Secure Model Integration for Confidential Predictive Maintenance

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing predictive maintenance models for semiconductor manufacturing devices lack accuracy due to limitations in sharing confidential parameters across multiple manufacturers, preventing the creation of more accurate models through collective analysis.

Innovation Solution

A secure computing system that receives and integrates concealed parameters from multiple semiconductor manufacturers using secure computation, outputting the results in a concealed form to maintain confidentiality while enhancing predictive maintenance accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If models from multiple semiconductor manufacturers are integrated to improve predictive maintenance accuracy, then the accuracy of the predicted data is improved, but the confidentiality of the manufacturers' parameters is compromised

Engineering Contradiction:

Improveaccuracy of predicted dataVSAvoidconfidentiality of model parameters

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

A secure computing system acts as an intermediary between multiple semiconductor manufacturers, enabling the integration of their predictive maintenance models without exposing their confidential parameters. The secure computing system processes and integrates the models while maintaining parameter confidentiality, thus resolving the contradiction between improving prediction accuracy through model integration and protecting manufacturers' proprietary information.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Loss of information

If a single prediction model is used for predictive maintenance, then the confidentiality of manufacturer parameters is maintained, but the accuracy of the predicted data is limited

Engineering Contradiction:

Improveconfidentiality of model parametersVSAvoidaccuracy of predicted data

Core Design Contradiction:

Loss of informationVSMeasurement precision

Solution Approach 1:

The secure computing system serves as a trusted intermediary that enables multiple manufacturers to contribute their models to a collective predictive maintenance system without directly sharing confidential parameters. This intermediary approach allows accuracy improvement through model integration while maintaining parameter confidentiality.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

Instead of sharing original confidential model parameters, the system uses copies or representations of the models that can be integrated for prediction purposes without revealing the underlying proprietary information. This allows multiple models to be combined while protecting the source confidentiality.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20240353832A1Secure computing system, information processing system, secure computing method, and recording medium

Publication Date: 2024.10.24 NEC CORP

AI summary

A secure computing system according to the present disclosure is used in an analysis pertaining to predictive maintenance of a semiconductor manufacturing device, and comprises: a parameter reception means that receives concealed parameters of a plurality of models generated by each of a plurality of semiconductor manufacturers; a secure computing means that integrates the plurality of concealed parameters by secure computing; and an output means that outputs, in a concealed format, the parameters integrated by the secure computing means.