An automated intelligent feature-based maintenance plan generation method

Great efforts have been made to make maintenance processes intelligent and automated. Data analysis, assembly-disassembly sequence generation, maintenance optimization models and knowledge-based systems are some of the research areas related to maintenance with significant contributions. Until today, there is no robust plan generation method available for the scientific and industrial community. The available methods are mostly human-dependent and are lacking in automatically assist maintenance operators to perform maintenance tasks. This paper presents a method to create an intelligent system capable of automatically provide maintenance instructions at a product and component level using feature-based product identification. This paper introduced a framework that integrates different algorithms for a knowledge-based decision, robust reverse engineering, CAD (computer-aided design) model feature recognition, product identification, and maintenance plan generation. The method is able to identify the product from CAD files generated through the 3D points-cloud data. The identified product is then linked to the knowledge-base, which provides an intelligent plan for the maintenance procedure. The method not only provides a product level maintenance procedure but also generates the shortest path for proper inspection and repair at the component level. The automated and intelligent tool contributed by this paper supports the naïve operator to execute tasks efficiently without relying on their expertise. The method is generic and able to accommodate and integrate new modules to support other applications in the future.