Understanding Poor Forecast Results, Bad Forecast Results, or Confusing Forecast Results generated by the Demantra Analytical Engine (Doc ID 744602.1)

Last updated on SEPTEMBER 11, 2016

Applies to:

Oracle Demantra Demand Management - Version 7.0.1 and later
Information in this document applies to any platform.
***Checked for relevance on 30-MAR-2016***


Goal

This note was developed to help the Client, Implementer and/or Oracle Support understand any concerns with the forecast that the Demantra Analytical Batch Engine (referred to in this note as 'the Demantra Engine') has generated.  This note is not meant to form the basis of a full forecast tuning and is instead meant to assess the validity of comments made by users along the lines of 'this forecast is bad' or 'I don't understand how this forecast line was generated by the Demantra Engine'.  The Demantra Engine is a powerful tool which, out of the box generally provides a quality forecast when provided with even minimal demand data.  However, utilizing the full extent of the the Demantra analytics (including the effects of causal factors on the business) might require a tuning exercise or at least further understanding of the Demantra Engine and its capability.  This note will help both support and the client better understand what the engine has to work with and what might be possible.

 It should be made clear that when assessing forecast quality you should not concern yourself with how the forecast looks in historical time buckets.  Demantra is not a 'best-fit' forecasting tool and thus the only concern we should have is how the forecast looks in future time buckets.  This concept of historical fit is discussed in <Note 602242.1>.

It should be noted that this document is geared towards Demantra Sales and Operations Planning and Demantra Demand Management installations .  Clients using Demantra Predictive Trade Planning, which utilize more sophisticated promotion modeling tools, might not find this document as useful although a lot of the concepts have very close similarites even if the data streams are different. 

Solution

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms