Prefabrication has been an architecture industry obsession for decades, a promising technique in the race to build better, faster, and stronger. In the dogged quest to use this technique, architects and developers have run into problem after problem, yet still remain unflinchingly confident that prefab is the path to the building of the future—as shown at 461 Dean, the 32-story modular high-rise designed by SHoP Architects and built by developer Forest City Ratner Corporation (FCRC).
Though it made headlines for aspiring to be the tallest prefab tower in the world, the building became a boondoggle, plagued by lawsuits, allegations of a flawed design, and stop-work orders that put the building years behind schedule. The project finally wrapped up construction in November, a staggering two years behind schedule.
After so many problems and delays, do the architects and developers believe prefab is still the panacea for our urban housing crisis? Co.Design spoke with SHoP and FCRC for a post-mortem on the project—and to find out what the future holds for modular design in the context of tall buildings.
It's an overly simplistic analogy, but constructing a modular building is a lot like snapping together Lego bricks. In a traditional structure, contractors erect the structure from the ground up on the building site; in a modular design, the structure is broken down into smaller sections—modules, which can be different sizes and shapes—that are built and finished in a factory, slotted into a brace frame, and bolted together. Then the units are all wired for mechanical, electrical, and plumbing systems.
"Modular design as a concept is sort of the holy grail for architects," says Susi Yu, executive vice president of residential development for FCRC.
If done correctly, modular prefab should slash construction times since site work—like building the foundation—and module fabrication can happen simultaneously. Normally you'd finish one step before embarking on the next in a traditional building. And since the majority of prefab construction work takes place in a factory, it's supposedly easier to perform quality control checks, and weather—like rain or snow—won't contribute to construction delays. It boils down to time and predictability, which both factor into the bottom line.
"There’s a history here, there’s a lineage—it’s not just about buildings," says architect Christopher Sharples, cofounder of SHoP. He cites how the aviation, automotive, and ship-building industries have all changed their manufacturing processes over the last century: Henry Ford pioneered the assembly line so he could produce cars at a faster rate than before. Industrialist Henry Kaiser found a way to build WWII Liberty ships—which often took over 200 days—in just 10 days. "Why can’t we innovate in the same way?"
The construction industry as a whole is notoriously slow to change. In the United States, it is slowly beginning to adopt new techniques; for example, architects are increasingly experimenting with mass timber tall buildings. Modular prefab has been used extensively for single-family homes, college dorms, low-rise affordable housing, and disaster-relief housing. Skyscrapers haven't entered the picture until recently.
The problem is rooted in the simple fact that research is expensive—and it's often challenging to convince clients to go the uncharted route since it can often be more expensive and laden with uncertainty. "Building anything over 10 stories in modular is something no one has wanted to do because you have to invest in research and development," Yu says. "There’s science behind it that you need to figure out."
FCRC viewed 461 Dean as a test case for how the entire development around it, today known as Pacific Park, could potentially be built. After refining the first building, each subsequent building would theoretically be easier and more affordable to build—a product of economies of scale. "It was only through looking at that pipeline of seeing how big the project is, that we could invest that energy and effort up front and spend what was frankly a lot of money just researching and figuring it out," says Adam Greene, vice president of residential development at FCRC. "The building itself couldn’t afford that R&D."
In a city renowned for its skyscrapers, Brooklyn has escaped the drive to build ever taller—until now. While the borough's urban fabric has been defined by low-rise brownstones and row houses, it's now undergoing a growth spurt as New York's real estate market searches for ways to add density and infill. This year, the borough approved its first 1,000-foot-tall skyscraper and there are now 19 residential towers under construction or recently completed—including 461 Dean—along Flatbush Avenue, one of the main thoroughfares cutting through the central business district. It's a dramatic change, considering that the tallest building in the area up until 2010 was a 37-story bank built in 1929.
461 Dean is part of Pacific Park, a master-planned, $4.9 billion development that will eventually include 16 residential towers holding 6,400 apartments. The 22-acre project was initially called Atlantic Yards—so named for the railyard it's constructed over—and began in the mid-2000s with architect Frank Gehry attached to the project. The land was acquired through eminent domain, which was a highly controversial move in the neighborhood; community groups sued and vocal residents refused to vacate their apartments. By 2009, Gehry was out as architect and SHoP was named to design the Barclays Center—the arena that anchors Pacific Park.
