Machine Learning Formula Suggestion System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing product development processes, particularly in industries like cosmetics and food, are laborious, costly, and time-consuming due to the need for extensive research, multiple trials, and compliance checks, often requiring skilled resources and involving lengthy approval processes, which can be challenging for small and medium enterprises.
Innovation Solution
A processor-implemented method using machine learning models to automatically suggest formulas for product categories by obtaining user preferences and desired functions, suggesting ingredients, and refining concentrations, thereby reducing the need for physical trials and accelerating the development process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional product development processes are used with extensive research, multiple trials, and compliance checks, then product safety and compliance are ensured, but development time and cost increase significantly
Solution Approach 1:
The patent applies preliminary action by conducting virtual prototyping and simulations before physical product development. Machine learning models predict ingredient interactions, stability, and safety outcomes in advance, allowing developers to identify and correct issues before committing to physical trials. This preliminary computational assessment ensures safety and compliance requirements are met while reducing the need for extensive physical testing.
Solution Approach 2:
The patent uses copying by creating virtual replicas of physical products and testing environments through digital twins and simulation models. These virtual copies allow comprehensive testing of product formulations, ingredient combinations, and compliance scenarios without consuming physical materials or requiring actual product manufacturing. The virtual prototypes are then refined iteratively before physical production begins.
2Manufacturing precision
If traditional product development processes are used with multiple trials and extensive testing, then product performance and quality are optimized, but development cost increases significantly
Solution Approach 1:
The patent replaces mechanical and physical testing systems with computational and machine learning-based virtual testing systems. Instead of conducting numerous physical trials to optimize product quality, the system uses AI models trained on historical data to predict optimal formulations, ingredient concentrations, and product performance characteristics. This substitution dramatically reduces material consumption, laboratory resources, and overall development costs while maintaining or improving product quality outcomes.
3Reliability
If skilled resources and multiple department approvals are involved in product development, then comprehensive expertise is applied, but process complexity and communication overhead increase
Solution Approach 1:
The patent applies universality by creating an integrated digital platform that consolidates multiple specialized functions into a single system. The machine learning framework simultaneously performs ingredient compatibility analysis, stability prediction, safety assessment, compliance verification, and optimization recommendations. This multi-functional system replaces the need for separate expert reviews and sequential approval processes from multiple departments, reducing process complexity while maintaining comprehensive expertise application.
4Measurement precision
If physical trials are conducted extensively to validate product formulas, then formula accuracy and performance are confirmed, but time and material resources are consumed
Solution Approach 1:
The patent applies preliminary action by using machine learning models to predict formula accuracy and performance characteristics before physical trials. The system analyzes historical data, ingredient properties, and interaction patterns to forecast optimal formulations with high confidence. This preliminary computational validation identifies the most promising formulas that are then tested physically, dramatically reducing the number of physical trials needed and consequently reducing material consumption while maintaining formula accuracy.
Data Source
AI summary
The embodiments herein relate to a system for automatically suggesting a formula for a product category. The system includes a user device, and a formula suggesting server. The formula suggesting server is configured to (i) obtain, by the user device, a primary formula and a desired function from a user or a first machine learning model (ii) suggest, by a second machine learning model, a list of related ingredients that have a causal relationship with the primary formula, (iv) processing a selection, of a second ingredient from the list of related ingredients by the user, to obtain a secondary formula that includes either a first ingredient or one or more ingredients, along with the second ingredient, (v) suggesting, using a third machine learning model, a concentration for the secondary formula that performs the desired function for the product category.