Manufacturing Data Analysis with Adaptive Sampling Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the manufacturing industry, identifying the cause of product anomalies is challenging due to the vast amount of manufacturing data generated, particularly with the rise of IoT technology, making it difficult for manual monitoring and analysis, especially when the number of data samples is small, leading to unreliable analysis results.
Innovation Solution
A manufacturing data analysis device that acquires initial manufacturing data under specified conditions, determines alternative acquisition conditions to reduce bias, and analyzes the relationship between manufacturing condition data and quality data to improve the reliability of analysis results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the user specifies a narrow acquisition condition for manufacturing data, then the analysis focuses on specific conditions, but the number of data samples becomes small leading to unreliable analysis results
Solution Approach 1:
The system dynamically adjusts the acquisition condition based on the initial analysis results. When the number of data samples is insufficient for reliable analysis, the acquisition condition is automatically expanded to include broader ranges of manufacturing conditions, thereby increasing the data sample size while maintaining analysis relevance
Solution Approach 2:
The system changes the parameters of the acquisition condition (such as time period, manufacturing parameters ranges) based on the initial analysis. By modifying these parameters to expand the data collection scope, the system ensures sufficient data samples are obtained for statistically reliable analysis results
2Quantity of substance
If the user specifies a broad acquisition condition for manufacturing data, then the number of data samples increases, but the analysis result becomes less reliable due to increased bias
Solution Approach 1:
The system dynamically optimizes the acquisition condition by initially using a broad condition to gather sufficient data samples, then adjusting the condition based on analysis results to reduce bias. This dynamic adjustment ensures both adequate sample size and analysis reliability
Solution Approach 2:
The system uses feedback from the initial analysis results to refine the acquisition condition. By evaluating the quality and distribution of initial data samples, the system adjusts the acquisition condition to reduce bias while maintaining sufficient data volume for reliable analysis
3Ease of operation
If manual monitoring of manufacturing data is performed, then data can be analyzed, but it becomes difficult and time-consuming with the large volume of data generated by IoT devices
Solution Approach 1:
The system performs self-service by automatically acquiring manufacturing data, determining acquisition conditions, and conducting analysis without requiring continuous manual intervention. The system autonomously handles the entire workflow from data collection to analysis result generation
Solution Approach 2:
The system replaces manual mechanical monitoring with automated computational analysis. By using computers and algorithms to process manufacturing data automatically, the system eliminates the time-consuming nature of manual data monitoring and analysis
Data Source
AI summary
According to one embodiment, a manufacturing data analysis device includes processing circuitry. The processing circuitry acquires, from manufacturing data related to a plurality of products, first manufacturing data under a first acquisition condition. The first manufacturing data includes manufacturing condition data related to a manufacturing condition for each of the products and quality data related to quality for each of the products. The processing circuitry determines a second acquisition condition different from the first acquisition condition based on the first manufacturing data. The processing circuitry acquires second manufacturing data including the manufacturing condition data and the quality data from the manufacturing data under the second acquisition condition. The processing circuitry calculates an analysis result of a relationship between the manufacturing condition data and the quality data by analyzing the second manufacturing data. The processing circuitry generates output data including the analysis result.