A Study on Sampling Strategies to Determine the Variability of Parametric History-Based 3D CAD Models

Abstract

Design reusability largely depends on the parametric quality of its associated digital product data. In this regard, the quality of the master model (typically a history-based parametric model) is crucial. However, no quantitative metrics exist that can provide an accurate assessment of parametric complexity and model reusability. In this paper, a set of 370 parametric 3D CAD models of various geometric complexities were analyzed to assess their robustness when undergoing alteration. Three indicators for estimating the modification ability of the model are proposed: Ratio for Exhaustive Modification, Ratio for Selective Exhaustive Modification, and Ratio for Weighted Exhaustive Modification. Correlations between these indicators as well as other geometric complexity metrics are studied. The geometric complexity metrics considered in our study include number of faces, surface area to volume ratio, sphericity, and convexity. Our experimental results with the proposed indicators provide new insights on the quantitative assessment of parametric complexity and support their use as reliable indicators of CAD model reusability.