Other Datapred use cases

What is your decision management challenge?

Industrial and consulting companies are using Datapred to solve complex decision management challenges combining predictions and constraint optimization.

 

Contact us

Industrial optimization

The dirty little secret of "predictive maintenance" is that outright asset failures have all but disappeared from modern industrial processes. So what is a data scientist to do?

The answer of Datapred clients is to focus on continuous asset performance monitoring:

  • Real-time multi-horizon performance predictions.
  • Identification of performance and risk factors.
  • Performance simulations.

Datapred's modeling engine is ideally suited for that, with use cases including industrial freezers, rolling mills and heavy-duty electric engines.


Demand prediction

Datapred is an excellent option for demand prediction challenges, for multiple reasons:

  • Sales data is the quintessential time series.
  • Sequential aggregation (the dynamic combination of multiple predictive models) works well on sales data (much better than Recurring Neural Networks for example).
  • Datapred's modeling engine automates hierarchical reconciliation - a key capability for finding where prediction accuracy hides up and down a product hierarchy.
  • Datapred's ability to assess the predictive power of every feature is convenient for assessing the effectiveness of demand management initiatives.

And more

Datapred's modeling engine will perform well in any context where:

  • Time series are prevalent.
  • The time factor is key.
  • Combining diverse modeling techniques makes sense.
  • You are working for production (not just exploring).

Contact us for a discussion on how we could help.

Features

  • Data sequentialization
  • Stationarization and filtering
  • Built-in models
  • Compatibility with ML libraries
  • Aggregation and stacking
  • Custom cost functions
  • Parallelization and distribution
  • Connectors
  • Graph structure
  • Performance monitoring
  • Checkpoints and backups
  • Continual improvement