Industrial Plant GUI Generation Using Natural Language Prompts

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The complexity of operating industrial plants is hindered by the need for extensive manual study and engineering skills to generate tailored graphical user interfaces, leading to inefficient production processes and unnecessary downtimes.

Innovation Solution

A method utilizing a pre-trained and fine-tuned generative large language model to generate a graphical user interface based on operator input, allowing for machine-guided operation of the industrial plant.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional engineering systems and manual interface generation are used, then graphical user interfaces can be created, but the process requires extensive manual study and engineering skills, leading to increased complexity and time consumption

Engineering Contradiction:

Improveease of generating graphical user interfaceVSAvoidcomplexity of engineering system

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent replaces manual mechanical engineering processes with an AI-based natural language processing system. Operators can generate graphical user interfaces by simply typing natural language descriptions, eliminating the need for manual configuration, engineering skills, and complex system interactions. The AI model automatically translates textual requirements into functional interfaces, substituting the traditional mechanical engineering workflow.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service interface generation where the AI model autonomously creates graphical user interfaces based on operator input without requiring manual engineering intervention. The model independently processes natural language requirements, generates appropriate interface configurations, and implements the interface, allowing operators to serve themselves without needing specialized engineering knowledge or manual system configuration.

Inventive Principle:

Principle #25Self-service

2Adaptability or versatility

If manual engineering and interface customization are required, then tailored graphical user interfaces can be obtained, but the process is cumbersome and leads to unnecessary downtimes and less-optimum production output

Engineering Contradiction:

Improveadaptability of graphical user interfaceVSAvoidproduction output

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by pre-training the AI model on extensive engineering knowledge, interface design patterns, and plant operation data before actual use. This preliminary training enables the model to immediately generate adaptive interfaces without requiring manual engineering work during production operations, thus maintaining both adaptability and productivity without downtime.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent substitutes manual interface customization processes with automated AI-generated interfaces. The system rapidly adapts to different operational requirements by processing natural language descriptions and generating appropriate interfaces in real-time, eliminating the cumbersome manual engineering process that caused production delays while maintaining full adaptability to specific technical tasks.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If operators study manuals and manually engineer interfaces, then they can operate the industrial plant, but the process is time-consuming and not carried out in an optimal manner

Engineering Contradiction:

Improveoperability of industrial plantVSAvoidtime for studying manuals and engineering

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The patent replaces manual study and engineering processes with an AI system that automatically generates operational interfaces through natural language processing. Operators no longer need to study extensive manuals or perform manual engineering tasks; instead, they simply describe their needs in natural language, and the AI model generates the appropriate interfaces, dramatically reducing the time required while maintaining full operability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system provides self-service operational capabilities where the AI model automatically generates and configures interfaces based on operator input without requiring manual engineering intervention. This eliminates the time-consuming manual processes while ensuring operators can effectively operate the plant through intuitively generated interfaces tailored to their specific needs.

Inventive Principle:

Principle #25Self-service

Data Source

PatentEP4614258A1Method of and device for operating an industrial plant

Publication Date: 2025.09.10 SIEMENS AG

Figure 1~2
Figure 3~4
Figure 5~6

AI summary

A method of operating an industrial plant (2) comprises: obtaining (S1) an entry (82) of a topic of interest in natural language from an operator of the industrial plant (2); generating (S2) a prompt for a pre-trained and fine-tuned generative large language model (61) based on the obtained entry (82); inputting (S3) the generated prompt to the generative large language model (6) so as to obtain, from the generative large language model (61), a specification of a graphical user interface (9) related to the entered topic of interest; rendering (S4) the graphical user interface (9) based on the obtained specification and based on current data obtained from the industrial plant (2); and operating (S5) the industrial plant based on an interaction of the operator with the rendered graphical user interface. Machine-guided operation of the industrial plant is enabled.