What kind of reports does the Predictive Maintenance tool generate?

It generates professional, interactive HTML reports featuring Plotly visualizations for FFT analysis, envelope analysis results, ISO 20816-3 compliance assessments, and ML-driven anomaly detection insights.

What types of analysis can the tool perform on industrial machinery?

It conducts advanced diagnostics including FFT spectrum analysis, envelope analysis for bearing faults, time-domain feature extraction, ML anomaly detection, and ISO 20816-3 vibration severity assessment.

What is the core function of the Predictive Maintenance tool?

This tool integrates Large Language Models (LLMs) with industrial predictive maintenance workflows, enabling AI-powered vibration analysis, bearing fault detection, and diagnostics through natural language conversations.

How does this tool leverage Large Language Models (LLMs)?

Utilizing the Model Context Protocol (MCP), it empowers LLMs like Claude to process raw vibration data, interpret machine manuals, and automate complex diagnostic workflows, providing expert-level insights.

Predictive Maintenance: LLM-Powered Vibration Analysis & Fault Diagnosis

Predictive Maintenance is an open-source framework that empowers Large Language Models (LLMs) like Claude with industrial machinery diagnostics capabilities through the Model Context Protocol. It transforms raw vibration sensor data and machine manuals into actionable insights, enabling AI-powered vibration analysis, bearing fault detection, and predictive maintenance workflows via natural conversation. This system helps democratize access to expert-level machinery diagnostics, augmenting human engineering work and making industrial AI more accessible for Industry 4.0 applications.

Key Features

010 GitHub stars

02Analyzes real bearing fault vibration data with 15 production-quality signals included.

03Offers advanced diagnostics including FFT spectrum analysis, envelope analysis for bearing faults, and time-domain feature extraction.

04Assesses vibration severity against the industry-standard ISO 20816-3 compliance.

05Performs ML anomaly detection using unsupervised/semi-supervised models (OneClassSVM/LOF) on healthy baselines.

06Generates professional, interactive HTML reports with Plotly visualizations for FFT, envelope analysis, and ISO compliance.

Use Cases

01Automate bearing fault detection and diagnosis from vibration signals with LLM assistance.

02Evaluate machine vibration levels against ISO 20816-3 standards to assess operational health.

03Integrate machine manuals to extract specifications, calculate bearing frequencies, and diagnose faults using conversational AI.