Overview (Optimization - Multifactor Elasticity)

Optimization - Multifactor Elasticity Accelerator deploys a powerful machine learning approach to assess elasticity from a forecasting model. These functions allow to forecast the quantity that would be sold at a given price.

Approach

Multifactor Elasticity Accelerator relies on several steps:

• First, map the data and include as many features as possible to give enough context and information to the forecasting model, including additional sources that could provide future values, such as special events.

• Build a first forecasting model and test outputs for a holding time frame in order to define the best model parameters and get metrics on the holding period to get a fair assessment of the model accuracy.

• Train the model with the latest available data in order to get full knowledge of latest trends.

• Predict quantity for next periods, including expected quantity with some price variations.

• Fit an elasticity curve to match the expected quantity for these price variations and store those elasticity parameters, made available for other part of the solution.

Outputs

The results of a Multifactor Elasticity model are forecasted quantity and elasticity parameters assessed by the model. Several charts are also available to check outputs, such as demand forecast per product, the elasticity curves built from these forecasts and an evaluation providing the elasticity parameters that can be used by another model such as Accelerate Shelf Prices Optimization or Accelerate Markdown Optimization.

Starting with version 1.2, the elasticity model can also output a fallback, called simple elasticity, defined at a higher level of aggregation, such as product category, which can then be used for new products or if there is not enough data to get an accurate elasticity assessment.

Limitations

• Attributes / features – Multi-factor elasticity relies on having the right features that impact the sold quantity. If those features are not available or the data are incomplete, output quality will be low. Also, additional features should either be product or data related, otherwise each additional feature would create a specific time series.

• Training periods – Training period should be a complete period, meaning if the last period only contains partial data, like half a week, the outputs will be biased and probably wrong.

• Sparse data – Multi-factor elasticity relies on sold quantity from past transactions to forecast coming sales. However, if some products are new or barely sold, forecasts might be less accurate. To work around this potential issue, some aggregations such as Accelerate Product Similarity are already in place, but some pre-processing or adjustment might be helpful in order to get accurate enough data.

• No predefined extension point – There is no out-of-the-box extension point defined for now. If you intend to add specific features, custom code should be written. (But then the accelerator becomes specific and it cannot be updated without extra effort to port those modifications.) Do not hesitate to report specific requirements and possible extension points to Pricefx.

• Data requirements – See Data Requirements (Optimization - Multifactor Elasticity).