Retail Demand Transfer Estimation Using Inventory Availability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing demand transfer estimation methods in retail fail to consider out-of-stock conditions, leading to inaccurate inventory management strategies, as they primarily focus on product assortment rather than inventory availability.

Innovation Solution

A processor-implemented method and system that integrates historical sales and inventory data to estimate product demand transfer by determining actual, missing, and expected shares of attribute values, calculating additional shares due to missed values, and computing demand transfer based on these shares, using a multiplicative model to account for inventory availability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing demand transfer estimation methods focus on product assortment, then assortment decisions can be optimized, but out-of-stock conditions are not considered leading to inaccurate inventory management

Engineering Contradiction:
Improveaccuracy of demand transfer estimationVSAvoidconsideration of out-of-stock conditions
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent segments the demand transfer estimation into two distinct components: assortment-based demand transfer and inventory availability-based demand transfer. By separating these factors, the model can independently analyze and quantify the impact of out-of-stock conditions on demand transfer, thereby improving estimation accuracy while maintaining adaptability to different retail scenarios.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces new parameters to the demand transfer model, specifically incorporating inventory availability metrics (in-stock probability, fill rate) alongside traditional assortment parameters. This parameter expansion allows the model to simultaneously consider both assortment decisions and inventory conditions, resolving the contradiction between estimation accuracy and adaptability to out-of-stock scenarios.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If demand transfer estimation includes multiple data sources (sales, inventory, assortment), then estimation accuracy improves, but system complexity increases

Engineering Contradiction:
Improvedemand transfer estimation accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent develops a unified demand transfer estimation model that serves multiple functions: it can estimate demand transfer due to assortment decisions, demand transfer due to out-of-stock conditions, and overall demand transfer. This multi-functional approach integrates multiple data sources (sales data, inventory data, assortment data) into a single coherent framework, improving measurement precision while managing system complexity through consolidation rather than separate analysis systems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces a probabilistic framework as an intermediary mechanism that connects multiple data sources to the demand transfer estimation. This probabilistic model acts as a mediator that systematically processes sales data, inventory data, and assortment data, transforming them into coherent demand transfer estimates. The intermediary framework organizes the complexity of multiple data sources into a manageable computational structure.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3879479B1System and method for inventory based product demand transfer estimation in retail
Publication Date: 2026.01.21 TATA CONSULTANCY SERVICES LTD
  • EP3879479B1 patent drawingFigure 1
  • EP3879479B1 patent drawingFigure 2
  • EP3879479B1 patent drawing

AI summary

This disclosure relates generally to a system and method for product demand transfer estimation in retail. Demand transfer happens across one or more products within a category of substitutable products. Product level demand transfer indicates the amount of demand that is going to be transferred from the unavailable product to another product within the category. An inventory availability is considered herein rather than assortment to ensure that the product is present or not for a particular point in time. The system is configured to estimate product level demand transfer value based on the attributes values present in the one or more products considered. Moreover, the embodiments herein further provide product level demand transfer using the attribute value and demand transfer of values of those attributes that are different between two products.