System can 3-D print an entire building

System can 3-D print an entire building
MIT researchers have designed a system that can 3-D print the basic structure of an entire building. The system consists of a tracked vehicle that carries a large industrial robotic arm, which has a smaller, precision-motion robotic arm at its end. Credit: Steven Keating, Julian Leland, Levi Cai, and Neri Oxman/Mediated Matter Group

The list of materials that can be produced by 3-D printing has grown to include not just plastics but also metal, glass, and even food. Now, MIT researchers are expanding the list further, with the design of a system that can 3-D print the basic structure of an entire building.

Structures built with this system could be produced faster and less expensively than traditional construction methods allow, the researchers say. A could also be completely customized to the needs of a particular site and the desires of its maker. Even the internal structure could be modified in new ways; different materials could be incorporated as the process goes along, and material density could be varied to provide optimum combinations of strength, insulation, or other properties.

Ultimately, the researchers say, this approach could enable the design and construction of new kinds of buildings that would not be feasible with traditional building methods.

The robotic system is described this week in the journal Science Robotics, in a paper by Steven Keating PhD '16, a mechanical engineering graduate and former research affiliate in the Mediated Matter group at the MIT Media Lab; Julian Leland and Levi Cai, both research assistants in the Mediated Matter group; and Neri Oxman, group director and associate professor of media arts and sciences.

The system consists of a tracked vehicle that carries a large, industrial robotic arm, which has a smaller, precision-motion robotic arm at its end. This highly controllable arm can then be used to direct any conventional (or unconventional) construction nozzle, such as those used for pouring concrete or spraying insulation material, as well as additional digital fabrication end effectors, such as a milling head.

Unlike typical 3-D printing systems, most of which use some kind of an enclosed, fixed structure to support their nozzles and are limited to building objects that can fit within their overall enclosure, this free-moving system can construct an object of any size. As a proof of concept, the researchers used a prototype to build the basic structure of the walls of a 50-foot-diameter, 12-foot-high dome—a project that was completed in less than 14 hours of "printing" time.

For these initial tests, the system fabricated the foam-insulation framework used to form a finished concrete structure. This construction method, in which polyurethane foam molds are filled with concrete, is similar to traditional commercial insulated-concrete formwork techniques. Following this approach for their initial work, the researchers showed that the system can be easily adapted to existing building sites and equipment, and that it will fit existing building codes without requiring whole new evaluations, Keating explains.

Ultimately, the system is intended to be self-sufficient. It is equipped with a scoop that could be used to both prepare the building surface and acquire local materials, such as dirt for a rammed-earth building, for the construction itself. The whole system could be operated electrically, even powered by solar panels. The idea is that such systems could be deployed to remote regions, for example in the developing world, or to areas for disaster relief after a major storm or earthquake, to provide durable shelter rapidly.

The ultimate vision is "in the future, to have something totally autonomous, that you could send to the moon or Mars or Antarctica, and it would just go out and make these buildings for years," says Keating, who led the development of the system as his doctoral thesis work.

But in the meantime, he says, "we also wanted to show that we could build something tomorrow that could be used right away." That's what the team did with its initial mobile platform. "With this process, we can replace one of the key parts of making a building, right now," he says. "It could be integrated into a building site tomorrow."

"The construction industry is still mostly doing things the way it has for hundreds of years," says Keating. "The buildings are rectilinear, mostly built from single materials, put together with saws and nails," and mostly built from standardized plans.

But, Keating wondered, what if every building could be individualized and designed using on-site environmental data? In the future, the supporting pillars of such a building could be placed in optimal locations based on ground-penetrating radar analysis of the site, and walls could have varying thickness depending on their orientation. For example, a building could have thicker, more insulated walls on its north side in cold climates, or walls that taper from bottom to top as their load-bearing requirements decrease, or curves that help the structure withstand winds.

The creation of this system, which the researchers call a Digital Construction Platform (DCP), was motivated by the Mediated Matter group's overall vision of designing buildings without parts. Such a vision includes, for example, combining "structure and skin," and beams and windows, in a single production process, and adapting multiple design and construction processes on the fly, as the structure is being built.

From an architectural perspective, Oxman says, the project "challenges traditional building typologies such as walls, floors, or windows, and proposes that a single system could be fabricated using the DCP that can vary its properties continuously to create wall-like elements that continuously fuse into windows."

To this end, the nozzles of the new 3-D printing system can be adapted to vary the density of the material being poured, and even to mix different materials as it goes along. In the version used in the initial tests, the device created an insulating foam shell that would be left in place after the concrete is poured; interior and exterior finish materials could be applied directly to that foam surface.

The system can even create complex shapes and overhangs, which the team demonstrated by including a wide, built-in bench in their prototype dome. Any needed wiring and plumbing can be inserted into the mold before the concrete is poured, providing a finished wall structure all at once. It can also incorporate data about the site collected during the process, using built-in sensors for temperature, light, and other parameters to make adjustments to the structure as it is built.

