Substrate Processing Abnormality Scoring for Inspection Priority
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In a scenario where multiple substrate processing apparatuses are installed in various locations worldwide, maintenance workers struggle to identify the cause of abnormalities in these apparatuses. This leads to delays in inspection and maintenance, as administrators must prioritize inspections based on the severity of abnormalities, which can be time-consuming due to the complexity of analyzing past time-series data and interviewing maintenance workers.
Innovation Solution
A substrate processing apparatus management system that includes an information collector to gather processing information from each apparatus, a score calculator to determine an abnormality score based on this information and invariant relationships, a request acceptor to handle inspection requests, a score acquirer to compile scores from multiple apparatuses, and a priority order presenter to quickly determine the inspection priority order based on these scores.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If administrators perform primary inspection including analysis of past time-series data and interviews with maintenance workers to determine inspection priority, then the inspection priority can be appropriately determined, but the time required to start actual inspection becomes long
Solution Approach 1:
The system performs preliminary calculation of abnormality scores based on operation information before inspection requests are made. The score calculator continuously computes abnormality scores using invariant relationships between multiple types of operation information, so when inspection requests arrive, the priority ordering is already prepared, eliminating the need for time-consuming primary inspections at the moment of request
Solution Approach 2:
The manual process of primary inspection including data analysis and interviews is replaced by an automated information analysis device that calculates abnormality scores based on predetermined invariant relationships. This substitution of automated computation for manual inspection processes dramatically reduces the time required while maintaining determination accuracy
2Productivity
If administrators inspect multiple substrate processing apparatuses in parallel when multiple inspection requests are received, then the response efficiency improves, but the complexity of determining inspection priority order increases
Solution Approach 1:
The system transforms the complex qualitative assessment of abnormality severity into a quantitative abnormality score parameter. By changing the parameter representation from descriptive analysis to numerical scoring based on invariant relationships, the system enables straightforward comparison and priority ordering of multiple inspection requests without increasing administrative complexity
3Reliability
If maintenance workers request inspection for substrate processing apparatus with abnormalities, then the abnormality can be inspected in detail, but the time to identify cause and request inspection increases
Solution Approach 1:
The system provides continuous feedback to maintenance workers in the form of abnormality scores calculated from operation information. This feedback mechanism enables maintenance workers to quickly identify apparatus requiring inspection without manual analysis, reducing the time to request inspection while maintaining reliable abnormality detection through the automated scoring system
Data Source
AI summary
A plurality of processing information pieces representing operations or states relating to a process for a substrate are collected from each of a plurality of substrate processing apparatuses. In regard to each substrate processing apparatus, a degree of abnormality of the substrate processing apparatus is calculated as an abnormality score based on a plurality of processing information pieces collected from the substrate processing apparatus and invariant relationships in regard to a plurality of processing information pieces. A management device accepts an inspection request in regard to an abnormality transmitted from each substrate processing apparatus. When two or more inspection requests are accepted, the abnormality scores of the substrate processing apparatuses corresponding to these inspection requests are acquired. Based on the plurality of acquired abnormality scores, priority order information relating to the priority order for two or more inspection requests to be responded is presented to an administrator.