Simplify Maximo Predict out-of-the-box capabilities by providing a guided UI for configuring and running predictive models instead of requiring customers to run Jupyter Notebooks in Cloud Pak for Data.

One of the biggest challenges with Maximo Predict is that its out-of-the-box (OOB) predictive capabilities are heavily dependent on Jupyter Notebooks and Cloud Pak for Data (CP4D). To use features such as Anomaly Detection, Failure Probability, Remaining Useful Life (RUL), and End-of-Life Prediction, implementation teams must configure CP4D projects, upload datasets, manage credentials, and manually execute notebooks.

Many Maximo consultants, business partners, and customers do not have a data science background, making it difficult to consume, demonstrate, and implement these capabilities. As a result, customers who are interested in predictive maintenance often find it challenging to adopt Maximo Predict.

It would be beneficial if the standard OOB predictive capabilities were available through a guided UI in Maximo Predict. For example, the UI could allow users to:

• Upload the required datasets.
• Validate the uploaded data.
• Select the required sensor attributes or input features.
• Configure basic model parameters.
• Run the prediction directly from the UI.
• View prediction results and visualizations without manually executing notebooks.

The Jupyter notebooks and CP4D workflows can continue to remain as the backend execution engine for these standard capabilities.

For advanced or customer-specific use cases, where customers want to build new notebooks, develop custom ML models, or implement their own data science workflows, they can continue using Jupyter Notebooks and Cloud Pak for Data.

This approach would make Maximo Predict much easier to implement, reduce the dependency on data science expertise for standard OOB features, improve partner enablement, and significantly increase customer adoption while still preserving the flexibility for advanced customizations.