HOW, FINALLY, TO FIX CONSTRUCTION
“You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”
-Buckminster Fuller
Throughout the 20th Century the AEC industry struggled to deliver solutions that simultaneously address two of our greatest challenges:
- We must build desirable housing, faster, and more economically, and
- Our industry has to reduce its impact on the environment.
Something needs to change; but the construction industry is actually getting less productive, not more (McKinsey & Co.). Without comprehensive innovation in process, designs, materials, and systems, construction stands little chance of addressing either the housing or climate crises.
Against that backdrop, it’s no surprise that the innovators within AEC space are increasingly looking to advanced manufacturing techniques, which have delivered dramatic improvements to the efficiency and economics of many American industries. Despite a complicated history, utilizing manufacturing practices to deliver a prefabricated architectural design remains the clearest path to increasing productivity, even though there are skeptics about how well it can be applied.
The most serious attempts at prefabrication have been focused on delivering single family housing to suburban developments, components to replace light frame construction, and manufactured homes (including tiny homes and Accessory Dwelling Units). For the purposes of this post, I am not alluding to theoretical paper architecture, spec projects, or exhibitions, but projects that took the form of businesses — ones that had to operate within the economic, regulatory, and social constraints of a commercial enterprise.
For some reason there is still some misconception that the prefabrication of buildings is a new idea and untested, which is not the case. John Manning first sold his “Portable Colonial Cottage” in 1833 in Australia. Famously inspired by Henry Ford’s Model T, William Levitt and sons developed their “reverse assembly line” to build Levittown with a 27-step process executed by unskilled labor that output 30 houses per day. In the Post-War era both the Lustron Corporation and General Panel Corporation were both well-funded and well-supported endeavors at creating fully prefabricated systems, but they never became cost competitive with cheap materials and on-site labor. Many contemporary homebuilders are prefabricating components of their buildings to a certain extent, only it is considered part of the process and not necessarily advertised as such. I hope one day to go into a deeper dive of the nuanced successes and failures of these endeavors, particularly related to impacts on climate and racial disparity. However, the point that I am making today is that low-density housing is limited in its ability to address our challenges, and the efficiencies of manufacturing have been tested and calibrated over the past 150 years, and are not going to be able to increase productivity much further. As long as the construction industry is subsidized by low wage labor, adding material and steps to the process in order to make it prefabricated will not be cost competitive.
In order to meet housing demands, while limiting our impact on, if not contributing solutions to climate change, we must build for density and we must design for performance. This means our efforts need to be focused on mid and high-rise buildings, not single family housing. The quantity of prefabricated projects in this market that have been built are miniscule relative to the overall market size, and the re-use of those systems on more than one project (which is where the value is created and benefits achieved) is practically non-existent. Despite the fact that conventionally constructed projects run into cost overruns, supply chain delays, and contractual disputes regularly, when those same failures occur on prefabricated projects, the use of the new system is scapegoated as the reason, restricting its further adoption before its been executed to its full potential.
I understand why it is difficult to believe how principles of industrialized architecture can be applied to this market, as it was, until the past few decades, the complexity and technical requirements of taller buildings had too many variables for practices of mass production to accommodate. However, I would like to further the argument that this thinking is outdated, and that new tools in computational design and information management make mid- and high-rise markets the most appropriate for applications of prefabricated architectural systems. It is because they have more regulations and more technical designs, with higher cost skills, materials and labor, that, with the larger pipeline, they can become cost competitive to move into a manufacturing facility and make the investment on pre-planning and pre-design.
Architects and façade fabricators have been utilizing principles of mass-customized design and manufacturing for complex enclosures for decades now and it is no surprise that facades are one of the systems that have improved the most in performance, flexibility, and precision in those same years. It begs the question, if we can create sophisticated, highly parametric designs for the enclosures, why can’t we deploy those exact practices on the systems and components of the building that are less subject to change, less visible, and can be re-used on multiple projects? Then we may find opportunities for efficiency and innovation to capture value from the extensive framework of 3D modeling, computational modeling tools, document generation, and data capture that high profile firms often set up on a per project basis for the more decorative elements.
