Pipelines scheduled automation

Our machine learning models need to adapt to evolving data patterns to maintain their accuracy and relevance. By scheduling our pipeline to run automatically every 10 minutes, we ensure that both the Stress Detection and Time to Failure Prediction models are continuously retrained with the latest battery data collected from the BMS application. This automated approach eliminates manual intervention, reduces the risk of model drift, and guarantees that our predictive models always reflect the most current battery behavior patterns.

Schedule recurring execution

Now let’s configure the pipeline to execute automatically:

  1. Now, let’s go back to the Pipelines view in the left-hand menu.

  2. Click again on the Three dots at the far right of our Model Retraining pipeline.

    Start Pipeline schedule

  3. Select Create schedule from the dropdown menu.

  4. Fill in the schedule configuration form:

    • Name: Scheduled run

    • Trigger type: Select Periodic

    • Run every: 10 minutes

  5. Scroll down and verify that the Model Retraining pipeline has been automatically selected.

  6. Check that the Parameters match the following values:

    Parameter Value

    aws_access_key_id

    minio

    aws_s3_bucket

    inference

    aws_s3_endpoint

    http://minio-microshift-vm.microshift-001.svc.cluster.local:30000

    aws_secret_access_key

    minio123

    influxdb_bucket

    bms

    influxdb_org

    redhat

    influxdb_token

    admin_token

    influxdb_url

    https://influx-db-microshift-001.{openshift_cluster_ingress_domain}/

  7. When all parameters are correctly configured, click Create schedule.

The pipeline will now execute automatically every 10 minutes, continuously improving your models with fresh battery data. Come back after 10 minutes and the first scheduled run should be underway.