IoT Device Onboarding via Metadata-Driven Meta-Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for integrating IoT devices into enterprise processes are inefficient, relying on legacy systems that lack real-time data connectivity and require manual onboarding, which becomes impractical for large numbers of devices, and do not support holistic integration scenarios like predictive maintenance.
Innovation Solution
The implementation of a meta-model that supports the modeling of real-world IoT entities, allowing for automated onboarding and integration of IoT devices through a metadata-driven approach, enabling the creation of object definitions and process variant configurations that model real-world processes and interactions, stored in a metadata repository.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If legacy systems are used for IoT device integration, then existing infrastructure can be maintained, but manual onboarding becomes impractical for large numbers of devices and real-time data connectivity is lost
Solution Approach 1:
The patent uses digital twins (virtual copies) of physical IoT devices to enable automated onboarding. Instead of manually configuring each physical device, the system creates and manages digital representations that can be automatically instantiated, configured, and integrated into enterprise processes, thereby dramatically improving onboarding efficiency for large numbers of devices
Solution Approach 2:
The patent introduces a metadata repository and object model as intermediaries between physical IoT devices and enterprise systems. This metadata layer enables automated translation and integration, allowing devices to be onboarded through standardized metadata definitions rather than manual device-by-device configuration, thus enabling automation at scale
2Productivity
If manual onboarding processes are used, then system infrastructure remains simple, but the process becomes impractical for thousands or millions of devices
Solution Approach 1:
The patent implements a universal object model and metadata repository that can handle multiple types of IoT devices through a single standardized framework. This universal approach allows the same system infrastructure to onboard diverse devices (sensors, actuators, smart devices) without requiring device-specific manual configuration processes, enabling scalable integration across thousands of device types
Solution Approach 2:
The patent uses configurable metadata parameters and variant types to adapt the standardized onboarding process to different device specificities. By changing metadata parameters rather than restructuring the entire system, the patent maintains simple core infrastructure while accommodating device diversity, avoiding complexity escalation
3Reliability
If legacy document-based interfaces are used, then existing user interfaces can be maintained, but real-time data connectivity and automated process triggering are lost
Solution Approach 1:
The patent replaces manual document-based interfaces with automated event-driven architecture. Instead of users manually creating service orders through document interfaces, the system automatically triggers processes based on real-time device events and metadata definitions, substituting mechanical manual operations with automated electronic processes that enable real-time integration
4Ease of operation
If standardized object models are used, then key users can adjust business models without coding, but flexibility for custom device-specific implementations may be reduced
Solution Approach 1:
The patent implements a dynamic metadata model where object definitions and process configurations can be modified at runtime without requiring system reconfiguration or coding. Key users can adjust business models by changing metadata parameters, and the system dynamically adapts to these changes while maintaining device-specific functionalities through variant types, thus providing both ease of operation and flexibility
Data Source
AI summary
The disclosure generally describes methods, software, and systems, including a method for using an object definition from which object documents are instantiated defining real-world variants of a physical object and including a meta-model identifying nodes, fields, and associations with other object definitions. The object definition includes, at instantiation of a given object document marking an existence of the physical object, an object identifier node, including an object identifier, and at least one variant type node, including a variant type identifier identifying the real-world variant of the physical object and including variant extension nodes and fields extending the object definition to a variant object definition. For each variant type node, process variant configuration information is received that models real-world processes for the real-world variant of the physical object, including process and data interactions between the real-world variant of the physical object and the other physical objects.