Surrogate Part Model for Cost-Durability Manufacturing Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional manufacturing processes are unable to optimize the earnings curve of a product without compromising durability, as they lack the capability to favorably alter costs without impacting performance.

Innovation Solution

A system and method that utilize a surrogate model generated through machine learning algorithms, incorporating manufacturing parameters, environmental factors, and customer de-rate, to optimize manufacturing processes for both cost and durability, allowing for strategic deployment of components and improved part longevity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Duration of action of stationary object

If typical manufacturing processes are optimized for maximum durability, then part longevity is improved, but manufacturing cost increases

Engineering Contradiction:
Improvepart longevityVSAvoidmanufacturing cost
Core Design Contradiction:
Duration of action of stationary objectVSEase of manufacture

Solution Approach 1:

The patent applies parameter changes by utilizing machine learning algorithms to analyze historical data and identify optimal manufacturing parameters that balance durability and cost. The system dynamically adjusts manufacturing parameters based on learned patterns from fielded part performance, enabling optimization without compromising longevity

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent creates a digital twin or virtual model of the manufacturing process and part performance through machine learning. This digital copy allows simulation and optimization of manufacturing parameters before actual production, reducing trial-and-error costs while maintaining durability targets

Inventive Principle:
Principle #26Copying

2Ease of manufacture

If manufacturing cost is reduced, then profitability is improved, but part durability deteriorates

Engineering Contradiction:
Improvemanufacturing costVSAvoidpart durability
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where performance data from fielded parts is continuously collected and fed back into the machine learning model. This feedback loop enables the system to learn from actual durability outcomes and adjust manufacturing recommendations to maintain durability while optimizing cost

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary analysis and prediction using machine learning models before manufacturing decisions are made. By predicting which cost-reduction measures will maintain durability thresholds, the system enables proactive optimization rather than reactive corrections

Inventive Principle:
Principle #10Preliminary action

3Device complexity

If traditional manufacturing processes are used, then process simplicity is maintained, but capability to optimize earnings curve is lost

Engineering Contradiction:
Improveprocess complexityVSAvoidearnings curve optimization capability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent replaces traditional mechanical or manual manufacturing process control with machine learning-based intelligent systems. This substitution enables automated analysis of complex data patterns and dynamic optimization of manufacturing parameters, providing earnings curve optimization capability without requiring complex manual intervention

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4167034A1Method and system for optimizing a manufacturing process based on a surrogate model of a part
Publication Date: 2023.04.19 GENERAL ELECTRIC CO
  • EP4167034A1 patent drawingFigure 1
  • EP4167034A1 patent drawingFigure 2
  • EP4167034A1 patent drawingFigure 3

AI summary

There is provided a method for optimizing a manufacturing process of a new part. The method includes executing, by a system configured to drive the manufacturing process, a set of manufacturing functions. Executing these functions include receiving data associated with one or more field parts similar to the new part, and generating, based on the data, a forecast representative of a longevity of the one or more parts. The method further includes generating a digital thread forming a surrogate model for the new part, based on the forecast. Further, the method includes creating the set of manufacturing functions based on the surrogate model and manufacturing the new part according to the set of manufacturing functions.