relysam - Design Reliability In. Predict Failures Before They Happen.

relysam is a Core Reliability Engineering platform with AI/ML enhancements designed for engineers, organizations, and learners who understand that reliability is designed in, not maintained in. This comprehensive open-source platform combines traditional reliability engineering methodologies with advanced artificial intelligence to provide powerful tools for predictive maintenance, failure analysis, and risk assessment.

Reliability-First Philosophy: Systems built with early reliability analysis have better maintainability and availability. relysam provides 30+ reliability engineering tools vs 1 maintenance template - because reliability is the cause, maintenance is the outcome. Invest in the cause, not the symptom.

Key Features: Design for Reliability (DfR) Assessment, FMEA & Root Cause Analysis with AI-powered recommendations, Statistical Analysis (Weibull, Monte Carlo, life data analysis, MTBF/MTTR), Fault Tree Analysis, Event Tree Analysis, Component Derating (MIL-HDBK-217, NSWC-11), Risk Assessment, 9 HRA methods (THERP, HEART, SPAR-H, CREAM, SLIM-MAUD, ATHEANA, JHEDI, SHERPA, MERMOS), 10 AI/ML models for predictive analytics with Custom Data Training, Knowledge Base with unified search, Database integrity monitoring, AI model training status, Safety & ethics standards, Real-time WebSocket monitoring, Extensive JSON reporting, Maximum offline operation, Automated test suite (105 pytest tests with pre-commit integration).

NEW in v1.1.0: Custom Data Training - Train AI models with your own domain-specific data! Features web interface for dataset upload, CLI tools for advanced users, automated scheduler, 11 REST API endpoints, and sample automotive industry datasets.

Who Can Use relysam: (1) Learning Reliability Engineering - guided workflows with Learn More modals on every template, progressive complexity from basic to advanced tools; (2) Reliability Engineers & Professionals - comprehensive tool suite, cross-training on multiple methodologies, AI/ML insights to complement expert judgment; (3) Trainers, Consultants & Experts - training aid for workshops, demonstration platform, template customization, cost-effective solution; (4) Technical Management - understand reliability capabilities, evaluate strategies, review AI-powered insights, implement reliability programs.

Industries Served: aerospace & defense, automotive, electronics, medical devices, energy, manufacturing, telecommunications, marine, general reliability engineering.

Technical Specifications: Compatible with FreeBSD 13.0+ and Nomad BSD 14.1+, powered by Python 3.11, 30+ reliability engineering tools, 9 complete HRA methods, 10 specialized AI/ML models, 100 database tables, 29 views, 7 triggers, 156 indices. Built specifically for FreeBSD environments with optimal performance and GPL compliance.

Open Source Compliance: Dual-licensed under GNU General Public License v3.0 (code) and GNU Free Documentation License v1.3 (documentation). All source code includes GPL headers, all documentation includes GFDL notices, comprehensive compliance checking tools included, license headers verified automatically. The project has been submitted to FreeBSD Bugzilla for review as a candidate for the FreeBSD Ports Collection.

Download & Community: Download relysam from Codeberg at https://codeberg.org/0ai/relysam. Join the community of reliability engineers using AI-enhanced tools for improved safety and efficiency. Contribute code, documentation, bug reports, feature suggestions, or simply use relysam and share your experience. Community growth takes time - start with awareness, progress through learning, then practice and implementation.

Installation: Clone repository, install Rust (pkg install -y rust), install system packages (py311-fastapi, py311-uvicorn, py311-pydantic, py311-SQLAlchemy, py311-numpy, py311-pandas, py311-scipy, py311-scikit-learn, py311-matplotlib, py311-seaborn, py311-pytorch, py311-tensorflow, py311-transformers, py311-mlflow, py311-flask, py311-httpx, py311-requests, py311-cryptography, py311-PyYAML), create virtual environment (uv venv --system-site-packages .venv), install 19 venv packages from requirements.txt, generate AI models and setup database (AUTOMATIC with install.sh - python scripts/regenerate_all_binaries.py and python scripts/relysam_database_setup.py --fresh --sample), start application (python -m app.main). Access at http://localhost:8000 with credentials admin/admin123 or guest/guest123.