Because of the development's scale and the rising construction costs in New York, FCRC began looking for ways to systematize and streamline the construction process. The same year, FCRC and SHoP revealed the first renderings of 461 Dean Street, which was then known as the B2 tower, and announced that Skanska, a prefab builder, would be contracted to manufacture the modules. The developer and builder set a projected completion year of 2012.
Stepping inside the tower, you wouldn't know that the building was prefabricated due in part to variations in the design, which contributed to the fabrication challenges. Because of zoning, the facade has a number of setbacks, which resulted in 23 different apartment configurations. Working with the engineering firm Arup, SHoP developed a brace frame system that secures the modules in place, which was the tower's big architectural innovation. In terms of the modules, SHoP needed to make sure that each part of the building could fit on the back of a semitruck, and developed a virtual model to understand how they would all fit together. After figuring out all of the calculations, SHoP and Arup handed over the virtual model to the contractor who then translated it into 3D construction documents to Skanska. The modules themselves were made from steel frames, fabricated in Virginia, and fitted with facade panels that came from Minnesota. At a factory in the Brooklyn Navy Yard, the contractors then completely built out the interiors—kitchens, bathrooms, closets, windows, etc.—and then craned them into place. Only the floors for each apartment were installed on-site; everything else arrived complete.
Yet despite the technical innovations developed for the project, things did not go as planned, and in 2014 Skanska halted construction completely as the stakeholders traded lawsuits. Skanska alleged that the design was flawed—for example, the tolerances were so small that tiny errors compounded to cause alignment issues—and that it could not guarantee its performance. FCRC counter-sued, saying that the builder did not have the right expertise to complete the job. Skanska terminated its contract with FCRC in September 2014. Eventually FCRC took over the prefab factory that had been built for the project in the Navy Yard, and began acting as both the developer and contractor.
In October 2016, FCRC announced it had sold off the prefab factory to a new company called Full Stack Modular, founded by Roger Krulak—formerly senior vice president of modular construction at FCRC—effectively ending its involvement with the grand prefab experiment. Now, a decade after New York City approved Pacific Park, all that's resulted is 363 units, half of which are market rate and half are designated to be affordable. It's not much to show, considering myriad obstacles.
What went wrong? Examining the relationship between architect, developer, and builder holds some insight. It seems that prefab is not just about designing a new system and developing new technology: The whole industry needs to change its process. Additionally, choosing which projects and sites are best suited for prefab is crucial.
While SHoP and ARUP worked closely on the digital model for the building, it was up to the contractor to turn that into a set of building instructions. There was likely information lost in translation. SHoP had previously used modular construction for Barclays Center—the arena adjacent to 461 Dean—and was responsible for both the design and creating the factory tickets. That went relatively smoothly. Since the architect, developer, and contractor were not involved during the entirety of the build at 461 Dean, that may have added to the challenge. "If Forest City and SHoP did the project, it probably would’ve been a different story," Sharples says. "I’m very happy with the quality and the look of the building. And at the end of the day, it’s not about product innovation, it’s about process innovation."
Skanska also says that there ought to be more communication between design and the factory floor. In a written statement, a spokesperson from the modular builder said:
The challenge with a modular building is that all of the building’s major components, whether constructed in the field or the factory, must come together and be assembled at one location much like the frame and components of a car comes together for assembly on the factory floor. Workers on a factory floor should not be improvising or force fitting components. Accordingly, the designer of a modular building, like the designer of car, must not only consider how each component will be designed, but how all of the components will fit together and ultimately be assembled into one, integrated whole. As a result, the modular building designer has a much greater obligation than the designer of a conventional building that simply includes some prefabricated elements.
When Skanska made its agreement with Forest City, Forest City had claimed that it had "Cracked the Code" of modular design and construction. Skanska had no design responsibility for the B2 Project. In fact, Forest City agreed that Skanska was entitled to rely upon the information supplied to it by and on behalf of Forest City. Thus, Skanska expected to assemble a properly designed modular building, not engage in an iterative R&D experiment.
