@chrisjcc has used the SHap library (https://github.com/slundberg/shap) to create displays showing why a machine learning model suggests the action that it does.
We can include this as part of the display for the operator to understand why the suggestion is what it is.