Aqueous Polymer Coating Odor Prediction Using VOC Decision Trees

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional odor evaluation of aqueous polymer compositions and coatings is laborious, subjective, and hazardous for human panelists, and existing computer-implemented methods are not applicable to the complex chemical compositions found in aqueous polymer systems.

Innovation Solution

A method using a decision tree ensemble trained with concentration data of volatile organic compounds (VOCs) to predict odor intensity, utilizing analytical characterization and machine learning for standardized and automated odor prediction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If human sensory panels are used for odor evaluation, then odor intensity can be assessed, but the process becomes laborious, time-consuming, and subjective

Engineering Contradiction:

Improveodor intensity assessmentVSAvoidevaluation time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the human sensory panel system with a computer-implemented machine learning model. The system uses a decision tree ensemble algorithm that processes volatile organic compound concentration data to predict odor intensity, eliminating the need for human panelists and their associated time consumption and subjectivity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a computational model that copies and simulates the human odor perception function. The machine learning model is trained on data from human panelists and replicates their odor intensity assessments, allowing for rapid, objective predictions without requiring actual human exposure to the samples.

Inventive Principle:

Principle #26Copying

2Measurement precision

If human sensory panels are used for odor evaluation, then odor intensity can be assessed, but the process becomes subjective and inconsistent

Engineering Contradiction:

Improveodor intensity assessmentVSAvoidtest consistency

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent implements a feedback mechanism where the machine learning model is continuously trained and validated against known odor intensity data from human panels. This feedback loop allows the model to learn from actual human assessments and improve its prediction accuracy, ensuring consistent and reliable results.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If human sensory panels are used for odor evaluation, then odor intensity can be assessed, but it poses safety hazards to panelists

Engineering Contradiction:

Improveodor intensity assessmentVSAvoidhazardous exposure

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an intermediary computational model that acts as a mediator between the odor sample and the assessment. Instead of humans directly exposing themselves to the odor, the machine learning model processes concentration data and predicts odor intensity, serving as a safe intermediary that eliminates direct human exposure to hazardous substances.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Extent of automation

If conventional computer odor prediction methods are used, then automation is achieved, but they are not applicable to complex aqueous polymer compositions

Engineering Contradiction:

Improveodor prediction automationVSAvoidapplicability to aqueous polymer systems

Core Design Contradiction:

Extent of automationVSAdaptability or versatility

Solution Approach 1:

The patent adapts the machine learning model to handle the specific parameters of aqueous polymer compositions. The decision tree ensemble is trained on concentration data of volatile organic compounds found in these complex systems, allowing it to accurately predict odor intensity for aqueous polymer compositions with multiple VOC types at varying concentrations.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent develops a universal machine learning model that can handle both simple and complex chemical compositions. The decision tree ensemble is designed to process diverse VOC profiles from aqueous polymer systems, making it applicable to a wide range of coating formulations with different chemical compositions and complexity levels.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20260018256A1An odor prediction method for aqueous polymer composition

Publication Date: 2026.01.15 DOW GLOBAL TECHNOLOGIES LLC

AI summary

A method and a system (400) for predicting odor of an aqueous polymer composition, such as a polymerising coating, comprising: analytically characterizing the aqueous polymer composition with a detector (4013), thereby generating concentration data for volatile organic compounds in the aqueous polymer composition from the analytical characterization; inputting the concentration data to a decision tree ensemble configured to predict an odor intensity of the aqueous polymer composition after polymerisation based on the concentration data; and outputting a predicted odor intensity of the aqueous polymer composition from the decision tree ensemble.