TABLE OF CONTENTS
Publish NPI
Publishing NPI results to Demand Planning
Once the NPI (New Product Introduction) results are generated, planners can publish them to Demand Planning. To do this, click the Publish button in the NPI screen. The application will display a confirmation message before proceeding.
When confirmed, the system publishes the final NPI forecast results to the Demand Planning module. This means the computed NPI forecast values are written into a specific time series (for example, Demand Planning Units), which is defined during the implementation phase and configured in the Admin App under Demand Planning settings.
As NPI forecasts are generated at the Product All–All–All level, publishing them copies these values to the corresponding Demand Planning time series at the same level (Product All–All–All). This ensures consistency between the NPI output and the demand planning baseline.
Disaggregation of NPI results
During NPI plan creation, the planner must select a reference item that the system will use to perform disaggregation after the NPI plan is generated. This selection happens within the same workflow — it is not a separate step after the NPI process is completed. (Refer the screenshot below for example)
The selected reference item serves as the base for determining how the NPI forecast should be distributed across lower-level relationships, such as Product–Customer–Location. When the NPI results are published, the application automatically:
Copies the relationships (customer and location combinations) from the reference item to the new NPI product.
Uses the reference item’s demand pattern to proportionally distribute the total NPI forecast values across these relationships.
For example, if the reference item chosen during NPI plan creation is ThinkPad, the system identifies all existing Product–Customer–Location combinations for ThinkPad and mirrors those combinations for the new NPI product. The total NPI forecast (at Product All–All–All level) is then proportionally disaggregated across these combinations based on ThinkPad’s historical demand share.
This approach ensures that the newly introduced product inherits realistic demand distribution patterns aligned with the reference product’s market footprint and proportional demand contribution.
Let’s consider the following example to understand the publish & disaggregation logic
The NPI item is eTab 300 V1. For generating the NPI forecast, two like products — eTab 100 V1 and eTab 200 V1 — are used with product weights of 60% and 40%, respectively.
The input measure considered is Statistical Forecast Units
The data used for computation spans from January 2024 to June 2024.
The system combines the statistical forecasts of the two like products using the assigned weights to generate the NPI forecast for eTab 300 V1.
The reference item for disaggregation is ThinkPad, which the system uses to distribute the NPI forecast proportionally across customer and location combinations based on ThinkPad’s existing relationships.
NPI Parameters
NPI Item → Etab 300 v1
Input measure → Statistical forecast units
Like item for disaggregation → Think Pad
Ramp up curve → 100 %
Publish Final NPI Forecast to → Demand Plan units (as per configuration)
The table below shows the like items (Etab 100 v1 AND Etab 200 v1) data at Product | Customer | Location | All and Product | All | All | All levels
The table below shows the Final NPI Forecast generated for Etab 300 v1 by considering the Like Products inputs at the Product All All All level. Notice, it has considered the Like products and respective product weights to generate the Base NPI Forecast. As the ramp-up curve is 100%, the Base NPI Forecast is copied to the Final NPI Forecast
Once clicking on the Publish button, the application copies the Final NPI Forecast generated at the Product All All All level to the Demand Plan Units at Product All All All level
The table below shows that the Final NPI forecast is copied to Demand Plan units for Etab 300 v1.
The application then disaggregates the values of Demand Plan units from Product | All | All | All level to → Product | Customer | Location | All Level,
The table below shows the disaggregation reference item (Think Pad) data at Product | Customer | Location| All and Product | All | All| All levels
The application then copies the relationships of the disaggregation item to the NPI item. The table below shows the same.
The Demand Plan units at the Etab 300 v1 | All| All | All level will be disaggregated proportionally at respective Customer location combinations. The example below shows how it is proportionally disaggregated.
For each period:
(Reference Item value at lower level) × (Final NPI Forecast) / (Reference Item total at product level)
I.e. For Think Pad | Walmart | Dallas (2100) x Etab 300 V1 | All | ALL (4406)/ Think Pad | All | All (5490)
Example – Jan-24
Total = 4406 → matches aggregate total
The same logic applies to all future months.
Totals at the disaggregated level match the original Final NPI Forecast.
Key Notes
Disaggregation relationships always come from the selected like item.
The Reference item used for disaggregation does not need to be one of the like items used in forecasting.
If no relationships exist for the like item, disaggregation cannot proceed.
Demand Plan Units must be configured at the detailed calculation level (Product | Customer | Location | All).
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article