Case Study

SUV Product Tuning

The previous SUV Market workflow CS-132 SUV Market 2015 Feature Tuning (similar to the Cola Market workflow CS-123 Cola Market 2015 Feature Tuning) tuned the Differentiating Features for the SUV Market. But Feature-level Tuning may not eliminate all of the error in the Market Simulation.

Additional Product-level Tuning can help a Market Simulation more accurately predict real-world scenarios.

Product Tuning works on the final Willingness To Pay (WTP) Matrix generated by the Feature Tuning workflow. Details of the individual Features are lost as Product Tuning works to adjust only the WTP Customer Distribution of each Product.

This Case Study provides a high-level overview of the workflow without detailed explanation. It assumes you are already somewhat familiar with KNIME and Market Simulation. If not, start by reviewing the Building Blocks and Community Nodes.

Product Tuning

The Product Tuning workflow is far less complicated than the earlier Feature Tuning workflow as all the details of the individual Features have been removed.

#1 Tune WTP Matrix

Tune Market

As with the Cola Market, the ‘Tune Market’ node in the SUV Product Tuning workflow is designed to take a partially-tuned Willingness To Pay (WTP) Matrix and adjust the Mean and Standard Deviation (SD) of the Customer Distribution for each Product until the predictions from the Market Simulation accurately reflect real-world observations.

Note that the ‘Tune Market’ node is only available as part of the Scientific Strategy Premium Edition. It is not available within the Free Community Edition of Scientific Strategy.

The Input Product Array (top-port) contains the real-world Prices and Quantities for each Product. In this case, the Quantity and Price data was gathered from the Chinese sales results for 2013. This data is read into the workflow from an Excel spreadsheet.

The Input WTP Matrix (bottom-port) contains the final results from the previous Feature Tuning workflow which were saved in an Excel spreadsheet. The WTP Matrix is a set of Distributions that reflect each Customer’s Willingness To Pay (WTP) for each Product.

After Product Tuning, the Sum of the Total Error is reduced down to 4,019 Customers whose purchase decision cannot be explained by the model. That’s about 0.3% of the 1.49 million total Customers who purchased an SUV in China that year.

Click on an icon below to see and scroll through the enlarged version of the images.