MSC ODYSSEE A-Eye 2025.1

MSC ODYSSEE A-Eye 2025.1

MSC ODYSSEE A-Eye is an artificial intelligence (AI) and machine learning (ML) application within the broader Hexagon ODYSSEE platform. It is specifically engineered for industrial applications, enabling engineers to build, train, and deploy AI models that can interpret complex sensor data, detect anomalies, predict failures, and optimize performance. Unlike generic data science tools, A-Eye is designed with engineering workflows in mind, allowing seamless integration of physics-based simulation data with real-world operational data to create hybrid (physics-informed) AI models and accurate digital twins. It transforms time-series data from tests or operations into actionable insights for predictive maintenance and performance optimization.

It bridges the gap between traditional engineering simulation and the new world of industrial AI and IoT analytics.

Primary Users: Data Scientists, Condition Monitoring Engineers, Reliability Engineers, and Digital Twin Specialists in manufacturing, energy, and aerospace for building AI-powered predictive models from sensor and simulation data.

Key Features & Capabilities

A-Eye focuses on making AI/ML accessible for solving industrial engineering problems.

Core AI/ML Functionality for Engineering:

1. Automated Machine Learning (AutoML):

   • No-Code/Low-Code Model Building: Guided workflows for engineers to create AI models for classification, regression, and anomaly detection without deep coding expertise.

   • Feature Engineering: Automated extraction of meaningful features from time-series sensor data (statistical, frequency-domain, time-domain features).

2. Hybrid Modeling & Digital Twin Enablement:

   • Physics-Informed AI: Integrate physics-based simulation data (from MSC Nastran, Adams, etc.) with operational data to create more robust and generalizable models that respect the underlying physical laws.

   • Reduced-Order Models (ROMs): Use A-Eye to create fast-running, AI-based surrogate models from high-fidelity CAE simulations for real-time digital twin applications.

3. Anomaly Detection & Predictive Maintenance:

   • Unsupervised Learning: Automatically identify abnormal behavior in asset data without pre-labeled failures.

   • Remaining Useful Life (RUL) Prediction: Build models to forecast when a component or system will fail, enabling condition-based maintenance.

4. Deployment & Integration:

   • Model Deployment: Package trained models as lightweight, executable “Smart Components” that can be deployed at the edge (on IoT devices) or in the cloud for real-time inference.

   • ODYSSEE Platform Integration: Part of a full lifecycle from data ingestion (via ODYSSEE Data Hub) to visualization (in ODYSSEE Dashboard).

🖥️ System Requirements

Requirements depend on the deployment model.

For Cloud (SaaS) Users:

• Client: A modern web browser (Chrome, Edge, etc.) and an internet connection. No local compute power required.

• Server/Compute: Fully managed by Hexagon in their cloud.

For On-Premises/Private Cloud Deployment:

• Server Infrastructure: Kubernetes-based cluster or virtual machines as specified by Hexagon for the ODYSSEE platform.

• Compute Resources: Significant CPU/GPU resources for model training, scaling with data volume and model complexity.

• IT Support: Requires dedicated IT/DevOps for platform management