AI Classification of Manufacturing Parameters for Corrective Action
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual categorization of manufacturing system parameters and components by subject matter experts is time-consuming and expensive, and existing methods fail to efficiently identify sensors and parameters well correlated with substrate properties, leading to inefficiencies and increased costs in manufacturing processes.
Innovation Solution
Utilize prompt engineering with a trained machine learning model, such as a natural language processing model, to categorize manufacturing system parameters and components into relevant categories, including criticality and subsystem assignment, and perform corrective actions based on these categorizations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual categorization by subject matter experts is used, then categorization accuracy is maintained, but time consumption and cost increase significantly
Solution Approach 1:
The patent replaces the manual mechanical categorization process performed by subject matter experts with an automated machine learning model. The model processes parameter names and descriptions to generate categorizations automatically, eliminating the need for human experts to manually review and categorize each parameter while maintaining high accuracy through trained algorithms.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between the raw manufacturing parameters and the final categorization output. This intermediary automatically processes and categorizes parameters based on learned patterns from training data, serving as a bridge that eliminates the need for direct human intervention in the categorization process.
2Measurement precision
If manual categorization by subject matter experts is used, then categorization quality is maintained, but manufacturing cost increases
Solution Approach 1:
The patent replaces expensive manual expert categorization with an automated machine learning system. Once the model is trained, it can categorize parameters at minimal computational cost, eliminating the recurring labor costs associated with hiring and paying subject matter experts for each categorization task.
Solution Approach 2:
The patent performs preliminary training of the machine learning model using historical categorization data and expert knowledge before deployment. This preliminary action captures expert wisdom in the model's parameters, allowing the system to perform high-quality categorizations automatically without requiring ongoing expert involvement, thereby reducing long-term costs.
3Quantity of substance
If existing methods are used to identify sensors and parameters, then comprehensive coverage is achieved, but efficiency decreases due to inability to identify well-correlated items
Solution Approach 1:
The patent changes the approach from exhaustive manual review of all parameters to targeted identification using the machine learning model. The model analyzes parameter characteristics and identifies those most relevant to substrate properties by comparing against training data, efficiently filtering and prioritizing parameters without needing to manually examine every single parameter in detail.
Solution Approach 2:
The patent replaces inefficient manual parameter identification with automated machine learning-based identification. The model quickly processes and evaluates parameters to identify those well-correlated with substrate properties, dramatically improving identification efficiency while maintaining comprehensive coverage of relevant parameters.
Data Source
AI summary
A method includes generating or receiving an input for an AI model. The input includes a description of a set of categories to which a parameter or component of a manufacturing system may belong. The input further includes a number of examples each including a parameter or component name and an indication of which category of the set of categories the parameter or component belongs to. The input further includes a name of a target parameter or component to be categorized according to the set of categories.The method further includes processing the input using the AI model to generate an output including a category associated with the target parameter or component. The method further includes performing a corrective action in view of the category associated with the target parameter or component.