In school, engineers are taught to solve design challenges by applying physics and mathematical principles coupled with open-ended problem solving skills to determine a viable solution: defining the performance criteria, design constraints and even the fabrication process. They graduate and then get a job in the real world. Then they are siloed in their respective departments, are given a CAD system, and end up spending way too much time and intellectual capacity figuring out how to capture their designs and ideas as valid geometric data in CAD.
CAD was supposed to be the ultimate engineering productivity tool…how did it become a gate to productivity, placing more emphasis on pressing buttons on a computer screen, minimizing the importance and value of what an engineer has been taught to do?
Current product development methods and practices place significant focus on the development of a complete 3D CAD model. CAD modeling is a high priority as performance criteria cannot be validated without it. Creation of design topology becomes paramount without knowing beforehand if it truly satisfies the performance criteria, potentially locking in the product development project into a predetermined, and perhaps less than optimal path.
During the model development process, parametric CAD forces engineers to think like a CAD system to define a valid model. Instead of focusing on design performance, they begin to think and worry more about interface picks and clicks. They worry about constraints, relationships, the order of features, dependencies, and proper history tree structure to accommodate change. It becomes a goal of first capturing a CAD model that can then finally be tested to validate design and performance criteria. That is not how designers or engineers are taught to think; it is how a CAD system makes one think. Over time, many grow to become great CAD users instead of great engineers.
Although parametric CAD can be powerful, it is also a limiting factor to ideation because of the time and effort that is required to develop a valid model. Because of the time and effort required, only a few design alternatives can be developed for evaluation. This limits the options that can be explored. During the testing phase, performance constraints are finally added to the CAD model for computer simulation (FEA). Alternatively, prototypes are fabricated for physical testing. These models are tested, modified, and tested again and again in an iterative loop until the product development team gets close to meeting the design criteria or runs out of time. The limited number of design alternatives explored becomes an issue when the path chosen is less than optimal. This forces trade-offs which diminish design value.
Finally, the design is handed off to manufacturing, where it must then be determined how to affordably produce the design within the cost targets. By this time, manufacturing costs may be locked in by the fundamental aspects of the design. Changes required at this stage of the process to accommodate fabrication and assembly methods can become cost prohibitive.
Enter Generative Design
But what if there was a better way… a new design and engineering practice that prioritizes the training, knowledge and experience of an engineer instead of CAD model creation? Generative Design is a new approach and method of design that runs counter to current design practices and methods. Generative Design increases the value of the engineer and good engineering decisions. This is more closely aligned with the product development process engineers are taught to approach design challenges.
Instead of CAD modeling, the engineer spends his time defining the engineering challenge, specifying performance criteria and constraints that define the design and manufacturing goals. With generative design, the designer or engineer defines only minimal geometry – connection points, hold out areas and space claim envelopes, not a complete CAD model. The generative design engine (not the engineer) then creates all valid CAD model topology options based on the specified constraints.
The Value of Generative Design
Generative design methods promote experimentation and synthesis, a major principle of good design. The generative design engine delivers a wide range of potential design options for consideration and trade-off studies in less time than a human can develop and evaluate 1 or 2 alternatives using existing methods. The value of generative design is its ability to expose the design team to a greater number of potential solutions to a specific set of design constraints. This will save time and more efficiently leverage engineering resources.
Generative design methods encourage economy and simplicity, also a major principle of good design. As the engineer defines the design challenge with nominal CAD geometry, the generative design engine exposes all options, returning alternatives that represent the minimum mass and topology required to satisfy the performance criteria and design constraints. The resulting material reduction leads to direct cost savings. The generative design engine may also return alternatives that result in part consolidation. This will simplify the assembly process and streamline the supply chain.
As detailed above, generative design practices subscribe to another principle of good design: form follows function. Many of the design alternatives returned from the function and performance criteria are topology options. These one wouldn’t have otherwise imagined and may be quite organic in form. This creates new opportunities for innovation…to explore these options as inspiration for aesthetic new shapes and form.
Finally, generative design methods foster yet another principle of good design: acknowledge the manufacturing process. The generative design engine considers the manufacturing process as a constraint. This leads to solutions that are manufacturing-aware, and provides an opportunity to review these solutions early in the process, when decisions for materials, fabrication, and assembly can still be made cost effectively.
So…what do you aspire to be…great at CAD, or a Great Engineer?
Give Fusion 360 a try today!