Demand Model Using Truncated Poisson Distribution for Retail Price Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Retailers face challenges in demand modeling when sales data includes time periods with zero unit sales, as it is unclear whether these periods represent out-of-stock events or zero sales, and they struggle to optimize prices while limiting the number of price changes due to business constraints.
Innovation Solution
The system models shopper demand using a truncated Poisson distribution and Newton-Raphson method, ignoring zero unit sales events to generate a demand model, and optimizes prices within each network by constraining the number of price changes while maximizing revenue and profit, using a combination of mathematical tradeoffs and business rules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional demand modeling includes zero unit sales events, then more complete sales data is utilized, but the model accuracy deteriorates due to inability to distinguish out-of-stock events from zero sales
Solution Approach 1:
The patent extracts and removes zero unit sales events from the sales data before applying demand modeling. By separating these ambiguous data points from the modeling process, the system prevents them from degrading model accuracy while still utilizing all positive sales data for training the demand model.
Solution Approach 2:
The patent segments the sales data into two distinct categories: zero unit sales events and positive unit sales events. This segmentation allows the demand model to be trained exclusively on reliable positive sales data, while zero sales events are handled separately through imputation methods that do not compromise model precision.
2Power
If the number of price changes is increased to optimize revenue and profit, then better financial performance is achieved, but business rule compliance deteriorates due to limited personnel for price changes
Solution Approach 1:
The patent transforms the price optimization problem from selecting individual price changes to determining an optimal number of price changes parameter. By treating the number of price changes as a controllable parameter with an associated penalty cost, the system automatically balances revenue optimization against operational constraints without manual intervention.
Solution Approach 2:
The patent implements a feedback mechanism where the actual number of price changes executed is compared against the planned number, and penalty costs are applied accordingly. This feedback loop allows the optimization model to learn from past executions and adjust future price change recommendations to better align with operational capabilities.
Data Source
AI summary
The disclosed technology improves the process of generating recommended prices for retail products. First, the present technology makes it possible to model shopper demand when sales data includes time periods with zero unit sales without hypothesizing whether the time periods are out-of-stock events or zero sales. This can be accomplished by applying a truncated Poisson distribution and the Newton-Raphson method to the non-zero unit sales to generate a coefficient vector that maximizes the likelihood of the observations in the sales data. Second, the present technology can be used to generate recommended prices for a group of products that optimize revenue and profit while limiting the number of products that require price changes to a predefined threshold value. This can be accomplished by iteratively replacing a current best value solution with a next best value solution across a collection of product networks until an acceptable number of unchanged prices is achieved.