Co.Design spoke to Krulak after the Full Stack Modular announcement, and while he remains optimistic about the future of modular construction in an urban context, he emphasized how important it was for the design, development, and construction teams to all be involved at the very beginning of a project—which is different from how 461 was conceived. "They’re three distinct silos that interface at different points in the building process," Krulak says. "It needs to happen at the beginning and it has to be a complete design build process."
Sharples agrees that moving from a linear to an integrated design-development construction process is important for the future of construction. The firm is currently working with the tech company Dassault Systémes to develop 3D software that solves this challenge. "We see this as an iterative build process," Sharples says. "It's how we take these technologies and see them as a means to be able to collaborate and move from this linear process where you do the design and you hand it off to your engineers, your engineers hand it off to the builder, and the builder hands it off to the contractor."
At the project's outset, it took the factory (managed by Skanska at the time) two to three weeks to build a module. By the end, under FCRC's management, the builders cut that down to six days. "The project took a little longer than expected and cost a little bit more than expected because we started the project with the wrong contractor," Greene says. "That’s what it comes down to: A dispute with them about how the project should have been built and how they were performing. Since they left the factory and we took over, we've achieved incredible cycle times ourselves."
Beyond that learning curve, some of the challenges with 461 Dean originated in its location on an irregularly sized lot. The site tapers into a sharp angle—the apartments with the best views are actually located in this section—and zoning setbacks dictated a stepped facade. "Next time we would evaluate [using] modular construction on a more straight-up site, meaning it would have less extraordinary geographic conditions, like the extreme wedge," Yu says. "I think we would be a little more thoughtful about stacking and not doing so many floor-plate changes because every time you have to change a line of apartments and you change the design, you have to transfer mechanical, electrical, and all of your building systems. Looking at a site where you can maximize the efficiency of modular is really the way to go in the future."
So while FCRC believes that it's iterated and solved many of the fabrication challenges that plagued the project, it doesn't look like Pacific Park will have another prefab high-rise. In 2014, the developer sold a 70% stake to Greenland Group, a Shanghai-based developer, which likely will not use modular construction in the remaining towers.
FCRC is still reeling from its grand experiment in prefab. On a November 2016 earnings call, FCRC CEO David LaRue announced a $430.9 million quarterly loss, largely as a result of Pacific Park.
Yet though 461 Dean turned out to be a failure from the standpoint of building faster and more cost-effectively, there may be a future in modular "semi-tall" buildings. ARUP and SHoP's model showed that their structural system could support a modular tower up to 60 stories, but Sharples thinks shorter structures might actually go further in the context of New York City. After 15 stories, extra brace framing is needed to support the building. Stay under that height, and the building's complexity is significantly less. "So if you keep the building under 15 stories, there’s a good case for that," Sharples says. "When you’re talking about the five boroughs, that’s not a bad height to be chasing."
Smaller-scale multi-family developments in New York have successfully been built with modular techniques. Capsys—which manufactured the modules for Carmel Place, a 10-story 55-unit micro-apartment building—has successfully built affordable housing in Brooklyn using prefab modules. Manhattan recently welcomed its first entirely modular apartment building in the seven-story, 22-unit structure by Gluck+, called the Stack.
"I think everyone is waiting to see how the market responds," Yu says. "This is a very disruptive technology within a field that is very traditional. I think to really change the industry and the way people have been building, I think it's going to take a while, but we’ve been having conversations with cities, other developers, and other construction companies who are very interested in what we’re doing."
As the 461 Dean saga shows, there's still plenty of headway to be made with modular building, starting with understanding the process as a collaboration between all stakeholders and evaluating the right techniques to use for a given site. If cities struggling with housing shortages and affordability are going to solve their problems, finding less expensive ways to build and streamlining the process will be essential. SHoP and FCRC still remain optimistic in prefab's potential to change how we construct buildings, but there's also a lesson to be learned: Skyscrapers might not be the best use-case for prefab—and just because you theoretically can build that way doesn't mean you should.
On 12/13/16 this story was updated to include comments from the contractor Skanska