Failure Prediction

Failure prediction research aims to anticipate malfunctions across diverse systems, from computer hardware and industrial equipment to robots and cloud infrastructure, ultimately improving reliability and efficiency. Current efforts focus on developing robust predictive models using machine learning techniques, including transformers, neural networks, and ensemble methods, often incorporating data imputation strategies to handle incomplete datasets. These advancements are crucial for optimizing maintenance schedules, enhancing safety in critical applications, and reducing downtime and associated costs across various industries.

Papers

June 13, 2022

Anomaly Detection and Inter-Sensor Transfer Learning on Smart Manufacturing Datasets
Mustafa Abdallah, Byung-Gun Joung, Wo Jae Lee, Charilaos Mousoulis, John W. Sutherland, Saurabh Bagchi
Anomaly Detection Smart Manufacturing Failure Prediction Cross Sensor Manufacturing Datasets

April 28, 2022

Digital Twin Framework for Time to Failure Forecasting of Wind Turbine Gearbox: A Concept
Mili Wadhwani, Sakshi Deshmukh, Harsh S. Dhiman
Digital Twin Time Matter High Impact Concept Failure Prediction Wind Turbine Blade Wind Turbine Gearbox

February 11, 2022

Failure Prediction with Statistical Guarantees for Vision-Based Robot Control
Alec Farid, David Snyder, Allen Z. Ren, Anirudha Majumdar
Vision Based Statistical Guarantee Failure Prediction Safety Critical

November 19, 2021

Data imputation and comparison of custom ensemble models with existing libraries like XGBoost, Scikit learn, etc. for Predictive Equipment failure
Tejas Y. Deo
Data Set Ensemble Model XGBoost Model Data Imputation Scikit Learn Failure Prediction Imputation Model

Failure Prediction

Papers

Anomaly Detection and Inter-Sensor Transfer Learning on Smart Manufacturing Datasets

Digital Twin Framework for Time to Failure Forecasting of Wind Turbine Gearbox: A Concept

Failure Prediction with Statistical Guarantees for Vision-Based Robot Control

Data imputation and comparison of custom ensemble models with existing libraries like XGBoost, Scikit learn, etc. for Predictive Equipment failure