Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AiBAT: Artificial Intelligence/Instructions for Build, Assembly, and Test

Benjamin Nuernberger, Anny Liu, Heather Stefanini, Richard Otis, Amanda Towler, R. Peter Dillon

TL;DR

An overview of the AiBAT system, a novel system for assisting users in authoring IBATs, which works by first analyzing assembly drawing documents, extracting information and parsing it, and then filling in IBAT templates with the extracted information.

Abstract

Instructions for Build, Assembly, and Test (IBAT) refers to the process used whenever any operation is conducted on hardware, including tests, assembly, and maintenance. Currently, the generation of IBAT documents is time-intensive, as users must manually reference and transfer information from engineering diagrams and parts lists into IBAT instructions. With advances in machine learning and computer vision, however, it is possible to have an artificial intelligence (AI) model perform the partial filling of the IBAT template, freeing up engineer time for more highly skilled tasks. AiBAT is a novel system for assisting users in authoring IBATs. It works by first analyzing assembly drawing documents, extracting information and parsing it, and then filling in IBAT templates with the extracted information. Such assisted authoring has potential to save time and reduce cost. This paper presents an overview of the AiBAT system, including promising preliminary results and discussion on future work.

AiBAT: Artificial Intelligence/Instructions for Build, Assembly, and Test

TL;DR

An overview of the AiBAT system, a novel system for assisting users in authoring IBATs, which works by first analyzing assembly drawing documents, extracting information and parsing it, and then filling in IBAT templates with the extracted information.

Abstract

Instructions for Build, Assembly, and Test (IBAT) refers to the process used whenever any operation is conducted on hardware, including tests, assembly, and maintenance. Currently, the generation of IBAT documents is time-intensive, as users must manually reference and transfer information from engineering diagrams and parts lists into IBAT instructions. With advances in machine learning and computer vision, however, it is possible to have an artificial intelligence (AI) model perform the partial filling of the IBAT template, freeing up engineer time for more highly skilled tasks. AiBAT is a novel system for assisting users in authoring IBATs. It works by first analyzing assembly drawing documents, extracting information and parsing it, and then filling in IBAT templates with the extracted information. Such assisted authoring has potential to save time and reduce cost. This paper presents an overview of the AiBAT system, including promising preliminary results and discussion on future work.
Paper Structure (23 sections, 6 figures, 2 tables)

This paper contains 23 sections, 6 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Document Understanding
  4. Assistive Document Authoring
  5. AI/ML for Industrial Use Cases
  6. IBAT Authoring and System Overview
  7. Information Extraction
  8. DocVQA Testing
  9. Custom Approach
  10. LLM Note Parsing and Final Step Generation
  11. Note Parsing
  12. Final Step Generation
  13. Experimental Evaluation
  14. Information Extraction Results
  15. LLM Setup and Metrics
  16. ...and 8 more sections

Figures (6)

  • Figure 1: A printed wiring assembly (PWA) for the Europa Clipper mission phillips2014europa is built, assembled, and tested using IBAT documents authored from assembly drawings.
  • Figure 2: The conceptual workflow of AiBAT. Here, a drawing note is parsed by the LLM into a set of operations, components, material, and items. The IBAT step template is then populated via the extracted information; here, the action "INSTALL" with the reference designator "U4" is inserted into the table.
  • Figure 3: AiBAT system architecture diagram. We first use a custom image processing approach to extract the assembly drawing notes (see Section \ref{['sec:information_extraction']}). We then call the LLM twice, first to parse the notes into actions, information, and entities, and then to generate the final IBAT steps via using the golden IBAT template steps.
  • Figure 4: A cropped screenshot of a flagged note. Flagged notes are indicated by the triangle shape around the note number.
  • Figure 5: An abridged, example few-shot prompt for parsing a drawing note into a list of actions, information, and entities.
  • ...and 1 more figures