Railway Maintenance Planning Using LLMs for Legacy Documentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing railway maintenance systems face challenges in automating maintenance planning due to the effort required to capture and process inconsistent historical documentation, leading to economic inefficiencies and limited implementation of automated solutions.
Innovation Solution
A method utilizing a computer-aided planning module with a Large Language Model (LLM) to evaluate diagnostic and analytical data, generating maintenance requirements and measures in human language, and outputting them via an interface, allowing semi- or fully automated maintenance planning without extensive initial implementation effort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated maintenance planning is implemented using traditional methods, then maintenance planning efficiency is improved, but the effort and cost for capturing and processing historical documentation increases significantly
Solution Approach 1:
The patent replaces traditional mechanical documentation processing methods (manual capture and processing of historical text documentation) with an artificial intelligence-based language model system. The LLM automatically processes historical documentation and generates maintenance plans, eliminating the need for complex manual data capture and processing infrastructure while significantly improving maintenance planning efficiency.
2Productivity
If full automation of maintenance planning is implemented, then labor costs are reduced, but the initial implementation effort and cost increase
Solution Approach 1:
The patent implements a hybrid automation approach where the LLM generates maintenance plans automatically, but human experts review and approve them. This partial automation strategy reduces initial implementation complexity and time requirements compared to full automation, while still achieving significant labor cost reductions. The system can be deployed incrementally, starting with LLM-generated plans that are gradually refined through human feedback.
3Measurement precision
If historical documentation is processed to improve automation, then maintenance planning accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent uses a language model to automatically process historical documentation and extract relevant maintenance information, replacing complex data processing systems. The LLM's natural language understanding capabilities enable it to accurately interpret unstructured historical documentation without requiring complex data cleaning, structuring, or preprocessing infrastructure, thereby improving maintenance planning accuracy while minimizing data processing complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Method for planning maintenance measures for a railway technical system The invention comprises a method for planning maintenance measures for a railway technical system (BTS), comprising a functional component (FK) that performs a function belonging to railway operation, a monitoring device (UV) that monitors the function of the functional component (FK) and generates monitoring data (UD) describing this function, a diagnostic device (DGN) that receives messages containing monitoring data (UD) generated by the monitoring device (UV) and generates computer-aided diagnostic data (DD) that describe a need for maintenance measures.A computer-aided planning module (PLM) is used, which employs a Large Language Model (LLM) to evaluate messages containing diagnostic data (DD) and to generate analysis data (AD) that describe the diagnostic data (DD) in human language. The evaluation is performed with regard to maintenance requirements and/or maintenance measures. A description of the maintenance requirements and/or maintenance measures is generated in human language. Output data (OD) representing the description of the maintenance requirements and/or maintenance measures is output via a first interface (S1). Furthermore, the invention comprises a railway engineering system, a computer program product, and a computer-readable storage medium. The advantage lies in the fact that various types of documentation describing the railway engineering system can be evaluated automatically.