Production Setup Using ML Models for Faster Stable Parameter Tuning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for setting up production facilities are complex, require high accuracy in measurement data, and are sensitive to process fluctuations, leading to high computational demands and inefficiencies.
Innovation Solution
A method utilizing machine learning (ML) models, specifically self-organizing maps (SOM) and modified self-organizing maps (mod-SOM), to automate or semi-automate the determination of setting parameters for production facilities, optimizing processes such as solder paste printing and injection molding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional control algorithms are used for setting up production facilities, then measurement accuracy and reliability are improved, but device complexity and computational requirements increase
Solution Approach 1:
The patent replaces traditional mechanical/control-based parameter determination with a machine learning model that processes measurement data to automatically determine optimal setting parameters. This substitution eliminates complex control algorithms while maintaining or improving measurement accuracy through data-driven insights.
Solution Approach 2:
The machine learning model enables the production facility to self-adjust setting parameters based on measurement data without requiring complex external control systems. The system serves itself by automatically learning from data and determining optimal parameters, reducing algorithmic complexity.
2Stability of the object's composition
If traditional control algorithms are used for setting up production facilities, then process stability is improved, but productivity and setup time increase
Solution Approach 1:
The machine learning model is trained in advance on historical measurement data and setting parameters, so that during actual production setup, it can quickly determine optimal parameters without requiring time-consuming traditional control algorithm execution. This preliminary training action enables fast, stable parameter determination.
Solution Approach 2:
The patent replaces time-consuming traditional control algorithms with a trained machine learning model that processes measurement data much faster, thereby increasing setup speed while maintaining process stability through consistent, data-driven parameter selection.
3Ease of operation
If manual determination of setting parameters is used, then ease of operation is improved, but loss of time and productivity increase
Solution Approach 1:
The system performs self-service by automatically determining setting parameters through the machine learning model based on measurement data, eliminating the need for manual operator intervention. This automation maintains operational simplicity while dramatically reducing setup time and increasing productivity.
Solution Approach 2:
The machine learning model is pre-trained on historical data, enabling it to quickly and accurately determine setting parameters during operation without requiring manual analysis or adjustment by operators, thus reducing setup time while maintaining ease of use.
4Manufacturing precision
If traditional setup methods are used, then manufacturing precision is maintained, but productivity and production capacity decrease
Solution Approach 1:
The patent replaces traditional manual or algorithm-based setup methods with a machine learning model that determines optimal setting parameters faster and more consistently, thereby maintaining manufacturing precision while significantly increasing production capacity and throughput.
Solution Approach 2:
The machine learning model learns from historical production data in advance, enabling it to quickly determine optimal setting parameters for each production run. This preliminary learning maintains product quality while reducing setup time and increasing overall production capacity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to a method for setting up a production facility (110, 120), wherein a setting parameter set (620) for the manufacture of a product by the production facility (110, 120) and a measurement parameter set (610) relating to the product manufactured by the production facility (110, 120) using the setting parameter set (620) and/or relating to the manufacture of this product are present, characterized in that the method comprises the following process steps: a.) Training an ML model (220, 320) using training data comprising the setting parameter set (620) and/or the measurement parameter set (610), b.) Automated or semi-automated determination of a further setting parameter set (620) for the production facility (110, 120) using the ML model (220, 320).