In 1910, the French artist Villemard created a series of illustrations imagining life in the year 2000. In one of his drawings, an architect sits in a booth pushing buttons on a console to manipulate a series of machines operating in the usual debris of a construction site. The various machines cut, shape, lift, and place stone blocks to build a house. There are no human laborers in his projection — mechanization has made them obsolete.
Villemard’s vision has not panned out, however. On the contrary, industry observers routinely deride the lack of technological sophistication in the construction industry, and have pigeon-holed it as old-fashioned and lagging behind more forward-looking and purposeful industries such as manufacturing.
This story has been told and told again. In the wake of the post-WWII housing boom, the editors of Fortune published a 1947 article titled “The Industry Capitalism Forgot,” in which they mocked homebuilding’s “feudal character” and “picayune scale.” In 2006, MIT Professor John Fernandez summed up the conventional wisdom when he wrote: “It is widely believed that construction is the slowest of all industries of such scale in implementing proven, scientifically sound technological innovation.” A decade later, McKinsey consultants continued the drumbeat, blaming limited productivity improvements on “poor project management and execution … underinvestment in skills development, R&D, and innovation.” In early 2023, University of Chicago economists Austan Goolsbee and Chad Syverson published an article titled “The Strange and Awful Path of Productivity in the U.S. Construction Sector” in which they concluded that aggregate data “demonstrate a large and decades-long decline in construction sector productivity.”
Anecdotes about the opposition to change paint a similar picture. Regardless of the era, workers who worried that labor saving devices might deepen the inherent insecurity of life in the trades sometimes frowned on the introduction of new machinery on the construction jobsite. Carpenter Joseph Emanuello recalled the first time he encountered electric circular saws: “One guy took the Skilsaw on the roof of the building and threw it off. He said, ‘The saw’s too fast, I’m going to cut by hand.’” The hesitancy to adopt innovation extended to the contractor’s office. When John Tocci, fresh out of engineering school, bought an expensive mainframe computer for his multi-generational family general contracting firm in 1983, his father — the owner — looked at the bulky item and grumbled, “When that freaking thing can lay brick, I’ll learn how to use it.”
But it is easy to exaggerate the level of resistance. In 1993, a researcher conducted interviews with tradesmen about their attitudes toward technological innovation. With few exceptions, they welcomed new tools that made the work safer and less physically demanding and, for the most part, their pride in increased production outweighed their fears of displacement. In general, construction has adopted an evolutionary, not revolutionary, pace of change.
Not everyone wants it to stay that way, however: Villemard has contemporary followers who continue to envision a fully mechanized future for construction, just over the horizon. A 2019 report by Tractica, a technology market intelligence firm, declared that construction is “ripe for disruption” due to its history of technological underdevelopment and predicts that the purchase of construction robots will increase tenfold by 2025. Gaurav Kikani, vice president of Built Robotics, suggests that there will be “an explosion of robotics” that will replace the eight-hour workday with automated activity around the clock. A recent global survey indicated that 81% of construction businesses will introduce or increase their use of robotics and automation in the coming decade. In a 2020 report, McKinsey & Co. futurists predict that 45% of the industry’s value chain will shift from traditional processes in the next 15 years.
What both the optimists and pessimists struggle with, however, is why some technologies take root in the industry — often creating significant gains in productivity — and why others, such as worksite robots, don’t. Understanding this point will help construction companies and the firms that supply them predict which investments will pay off and which will end up as curiosities that sit gathering dust.
Why Certain Tech Doesn’t Take Hold
Many of the inventions breathlessly touted by the industry press remain on the fringes of the industry. State-of the-art six-axis articulated robots may have transformed the auto industry, but their stationery character does not work on a construction site. A robot that is immobile and cannot adjust to the rough terrain and multi-story nature of a building project is functionally useless. Autonomous earth-moving machinery can dig out and prepare foundations. Robots can do simple layout. Drones can navigate job sites and record daily progress. Exoskeletons can relieve the burden of lifting heavy items and YouTube videos portray the wonders of 3D printing. Yet many of these advances remain novelty items, available only to a minority of firms that have the resources and inclination to experiment with equipment and systems that have not consistently been proven to be quality- or cost-effective.
“People think robots are intelligent. They’re getting better, but they’re still pretty dumb,” says Jeremy Hadall, chief technologist for robotics and automation at the United Kingdom’s Manufacturing Technology Centre. “Robots have a place in building productivity in the construction industry, but you have to be realistic about what they can do. Are we going to see completely automated robotic building sites in 10 years? No.” Similarly, Erin Bradner, Director of Robotics at the Autodesk Robotics Lab suggests the current level of sophistication of co-bots may be exaggerated: “They’ll stop when they encounter an obstacle, but you don’t want your project stopping every time a person sneezes.” Reza Akhavian, a professor of engineering at San Diego State University, received a $691,000 National Science Foundation grant to study robotics in construction. “Currently and for the foreseeable future, the field of robotics in general, and construction robotics in particular, are not even close to a state in which robots can replace human workers,” Akhavian argues.
