Predictive Maintenance Data Negotiation With Policy-Based Anonymization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current predictive maintenance solutions face issues of data imbalance and lack of control over information exchange between companies, leading to labor-expensive and error-prone manual setups, with limited control over the amount and quality of shared data.
Innovation Solution
A method and system for anonymization and negotiation in predictive maintenance, utilizing an aggregation service module to generate aggregated data, an anonymization module to anonymize data using user-definable policies, and a transmitting module to send anonymized data to a condition prediction service, with dynamic adjustment of policies for optimal data trade-offs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual project-specific setup is used to define information exchange, then control over data exchange can be achieved, but labor cost increases and error-prone processes occur
Solution Approach 1:
The system enables automated self-configuration of data exchange parameters between companies. The negotiation mechanism allows systems to automatically agree on data formats, quality requirements, and exchange protocols without manual intervention, while still maintaining controlled access through predefined authorization frameworks.
Solution Approach 2:
The patent transforms fixed manual configuration into dynamic parameter negotiation. Data exchange parameters such as quality thresholds, formats, and authorization levels are changed from static manually-set values to dynamically negotiated parameters that adapt to specific project requirements automatically.
2Reliability
If unlimited access to operational data is granted to device manufacturer, then predictive maintenance quality improves, but data security and commercial secret protection deteriorate
Solution Approach 1:
The system applies different authorization levels to different data elements. Sensitive operational data receives higher protection levels while less sensitive maintenance-relevant data allows broader access. This enables the device manufacturer to access sufficient data for accurate predictions without exposing critical commercial secrets.
Solution Approach 2:
An authorized intermediary system manages data access between the operator and device manufacturer. This intermediary enforces authorization policies, filters sensitive information, and provides controlled access to maintenance data, ensuring both prediction quality and data security.
3Adaptability or versatility
If device knowledge is shared across multiple operators, then maintenance expertise improves, but competitive advantage is lost
Solution Approach 1:
Device knowledge is segmented into different authorization levels and access categories. General maintenance knowledge can be shared across operators, while proprietary device-specific knowledge remains restricted to authorized personnel only. This segmentation enables collaborative maintenance improvement without losing competitive advantages.
4Adaptability or versatility
If third-party systems extract and pass data outside infrastructure, then data exchange flexibility improves, but control over information flows is lost
Solution Approach 1:
The system implements feedback mechanisms that monitor and report data extraction and transmission activities. When third-party systems access data, the authorization service receives feedback about the data flow and can enforce policies, log activities, and control information exchange to maintain security while allowing flexibility.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method for anonymization and negotiation for predictive maintenance, the method comprising: Aggregating (S11), by an aggregation service module (110), runtime data and/or context data using an aggregation policy to generate aggregated data; Anonymizating (S12), by an anonymization module (120), the aggregated data using a user-definable internal anonymization policy to generate anonymized data; and Sending (S13), by a transmitting module (130), the anonymized data to a condition prediction service module (210).