Initiating Geometric Dimensioning and Tolerancing using Knowledge-Based Engineering : Supporting the annotation process for Model-Based Definition

Sammanfattning

The transition from 2-D drawings to Model-Based Definition (MBD) promises a single source of truth for product manufacturing information, yet the manual creation of Geometric Dimensioning and Tolerancing (GD&T) annotations remains time consuming and prone to error. This master’s thesis investigates how Knowledge-Based Engineering (KBE) can be employed to streamline the annotation workflow for MBD models at Saab AB. The work aims to demonstrate the feasibility of an automated solution that connects to CATIA V5 via Visual Basic for Applications with the goals of improving design efficiency and increasing design consistency. A methodology derived from the MOKA framework was adapted to the project’s context, comprising an Identify and Justify phase, a Capture phase, and an iterative Formalize-Package-Activate loop to develop a prototype application. The resulting tool, the GD&T Annotation Support Application, consists of a main GUI, a Template Tool (four modes for parts, assemblies, ARM, and IRM) and an Editing Tool that enables creation and update of datums, tolerances, dimensions, views, captures, and text notes. Testing with Saab design engineers on sheet-metal, machined parts and assemblies showed a 45% reduction in total annotation time compared with the manual process, with the part-mode template alone achieving a 75% time saving for the first stage of the workflow. Survey results indicated high user satisfaction, perceived ease of use, and confidence in the application’s output. The thesis concludes that KBE can substantially accelerate MBD annotation while enhancing consistency, provided that the underlying CAD model follows a standardized specification tree structure. Future work should further develop the ARM and IRM mode, broaden the testing of the modes and incorporate richer knowledge capture to achieve higher grade automation.