ENEL is trying to develop predictive models to be used in forecasting processes
This use case is focused on one of the main processes:
Forecast external costs based on historical economical results
Main challenges
External costs should be forecasted at several levels of aggregations (e.g., currency, legal entity, activity)
Historical data has been anonymized, preventing external knowledge from being added to the data
Historical data contains some issues in terms of data quality that can decrease the performance of the model and need to be addressed and fixed (e.g., unpredictable events, closed plants)
