Most architects wonder about the advantage of generative design in architecture. Despite its growing popularity in architecture, generative design has its pros and cons. In our last post, we explored the basics of generative design and understood it vis-à-vis parametric and computational design. Using generative design, experienced architects can optimize, explore, and shortlist highly creative design options in a matter of minutes.
But firstly, what is generative design? It is an iterative design process used in architecture and manufacturing. The process creates unlimited unique designs basis specified constraints and parameters. Some examples of software that provide an option for generative design are Autodesk Revit, Infurnia, Creo, and Rhino. In our post today, we will explore its advantages and disadvantages.
Unlimited Choices: Backed by the storage capacity of cloud, an advantage of generative design is that the process produces hundreds of design iterations despite specified constraints. This is well beyond what an architect or designer can manually create. Given a plethora of options, it is then easy to filter out and choose those options which one likes.
Overcoming Barriers to Creativity: As the design options produced by generative design tools are way beyond what the human mind can envisage, it helps one overcome creative barriers. Creative processes in professional setups usually end up being a matter of chance. Pressed by time constraints, it is difficult to come up with highly creative solutions. An advantage of generative design helps overcome such creative blocks.
Automated Adjustments: It is not surprising that generative design which is a step ahead of parametric modeling, is capable of automated adjustments. The iterative process produces designs basis the constraints and metrics defined by the user. Changes reflect as automated adjustments and are processed very quickly on cloud.
Shorter Lead Time: Last-minute additions to construction projects create undue pressure. Yet, they need to be creative. This could be the design of a hotel lobby or a marquee structure in a commercial establishment. The advantage of generative design is that it helps create multiple creative options within a short period- a task which is humanly impossible.
Lower Costs: Designs produced by generative modeling do not require additional simulation or testing. The designs adhere to safety norms and are ready for execution because simulation is built into it as a feature. Thus, another advantage of generative design is that it leads to lower costs and frees up budget for other processes.
Practical Creativity: Compared to manually created designs, which sometimes have issues of adherence, the highly creative options created through the process, strictly adhere to constraints. There is a ‘self-check’ involved and errors are avoided. Whereas in other processes, there are chances of human error, despite creativity.
Free Up Time: An advantage of generative design is that it helps architects skip the need of coming up with designs and manually navigating the design options. It frees up their time so that they can focus on more human and practical engagements. It allows them to take up a more strategic role.
Making Humans Redundant: While embraced by a large part of the AEC community, generative design and its reliance on artificial intelligence has raised concerns by a section of the industry. Architects wonder if the machine learning-enabled software feature would replace the need for designers & architects in the future. It is indeed a fair disadvantage of generative design as AI & machine learning has instilled fear in other industries such as manufacturing too. Generative design proponents believe such fears are unfounded. They believe that architects need to expand their skill sets. Generative design is just another tool in the toolbox. After all, there is a lot more to the architectural profession than constraint-based design.
Shortlisting Options: Interestingly, a disadvantage of generative design is that it throws up too many options. But there are software designed to filter options basis cost or other parameters. But there many software that do not provide an advanced filtering option. Most architects do not have the time or interest to tediously filtering out designs. Thus, sometimes, it can get a little too overwhelming. However, with machine learning; many programs can shortlist designs based on earlier choices made by the architects.
Manual Intervention: Though largely automated, a disadvantage of generative design is that it does require the architect to be capable of leveraging machine learning and AI to his advantage. Otherwise, the resulting designs may not stand up to the high expectations we have, out of generative design.
As one can see above, there are multiple advantages and disadvantages of generative design. Experienced architects who adopt generative design, benefit from saving in time and energy. They can use the gain to make key strategic decisions. Of course, there are large gains in terms of creative output too. The process may not replace parametric modeling anytime soon. But generative design is surely a must-have to enhance one’s architectural practice.