Uncertainty Aware Functional Behavior Prediction and Material Fatigue Assessment for Circular Factory

2026-06-03 · Source: Artificial Intelligence · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, quick

Summary

A novel framework addresses the challenge of assessing returned products with heterogeneous degradation in circular factories by combining uncertainty-aware functional prediction with component-level fatigue assessment. Developed for an angle grinder, the system integrates current tool state and recent force-torque usage data. It employs a convolutional encoder for loading patterns and an LSTM backbone to predict nine functional variables, providing Gaussian mean and variance estimates. Concurrently, the same loading history informs output-shaft fatigue through finite-element stress reconstruction, S-N/Miner damage evaluation with Haibach extension, and Paris-law crack-growth analysis. A streaming replay algorithm merges these into functional, material, and system reliability trajectories. Held-out tests demonstrate a mean 2%-tolerance accuracy of 0.9652 across all nine outputs, with thermal variables predicted almost perfectly. Drive motor current and load speed, with R^2 values of 0.9750 and 0.9924 respectively, are the most dynamic outputs, where torque history is particularly important.

Key takeaway

For Manufacturing Engineers or Circular Economy Managers assessing returned products for reuse, this framework offers a robust method to predict future functional fulfillment and component integrity. By integrating uncertainty-aware functional prediction with material fatigue assessment, you can make more informed, instance-specific decisions on product disposition, reducing waste and optimizing resource utilization. Consider implementing similar combined prognostic approaches for complex electromechanical systems.

Key insights

The framework links functional prediction and material fatigue for instance-specific reliability in circular factories.

Principles

• Combine functional and material prognostics.
• Account for uncertainty in degradation states.
• Instance-specific reliability requires usage history.

Method

The framework uses a convolutional encoder for loading patterns, an LSTM for functional variable prediction, and finite-element analysis for fatigue assessment, consolidated by a streaming replay algorithm.

In practice

• Assess returned products for reuse decisions.
• Predict angle grinder component integrity.
• Monitor system reliability trajectories.

Topics

• Circular Manufacturing
• Predictive Maintenance
• Material Fatigue
• Functional Prognosis
• LSTM Networks
• Uncertainty Quantification

Best for: AI Scientist, Machine Learning Engineer, Research Scientist

Related on AIssential

• Uncertainty Aware Functional Behavior Prediction and Material Fatigue Assessment for Circular Factory — Jointly assessing functional behavior and material fatigue enables instance-specific redeployment decisions for returned products.
• Finite Element-Based Material Learning via Automatic Differentiation: Learning constitutive neural network models from full-field deformation data — FE-MAD uses automatic differentiation within a JAX-FEM solver to learn constitutive neural network models from full-field deformation data.
• Machine learning for sustainable geoenergy — Machine learning can enhance sustainable geoenergy projects by addressing data scarcity and uncertainty through physics-informed methods.
• News - NSWCPD Applies AI and Machine Learning to Predictive Machinery Health - DVIDS — AI/ML-driven vibration analysis can predict mechanical failures, enhancing condition-based maintenance for naval platforms.
• CatBoost + Conformal: production-grade uncertainty for tabular ML — Integrating CatBoost with Conformal Prediction provides robust uncertainty quantification for tabular ML.
• A Multi-Plant Machine Learning Framework for Emission Prediction, Forecasting, and Control in Cement Manufacturing — Data-driven ML models can significantly reduce cement plant NOx emissions and costs without hardware changes.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence.