The company is designing a tool that can generate billions of planning scenarios and evaluate the impacts of different scenarios on key quality-of-life measures.
Sidewalk Labs has announced a new generative design tool that uses machine learning and computational design to generate “millions of comprehensive planning scenarios”.
The Alphabet company’s product manager, Violet Whitney, and designer Brian Ho are developing the tool to help planning teams to fully evaluate and understand all the options available to them. The aim is for planners, architects and developers to make choices that best reflect local priorities.
In a blog post on Medium, Whitney points out that the various experts on a planning team often run separate analyses to produce a neighbourhood design. For example, an architect uses one type of software to simulate sunlight, an engineer uses another to plan streets, a real estate developer models economics in a spreadsheet, and so on.
The time and cost needed to coordinate all these competing elements often mean a project can only afford to develop a handful of designs for the team, with limited insight into how these options will impact the community.
Whitney stresses what the generative design tool doesn’t set out to do is automate the urban planning process or eliminate the need for human-driven design. Instead, it provides a set of features that can empower planning teams to do their job even better.
What the tool can do is “make the design process more holistic and efficient, helping planners and the community make the most informed decision possible”, she said.
“If generative design does its job, it can result in neighbourhoods that truly reflect the needs and priorities of the communities they serve.”
The tool starts with a set of foundational information that can include a geographic area, physical or regulatory qualities of the place, and (if available) existing development plans.
Guided by a designer’s input, it can then draw from environmental (non-personal) data that’s commonly used by engineers, architects and developers to help plan a neighbourhood: things like street layouts; block orientations; real estate economics; weather patterns; and building heights.
Using all that information, the tool can generate a series of possible scenarios – along with their expected performance – for planners to consider using or refining as the design process evolves.
According to Whitney, the tool often arrives at designs that traditional planning methods might not have found and while generative design isn’t currently designed as a community engagement tool, “it’s not hard to imagine that usage being developed down the line”.
The tool also has the ability to get smarter as time goes on. With machine learning, a generative design simulation can not only understand trade-offs between various objectives like daylight and density, but also learn what’s worked (and what hasn’t) from years of existing neighbourhood designs. Today, that type of institutional knowledge is trapped inside professionals’ heads, and difficult to pass on to communities, Whitney noted.
“In the end, if generative design does its job, it won’t just make the planning process more accessible – it can result in neighbourhoods that truly reflect the needs and priorities of the communities they serve.”
You might also like: