An Intelligent Framework for Automatic Maintenance Plan Generation based-on Product Features Recognition

• Author / Creator

  Yepez Herrera, Pedro

• Maintenance takes care of the health of equipment and/or mechanical system. Currently, maintenance is considered as one of the most labor-intensive and expert-centered manufacturing process, since disassembly/assembly procedures and its level of difficulty, makes it very difficult to automate the process. Therefore maintenance planning for naïve operators represents a difficult task, sometimes even for experienced operators. An extensive and structured maintenance knowledge is required to be able to plan for maintenance execution at large companies as the number of equipment and systems are high. Sometimes old equipment information is not available, therefore maintenance planning and execution only depends on experienced operators, which ultimately increases maintenance costs.The development of a framework that is able to generate a maintenance procedure automatically by feature-based product recognition as well as providing intelligent guidance to operators represents a big step towards maintenance process automation. The proposed framework integrates different algorithms for knowledge-based decision, robust reverse engineering, CAD (computer-aided design) model features-recognition, product identification, and maintenance plan generation. A method recognizes the selected product from a CAD model by extracting geometrical information and then associating it to a knowledge-base to identify the product using a feature recognition algorithm. An adapted robust reverse engineering module is able to reconstruct the CAD model from points-cloud data generated through a laser scanner or Time-of-Flight (ToF) sensor. Once the product is identified based on features-recognition from the CAD model, it is automatically linked to a knowledge-base to provide existing maintenance procedure or generate a new maintenance plan from similar a case-base. Corrective maintenance procedure currently depends on human knowledge and decision in order to locate the damaged component, repair it and reduce downtime. For these cases, a component level module is developed and integrated, which uses disassembly precedence graph and Failure Mode and Effect Analysis (FMEA) data from a specific product to generate the efficient (shortest) disassembly path for inspection and repair. The method supports naïve operators to efficiently execute disassembly tasks without relying on their expertise. The presented framework is a generic in nature, as it is designed to accommodate new modules to support other applications, for example remanufacturing planning.

• Subjects / Keywords

  • Features Recognition
  • Maintenance Automation
  • CAD
  • Knowledge-based system

• Graduation date

  Spring 2019

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-cw2q-mb12

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.