Digital Construction: An Integrated Digital Approach to Architectural Processes
This post picks up on a previous article I had written, a presentation for a Green Week lecture titled,Innovations in Construction which goes into further detail on the “Digital Construction” processes discussed in the ending:
There are currently three trends occurring in the architecture-construction industry which will have long-lasting implications on the way buildings are designed, fabricated, and constructed.
The first of these trends deals with waste, where a large amount of waste is generated during the construction phase. Linking digital models to fabrication processes have already begun seeing a reduction in the amount of waste during the early stages of a building, however, unlike fabrication, construction remains relatively far behind in terms of its reduction of waste.
Waste, it is synonymous with human culture and our attempts to create a sustainable future for civilization. But in the age of information so too are concepts like networks, intelligent systems, and big data identifiable with human civilization. What we currently lack in our efforts to achieve a sustainable human habitat we are only beginning to touch upon with ideas like lifecycle building, integrated project delivery, building information modeling, digital fabrication, and energy consumption analysis. Before the rise of the internet [as we know it today] we lacked the ability to rapidly pull information from anywhere in the world and to be able to visualize vast amounts of data to problems that were previously unbeknownst to us as we looked at only a subset of what was actually happening in a larger network of events. Presently we are talking about energy analysis on digital building models, post-construction energy analysis, and building information modeling, but what we lack is a system that begins to have each of the components along the path of construction, the assimilation of matter from the time it leaves the ground until it is finally congealed into a structural form, speak to one another and to be able to visualize across a larger field of study the cause and effects relationships each has on the other. In order to be able to curb waste out of our production cycle for building we must be able to collect and visualize the information necessary along its path.
In order to begin to mitigate waste within the construction industry we must look at the highest efficacy value—the highest being that of prevention. To date the main conversations are surrounding topics of minimization, reuse and recycling, moving away from concepts of energy recovery and outright disposal. Prevention implies a “smart” system, a network that tracks materials [for construction] from the moment material is pulled from the Earth, to its virtual manifestation and configuration inside of a digital model, the amalgamation of a built entity, to the re-use or recycling of its constituent parts after the structure has been decommissioned. Bruce Sterling, a technology theorist and science fiction writer, refers the neologism of the Spime, where unique and identifiable materials and goods are able to be tracked throughout their lifecycle using such technologies as RFID tags. At any time these materials geolocation and other tagged information may be queried. This begins to build a database of known materials in the global production cycle with the ability to track its movement throughout the cycle of production. For instance, within a Building Information Model (BIM), items within the projects digital model may be searched for, giving their precise location either in route to the construction of the project or post-construction within the building. This information is retained throughout the life of the product. The materials catalogued become part of a larger repository of searchable physical objects in the world where we are able to keep tabs on the amount of material within the production cycle and built environment.
Production: Construction v. Manufacturing
The second phase deals with the condition of productivity and efficiency. While the manufacturing industry has continued to climb, namely as a result of increased efficiencies in production due to mechanization and computation, the construction industry has stagnated, flat-lined, and in some circumstances declined over time. As construction and labor costs continue to rise people are looking to the manufacturing industry for ways to cut costs in terms of labor and delivery of the product to their clients. Architectural fees continue to diminish under the weight of increased competition and greater building costs, as firms continue to seek more efficient means of delivering the architectural product. Already this has led many to skip the traditional role of sub-contractors to working directly with the fabricator, where the building model is used directly for fabrication purposes in an effort to save money, control quality, and decrease unnecessary steps of a middle-man. This is leading some to look at what has enabled the manufacturing industry to continue to efficiently increase production at an almost exponential rate – automation and robotics. Today robots are used in manufacturing for material handling, processing operations, assembly and inspection. The adoption of robotics in manufacturing has increasingly made processes more efficient and productive over time. These processes can be directly related to the construction industry as the applications are strikingly similar. Architects are increasingly turning to a robotic workforce to fabricate increasingly complex building facades that are directly tied to their digital models. Similarly, plans are on the horizon to make such things as vehicular shipping and driving more efficient and productive with algorithms by determining the best routes for trucks to take; the military and Google are already utilizing such autonomous vehicles in their operations.