Documentation: User Manual (manuals/relysam_freebsd_user_manual_complete.md), FAQs (manuals/relysam_freebsd_faqs.md), Installation Guide (INSTALLATION.md), Developer Manual (manuals/relysam_freebsd_developer_manual.md), Project Structure (manuals/relysam_freebsd_project_structure.md), CONTRIBUTING.md, AI Models Complete Guide (docs/AI_MODELS_COMPLETE_GUIDE.md) - Comprehensive guide to all 10 AI models, Custom Data Training Guide (docs/CUSTOM_DATA_TRAINING_GUIDE.md) - Complete guide to training AI models with your data (NEW!), HRA Methodologies (docs/hra_methodologies_documentation.md), Database Regeneration Guide (docs/DATABASE_AND_AI_MODELS_REGENERATION_GUIDE.md), Architecture Overview (docs/architecture_overview.md), FreeBSD Ports Guide (docs/FREEBSD_PORTS_GUIDE.md), Open Source Compliance (docs/OPEN_SOURCE_LICENSE_COMPLIANCE.md), Security (docs/SECURITY.md), CHANGELOG, VERSIONING, Compliance Daemon (docs/compliance_daemon.md), File Registry (docs/file_registry_system_documentation.md), Offline/Online Capabilities (docs/offline_online_capabilities.md), Servers and Services (docs/servers_and_services.md).

⚙️ relysam - Core Reliability Engineering Platform

A comprehensive open-source platform for reliability engineering with AI/ML enhancements

🏗️ Platform: FreeBSD 🐍 Python: 3.11 🔢 Version: 1.1.1 🔓 Open Source
🌟 View Repository on Codeberg
🛡️ License: GPL-3.0 📄 License: GFDL-1.3

📋 Overview

relysam is a comprehensive relysam is a Core Reliability Engineering platform with AI/ML enhancements designed for engineers and organizations involved in reliability engineering, safety analysis, and risk assessment. This platform combines traditional reliability engineering methodologies with advanced artificial intelligence to provide powerful tools for predictive maintenance, failure analysis, and risk assessment.

Repository:
License: GPL-3.0 and GFDL-1.3
Platform: FreeBSD optimized with Python 3.11 compatibility

Version 1.0
FreeBSD
Python 3.11

🚀 Key Features for Reliability Engineers

🔧 Reliability Analysis Tools

📊 Weibull Analysis

Advanced life data analysis with multiple distribution fitting

🏗️ MIL-HDBK-217F

Military reliability prediction standards implementation

⚙️ NSWC-11

Mechanical reliability prediction methodology

⏱️ MTBF/MTTR Calculations

Mean Time Between Failures and Mean Time To Repair

📈 Life Data Analysis

Censored and uncensored data analysis

🎲 Monte Carlo Simulation

Probabilistic risk assessment

🌳 Fault Tree Analysis

Top-down failure analysis

🌿 Event Tree Analysis

Sequential accident modeling

⚠️ Failure Mode Analysis

🔍 FMEA

Failure Mode and Effects Analysis: Comprehensive failure mode identification

🔎 Root Cause Analysis

Systematic investigation of failure causes

🔄 Mixed Failure Analysis

Combined failure mode assessment

👤 Human Reliability Analysis (HRA)

Method Category Description
🎯 THERP First Gen Technique for Human Error Rate Prediction
❤️ HEART First Gen Human Error Assessment and Reduction Technique
⚙️ SPAR-H First Gen Standardized Plant Analysis Risk - Human
🧠 CREAM Second Gen Cognitive Reliability and Error Analysis Method
📏 SLIM-MAUD First Gen Success Likelihood Index Method
🌡️ ATHEANA Second Gen A Technique for Human Event Analysis
📋 JHEDI Specialized Justification of Human Error Data Information
🔬 SHERPA Second Gen Systematic Human Error Reduction and Prediction Approach
📋 MERMOS Specialized Méthode d'Evaluation de la Réalisation des Missions Opérateur pour la Sûreté