Keating says the team's analysis shows that such construction methods could produce a faster and less expensively than present methods can, and would also be much safer. (The construction industry is one of the most dangerous occupations, and this system requires less hands-on work.) In addition, because shapes and thicknesses can be optimized for what is needed structurally, rather than having to match what's available in premade lumber and other materials, the total amount of material needed could be reduced.

While the platform represents an engineering advance, Oxman notes. "Making it faster, better, and cheaper is one thing. But the ability to design and digitally fabricate multifunctional structures in a single build embodies a shift from the machine age to the biological age—from considering the building as a machine to live in, made of standardized parts, to the building as an organism, which is computationally grown, additively manufactured, and possibly biologically augmented."

"So to me it's not merely a printer," she says, "but an entirely new way of thinking about making, that facilitates a paradigm shift in the area of digital fabrication, but also for architectural design. ... Our system points to a future vision of digital construction that enables new possibilities on our planet and beyond."

Explore further

Modern construction using long-forgotten techniques

More information: Steven J. Keating et al, Toward site-specific and self-sufficient robotic fabrication on architectural scales, Science Robotics (2017). DOI: 10.1126/scirobotics.aam8986
Journal information: Science Robotics

Citation: System can 3-D print an entire building (2017, April 27) retrieved 20 October 2019 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments

Apr 27, 2017
There go the construction jobs. Unless they can become construction robot operator/maintainer/developer jobs. Like Japanese auto workers did in the early 1980s and USA auto workers largely didn't until much later.

Robots can increase human labor productivity instead of just replacing human labor. It's up to the humans.

Apr 27, 2017
I think they should dial down the construction speed and up the quality. The walls seem awfully rough and the step width very coarse.

Integrating insulation and plumbing into the walls seems a natural first step (Especially plumbing. If it's printed solidily into the wall structure then there's no real point where it could break or leak)
Add easy access tunnels for power/data lines to the print file and you have something there.

Next step would be a software that lets you build your dream home from basic room building blocks where shapes and sizes are changeable and the software automatically recalculates the layout of plumbing/tunnels/insulation.

From an engineering standpoint I'd put the printer on the outside (possibly able to cling to the walls its printing or just on a crane setup). Especially once you want to be able to print floors/multi-storey buildings that's a must.

Apr 27, 2017
The reporter, David Chandler, says "the intention" and "the idea" are to operate in remote areas, in developing economies, after disasters. If that's the MIT Mediated Matter group's intention and idea not just Chandler's, then speed will be prioritized over quality. At least for a while, as the early machines will be very expensive and the building users able to pay (even through social subsidies like taxes or charities) very little. I expect the military will be an early user, especially as developed by MIT, so again speed rather than quality.

But those are the early adopters beta testing the revolution. Once the machines are cheaper, more reliable and interoperate with standards their mass market versions will prioritize quality over speed. Because with minimal human labor the speed won't be as important. Though if these building printers make the process really cheap, durability might no longer be as important.

I think they should dial down

Apr 27, 2017
Yes, integrating insulation, plumbing, wiring (power and network) is an important step. Each construction trade will be automated. This machine doesn't seem to print any finished interior or exterior layer (carpeting, paneling, paint, etc), but those will come, along with fixtures like lighting, once the structural process is proven economical.

Integrating solar will make it very cheap. The onsite labor is now something like 75%+ installed cost, slowing the remaining decline of consumer pricing. Scaling to 10TW by 2030 requires this kind of automation.

The construction industry is very conservative in techniques, making it resist change, but also ripe for disruption. But even the 3D printing "industry" is barely started integrating multimaterials into printing, like plastic + metal. It'll be 10 years before this tech is ready, and another 10+ before it's mainstream construction.

Integrating insulation

Apr 27, 2017
One downside of this tech is eliminating wood from construction. Wood in buildings sinks a lot of CO2e that would otherwise rot back into the atmosphere or never get photosynthesized. Also wood is comfortable for human occupation, moreso than concrete.

I hope a spraying machine like this one gets hybridized with processes that assemble sticks and sheets while retaining its efficiencies and effectiveness.

It would be interesting to see the Greenhouse impact of switching to this tech in the mainstream large scale, netted against reducing the sinking of lumber.

Apr 29, 2017
"Integrating insulation and plumbing into the walls seems a natural first step (Especially plumbing. If it's printed solidily into the wall structure then there's no real point where it could break or leak) "

"Yes, integrating insulation, plumbing, wiring (power and network) is an important step."

Integrated power and plumbing will also be unmaintainable, so when e.g. the network wiring becomes obsolete you can't easily replace it without harming the integrity of the structure i.e. taking a jackhammer to it. Plumbing, same deal, plastic pipes degrade, metal pipes corrode, or both can and do silt over from mineral deposits and become unusable in some decades of use.

Even the plastic insulation of electrical wires will degrade and the wiring needs to be re-done eventually.

