API Wrapper for Scalable Analytic Model Deployment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional systems for analyzing and predicting the operation of real-world physical systems, such as electro-mechanical systems, face inefficiencies and errors due to the time-consuming process of re-writing and debugging models when transferred from a local computer to another system, and managing complex industrial-grade systems with numerous dependencies is challenging.

Innovation Solution

A system and method that includes an analytic model, an execution platform, and an API wrapper for each, allowing for the transmission and deployment of models between platforms with minimal setup time, enabling self-aware and self-registering components that can run independently and seamlessly integrate across systems, facilitating rapid development and deployment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a model is transmitted to another computer for execution, then the model can be run on a different system, but re-writing and debugging the model program is required which is time consuming and error-prone

Engineering Contradiction:
Improvemodel portabilityVSAvoiddeployment time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent creates a digital twin - a virtual copy of the physical system that includes replicated geometry, materials, and physics properties. This digital model can be transmitted and executed on different computers without modification, eliminating the need for re-writing and debugging. The digital twin serves as a portable replica that maintains fidelity across different execution environments.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent transforms the physical system into a digital representation by changing the parameters from physical attributes to computational parameters. The geometry, materials, and physics are converted into data structures and mathematical models that can be executed virtually. This parameter transformation enables the model to run on different systems without requiring program rewriting.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If conventional methods are used to manage complex industrial-grade systems, then system functionality is maintained, but the complexity of managing numerous dependencies makes the system challenging to manage

Engineering Contradiction:
Improvesystem functionalityVSAvoidsystem management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the complex industrial system into distinct digital components including geometry models, material properties, physics definitions, and boundary conditions. Each segment can be independently managed, modified, and validated. This segmentation reduces management complexity by breaking down the monolithic system into manageable modules while maintaining overall system functionality through their integrated simulation.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS10628145B2Scalable and secure analytic model integration and deployment platform
Publication Date: 2020.04.21 INNOVATEPRO MANAGEMENT USA LLC
  • US10628145B2 patent drawing
  • US10628145B2 patent drawing
  • US10628145B2 patent drawing

AI summary

According to some embodiments, system, apparatus and methods are provided comprising an analytic model for an installed product; an execution platform configured to execute the analytic model; an application programming interface (API) wrapper associated with each of the analytic model and the execution platform, the API wrapper including input information, output information and a technique; and a storage in communication with the analytic model and the execution platform and storing program instructions to perform the functions as follows: transmitting information between the analytic API wrapper and the execution platform API wrapper; and deploying the analytic model to the execution platform based on the transmitted information. Numerous other aspects are provided.