It was my privilege for many years to be one of the many people at SHoP Architects who worked on developing these practices and it was through our direct to fabrication projects that we came to the collective realization that the benefits of new technologies were reduced and occasionally eliminated when applied on limited scopes of work within a project instead of across its entirety. Assembly OSM is a culmination of those repeated efforts and lessons learned:
· The same tools and design strategies that allowed the SHoP team to model and deliver just under 3,000 highly parametric frames for the Botswana Innovation Hub can easily generate a model and the fabrication data for structural frames of a building.
· Computational codes that we developed to check for missing manufacturing data or abnormal tolerances across the bill of material for thousands of parts can be designed out of the system and applied across components used on every project.
· The weeks spent collaborating with the façade manufacturer and their team outlining processes and coordinate how to pull the information they need from our model into the machine, along with helping to refine training on the system, won’t be shelved at the end of the project but will continue to be useful in improving every single project after they are onboarded to our process as qualified suppliers.
· Structural frames, fire protection, mechanical, electrical, and plumbing have clear performance requirements and more consistent rules which become cost effective to codify when reused.
· The digital twin can reduce redundancies in documentation and speed up design resolution by populating the design with known data and geometry in the earliest phases of the project.
· Augmented and virtual reality applications to communicate assembly details increase in value when they are applicable to details that occur in every project.
· Validating process and performance of assemblies and systems through more detailed prototypes and performance monitoring during occupancy become sources of saving costs by way of continuous improvement.
These are only a few of the applications that can be taken directly from our experience in high precision façade design and collaboration with manufacturers and applied to an entire building system. However, there are other significant changes required along with these technologies for them to be successful.
First, there must be a comprehensive solution through the entire process. It may seem overly ambitious for Assembly OSM to take responsibility for the full contract of delivering a building, but the evolution of this industry cannot be fragmented across multiple stakeholders with differing objectives. The multitude of teams, methods of communication, processes, and softwares used across a single project mean that any innovations adopted by one team do not have a significant impact on the overall project performance and are hard to measure, reuse and adopt best practices. There needs to be a unified vision and tightly-coordinated execution, which we are excited to take the lead on.
Second, we must find the right balance of standardization and customization. It is imperative that we align the variable parameters of our subassemblies with the capabilities of the technology and provide custom options in a controlled way. For the client this means enabling customization where it matters most, the massing, enclosure, and interior finishes, but constraining options for the engineered systems within. For our manufacturing partners it means working with them to determine where flexibility is readily handled through computers in machines, i.e., part lengths, and standardizing and optimizing components that require a high level of processing. These design choices require months of work and collaboration and therefore need to be utilized on every project to distribute the costs across the entirety of the pipeline.
Third, there is a standardization of protocol, documentation and data collection that is critical to execute and employ within the technology of a database and code, to reduce variables. This type of standardization is often met with a resistance in our profession as it either can interfere with the artistic sensibilities of architects or the expertise of the builder. I am not one who thinks our professions will easily be replaced by computers or machines, but we need new ways of collaborating with them, and must embrace this to stay relevant and best provide our services and critical thinking where most needed.
Finally, and of critical importance, the creation of the team who delivers on this vision needs to cast a wider net than we are used to seeing in our profession. Architects, engineers, and contractors are trained to deal with a wide variety of conditions and materials and to continuously learn new things for each project. Therefore, the application of many of these technologies and design for manufacturing principles have previously been applied to our industry by those trained in one of these fields but are hobbyists in the ideas and applications of the others. I fully own up to being one of those people. We need to hire and empower manufacturing engineers to outline our assembly processes, computer engineers to write our codes, and mechanical engineers to design our details. This is not to say there is not a role for architects, engineers who work in this industry, and contractors in the future of the profession, nor that our knowledge and experience is no longer relevant. We are still the professionals responsible for the health, safety and welfare of the building occupants, the wider community, and to design a project worthy of the resources and permanence within our cities. I am only saying we need to widen the pool of collaboration and leverage the deeper knowledge that is applicable to this new way of working.
It is only when our industry embraces new models of practice that address how we work holistically that we can begin to mirror the gains in productivity that advanced manufacturing and information technology have delivered elsewhere.
Rebecca Lorenz is Director of Research and Development. She has worked in architecture for 15 years and has over a decade experience in advance fabrication in architectural applications. She was on the SHoP team that developed the concept that is now Assembly OSM.
Assembly is hiring! Check out our jobs page!