🤖 AI/ML-Powered Features

🧠 10 Integrated AI Models

Anomaly detection, failure prediction, maintenance optimization

🔮 Predictive Analytics

Machine learning-driven reliability forecasts

💡 Intelligent Recommendations

AI-powered improvement suggestions

🔍 Pattern Recognition

Automated failure pattern identification

⚡ Automated Assessments

AI-enhanced reliability evaluations

🏭 Industry Applications

Industry Applications Benefits
✈️ Aerospace Flight safety, avionics reliability Enhanced safety margins
🚗 Automotive Vehicle safety systems, component reliability Improved vehicle safety
🔌 Electronics Circuit reliability, semiconductor failure analysis Reduced failure rates
🏥 Medical Devices Patient safety, device reliability Higher safety standards
⚡ Energy Power plant safety, grid reliability Improved system stability
🏭 Manufacturing Production equipment reliability Reduced downtime
📡 Telecommunications Network reliability, service availability Better service quality
🚢 Marine Ship safety systems, navigation reliability Enhanced maritime safety

📸 Application Screenshots

Main Dashboard
Main Dashboard
relysam main dashboard provides access to all core reliability engineering tools with AI/ML enhancements.
Reliability Analysis Tools
Reliability Analysis Tools
Advanced reliability analysis 3D visualization of aviation Fuel Management System simulation.
Enhanced Life Data Analysis with AI Results
Enhanced Life Data Analysis with AI Results
AI-powered insights and recommendations for Comprehensive life data analysis.
HRA Methodologies Interface
HRA Methodologies Interface
HRA methods comparison - Capability analysis radar chart.
HRA THERP Learn More Modal
HRA THERP Learn More Modal
Learn more modal for detailed information on THERP (Technique for Human Error Rate Prediction) methodology.
AI Models Training Status
AI Models Training Status
AI/ML model training status and performance metrics, with real time monitoring and automatic retraining.
Knowledge Base Interface
Knowledge Base Interface
Unified search and access to reliability engineering best practices, standards, and failure modes.
Reports and Comparisons
Reports and Comparisons
DfR assessment reports comparison in comprehensive reporting infrastructure.
SQLite Database Integrity Analysis
SQLite Database Integrity Analysis
Database integrity monitoring and analysis tools for ensuring data consistency and reliability.

🛠️ Installation

For detailed installation instructions, please refer to our INSTALLATION.md guide.

📋 System Requirements

  • Operating System: FreeBSD 13.0+ (also supports Nomad BSD 14.1+)
  • Python Version: Python 3.11 (default Python on FreeBSD 15.0)
  • Permissions: Root access (sudo privileges) to install system packages
  • Storage: Approximately 4GB free disk space for installation
  • Memory: 8GB RAM recommended for AI model processing

⚙️ Installation Steps

1️⃣ Clone the Repository

git clone https://codeberg.org/0ai/relysam.git
cd relysam

2️⃣ Install Rust (Required for Cryptography)

# Install Rust compiler (required for cryptography package)
sudo pkg install -y rust

3️⃣ Install System Packages

# Core web framework
sudo pkg install -ny py311-fastapi py311-uvicorn py311-pydantic py311-SQLAlchemy

# Data science
sudo pkg install -ny py311-numpy py311-pandas py311-scipy py311-scikit-learn py311-matplotlib py311-seaborn

# ML frameworks
sudo pkg install -ny py311-pytorch py311-tensorflow py311-transformers py311-mlflow

# Additional packages
sudo pkg install -ny py311-flask py311-httpx py311-requests py311-cryptography py311-PyYAML

# Note: For complete list of 12 package groups, see INSTALLATION.md

4️⃣ Create Virtual Environment

# Install uv package manager (if not already installed)
pip install --upgrade uv

# Create virtual environment with system site packages
uv venv --system-site-packages .venv
source .venv/bin/activate

5️⃣ Install Virtual Environment Packages

# Install the 19 packages from requirements.txt
# ⚠️ NOTE: pillow is NOT installed here - use system py311-pillow
uv pip install -r requirements.txt

6️⃣ Generate AI Models & Setup Database (AUTOMATIC with install.sh)

# Generate 10 AI models (5-10 minutes)
python scripts/regenerate_all_binaries.py

# Setup database with sample data (2-3 minutes)
python scripts/relysam_database_setup.py --fresh --sample

# [Optional] Import full reference data (~4,000 rows, 2-3 minutes)
python scripts/relysam_database_setup.py --seed

# Note: If using install.sh, this step runs automatically!
# The script will prompt you if you want to import reference data.