By automating and tracking vehicles (resources), process such as these may be tied directly into a building model with the ability to instruct their movements and operations. As military research and technology trickles down to become consumer products, technologies like Unmanned Aerial Vehicles (UAVs) will become increasingly more prevalent. Combined with advanced cameras accurate digital terrains may be modeled with the ability to tie directly into the digital model. The entire process of site operations becomes automated and tied directly to the digital model, where the UAV monitors in real-time the transformation of the site to create a repository and history of site operations. Already universities across the globe are adopting technologies with the ability to automate fabrication processes, in most cases researching the operations of multiple robots at any given time. As students become increasingly more familiar with these technologies and they are further implemented in education and fabrication stages, students will drive the future market for a proliferation of digital/robotic control. With tools such as these we are able to gain more precision between what is modeled and what is fabricated and constructed. The project titled, Wave Pavilion by graduate students at the University of Michigan, utilized one robotic arm to bend steel rods according to splines in a digital model. Another robot could be used to place in space and spot weld the sections together. What only took this team of robots and students a few days to fabricate and construct, would have taken weeks to accomplish given the complexity of the digital model. It becomes increasingly inefficient at that point to break the chain of automation, stopping at fabrication, and returning to a manual labor workforce for construction purposes.
Matter Lifecycle Management
That the evolution of labor in human society toward that of mechanical-autopoesis is perhaps nowhere less evolved than in that of the construction industry. While manufacturing and fabrication industries have already [in large part] moved toward systems of automation, building construction is by and large a manual operation of human labor. As an increasingly greater portion of the architectural field is moving toward digital processes it makes less sense to break the chain of these processes to be interpreted by a manual workforce reliant on the unknown expertise and interpretive skillset of local laborers. Currently, the means of building construction utilize methods which are falling behind digital capabilities and methods [or processes] displayed during the beginning and interstitial stages of architectural development; i.e. the digital modeling and fabrication stages. Already we are seeing once human labor processes in the chain of production being handed over for mechanized processes –in terms of fabricated elements that were once reliant on the skill of craftsmen. Architects and designers can now send their [increasingly complex] digital models to a mechanized workforce to be crafted/fabricated rather than relying on the unknown skill of local laborers or craftsmen with greater control over the process and precision.
The automotive industry is a prime example, as vehicles not only are assembled, fabricated, and constructed by a robotic workforce, but create a vast information network through the monitoring of materials as they traverse the systems of production in a mostly automated environment. Like the automotive industry processes, not only fabrication but construction too has transitioned toward robotic processes—where vehicles are digitally modeled, fabricated, and constructed using computational means. Similar to fabrication processes which utilize robots, by adding construction to this chain of events driven by digital models, allows the designer to directly influence, instruct, and automate the timing and overall chain of events from design, fabrication to construction. With the utilization of 4D BIM modeling it is not a leap to imagine the construction industry moving in the same way as fabrication [for design] is today, or the processes which the automotive industry have already adopted. What the automotive industry has learned in terms of precision and efficiency the architectural industry can similarly seek to gain. Aviation too has a similar chain of production with complex digital files and simulation models that far exceed the nature of the typical architectural process. Now we are seeing aviation models that are not only monitored during the fabrication and construction phases but have grown to include the monitoring of systems post-construction. This year the large aviation manufacturer, Boeing, will no longer sell jet engines; rather they will lease them, so that overtime they may continually monitor the performance of the jet engines within a larger data-set as opposed to the monitoring of an individual engine—where little information can be gathered compared to the emergent information that will be seen from looking at the entire field of engines operating within the world. Their monitoring not only allows them to see in real-time the performance of their product/design, but how to better re-design future engines, where second-order problems might arise that were not previously witnessed. These precedents share aspirations that are currently on par with those of the architectural community.
