Region-Based MTBF Prediction
A Machine Learning-based application that calculates the optimal failure rate for each part at a region. It uses both the regional and the global attributes to find the best failure rate which can be potentially any combination of the regional or the global failure rate. These are some of the key benefits:
• An optimal failure rate is calculated for each part at each region.
• A comparison of Servigistics failure rate and Machine Learning-based failure rate via MAPE, MAD, RMSE, Forecast Accuracy calculation between the two.
• $Bias ML, $Bias SPM and Delta Bias for getting the high financial impact Parts.
 |
The Price used to calculate $ values for each Part is average of prices for all locations in a region.
|
• A dashboard to see the summary of all parts including the key attributes.
• A detailed dashboard to analyze a part at any region
• The Feature Importance metrics to see what features are critical for each prediction. This helps in understanding the impact of regional or global features in determining the failure rate for a part for a region.
Related dashboards
ETLs
Use these ETLs to get output from Zeppelin Notebooks, transform data if required, and save results in the Servigistics schema.
Input
◦ QO_ETL_PAI_ML_RMTBF_SKUStreams_MSSQL
◦ QO_ETL_PAI_ML_RMTBF_SKUStreams_ORACLE
| | PO_PAI_ML_RMTBF_FSHostID can be used to select Host Name for Input |
Output
◦ QO_ETL_PAI_ML_RMTBF_TempPrediction_MSSQL
◦ QO_ETL_PAI_ML_RMTBF_Prediction_MSSQL
◦ QO_ETL_PAI_ML_RMTBF_TempPrediction_ORACLE
◦ QO_ETL_PAI_ML_RMTBF_Prediction_ORACLE