Generative Design is Transitioning from Manufacturing to also address Construction

Autodesk provides solutions to customers in three industries:

• Architecture, Engineering, and Construction (AEC)
• Product Design and Manufacturing (PD&M)
• Media and Entertainment (M&E)

So it's no coincidence that we have three industry collections:

It will be interesting the see the evolution of these industry collections as the industries are starting to converge:

• Autodesk customers who make buildings (AEC) are starting to behave more like customers who make things (PD&M). Whereas buildings used to be one-offs, more and more, parts of buildings (e.g., trusses) are being constructed offsite in environmentally-controlled warehouses, brought to the construction site, and assembled into position. AEC customers are suddenly concerned with mass production and quality control.
• Autodesk customers who make things (PD&M) are starting to make more bespoke items. Instead of setting up huge factories with static assembly lines to make thousands of identical items, manufacturers are becoming more agile, configuring microfactories to make small runs of personalized items  (more like one-offs of traditional AEC projects).
• M&E-based tools have been the mainstay for rendering and animation visualization for AEC and PD&M for years. Of late, both AEC and PD&M customers see the benefit in showcasing what they make via Augmented Reality and Virtual Reality.

It will also be interesting to see the evolution of our new technologies. For example, generative design was originally created for optimization of mechanical parts for the PD&M industry; however, it is starting to be applied to AEC. Project Discover is a workflow for generative design for architecture. It involves the integration of a rule-based geometric system, a series of measurable goals, and a system for automatically generating, evaluating, and evolving a very large number of design options. The result is a tool to explore a wide design space and get closer and closer to achieving all of the goals simultaneously.

Here is an example of generative design for office layout:

Project Discover showing 3 variables: daylight (how much in each office?), traffic (how often will employees be walking by?), and distraction (how likely are employees to congregate nearby?)

By starting with high-level goals and constraints and then using the power of computation to optimize for multiple criteria, Project Discover produces high-performing and novel designs that would not have been possible to create without this approach. Instead of an architect settling for the first design that meets all or most of the criteria, the architect can select the best design based on those criteria. With generative design, the computer becomes a partner in the design process instead of a documentation tool.

For more information about generative design in AEC, check out these articles:

• Autodesk's Project Discover Creates Floor Plans That Try To Please All Workers
• Generative Urban Design: A collaboration between Autodesk Research and Van Wijnen

For those who would like to be considered to have a hands-on experience:

Learn more about the BETA.

Thanks to the Autodesk Research team for some of the copy in this post.

Are you ready to take some of the tedium out of floor planning or community planning? What kinds of requirements does your planning need to satisfy? Let us know via email at [email protected] or share your comment to this posting.

Optimized architecture is alive in the lab.