Create Failure Probability Model

Maximo Predict comes with notebook templates to assist in streamlining data uploads to Maximo. This notebook will create the Failure Probability using provided csv files.

These instructions use the notebook named 6_PMI-Failure Probability-BinaryClassification-HPU file with the Substation Transformer for Health and Predict for Utilities Demo Assets. Note that this uses simulated Pump Data for the sensor readings.

In this exercise you will use Watson Studio and Health and Predict - Utilities to:

  1. Upload the and Run the Failure Probability Notebook using a template to Train and Display a Failure Probability value
  2. Confirm the Data Has been uploaded for your assets

Note

You must complete the previous exercise for Setup Watson Studio before you start this exercise.

Pre-requisites

Note

It is best to perform this lab in your own Watson Studio Project created using Setup Watson Studio instructions. If you are using a shared project, ensure you append each file uploaded with your initials and update the file paths in the notebooks to include that change.

Upload and Start the Predicted Failure Date Notebook

  1. Upload or open the Failure Probability template notebook to your Project. Use the steps from the previous exercise Add Notebook From File to a Watson Studio Project If you are using a shared project, rename the notebook template by prepending your initials to the template. If this is done, ensure any paths or file names within the notebook are updated as well. If you already have uploaded the notebook, open it with Watson Studio. Select the 6_PMI-Failure Probability-BinaryClassification-HPU notebook template.

  2. Open the notebook. Click on the pencil icon next to your notebook.

  3. If the notebook fails to start, restart it. Click on the i icon , Environment tab, Running status drop down select box and choose Restart

Restarting Environment

Run the Notebook

Install the Maximo Predict SDK

  1. Read the introduction to the Failure Probability Notebook.

  2. Run the first cell to install pyspark and set up the error logs.

  3. Run the second cell to define the requirements and some environment variables to run this notebook. This cell also checks that the precursor notebook has been run. Additionally, this cell sets the device_type and the asset_group_id from the stored device type and asset group id in the JSON file produced from the Fast Start Data Loader notebook. Ensure the printed values match your asset group and device.

Asset Group ID

  1. Run the next cell to uninstall the pmlib library. This is done to ensure the correct version is installed.

  2. Run the next cell to define the API keys used to call Maximo Predict from the Predict_Envs.JSON file

  3. Skip the next cell. It should be commented out, but it would be used to define the API and Predict Environment variables manually if the Predict_Envs.JSONfile was not available.

  4. Run the next cell to import the os, trim the provided base url to be used when downloading pmlib later in the notebook and used to contact the environment via API.

  5. Skip the next cell. It should be commented out. This cell is used to hardcode some API information that is not needed here.

  6. Run the following five cells to reinstall the pmlib and import sklearn and other necessary libraries.

Set up the Model Training Pipeline, Train, Register and Enable the Model

  1. Run the first to import pmlib library

  2. Run the second cell to define the sensor readings columns to be considered in the model

  3. Skip the next few cells until you reach the How to Custom Model section. In this section, we customize the pipeline settings and the models to be considered.

  4. Run the first cell to import the necessary models for the pipeline

  5. Run the next cell to transform the desired features and define some additional settings for the pipeline. Some of these settings include determining which models to consider and how to process the features

  6. Skip the next cell, those are some alternate settings.

  7. Run the next cell to define the pipeline and standard settings for the modeling process

  8. Run the next cell to train the model. This cell can take some time for some models.

  9. Once this process is complete, run the following cell to register the model to your asset group.

  10. Finally, Run the next cell to enable it and determine how often it will be run in monitor. This is the last cell to be run in this notebook.

Confirm Model Registration

  1. Navigate to Maximo Health and Predict for Utilities within your environment

  2. Use the left-hand menu to go into Predict Grouping Predict Grouping

  3. Select your asset group Predict Grouping

  4. Click into your asset group and ensure you have Failure Probability listed under Trained instances registered for this group and select an asset to go to the Health Dashboard

Predict Grouping

  1. Scroll down and expand the Predict section to ensure the Failure Probability, Failure Probability Trend and Factors that Contribute to Failures is visible.

Predict Results

Note

Recall in the Create Utilities Predict Group and Upload Sensor Data lab only some assets have sensor data. If an asset does not have sensor data, it will not have Predict data. Additionally, most the assets with data will have a Predicted Failure date of 0 \plusminus x The following assets should have a future failure date when the notebook is ran:

Congratulations you have created a Failure Probability model and associated it to your assets!