Product Assortment Planning System Cannibalization Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Retailers face challenges in determining the optimal product assortment as they struggle to accurately assess the impact of adding or removing products on overall performance metrics like sales volume and profits, often relying on inaccurate assumptions or estimations due to difficulties in measuring cannibalization effects.
Innovation Solution
A product assortment planning system that includes a data store for performance metric values, an equivalization unit to adjust these values, and a scaling unit to estimate cannibalization, allowing for the calculation of scaled performance metric values for each product, thereby determining the optimal product assortment to maximize performance metrics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If retailers assume all sales volume is lost when a product is deleted, then the calculation is simple, but the accuracy of performance metric estimation deteriorates
Solution Approach 1:
The patent introduces an intermediary computational model that processes product relationship data and historical sales data to estimate cannibalization effects. This intermediary layer between the simple assumption and the actual complex reality provides accurate estimates without requiring retailers to manually account for every complex interaction.
Solution Approach 2:
The patent replaces the mechanical assumption-based approach with a data-driven computational model. Instead of relying on simple assumptions about sales loss, the system uses algorithms that process historical data and product relationships to automatically calculate accurate estimates of cannibalization and sales impact.
2Device complexity
If retailers estimate sales volume loss based on product importance, then the calculation remains simple, but the accuracy of performance metric estimation deteriorates
Solution Approach 1:
The patent transforms the static parameter of 'product importance' into dynamic, data-driven parameters including historical sales data, product relationships, and cannibalization coefficients. By changing these parameters from subjective importance ratings to objective measured values, the system achieves accurate estimation without excessive computational complexity.
Solution Approach 2:
The system incorporates feedback loops where historical sales data and actual product performance inform the cannibalization models. This feedback mechanism allows the system to continuously improve its estimates based on real-world outcomes, maintaining accuracy while keeping the computational approach manageable.
3Measurement precision
If retailers accurately measure cannibalization effects, then the accuracy of performance metric estimation improves, but the complexity of the system increases
Solution Approach 1:
The patent segments the complex cannibalization measurement problem into manageable components: product relationship identification, historical data analysis, coefficient calculation, and impact projection. By dividing the system into these modular segments, each handling a specific aspect of the measurement, the system achieves high accuracy without becoming unmanageably complex.
Solution Approach 2:
The patent changes the approach from measuring every possible interaction to measuring key parameters such as cannibalization coefficients and product substitution rates. By focusing on these critical parameters rather than attempting to model all possible product relationships, the system achieves accurate cannibalization measurement with controlled complexity.
Data Source
AI summary
A product assortment planning system determines scaled performance metric values for an assortment of products. The system includes a data store storing performance metric values for an assortment of products including a target assortment of products and a source assortment of products, an equivalization unit and a scaling unit. The equivalization unit equivalizes the performance metric values for the source assortment of products. The scaling unit determines incrementality assumptions. The incrementality assumptions are an estimation of an amount of cannibalization that occurs for the target assortment of products as a result of combining the source assortment of products with the target assortment of products. Scaled performance metric values are calculated for each product in the assortment of products based on the equivalized performance metric values and the incrementality assumptions.


