Design intent is generally understood simply as a CAD model's anticipated behavior when altered. However, this representation provides a simplified view of the model's construction and purpose, which may hinder its general understanding and future reusability. Our vision is that design intent communication may be improved by recognizing the multifaceted nature of design intent, and by instructing users to convey each facet of design intent through the better-fitted CAD resource. This paper reviews the current understanding of design intent and its relationship to design rationale and builds on the idea that communication of design intent conveyed via CAD models can be satisfied at three levels provided that specialized instruction is used to instruct users in selection of the most suitable level for each intent.Keywords: design rationale, design intent, CAD model quality, CAD education.
INTRODUCTIONFeature-based parametric CAD is a commonly deployed 3D modeling technology that is widely used in industrial settings. In these systems, the 3D CAD model is created by gradually and sequentially adding geometric features through parent/child relationships, which creates an interconnected structure that, when defined properly, allows for more flexible and reusable models. This process is recorded in a structure known as a design tree, feature tree, or history tree. Parent/child interdependencies are the basic elements that facilitate CAD reusability and alteration of parametric models. When these dependencies are defined properly, changes in the artifact can be performed efficiently, as alterations propagate automatically from parent to child nodes. However, parent/child dependencies can also be the root of numerous regeneration problems, which often forces designers to rebuild the CAD model entirely, costing time and money.Previous researchers have determined that 48% of CAD models fail during design exploration [38] and according to the 2013 State of 3D Collaboration and Interoperability Report, 49% of engineers spend more than 4 hours per week repairing design data with 14% spending more than 24 hours per week [39]. The same report states that 32% of organizations miss deadlines due to design data problems [39]. Gerbino states that data exchange issues result from poor modeling strategies [28]. González-Lluch and colleagues echo these sentiments stating that erroneous CAD models that filter toward downstream applications require effort to rework the models to remove data corruption [29]. Poor understanding and/or communication of design rationale and design intent are commonly argued to cause most of those failures. But the concepts of design rationale and design intent are complex in themselves.Describing the purpose of a design and the justifications for specific decisions made when creating it are essential tasks for engineers and design professionals. Design rationale can be defined as the explicit documentation of the reasons behind the decisions made when designing a system or artifact [52]. Although d...