While the technology used in design has come a long way in recent decades, most design continues to be limited by the actual design process. Designers cannot avoid their biases created from past experiences, outdated assumptions, and the limits of their imaginations.  Resource constraints tend to prevent designers from being able to deeply explore all potential solutions to a design problem. Generative design gives the design community the ability to work beyond the limitations of traditional design and permits the exploration of design possibilities used in fields such as art, product design, and architecture.

Autodesk refers to generative design as being able to quickly create high-performing design alternatives from a single idea. Generative design presents many solutions for the designer to choose from. Generative design is an exploration process wherein designers use generative design software and input what is required of the design such as materials, cost constraints, manufacturing processes, performance requirements, or spatial constraints. The generative design software explores the potential solutions, quickly generates design alternatives, tests designs, and learns what works.

Generative design has also been referred to as “Algorithmic Design”.  Computers are able to explore a large number of design solutions to determine the most effective design. Generative design gives designers the ability to explore design solutions through simulation and harnesses the power of Artificial Intelligence (AI).  The design solutions offered by generative design need to be evaluated by a human designer. While 3D models are the typical output for generative design for architecture, the output can also be images, sounds, animation, and more.

As discussed in the PTC Blog, generative design output is dependent on the information provided by the designer, so the more information given, the better the solutions that will be provided. A major benefit to generative design is the number of design solutions that can be created in a short time, allowing products to make it to market quickly.

Additional benefits include the ability to achieve expert results without user expertise, the creation of novel products that do not rely on small changes to previous designs, the elimination of user bias for particular designs, and optimization of elements such as cost and durability. Entry-level engineers using generative design are afforded an advantage that allows them to quickly produce successful models. Additionally, all engineers using generative design are able to avoid trial and error to refine their designs and focus on what the design needs to accomplish.

Formlabs explored the application of generative design and identified many industries that face challenges that could be met by generative design. Aerospace, architecture, the automotive industry, consumer goods, and product manufacturing all face similar design challenges. Generative design allows engineers to explore solutions for complex engineering challenges such as reducing the weight of components and optimization of performance.

Generative design algorithms frequently create efficient organic shapes supported by latticework. These designs can be costly to create using traditional manufacturing processes such as injection molding or CNC machining. 3D printing can meet this challenge by providing a 3D model of design iterations. The cost-effectiveness of 3D printing increases as the complexity of design increases. 

While generative design is gaining popularity, it has its detractors. Daniel Davis has written on the topic and has proposed that the promises of generative design are empty, and refers to the generative design as the architecture industry’s white whale.  His reasons for pessimism include that as of the time of his writing (2020), the designer is required to create the algorithm which generates the various designs.  While generative design creates hundreds of design options, many of these options are not viable. The comparison of design options remains the responsibility of the human designer, and the challenge of comparing options grows with design complexity. Having too many options can freeze a designer. Architectural performance is complicated, and generative design may be prone to using arbitrary metrics or optimizing for easily measured elements. Davis also pointed out that generative design simplifies the design process into input, creation, and decision, which is an oversimplification of what designers do. Designers do not tend to follow a linear process but rather cycle between large and small changes to design, broad project objectives, and implementation. 

Autodesk offers generative design capabilities through the AEC collection of Revit 2021. They are promoting generative design by helping designs explore solutions with rapid testing and evaluating design iterations, thereby improving BIM’s capabilities early in the design phase. BIM has become integral to the design, architecture, and construction process because it can integrate all building data into an accessible digital representation that can be viewed in a 3D environment before construction. Generative design may play a large role in some aspects of design in the future.  The results of generative design capabilities in Revit will give us clues to how the adoption of this technology will impact the industry.  

If you are interested in discussing how CADdetails can help your company with your BIM needs, please review our BIM development services and reach out to learn more.


cover image Ⓒ unsplash