Remaining Useful Life Prediction Using Weibull (RUL)

What Does the Algorithm Do?

The Weibull algorithm calculates the distribution of asset lifetimes based on age. To perform this calculation, the algorithm needs to know the ages of assets when they failed, and the ages of assets that still work. The assets and their components need to be of the same type. The result is an estimate of the remaining useful life (RUL) of each component at the time of the calculation. The algorithm can also calculate the probability of failure of components within a certain time period.

Model Configuration

To configure a model for an RUL prediction with the Weibull algorithm, use the REST APIs or configuration UIs for the machine learning engine. For more information, see the chapters Managing Machine Learning Engine Using Configuration UIs and Managing Machine Learning Engine Using REST APIs in the guide Configuring SAP Predictive Maintenance and Service, on-premise edition 1.0

Data Preparation for Model Training and Scoring

Before training and scoring, data scientists need to configure a model. In the configuration, they need to specify the column that contains the age when an asset failed, or the current age of an asset that is still working.

The algorithm also needs information on whether an asset is still working or if it is not working anymore. The status value 0 means that the asset is still working, the status value 1 means that the asset failed and is not working anymore.

To calculate the probability of failure for a certain time period, this time period needs to be specified using the same time unit as used for the age of the asset.

To predict the probability of failure for a certain time period (in a week's time, for example), this prediction period needs to be specified as well using the same time unit as used for the age of the asset.

Model Training

To train a model for RUL using the Weibull algorithm, use the REST APIs or configuration UIs for the machine learning engine. For more information, see the chapters Managing Machine Learning Engine Using Configuration UIs and Managing Machine Learning Engine Using REST APIs in the guide Configuring SAP Predictive Maintenance and Service, on-premise edition.

Model Scoring

To score a model for RUL using the Weibull algorithm, use the REST APIs or configuration UIs for the machine learning engine. For more information, see the chapters Managing Machine Learning Engine Using Configuration UIs and Managing Machine Learning Engine Using REST APIs in the guide Configuring SAP Predictive Maintenance and Service, on-premise edition 1.0