Creativity is not a result of acquired skills and knowledge, but an art of mixing them with emotions and experience. The emergence of robotics and artificial intelligence meant smarter and better results in a short amount of time. Since they lack a sense of emotion, originality and creativity are absent in their productivity. But, over the past few years, these assumptions are being torn down as AI produces reasonable ideas for designing. A design process that’s gaining popularity in the AEC industry is generative design.
Generative design is the process of feeding quality data to softwares that will then create an algorithm. After it is trained the software produces results that could satisfy the user. This is an iterative design exploration process that uses an AI-driven software program to generate a range of design solutions that meet a set of constraints. A game-changer to the traditional design approach and design process for the better not worse.
Generative design has grown to be so popular among designers in all fields, from students to professionals. But not the correct notion about it is being spread. Many believe that an AI-developed result means a lack of thought process and originality. They think designers will soon be replaced by softwares that will do their work. Generative designing is a helping hand and not a replacement. It can be utilized to save-up time spent on making permutations and combinations of plans, solutions, and forms to use a space avoiding human errors. The saved time will produce a cost-effective project and can be invested in stages that require more attention.
So how is generative design utilized in architecture?
Since softwares can be changed and personalized, generative designing is flexible to adapt for various projects. It plays a major role in architectural space planning. This is done by feeding a program with pre-existing designs. These designs have to be of good quality and well-organized data. That will ensure reliable and predictable results. Generative design can be a stage of the design process. It doesn’t have to be the final output but a tool that helps put together a finer blueprint.
Many architecture tools and platforms based on generative design processes have been created and are being utilized, a few of which are:
1.Evolving Floor Plans:
An experimental research project by researcher Joel Simmons. He created a design algorithm that would make plans more efficient. His algorithm transforms a standard floor plan, to minimise the length of corridors and reduce walking time. This results in free-form layouts, with not a square or rectangular space insight. Simon describes the plans as “biological in appearance, intriguing in character and wildly irrational in practice”.
- Digital blue foam
DBF is a Singapore-based company started by architects and coders. They have created a web-based user-friendly sustainable design platform to help architects and designers design high-quality and carbon-negative buildings. The early stages of a design process are effortlessly done, thus relieving time spent on the main design. DBF allows its users to check optimum building orientation, compute massing options as per set project goals and constraints, and convert them into BIM.
3.Finch
A parametric design tool that generates floor plans according to site constraints. Finch was developed by Architecture studio Wallgren Arkitekter and Swedish construction company BOX Bygg. Purposed to help architects a site’s potential, helps save tons of time at the early stages. It works as a plug-in to Grasshopper and generates 2-D and 3-D plans.
4.ACPVLab
A research and development unit launched by Architecture firm Antonio Citterio Patricia Viel. ACPVLab has developed a “handcrafted” computer script that will allow architects and interior designers to explore thousands of different layout permutations instantly. “Through algorithms providing a series of optimal design outcomes in iteration sequences, the methodology will help architects and designers evaluate and refine designs,” the firm said.
Though generative design technology may be relatively new in architecture, companies have set out to construct real-life projects with it.
- Autodesk Technology Centre Toronto
Autodesk has its own generative design tool called Project Discovery and this office is a prime example of generative design in practice. To produce user-friendly designs, 250 employees were consulted about their office preferences. These included things like the distance between two tables, outside views, daylight, and such. All these parameters were inputted into the software that returned over 10,000 solutions and the designers pored over all of them to come up with the one most suitable. The project is a much-heralded blend of human knowledge and machine learning.
- The Heydar Aliyev Centre
Azerbaijan‘s famous landmark leaves all its visitor’s in awe. But within it lies the beauty of generative designing. Modern tools have made it possible to come up with such a shape that meets all design parameters. The design defied the odds because it was curvy and impossible to build on. This venue includes a conference hall, a library, and a museum. All these carefully planned using an algorithm that provided the perfect interior design.
- Stamhuis Stores
Stamhuis is famous for embracing modernity and was one of the first companies to work with generative design. Headquartered in the Netherlands, Stamhuis has a nearly century-long history of constructing and renovating retail spaces. Stamhuis has a wide array of clients from supermarket chains to small stores. This means that the company has to adapt quickly and work closely with each client. Sometimes, one client alone may require over 60 different projects. Fortunately, generative design works in both directions. This means Stamhuis can accept a large volume of work involving both large and small structures.
Generative design plays a growing central role in architecture and has immense potential to advance the development of this field. Companies are utilising it but many more have to start. Generative design is a tool and not the final design. The generated results must be reviewed and accordingly used and modified.
References
Generative Design 101. Formlabs. [online]. Available at: https://formlabs.com/asia/blog/generative-design/ [accessed on 28 March 2021]
Generative Design for Architectural Space Planning. Medium. [online]. Available at: https://medium.com/autodesk-university/generative-design-for-architectural-space-planning-9f82cf5dcdc0 [accessed on 27 March 2020]
Amy Frearson. Evolving Floor Plans is a research project that optimises building layouts. [online]. Available at: https://www.dezeen.com/2019/01/03/evolving-floor-plans-research-joel-simon/ [accessed on 28 March 2021]
Archistar. 4 Examples of Generative Design in Action – Archistar. [online]. Available at: https://archistar.ai/blog/4-examples-of-generative-design-in-action/ [Accessed on: 28 March 2021]
