Solid models are the critical data elements in modern
computer-aided design environments, because they describe the
shape and form of manufactured artifacts. Their growing ubiquity
has created new problems in how to effectively manage the many
models that are now stored in the digital libraries for large
design and manufacturing enterprises. Existing techniques from
the engineering literature and industrial practice, such as
group technology, rely on human-supervised encodings and
classification; techniques from the multimedia database and
computer graphics/vision communities often ignore the
manufacturing attributes that are most significant in the
classification of models. This paper presents our approach to
comparing the manufacturing similarity assessments of solid
models of mechanical parts based on machining features. Our
technical approach is threefold: perform machining feature
extraction, construct a model dependency graph (MDG) from the
set of machining features, and partition the models in a database
using a measure of similarity based on the MDGs. We introduce
two heuristic search techniques for comparing MDGs and present
empirical experiments to validate our approach using our testbed,
the National Design Repository.