LLM Service Workflow for Industrial Automation Troubleshooting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Service technicians face cumbersome and time-consuming processes when troubleshooting industrial automation systems, as they need to toggle between multiple sources of information to address problems, such as product manuals and troubleshooting guides.

Innovation Solution

An LLM-based service assistant generates a guided workflow for service technicians using supplemental data from industrial automation systems, including device characteristics and process information, to provide efficient problem-solving support.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If service technicians manually search through multiple sources of information (product manuals, troubleshooting guides), then they can find relevant information, but the process becomes cumbersome and time-consuming

Engineering Contradiction:
Improvecompleteness of troubleshooting informationVSAvoidtime to resolve issues
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent introduces an intermediary system (service assistant with LLM) that mediates between the technician and multiple information sources. The LLM aggregates data from work orders, device telemetry, manuals, and troubleshooting guides, then presents synthesized guidance through a mobile interface, eliminating the need for technicians to manually search multiple sources

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent combines multiple information sources (product manuals, troubleshooting guides, device telemetry data, work order information) into a single integrated service assistant interface. The LLM merges these disparate data sources to generate comprehensive troubleshooting workflows, reducing the need to toggle between multiple sources

Inventive Principle:
Principle #5Merging (Combining)

2Adaptability or versatility

If service technicians toggle between multiple sources of information, then they can access comprehensive data, but the operation becomes cumbersome

Engineering Contradiction:
Improveaccess to multiple information sourcesVSAvoidease of troubleshooting process
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The service assistant performs multiple functions within a single interface: it retrieves work order data, accesses device telemetry, searches manuals and troubleshooting guides, generates step-by-step workflows, and provides interactive guidance. This multi-functional system eliminates the need to switch between separate tools and sources

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The LLM-based service assistant acts as an intermediary that handles the complexity of accessing and synthesizing information from multiple sources, presenting a simplified unified interface to the technician that maintains adaptability while improving ease of operation

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If traditional troubleshooting methods are used with multiple separate tools and sources, then comprehensive information can be gathered, but productivity decreases

Engineering Contradiction:
Improveaccuracy of troubleshooting informationVSAvoidissue resolution speed
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by automatically gathering and synthesizing relevant information from multiple sources before the technician begins troubleshooting. The LLM pre-generates customized workflows based on work order data and device telemetry, so technicians receive ready-to-use guidance rather than starting from scratch

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The service assistant intermediary automatically correlates data from work orders, device telemetry, and knowledge bases to generate accurate troubleshooting workflows, maintaining reliability while accelerating the process by eliminating manual information gathering and synthesis steps

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS20250342448A1Systems and methods for large language model (LLM) generated service content
Publication Date: 2025.11.06 ROCKWELL AUTOMATION TECH INC
  • US20250342448A1 patent drawing
  • US20250342448A1 patent drawing
  • US20250342448A1 patent drawing

AI summary

A method includes receiving a work order associated with a problem experienced by an industrial automation system configured to perform an industrial automation process, receiving, from an industrial automation device within the industrial automation system, supplemental data including one or more characteristics of the industrial automation device, the industrial automation system, the industrial automation process, or any combination thereof, generating, via processing circuitry, using one or more large language models (LLMs), service content based on the work order and the supplemental data, wherein the service content includes a guided workflow for addressing the problem associated with the work order, and providing the service content to a mobile computing device for display via a user interface.