A methodology for automatic retrieval of similarly shaped machinable components

A methodology that can intelligently search a CAD database of 2.5D machinable components and retrieve components and sub-pieces with similar shape characteristics has been presented. The proposed methodology has the potential to reduce both the development time and cost of new designs by effectively re-using existing designs more efficiently. This is accomplished by searching a database of parts with their associated plans and extracting parts of those plans that are required for the query part. These plan parts can subsequently be merged to develop the entire plan for the query part. The methodology utilizes qualitative matrices to represent the spatial relationship between the maximal features of the query component. These feature-spatial relationship pairs are matched against those present in existing components within the database. The query can be relaxed based upon TAH structures that allow for generalizing of features and spatial relationships to allow for the best match to be found if no exact match is found.