If the structures are built to be torn down and remade every so often, 20-40 years or so, then the integrated solution makes sense. If they're built to last, not so much.

Apr 29, 2017
"The construction industry is very conservative in techniques, making it resist change"

That's because they're carrying huge liabilities, and the results of new "experimental" construction styles are only visible after decades. Some of the energy efficiency standards and regulations imposed on the construction industry in the 70's for example caused huge damp/mold problems to develop and become apparent in the 90's and the construction companies were held responsible for it while the real culprit was the regulators who told them to wrap everything in plastic to stop air leaks and make buildings like bottles.

While the politicians would like to try everything and regulate the heck out of anything, they're usually not considering the fact that people have to live in these buildings, and sick houses make sick people.

Apr 29, 2017
Integrated power and plumbing

This is false. Pex pulled through empty wall space typically lasts 40-50 years, or more if it's higher quality. Sealed inside closed cell foam for example it could last over a century. And when its lifetime starts to end it's easy to just install new piping, even into sprayfoam by easily routing a new channel then resealing it for the next century. Unless in 2117 there's a better alternative. Meanwhile locked into insulation it's even more energy efficient.

Wiring is even more suited to this treatment.

Where are you getting these myths?

Apr 29, 2017
That's false. Regulations don't specify airsealing any building. Everyone airsealing a building either installs mechanical ventilation and humidity control (typically integrated into single spot units or distributed through ductwork). When they don't it's because they're cheaping out either deliberately or incompetently. There's a lot of that because the builders are long gone by 20 years later. The obvious remedy to this is regulated building codes and inspections, because the building owners failed to protect themselves in the "free market".

Where are you getting these myths?

That's because[/q

May 02, 2017
"This is false. Pex pulled through empty wall space typically lasts 40-50 years"

That's completely besides the point. How useful are 20 year old network cables in the walls, when the standards have moved on?

"And when its lifetime starts to end it's easy to just install new piping"

If it's an integrated structure inside a concrete and foam wall, you have to break the wall to replace the pipes, or install new pipes on top of the wall.

" Regulations don't specify airsealing any building."

Building codes have long demanded vapor and air barriers due to energy efficiency, and the need to adhere to tighter energy standards needs more and more extensive sealing because letting air move through the walls and insulation increases the need for heating. The moisture problem is then "dealt" with by mechanical ventilation, but you still get damp spots in the structure where the ventilation is weaker and the houses eventually rot.

May 02, 2017
"The obvious remedy to this is regulated building codes and inspections, because the building owners failed to protect themselves in the "free market"."

Yes, more bureaucracy and cost on top of bureaucracy and cost. That'll fix it.

The problem is that nobody was competent in the 70's when these codes and regulations first came out, because wrapping buildings in plastic was a new thing, energy efficiency was a new thing, and nobody had any experience in how it would actually turn out - so 20 years later we got the results: it didn't work.

So it took another decade for anyone to admit that anything was wrong, and revise the codes, and still they haven't gotten it quite right...


May 02, 2017
The main issue is that the pipedream of a "zero energy home" which is kept warm simply by the body heat of the occupants and the appliances they use, quite literally demands you to live in a dewar/thermos bottle. The end result is air quality problems and mold, unless you somehow build ventilation ducts inside the walls to air out every corner and nook and cranny mechanically - yet, without losing any heat to the outside or spending any energy to push the air around.

The other option is massive overuse of materials and inefficient construction - like 5 foot thick straw bale walls, which you obviously can't use in an apartment building because the interior space would become approximately 5 square feet in turn.

So the clueless are telling the incompetent what to do, and the rest of us have to live within the results - that's why the construction industry, or at least the more honest companies, prefers to take things slow.

May 11, 2017

If old cables become obsolete, they can be replaced by routing a groove in the old wall. That's if future drilling robots aren't the solution, or something better.

For plumbing, the PEX will not become obsolete because of changing standards, but routing or drilling robots etc is just as viable.

Also, sprayfoam buildings don't rot, certainly not as quickly as wood buildings exposed to elements and (micro)biology do.

Please post some links that back up the outlandish claims you're making.

That's completely

" Regulations don't specify airsealing any building."

May 11, 2017
No, building codes are not on top of bureaucracy and cost, they're on top of the "free market" that didn't protect property well enough so it had to be forced on the market.

What problem? Please post links to unbiased data backing up your claim that in the 1990s many buildings had rotted quicker because they adhered to building codes. Not just because they were built cheaply or just stupidly while others were not. You are citing the Chicago Tribune's quote of someone from the Engineered Wood Association, which hypes products to resist mold, and a New Orleans lawyer who sues people for mold - speaking from vested interests. Your other story is a single building in 2001. Where's reliable data that indicts the codes rather than bad construction, not marketing?

And in any case, isn't your point simply that doing new things has risks from incomplete knowledge? But your lesson is that we shouldn't do new things, or regulate them?

Yes, more bureaucracy

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more