Predictive Maintenance using Machine Learning
Data preprocessing, Feature Engineering, Model Development, Evaluation & Selection for Predictive Maintenance in Manufacturing.
Developed by
License
Apache License 2.0 (Apache-2.0)
Part of the development of this asset was supported by the AI REGIO project, which is funded by the European Union Framework Programme for Research and Innovation Horizon 2020 under Grant Agreement n° 952003.
Main Characteristic
- Analysis on common data sources for predictive maintenance problems such as failure and maintenance history, Machine conditions and usage and telemetry data.
- Feature engineering which requires bringing the different data sources together to create features that best describe a machines's health condition at a given point in time. Several feature engineering methods are used to create features based on the properties of each data source.
- Label construction required in multi-class classification for predicting failure due to a problem.
-
Model Training, Validation and Testing: 3 different types of gradient boosting decision trees are examined
-
Model Selection taking into account various evaluation metric as well as other aspects such as required time for model training.
Business Categories
Manufacturing
Last updated
18.11.2022 - 11:15
Trustworthy AI
N/A
GDPR Requirements
N/A