The reluctance to embrace new technology is based on more than the irascible and retrograde attitudes of industry players. Robots and other forms of automation are costly and require an extended time frame before offering a satisfactory return on capital. Given the decentralized nature of the industry and the predominance of small and medium-sized firms, few owners have the financial capacity to purchase expensive paraphernalia that may only provide a profit in the very long term. And while large general contractors may have more cash on hand, subcontractors employ the bulk of the workforce. There is little motivation for one well capitalized company to invest in a technology that will ultimately benefit another smaller firm.
Finally, companies in other industries have traditionally been motivated to spend capital on automation and expensive new procedures to reduce escalating labor costs. Unfortunately, construction executives found simpler ways to cut labor costs through the misclassification of employees as independent contractors, cash compensation, and reduced wages and safety standards — all components of a successful crusade to undermine the union sector in many parts of the country. There is less incentive to buy robots to replace high-priced labor when the labor itself is not as pricey.
Construction’s Digital Transformation
Still, the digital revolution in the twenty-first century has propelled construction, particularly the architectural functions and project management systems. Building Information Modeling (BIM), a software process that transforms traditional blueprints into digital images, has introduced the critical factor of coordination into the previously siloed products from the various design disciplines — architectural, structural, mechanical, electrical, civil — and allowed the management team to essentially build a project twice: once virtually on the office desktops and tablets and a second time with the actual trades workers and materials in the field. Jeff Gouveia, Executive Vice-President of Suffolk Construction, described the pre-BIM work style to me in an interview: “Prior to modeling, coordination was done with a light table with a clear piece of plastic over it. You overlaid 2D drawing on top of 2D drawing on top of 2D drawing. You got them just right, taped them in the corner, made sure that they were in the right scale, and then you basically started circling conflicts, where a piece of ductwork or pipe went into a column or a beam.” Modeling allows for the advance visualization of an entire development from site-work to finish hardware.
Modeling may reduce total construction costs, but it requires upfront investments on a scale that is only available to larger general contractors and a narrow slice of elite subcontractors that can afford the software and, above all, the hiring and training of highly computer-literate staff. Institutional owners like universities and hospitals that expect their projects to last decades are more likely to accept the initial costs of modeling than developers who tend to focus on the bottom line, just surviving through the warranties and short-term flipping goals.
While BIM initially just involved the professionals, modeling has since filtered down to workers in the field to offer suggestions and critiques of the unfolding designs. Gouveia describes the transition on a standard task like laying out the sleeves for plumbing and electrical pipes through a concrete deck. “Before, you’re unrolling the drawings, the wind is blowing, you’ve got a piece of rebar on one side, and you’ve got a brick on the other side, and you’re trying to hold it down. Then you’re scaling it and going over and using your tape. Now, someone’s got a laptop or a tablet or an iPad, with the model on there, with that floor laid out. You’re laid out in a matter of 90 minutes, whereas before that would have taken maybe five hours.”
Though currently limited to larger projects, digitization and modeling will continue to penetrate further into the industry. The 2021 $1.2 trillion Infrastructure Investment and Jobs Act (ILJA) included $100 million for advanced digital construction management systems and related technologies. While the language in the bill is non-specific, the funds are expected to support advanced digital management tools, 3D modeling, drone visualization, and data analysis. In 2016, the use of BIM was mandated on public projects in the United Kingdom and the National Institute of Building Sciences (NIBS) hosted a meeting in early 2021 to consider adopting a BIM standard for the U.S. Modeling is expanding beyond three dimensions to incorporate 4D (scheduling), 5D (cost) and even 6D (facility management).
On the job site, foremen have company-issued tablets as part of their tool kit so plans and modifications can be transmitted instantaneously. The old reliable gang boxes that housed workers’ tools for overnight storage now can also have Wi-Fi, printers and flat screens on the underside of the lid to display the latest shop drawings electronically. Most union training programs have incorporated CAD and BIM into their curricula. New young entrants into the industry from diverse demographic and socio-economic backgrounds may be able to translate their home computer expertise into modeling proficiency in the field. “We are always going to need mechanics to assemble stuff,” John Tocci told me in an interview. “but the mechanic of the future will also be the video game guy.”
While Villemard’s vision may now seem like the stuff of science fiction, there have been steady if undramatic steps to transform the industry. It takes far fewer trades workers to erect a skyscraper than it did even a decade or two ago. Tools and materials are constantly upgraded and BIM represents a novel and exceptional method to design and build a project. Turbulent change is unlikely, however, in an industry that remains fundamentally decentralized. The ability to invest in game-changing technologies is limited to the upper tier of the contracting community. For the foreseeable future, the vast majority of small and medium-sized subcontractors and trades workers will adjust and adapt, not disrupt.
This article is adapted from the author’s book, The Way We Build: Restoring Dignity to Construction Work. It is the first part of a two part series.