How generative design can become a real game-changer for the AEC industry.
We recently launched Explore on the Spacemaker platform, the next generation of our generative design feature, which opens up many new opportunities for architects and urban planners. Explore marks a significant evolution of the software, and the journey to this point has been an interesting one. I sat down with Spacemaker’s co-founder and CTO, Carl Christensen, to discuss the importance of Explore, and how we got to this point.
Maria: Explore is one of our most significant updates to the platform this year. Can you tell us about the idea behind Explore?
Carl: One of the core ideas of Spacemaker is to empower planning teams to find the best trade-offs between different urban and site qualities while gaining confidence that they have explored a near-exhaustive list of options. Our Design and Analyze functions are crucial for establishing the best proposals for a building site. But it is the combination of artificial intelligence with computational design that puts a comprehensive list of scenarios into the hands of our users. This is what allows them to find a solution that maximises the potential of a neighbourhood for all the different stakeholders involved.
We have now connected the generative capabilities of the software more closely to the existing functionalities, which makes what is called parametric design overall far more accessible to the user, and enables faster and more responsive iterations.
Maria: What are the insights that led to this decision? How did the product team arrive at the version that we are releasing now?
Carl: In 2018, Spacemaker was the first company to make generative design available to “everyone”. Up until that point, it had been a niche technique reserved to very technical architects who understood computer programming. The launch was a great success in many ways, but we soon realized that we needed to simplify the workflow to unleash its true potential for architects and urban planners.
While users came up with designs that dramatically improved living qualities, we learned that we were not quite there yet when it came to fulfilling our vision of empowering users to have a fully agile, iterative process. Our generative feature was precise but not fast enough, and users did not feel that it was integrated effectively with our other features.
So we went back to the drawing board to redesign it from the ground up to make it truly useful. We first added a range of assisted design tools and then replaced the generative design tool completely. The result is a workflow that seamlessly blends analysis, assisted and generative design and collaboration.
Maria: Tell me how this works.
Carl: Essentially, Spacemaker helps generate different versions of a site proposal, and Explore is what allows users to “go wide”, getting inspiration and suggestions on new directions, and “go deep” - working on each of these proposals in more detail. They can swiftly zoom in and out of different perspectives and scenarios, make changes and immediately understand their impact on key quality-of-life measures. All this is happening in real-time, which means you can conduct feasibility studies and test different designs for your site in minutes, not days or weeks. That's game-changer number one, but what's more is that all of this is happening within one single, uninterrupted workflow, turning the entire planning process into a truly iterative one. Game changer number two.
Maria: It sounds like Explore could have a pretty fundamental impact on the way site planning is done by architects and urban planners?
Carl: Indeed! The fundamental problem we are trying to fix is the broken workflow of real estate development. What architects and urban planners know right now is the waterfall model, it is how sites have been planned for as long as can be remembered. This model relies on a linear method of working, breaking down project activities into sequential phases.
The issue with this approach is that you have to lock down an incredible number of parameters very early in the planning process, and you can then only move within these. Think of geographical and physical limitations of a site, but also regulatory requirements related to qualities such as wind, noise or sustainability, or simply even the preferences of urban dwellers themselves. The smallest change later down the line requires an unravelling and re-doing of the entire process up to that point. This is very cost-intensive, time-consuming and an inhibitor not only to creativity but to finding the most optimal solution for a neighbourhood.
Generative design in itself is a powerful tool but it also only allows the users to move within the parameters that were set at the beginning of the process. Yes, it enables iteration, but it does not fix the process. By integrating generative design with assisted design as well as with analysis and collaboration capabilities, we are delivering a process that is both iterative and agile.
Now planning teams can make changes to their site and parameters at any given point in time. Spacemaker will immediately show the impact of the changes and present different options.
The feedback that we have received from our users is overwhelmingly positive. Usually, generative design tools are hard to use, requiring advanced tech skills. We made it as accessible as possible to any user, regardless of their tech-savviness as we believe that a digitalised, data-driven and iterative workflow is the future of urban planning.