Over the course of the last century, urban planners and architects designed cities as top-down and permanent systems. In Soviet cities, even mid-level architects could fix locations and exact types of all services and shops in neighborhoods. Residents had no choice but to adapt to these plans.
Today we learn to perceive cities and spaces as dynamic, rather static and permanent systems. Let’s explore how new technologies could help us better design and run buildings and cities.
Design Stage: A/B Testing
It’s common to use 3D visualizations and renders to present architectural proposals. Unfortunately, there is a growing gap between sexy renders and actual building quality. Visualization has become a sales instrument rather than a tool for learning about future building performance.
Virtual reality (VR) may soon bridge this gap by helping clients and architects truly live in the potential buildings and test different options and scenarios. A number of firms are working on making design in VR as seamless as possible: IrisVR, ShapeSpark, and Floored.
Beyond applied design, VR creates opportunity for affordable behavioral studies and A/B testing of spaces. A/B testing is jargon for experiments in marketing or user interface development. Its goal is to compare user behaviors in different environments and maximize desirable outcomes. This way, we could rapidly learn without building physical copies.
Post-Occupancy: Real-Time Analytics
After a building is constructed, people have to adapt to it. As time passes, we need to learn whether a space is comfortable for people and is still used in an efficient way.
One of the first firms working on post-occupancy indoor analytics was DEGW (acquired by AECOM). The bottleneck of their approach was the consultancy business model based on human observations. It was simply not scalable.
Today a number of startups work on automation of post-occupancy space analysis:
- Robin leverages workers’ cellphones and beacons to improve the UX of space booking and suggest improvements in office scheduling and organization.
- Building Robotics and build.Science integrate into existing office infrastructure to aggregate and analyse data in a single software hub.
Learning about how a particular person, team, or a whole company uses their office will help create more comfortable and productive environments.
Post-Occupancy: Spacial Reconfiguration
Post-occupancy analytics will eventually suggest improvements to the spatial layout of spaces. These suggestions could change over time due to new tenants, company scaling, and multiple other new use scenarios.
Unfortunately, existing demountable wall technologies require messy construction. Reconfiguration is too expensive to make frequent changes economically feasible.
Frank Duffy, founder of DEGW, in his book “Work In The City” calculated that the costs of space plan reconfiguration could exceed construction costs over building lifetime.
At Dom, we’re working on a solution for this problem. We develop a new type of indoor modular system to make spaces seamlessly reconfigurable.
Coupled with post-occupancy analytics, the goal is to make buildings dynamically adapt to users, not vice versa.
A number of universities have dedicated research groups focused on experimental technologies for space reconfiguration: Changing Places Group at MIT, TUDelft, and Bartlett Interactive Architecture Lab.
Agile Planning & Big Data
At city scale, urban planners face similar problems. It’s extremely hard to forecast user behaviors before a space is built. Again, people have to adapt to spaces.
Over the last 20 years, the awareness of the need for better pre-planning spatial analysis gradually evolved. Space Syntax pioneered work on software for spatial analysis and visualization.
Even though these analytical approaches and tools were around for a while, they never became de facto standards in planning and design process. At the time, they required installation of special monitoring equipment for a particular project.
The increasing number of aggregated data from various sources (cellphone carriers, delivery companies, social media, banks, etc) has potential to make these analytical tools more sophisticated. Over time these tools will become more and more affordable because they leverage pre-existing infrastructure and data.
Long term, urban planning should adopt agile practices (by analogy to software development). Plans should become iterative and adjusted over time based on experiments and factual data.
Of course, a lot of urban infrastructure like subways, airports, or bridges will remain permanent. At the same time, more attention should be dedicated to temporary infrastructure and non-permanent uses of space.
A number of startups decided to focus on filling temporary vacant real estate; AppearHere, MeanwhileSpace, and 3Space connect vacant commercial real estate with pop-up retail shops or temporary art projects.
Big data makes agile planning feasible because it is becoming possible to catch neighborhood trends early on and prepare for citizens’ value changes. Urban big data can help conduct experiments at a neighborhood scale, consistently measure the results, and learn.
For example, Habidatum created Moscow emotional maps by analyzing social media. They were able to surface neighborhoods where residents felt comfortable, and by contrast, most insecure at different times during the day.
We no longer have to pretend that buildings and cities are permanent. Uncertainty and change are inevitable and interesting parts of urban life. Technology can help us make smarter design decisions, better test our assumptions before construction, learn to support natural human behaviors in spaces, and discover urban trends early on. Soon cities and buildings will dynamically adapt to people, not vice versa.