Learned Model for Injection Molding Setting Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing technologies for injection molding require repetitive adjustments of settings to achieve optimal product quality, consuming time and labor, and involve cumbersome processes for deriving settings to improve quality.

Innovation Solution

A learned model is developed that includes an input layer, intermediate layers, and an output layer, capable of performing machine learning based on detected results and measurement data from injection molding, to output setting information for improving product quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If repetitive adjustments of settings are made to achieve optimal product quality, then product quality is improved, but time and labor consumption increase

Engineering Contradiction:

Improveproduct qualityVSAvoidsetting adjustment time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent replaces the mechanical trial-and-error adjustment process with an AI-based system. The learning model processes input data (detection results, measurement values) through neural network layers to automatically output optimized setting values, eliminating the need for manual repetitive adjustments while maintaining high manufacturing precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the learning model to autonomously determine optimal settings without human intervention. The model learns from training data and automatically generates setting recommendations, making the system self-sufficient in the setting adjustment process and significantly reducing time consumption.

Inventive Principle:

Principle #25Self-service

2Manufacturing precision

If repetitive adjustments of settings are made to achieve optimal product quality, then product quality is improved, but labor consumption increases

Engineering Contradiction:

Improveproduct qualityVSAvoidlabor requirement

Core Design Contradiction:

Manufacturing precisionVSEase of operation

Solution Approach 1:

The patent replaces manual labor with an automated AI system. The learning model, implemented through software running on a computer, performs the setting optimization function that would otherwise require skilled technicians to perform through repeated trial adjustments, thereby reducing labor consumption while maintaining expertise in quality optimization.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system eliminates the need for human operators to perform repetitive setting adjustments. The learning model autonomously processes input data, learns from training examples, and generates optimized settings without human intervention, significantly reducing labor requirements while maintaining high product quality standards.

Inventive Principle:

Principle #25Self-service

3Manufacturing precision

If complex processes are used to derive settings for improving quality, then manufacturing precision is improved, but device complexity increases

Engineering Contradiction:

Improvequality improvementVSAvoidprocess complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent replaces complex manual derivation processes with a standardized AI learning model. Instead of requiring technicians to navigate complex decision trees or use multiple calculation methods, the system uses a single integrated neural network that automatically determines optimal settings based on input data, simplifying the overall process while maintaining manufacturing precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The learning model serves multiple functions within a single system: it processes various input data types (detection results, measurement values), learns from training data, and generates optimized settings. This multi-functional approach consolidates what would otherwise require multiple separate processes and tools into one unified system, reducing overall complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250117554A1Learned model, management apparatus for injection molding machine, and method of generating training data

Publication Date: 2025.04.10 SUMITOMO HEAVY IND LTD

AI summary

A learned model includes an input layer; intermediate layers connected to the input layer; and an output layer connected to the intermediate layer. The learned model causes a computer to function to perform machine learning based on first data and ground truth information, the first data indicating a detection result or a measurement result of the molding product, for each first value set or detected with respect to a predetermined item for producing the molding product by an injection molding machine, and the ground truth information indicating a setting of the predetermined item derived based on evaluation information of the molding product when the first value is set or detected, and output, from the output layer, information relating to a setting of the predetermined item, when second data is input from the input layer, the second data indicating a detection result or a measurement result of the molding product.