CAD model repair using knowledge-guided NURBS

This paper presents a methodology to repair flawed migrated models that are designed with knowledge-guided NURBS (KGN). Incompatible traditional CAD packages that are based on NURBS technology have been for many years carrying out the exchange of models and data through either neutral file formats (IGES or STEP) or proprietary formats. The exchange process most often produces poor quality data and results in errors in the migrated CAD models costing the industry billions of dollars every year. While previous studies by large focus on finding means to repair migrated models through patching, we propose to repair errors in CAD models which are generated using KGN through a reconstructing process instead of patching. We have implemented a prototype system to show that employing the right tools on the stored knowledge and data a designer can reconstruct a migrated model based on new local requirements of the receiving system without having to repair it, avoiding delay and corrective processes that only produce patched-up design bases and limiting the effective use of such models.