How A.I. Is Helping Architects Change Workplace Design

“I’ve been a workplace designer for the last 24 years,” said the architect Arjun Kaicker. “I’ve seen more change in the last 24 months than in the whole of my career.”

Mr. Kaicker co-runs Zaha Hadid Analytics + Insights, or ZHAI, a five-person team that uses data and artificial intelligence to design workplaces. The team is part of Zaha Hadid Architects, the firm founded by the influential architect Zaha Hadid in London in 1979.

“The pandemic has really supercharged innovation in the workplace,” Mr. Kaicker said in a recent video interview from Atlanta.

Before, “the majority of office buildings had a one-size-fits-all desk for everyone, and the same environment around them, the same everything,” he said.

Now that they’re back at their desks, “people are requesting more choice, more personalization and more mobility.”

To address the changing landscape of work, the firm has been turning to A.I. to help its architects design better office buildings, and spaces that cater to workers’ individual needs.

While plenty of architectural firms around the world are deploying digital data in this way — including Foster + Partners, also headquartered in London, and the U.S. firms HOK and NBBJ — Zaha Hadid Architects is among the few to have a dedicated in-house team.

Historically, offices have been designed based on organizational charts to “see who reports to whom and which departments might need to sit next to each other,” and based on observational studies or staff questionnaires, said Jeremy Myerson, co-author of “Unworking: The Reinvention of the Modern Office,” in a phone interview.

Today, with staff members often working from both home and the office, “businesses just can’t afford to have swaths of real estate that are underutilized for the week,” he said. Many businesses are using algorithms and machine intelligence “to have a much more real-time, dynamic reading of what’s happening in the space.”

Sensors track people and environmental conditions — temperature, air quality, noise levels, humidity, carbon dioxide levels and daylight. Architects and workplace designers then cross-reference that data to get a better picture of actual needs.

And they’re putting the data to use: relocating coffee spots and pantries to more popular corners, rearranging furniture and desks, redesigning lighting, seating people at desks that are better suited to their work and using partitions in smarter ways.

Why is workplace design not more common as an in-house department at architecture firms? Because the firms “think of it as commercial and corporate,” said Patrik Schumacher, who succeeded Ms. Hadid as principal of the firm. (Ms. Hadid died in 2016.) They prefer to design museums and housing, when in fact offices are “where wealth and prosperity are generated,” he said.

Since ZHAI, which the architect Ulrich Blum leads with Mr. Kaicker, was set up in December 2015, it has worked on more than 100 building projects, at least 60 percent of which have been offices.

At the firm’s headquarters in London, Mr. Blum explained that, unlike cars and electronic devices, a building in the 21st century is not as responsive and advanced as it could be.

While buildings often have state-of-the-art air-conditioning, lighting and security systems, “to make those systems all talk to each other seamlessly is still a challenge,” Mr. Blum said. ZHAI aims to change that using a host of new tools and technologies.

As he spoke, A.I.-generated office floor plans flickered on a large screen in front of him, with green and red dots representing the most and least desirable desk positions.

Mr. Blum said ZHAI had a computer tool that, in 27 hours, could come up with 100,000 designs for a building’s interior; an architect would have to produce 40 drawings a day for a decade to deliver that many options. He clicked open a flurry of diagrams for the fluid, futuristic Infinitus Plaza in Guangzhou, China, which the firm designed. A.I. was used to come up with options for positioning parts of the building’s core, such as pipes, staircases and elevator shafts.

Privacy is a significant concern when it comes to A.I.-assisted workplace design. If businesses are able to see who does what, where and when in granular detail, they could breach employees’ privacy and use the data against them.

Even though the data is being compiled and harvested anonymously, some worry that “there’s a supervisor inside your machine,” Mr. Myerson said, noting that people were reluctant to have their data collected, even when it related to how offices were being experienced collectively.

He pointed to an example: In January 2016, The Daily Telegraph in London installed desk monitors containing heat-and-motion sensors underneath each employee’s desk to improve energy efficiency in the office, but these were withdrawn immediately after complaints about privacy by the National Union of Journalists, the main journalists’ union in the United Kingdom.

Mr. Schumacher noted the need for safeguards in any tracking systems. “Firms have to be responsible,” he said. “We need to be sure that, when these systems are used in offices to draw conclusions and improve matters, they’re not a kind of alien control system where we’ve been tracking individuals to penalize them.”