Machine Learning Pricing for Demand-Driven Retail Inventory

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Retailers and manufacturers face challenges in determining optimal pricing and inventory levels for items, as setting prices too high can discourage sales while setting them too low can reduce overall revenue, and they struggle to predict demand accurately, leading to overstocking or understocking.

Innovation Solution

Utilizing trained machine learning processes, including time-series models and mixed-integer programming, to predict item demand and determine recommended prices based on historical sales data, allowing for accurate inventory management and pricing strategies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If the price of an item is set high, then the revenue per sale increases, but the quantity of items sold decreases

Engineering Contradiction:

ImproverevenueVSAvoidquantity of items sold

Core Design Contradiction:

Loss of energyVSProductivity

Solution Approach 1:

The system dynamically adjusts pricing parameters based on predicted demand, item category, and budget allocation. By changing price parameters in response to varying conditions, the system optimizes the balance between revenue per sale and quantity sold, resolving the contradiction between these two objectives.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The pricing system transitions from static to dynamic by using machine learning models to predict demand and determine optimal prices in real-time. This dynamic approach allows the system to adapt pricing strategies to maximize both revenue and sales quantity under different market conditions.

Inventive Principle:

Principle #15Dynamics

2Productivity

If the price of an item is set low, then the quantity of items sold increases, but the overall revenue decreases

Engineering Contradiction:

Improvequantity of items soldVSAvoidrevenue

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The system adjusts pricing parameters dynamically based on predicted demand and budget constraints. By modifying price parameters according to real-time conditions, the system achieves optimal balance between sales volume and revenue, preventing the loss of revenue that would occur with uniformly low pricing.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The dynamic pricing model enables the system to respond to changing market conditions by adjusting prices optimally. This resolves the contradiction by allowing the system to capture high-volume sales when appropriate while maintaining revenue targets through strategic price adjustments.

Inventive Principle:

Principle #15Dynamics

3Ease of operation

If retailers stock a large amount of an item, then the availability for sale increases, but the inventory cost and risk of unsold items increases

Engineering Contradiction:

Improveavailability of itemsVSAvoidinventory level

Core Design Contradiction:

Ease of operationVSQuantity of substance

Solution Approach 1:

The system performs preliminary demand prediction using machine learning models before setting inventory levels. By anticipating future demand based on historical data and trends, the system prepares appropriate inventory quantities in advance, ensuring availability while avoiding excessive stockpiling.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from sales data and demand predictions to continuously adjust inventory levels. This closed-loop approach allows the system to maintain optimal availability while minimizing excess inventory, as the system learns from past performance and adapts its stocking decisions.

Inventive Principle:

Principle #23Feedback

4Quantity of substance

If retailers stock a small amount of an item, then the inventory cost decreases, but the availability for sale and potential sales loss increases

Engineering Contradiction:

Improveinventory levelVSAvoidavailability of items

Core Design Contradiction:

Quantity of substanceVSEase of operation

Solution Approach 1:

The system performs preliminary demand prediction to determine optimal inventory levels before stocking decisions are made. By forecasting future demand accurately, the system ensures sufficient inventory availability while minimizing excess stock, resolving the contradiction between low inventory costs and maintained availability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors sales performance and demand patterns, using this feedback to adjust inventory levels dynamically. This ensures that inventory is maintained at optimal levels to prevent stockouts while avoiding unnecessary holding costs.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS12572954B2Methods and apparatus for determining item demand and pricing using machine learning processes

Publication Date: 2026.03.10 WALMART APOLLO LLC

AI summary

This application relates to employing trained machine learning processes to predict demands of items during future temporal periods, and to determining a recommended price for the items. For example, a computing device may obtain sales data for an item, and may generate features based on the obtained sales data. The computing device may input the generated features to a trained machine learning process to generate output data characterizing a predicted demand of the item during a future temporal interval. Further, the computing device may determine a recommended price for the item based on the predicted demand and a budget allocation that corresponds to the item. The computing device may store the predicted demand of the item and the recommended price for the item in a data repository.