This brings us to the third trend occurring in the architectural profession which is perhaps leading toward a digital construction process. BIM software packages may in the future have the ability to tie together the aforementioned concepts of material tracking and robotic fabrication with the aim of being more efficient in workflow. Tied together with energy modeling and simulation practices, we are able to insert more data and create a greater level of detail in our digital models. Like robotics for manufacturing, BIM is increasing productivity and efficiency in terms of labor (on the architectural side). However, our current BIM (Building Information Management) and BLM (Building Lifecycle Management) models only see a part of the project—it pulls them out of their global context of matter. It looks at building processes only in terms of when the building team starts and when the building is handed over to the project owner or operations management. But, what about the processes before and after construction? How are these processes linked through time to build a database, a larger picture, of what is occurring and how the project relates in its local, regional, and global context (the stream of matter-flows). In order to properly utilize BIM and BLM we need to increase the scope of what the building model does and can do; the amount of information it has access to and what information it is sharing. This includes material tracking, automation, fabrication, material delivery, and simulation. Drawings and details in the model become not just representations but to be interpreted by a laborer, but become actual instructional code for robotic production acting in physical space. In other words your model controls a fully automated workforce directly tied to real-world operations. Already in Japan we are seeing systems that are using automation in hi-rise construction. A rig outfitted with robotic armatures moves up each floor along with the construction of the building, controlled remotely by off-site operators and the digital model instruction code. They are able to monitor in real-time the status of the building and control its operations. This is likened more to our notions of manufacturing that that of our traditional industry. Technological advancements in stereo-lithographic printing (3D printing) are increasingly as more people have begun to adopt the technology. Already we are seeing systems such as D-Shape and Contour Crafting which are fully automating the construction process by 3D printing full-scale structures. This year there are already plans to 3D printing three full-scale houses, each vying for the rights to be called the first fully 3D printed livable structure. Last year and this year both American and European space agencies announced plans and partnerships with Universities and architects to 3D print full-scale structure on the moon as bases for astronauts. Further into data collection and automation processes tied to BIM models we see not only robotics playing a large role in changing how we think about design, but that of simulation and the monitoring of the building systems post-construction. These processes are increasingly being used to help shape the building in the design phase, call on appropriate building materials for the context, and monitor buildings throughout their lifecycle. This operability will likely increase in scope to include the simulation of not just climate data, but also that of actual machines on the construction site. Machinery already on site such as cranes and trucks will increase in scope to include unmanned aerial/ground vehicles and robotics. These may all be instructed by the BIM model (like 6-axis robotic arms), where they are initially simulated (for collision detection) within the modeling environment and then convert each operation and building detail in to machine code to be carried out by a robotic workforce.
What is being proposed here is a fully integrated digital model for the process of architectural construction. This includes both the monitoring of materials from geological sequestration to conception as well as a model fully instructed by a digital file that leads an orchestration of events in the process of building. As in fabrication, this translates to the architectural drawing as no longer being merely a representation but becomes actual instruction for a robotic workforce. This includes a host of robotic “species” performing such operations as; site surveying via unmanned aerial vehicles (UAVs), fabrication machines (6-axis robotic arms), on-site and in-site materials delivery (cranes and autonomous ground vehicles [AGVs]), additive and subtractive construction (6-axis robotic arms), material placement and connection (cranes, robotic arms, aerial agent-based robots), site work operations (AGVs), and post-construction autonomous robotics (utilized for maintenance and monitoring operations). What is also integrated is the monitoring of building material and post construction performance. This allows materials to be monitored from the time they are pulled from the ground in their raw element form, to their placement in digital/physical space, to their recycling—tracking and displaying the movement of matter throughout its full lifecycle.
This digitally controlled process would key in on-site deliveries of materials, management of robotic cranes and arms on the site to place components onto the building, and perhaps even control the 3D printing of concrete structures. The entire construction site would become an automated and carefully timed orchestration driven by the architect’s digital model. A large amount of the processes on the construction site today could perhaps be controlled with a greater influence put on the digital model and carried out by a computer-controlled robotic workforce. As our digital models grow increasingly more complex it will become increasingly inefficient to leave the chain of events broken from computationally driven models to manual labor. Buildings today with the use of BIM, parametric modeling, and digital fabrication may include a multitude of customized components that are growing increasingly complex and difficult for unskilled or even skilled workers to construct preceding fabrication (based on time of construction or complexity of the project [numbers of components and connections]). To aid in the precision and timing of these models, robotic/digital construction offers the ability to directly receive information from the digital model, fabricate, and construct the buildings in perhaps a more efficient and precise manner.
Questions that are blatantly obvious when pursing this research then arise as to the not only the technicalities of such a system but one of economics and ethics. Like Foxconn in China [who manufactures parts for Apple], who are gearing up to replace 1,000,000 workers with 1,000,000 robots; we must ask ourselves what happens when 1,000,000 robots displace 1,000,000 construction workers? What happens when these robotics begin to not only take the jobs nobody wants, but begin taking jobs that people may actually want. What does this mean for our economy? The building industry? What happens to the displaced workforce and how are job-re appropriated to work side-by-side in a robotic workforce.