Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

FMEA Builder: Expert Guided Text Generation for Equipment Maintenance

Karol Lynch, Fabio Lorenzi, John Sheehan, Duygu Kabakci-Zorlu, Bradley Eck

TL;DR

This system uses large language models to enable fast and expert supervised generation of new FMEA documents and shows that foundation models can correctly generate over half of an FMEA's key content.

Abstract

Foundation models show great promise for generative tasks in many domains. Here we discuss the use of foundation models to generate structured documents related to critical assets. A Failure Mode and Effects Analysis (FMEA) captures the composition of an asset or piece of equipment, the ways it may fail and the consequences thereof. Our system uses large language models to enable fast and expert supervised generation of new FMEA documents. Empirical analysis shows that foundation models can correctly generate over half of an FMEA's key content. Results from polling audiences of reliability professionals show a positive outlook on using generative AI to create these documents for critical assets.

FMEA Builder: Expert Guided Text Generation for Equipment Maintenance

TL;DR

This system uses large language models to enable fast and expert supervised generation of new FMEA documents and shows that foundation models can correctly generate over half of an FMEA's key content.

Abstract

Foundation models show great promise for generative tasks in many domains. Here we discuss the use of foundation models to generate structured documents related to critical assets. A Failure Mode and Effects Analysis (FMEA) captures the composition of an asset or piece of equipment, the ways it may fail and the consequences thereof. Our system uses large language models to enable fast and expert supervised generation of new FMEA documents. Empirical analysis shows that foundation models can correctly generate over half of an FMEA's key content. Results from polling audiences of reliability professionals show a positive outlook on using generative AI to create these documents for critical assets.

Paper Structure

This paper contains 9 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Solution Approach
  3. Dynamic Few Shot Prompting (DFSP)
  4. Structured responses
  5. Evaluation
  6. Generating Equipment Boundaries
  7. Generating Failure Locations
  8. User Feedback
  9. Outlook

Figures (2)

  • Figure 1: Flow of a single step (boundary generation) in our system: The step's input (1) is used to select candidate examples from the database (2). Examples confirmed by the user (3) appear in the prompt (4). The system parses the LLM's response (5) and optionally aggregates responses from multiple prompt, model variations (6) for presentation to the user (7).
  • Figure 2: Graphical interface for generating equipment boundaries: The user enters a short description of the equipment as free text. Examples of similar equipment from our database are presented for consideration. Equipment cards that are ticked serve as examples in the prompt to generate an equipment boundary.