Setup Time Anomaly Detection Using Probability Density Distributions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In modern manufacturing environments, where a variety of products are produced with fewer expertized workers, setup times exhibit significant variations, making it challenging to accurately identify and reduce abnormal setup times, which are a major contributor to increased process lead times and decreased productivity.
Innovation Solution
An information processing method and device that calculate the probability density distribution of setup times based on production performance data, determining whether these times are abnormal and identifying the cause of anomalies, thereby enabling targeted improvements in setup processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If setup times are reduced through conventional methods, then productivity improves, but measurement precision of abnormal setup times deteriorates due to significant variations in setup times
Solution Approach 1:
The patent changes the parameter representation from raw setup time values to probability density distributions. By calculating the probability density distribution of setup times and comparing it with the distribution of standard setup times, the system can identify abnormalities even when setup times vary significantly. This parameter transformation enables accurate anomaly detection despite the inherent variability in setup processes.
2Adaptability or versatility
If expertized workers are reduced to increase adaptability, then versatility improves, but setup time stability deteriorates due to significant variations in setup times
Solution Approach 1:
The patent implements a feedback mechanism by calculating the probability density distribution of actual setup times and comparing it with the distribution of standard setup times. The system provides feedback on whether the setup time is abnormal based on this comparison, enabling continuous improvement of setup processes even with non-expertized workers. The feedback loop allows for identifying and correcting deviations from standard setup patterns.
3Speed
If conventional anomaly detection methods are used, then detection speed improves, but measurement precision deteriorates due to inability to accurately identify abnormal setup times
Solution Approach 1:
The patent transforms the detection approach by changing from direct anomaly detection to probability density distribution comparison. By calculating and comparing probability density distributions, the system achieves both accurate identification of abnormal setup times and maintains efficient detection speed. The method leverages statistical properties to enable precise anomaly detection without requiring complex analysis of individual setup time values.
Data Source
AI summary
An information processing method includes: calculating a probability density distribution of a work time that is at least part of a setup time, based on production performance data read from storage, the setup time being time taken for a setup work that is performed between lots; determining whether the work time is anormal, based on the probability density distribution; and outputting a result of the determining.