Quantifying the ability of a digital design concept to perform a function currently requires the use of costly and intensive solutions such as Computational Fluid Dynamics. To mitigate these challenges, the authors of this work propose a deep learning approach based on 3-Dimensional Convolutions that predicts Functional Quantities of digital design concepts. This work defines the term Functional Quantity to mean a quantitative measure of an artifact's ability to perform a function. Several research questions are derived from this work: i) Are learned 3D Convolutions able to accurately calculate these quantities, as measured by rank, magnitude and accuracy? ii) What do the latent features (that is, internal values in the model) discovered by this network mean? iii) Does this work perform better than other deep learning approaches at calculating Functional Quantities? In the case study, a proposed network design is tested for its ability to predict several functions (Sitting, Storing Liquid, Emitting Sound, Displaying Images, and Providing Conveyance) based on test form classes distinct from training class. This study evaluates several approaches to this problem based on a common architecture, with the best approach achieving F Scores of > 0.9 in 3 of the 5 functions identified. Testing trained models on novel input also yields accuracy as high as 98% for estimating rank of these functional quantities. This method is also employed to differentiate between decorative and functional head-wear, which yields an 84.4% accuracy and 0.786 precision.Dering MD-17-1178 1 3-D scanners, or RGB-D Sensors such as the Microsoft Kinect) has further enabled the capture, sharing, and production of such artifacts around the world [1, 2].However, just because a design "looks good" does not mean that it will achieve its intended functions or behaviors.While simulation models such as CFD use advanced numerical methods and algorithms to evaluate the feasibility of designs, they require extensive modeling expertise, advanced computing resources and are time consuming [3,4]. While some of these publicly-available digital design models are decorative in nature, many of these artifacts are meant to serve a specific purpose or function that may be implied or explicitly stated by the designer. Formally, a function is defined as a purpose intended for an object. This work proposes a method that uses as input, a model file representing a possible artifact. A neural network analyzes this artifact, and as output, predicts how well it will perform each intended function. For example, some objects can be used for sitting. This network will predict how much force it will be able to withstand, when sat upon. These trained predictive models will provide insight from their latent variables as well. A latent variable is a variable contained within the model whose value may describe the forms and patterns that have the most predictive power.Typically, design defines a product by its form, function, and behavior [5]. However, this work is limited to a stud...