Machine Learning Product Formulation for Faster Attribute Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional methods for predicting attributes of products, such as ingredients, colors, and packaging, are time-consuming and complex.

Innovation Solution

A system utilizing machine learning models to receive and process financial characteristics and properties of sample products to determine desired attributes for potential products, enabling efficient production.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional methods are used to predict product attributes, then measurement precision may be maintained, but productivity deteriorates due to time-consuming processes

Engineering Contradiction:

Improvespeed of determining product attributesVSAvoidtime required for attribute prediction

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-collecting and storing financial characteristics, ingredient data, and product attributes in databases before actual product development. Machine learning models are pre-trained on historical data, enabling rapid prediction of product attributes without time-consuming conventional analysis during the product creation process

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces conventional mechanical/manual methods of attribute prediction with automated machine learning systems. The machine learning model automatically processes financial characteristics and ingredient compositions to predict product attributes, eliminating the need for manual analysis and significantly reducing determination time

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If conventional methods are used for attribute prediction, then measurement precision may be maintained, but device complexity worsens due to nontrivial processes

Engineering Contradiction:

Improveaccuracy of product attribute determinationVSAvoidcomplexity of prediction system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces machine learning models as intermediary components between financial characteristics/ingredients and product attributes. These models act as mediators that automatically process input data and generate predictions, simplifying the overall system architecture while maintaining measurement precision through algorithmic accuracy

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system segments the attribute prediction process into distinct modular components: data collection modules, machine learning model modules, and output generation modules. Each module handles specific tasks independently, making the complex system more manageable and easier to implement while maintaining overall accuracy

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20260037923A1Systems and methods for producing a product

Publication Date: 2026.02.05 COLGATE PALMOLIVE CO

AI summary

In one embodiment, a method includes receiving, for each sample product of sample products belonging to a product category, one or more identities of ingredients forming the sample product. For each of the sample products, a value for each of one or more respective properties of the sample product is also received. A machine learning model receives the values of the one or more respective properties of the sample products and the identities of the ingredients forming the sample products, as well as a desired first value for each of one or more respective properties for a first potential product. The machine learning model determines first identities of ingredients for forming the first potential product based on the desired first value for each of the one or more respective properties of the first potential product.