Manufacturing Cloud Data Brokering for Secure Multi-Tenant AI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud-based industrial solutions face challenges such as data accessibility across data centers, security issues, and data sharing limitations due to architectural limitations of existing cloud platform architectures, hindering wider implementation.
Innovation Solution
A multi-tenant Software-as-a-Service (SaaS) manufacturing platform with a manufacturing cloud system that collects and stores data from multiple customer entities, tags data with metadata for sharing, and uses AI to infer and render the status of manufacturing processes, while enforcing model-based security and brokering services across geographically distributed facilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If cloud-based computing platforms are used for industrial operations, then operational coordination and data sharing across multiple locations are improved, but data accessibility, security issues, and data sharing limitations arise due to architectural limits
Solution Approach 1:
The patent segments customer data into isolated data silos within the multi-tenant cloud platform. Each customer entity's data is stored in separate data structures with defined boundaries, preventing unauthorized access while allowing controlled sharing. The tagging component further segments data by adding metadata labels that identify data types and sharing permissions, enabling fine-grained access control across tenants.
Solution Approach 2:
The patent introduces a brokering component as an intermediary that mediates data sharing requests between customer entities. This intermediary evaluates sharing requests against defined policies and relationships, acting as a security layer that enables collaboration while maintaining data protection. The tagging component also serves as an intermediary by adding metadata that facilitates secure data exchange without direct access between tenants.
2Loss of information
If data is collected and stored from multiple customer entities in a multi-tenant system, then data sharing and analysis capabilities are improved, but data accessibility issues and architectural limitations occur
Solution Approach 1:
The patent implements a universal data model that can handle multiple customer entities, data types, and sharing scenarios through a single architectural framework. The tagging component provides multi-functional metadata that serves multiple purposes: identifying data types, controlling access permissions, and enabling AI analysis. This universal approach eliminates the need for separate architectural solutions for different data sharing scenarios.
Solution Approach 2:
The patent changes the parameter of data organization by introducing metadata tagging as a new dimension for data structuring. Instead of complex hierarchical storage, the system uses flat data silos with rich metadata parameters that define accessibility, sharing permissions, and data characteristics. This parameter-based organization simplifies the architectural complexity while improving data accessibility across the multi-tenant platform.
3Productivity
If AI is applied to infer manufacturing process status, then production optimization and supply chain management are improved, but data processing requirements and system complexity increase
Solution Approach 1:
The patent applies preliminary action by pre-tagging all industrial data with metadata before AI processing. This preprocessing step categorizes data by type, source, and sharing permissions, creating a structured dataset that is ready for AI analysis. The tagging component performs this preliminary organization continuously as data is collected, so when AI inference is needed, the data is already optimized for processing, reducing computational complexity.
Solution Approach 2:
The patent uses copying by creating standardized data representations with consistent metadata schemas that can be replicated across different customer entities and data sources. This standardized copying approach allows the AI component to process data from multiple sources using the same inference models without requiring complex custom processing for each data source, thereby reducing system complexity while maintaining high productivity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A multi-tenant, cloud-based Software-as-a-Service (SaaS) manufacturing platform offers a variety of industrial applications to end customers - including but not limited to MES, ERP, quality management, supply chain management, and customer relationship management (CRM) - and implements associated architectural features that address a number of issues relating to data sharing, security, scalability, and other concerns.