production downtime
• Increased maintenance costs
• Reduced operational efficiency
• Safety risks

Traditional maintenance approaches are reactive and inefficient.

---

Proposed Solution

TwinEdge AI combines:

• Edge AI
• Predictive Analytics
• Digital Twin Concepts
• Machine Health Monitoring

The system analyzes machine operating parameters and predicts potential failures in real time.

---

Dataset

AI4I 2020 Predictive Maintenance Dataset

Features used:

• Air Temperature
• Process Temperature
• Rotational Speed
• Torque
• Tool Wear
• Failure Indicators (HDF, OSF, PWF, TWF, RNF)

---

Technology Stack

Programming Language

• Python

Libraries

• Pandas
• NumPy
• Matplotlib
• Seaborn
• Scikit-Learn
• Joblib

Machine Learning Model

• Random Forest Classifier

---

Project Workflow

1. Data Collection
2. Data Preprocessing
3. Exploratory Data Analysis (EDA)
4. Feature Engineering
5. Model Training
6. Model Evaluation
7. Failure Prediction
8. Dashboard Visualization

---

Model Performance

Metric	Value
Accuracy	99.9%
Model	Random Forest
Prediction Type	Binary Classification

---

Key Results

• Achieved 99.9% prediction accuracy
• Identified major machine failure factors
• Generated feature importance analysis
• Developed dashboard-ready outputs

---

Project Structure

TwinEdge_AI/

├── data/

├── notebook/

├── model/

├── app/

├── screenshots/

├── presentation/

├── README.md

---

Business Impact

• Reduced unplanned downtime
• Improved machine reliability
• Lower maintenance costs
• Increased productivity
• Supports Industry 4.0 transformation

---

Future Scope

• Real-time IoT sensor integration
• Edge device deployment
• Cloud monitoring
• Advanced Digital Twin simulation
• Multi-factory implementation

---

Team STARBUCKS

Project developed for Tata Technologies InnoVent-27.