Promotion Layout Selection Using ML Demand Forecasting

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional promotion methods are time-consuming, inconsistent, and lack systematic optimization, often relying on ad-hoc decision-making and external suppliers with differing interests, leading to suboptimal outcomes.

Innovation Solution

A computer-implemented method and system using machine learning models to optimize promotion parameters, including optimization and forecasting models, to determine optimal promotional strategies based on historical data and user inputs, providing output analytics such as demand forecasts and promotional material configurations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional ad-hoc methods are used to create promotions, then personnel can make decisions based on best judgement, but the process becomes time-consuming and inconsistent

Engineering Contradiction:

Improveconsistency of promotion decisionsVSAvoidtime to create promotion

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces the mechanical human decision-making process with an automated machine learning system. The ML model analyzes historical data and current parameters to generate promotion recommendations, eliminating the time-consuming and inconsistent ad-hoc approach while maintaining reliability through systematic data-driven decisions.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service promotion creation by allowing users to input basic parameters (macroscopic objective, product selection, date range) and automatically receiving optimized promotion recommendations. The ML model performs the complex analysis and optimization work autonomously, reducing dependency on external help and manual processes.

Inventive Principle:

Principle #25Self-service

2Adaptability or versatility

If conventional ad-hoc promotion creation is used, then flexibility in decision-making is maintained, but the approach lacks systematic optimization and statistical selectivity

Engineering Contradiction:

Improvesystematic optimization capabilityVSAvoidcomplexity of promotion methodology

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system systematically analyzes and optimizes multiple promotion parameters including discount levels, product selection, date ranges, and target audiences. By changing and optimizing these parameters based on historical data patterns, the system achieves systematic optimization while maintaining adaptability to different promotion scenarios through configurable input parameters.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If external help from suppliers is used for promotions, then specialized expertise is obtained, but different supplier interests may conflict with business optimization

Engineering Contradiction:

Improvealignment with business goalsVSAvoidease of promotion implementation

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The system eliminates dependency on external suppliers by providing in-house promotion optimization capabilities. The machine learning model performs all analysis and recommendation generation internally, ensuring alignment with business goals while simplifying implementation through automated, standardized processes that don't require external coordination.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS12614206B2Method and system for generation of at least one output analytics for a promotion

Publication Date: 2026.04.28 KINAXIS INC

AI summary

There is provided a method and system for generating an output analytic for a promotion. The method includes training and instantiating a machine learning model comprising at least a Random Forest model, with a selection training set, the selection training set comprising the historical data and the one or more input parameters; selecting, by the processor, using the machine learning model a configuration and a layout for the one or more products on the promotional materials; outputting, by the processor, the promotional materials based on the selection of the configuration and layout.