7️⃣ Start the Application

# Activate the virtual environment
source .venv/bin/activate

# Start the application
python -m app.main

Access the web interface at: http://localhost:8000
Default accounts: admin/admin123, guest/guest123

Recommended: Use the automated installation script: sudo ./install.sh

This script handles all steps automatically, including an optional prompt to import full reference data (~4,000 rows).

For detailed installation procedures, see INSTALLATION.md.

💎 Why Choose relysam?

🔧 Comprehensive Toolset 30+ specialized reliability engineering tools in one platform
🤖 AI/ML Enhancement Advanced machine learning models for predictive insights
🔓 Open Source Fully transparent, auditable, and community-driven development
🏗️ Industry Standards Implements all major reliability engineering methodologies
⚡ FreeBSD Optimized Efficient performance on FreeBSD systems
🔐 GPL Licensed Freedom to use, modify, and distribute
🎓 Academic Friendly Suitable for research and educational purposes
🌐 Maximum Offline Operation All core functionality works without internet access

🏗️ Architecture

relysam uses a hybrid approach with FreeBSD system packages and virtual environment packages:

  • 📦 409 system packages (py311-*) accessed via --system-site-packages flag
  • 🐍 19 virtual environment packages installed in .venv directory
  • 💾 Single SQLite database (relysam_ai.db) with 100 tables, 29 views, 7 triggers, 156 indices
  • ⚡ FastAPI web framework for REST API and web interface
  • 🧠 10 integrated AI models for predictive analytics

🤝 Contributing

We welcome contributions from the reliability engineering community!

To contribute:

  1. 🐙 Fork the repository
  2. SetBranch Create a feature branch
  3. ✏️ Make your changes
  4. 📝 Add documentation for new features
  5. 📤 Submit a pull request

See our CONTRIBUTING.md file for detailed guidelines.

📚 Documentation

📖 User Manual

Complete user guide

🛠️ Developer Manual

Development guidelines

⚙️ Installation Guide

Complete setup instructions

📄 License

This project is licensed under the GNU General Public License v3.0 (GPL-3.0) and GNU Free Documentation License v1.3 (GFDL-1.3). See the LICENSE_GPL and LICENSE_GFDL files for details.

🆘 Support

❓ Support

Issues & Support

Pull Requests

Wiki

🔍 Keywords for Search

reliability engineering, FMEA, failure mode analysis, human reliability analysis, HRA, THERP, HEART, SPAR-H, CREAM, SLIM-MAUD, SHERPA, ATHEANA, JHEDI, MERMOS, Weibull analysis, MIL-HDBK-217, MTBF, MTTR, fault tree analysis, event tree analysis, predictive maintenance, risk assessment, safety engineering, open source, FreeBSD, machine learning, AI, predictive analytics, life data analysis, Monte Carlo simulation, aerospace safety, automotive reliability, medical device safety, power systems reliability, manufacturing reliability, telecommunications reliability

Common Questions About Relysam

What is relysam?

relysam is an open-source reliability engineering platform with AI/ML enhancements that provides tools for FMEA, Weibull analysis, HRA, fault trees, predictive maintenance, and risk assessment.

What industries use reliability engineering tools like relysam?

Reliability engineering tools are used in aerospace, automotive, electronics, medical devices, energy, manufacturing, telecommunications, and marine industries.

What HRA (Human Reliability Analysis) methods does relysam support?

relysam supports THERP, HEART, SPAR-H, CREAM, SLIM-MAUD, SHERPA, ATHEANA, JHEDI, and MERMOS methodologies.

Is relysam free and open source?

Yes, relysam is completely free and open source under GPL-3.0 and GFDL-1.3 licenses.