SD_PREDICTED_DELIVERY Predicted Delivery Processing Delay ML train failure

[yejinshim]

Hello,

I'm trying to use 'Predicted Delivery Processing Delay (PRDTDELIVPROCGDELAY)' and train the ML, but the train keeps failing saying ' We've noticed your training dataset has very few records'.

As I've already created more than 14,000 S/O and follow-up document such as outbound deliveries and PGI/billing were done, I don't think this is the dataset record issue. For Delivery Creation Delay Scenario which is under the same SD_Predicted_Delivery package, this scenario counts the dataset records between 14,000-15,000.

Is there anything I should do to train this ML scenario??

Thank you in advance