AI Engineering Framework for Automation Code Reuse

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The application of artificial intelligence (AI) in the engineering phase of industrial automation is underdeveloped due to issues such as scarce engineering data, short duration of the engineering phase, and difficulty in capturing human intent and knowledge, leading to longer development cycles and non-reusable codes.

Innovation Solution

The implementation of machine learning techniques, including code classification, semantic code search, and hardware recommendations, to automate engineering tasks by organizing and predicting code and hardware configurations, utilizing neural networks and data curation to improve productivity and code reusability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If machine learning techniques are applied to automate engineering tasks, then productivity is improved, but device complexity increases

Engineering Contradiction:
Improveengineering task automationVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces machine learning models as intermediary components between human engineers and automation systems. These models learn from historical engineering data and automatically generate code, hardware configurations, and system designs, acting as a mediator that translates engineering requirements into implementable solutions without requiring deep expert knowledge from users.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system creates reusable code templates and configurations by learning from past engineering projects. Once a solution is developed for a particular automation task, the machine learning model captures its structure and logic, then automatically generates similar solutions for future projects with analogous requirements, eliminating the need to recreate solutions from scratch.

Inventive Principle:
Principle #26Copying

2Loss of time

If code reusability is improved through machine learning, then loss of time is reduced, but manufacturing precision may worsen due to automated code generation

Engineering Contradiction:
Improvedevelopment cycle timeVSAvoidcode accuracy
Core Design Contradiction:
Loss of timeVSManufacturing precision

Solution Approach 1:

The machine learning model incorporates feedback mechanisms where generated code is automatically tested against validation criteria and historical performance data. The system learns from successful and unsuccessful code generations, continuously refining its output quality. User corrections and modifications to generated code are fed back into the training data, improving future generation accuracy while maintaining reusability.

Inventive Principle:
Principle #23Feedback

3Productivity

If hardware configuration accuracy is improved through machine learning, then productivity is improved, but difficulty of detecting and measuring increases

Engineering Contradiction:
Improvehardware configuration speedVSAvoidconfiguration verification
Core Design Contradiction:
ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The machine learning model performs preliminary validation and verification of hardware configurations before final deployment. It checks compatibility between selected components, validates configuration syntax, and ensures adherence to system requirements upfront, preventing errors from propagating to later stages and reducing the need for manual debugging.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3997530B1Automation engineering learning framework for cognitive engineering
Publication Date: 2023.11.08 SIEMENS AG
  • EP3997530B1 patent drawingFigure 1
  • EP3997530B1 patent drawingFigure 2
  • EP3997530B1 patent drawingFigure 3

AI summary

Applications of artificial intelligence (AI) in industrial automation have focused mainly on the runtime phase due to the availability of large volumes of data from sensors. Methods, systems, and apparatus that can use machine learning or artificial intelligence (AI) to complete automation engineering tasks are described herein.