Automation in construction is one of four scans that forms part of the Technological Innovation theme for RIBA Horizons 2034.
Automation in construction often sparks strong reactions. Some people are excited about its possibilities. Others are full of trepidation about its potential impact on human resources. Importantly, we tend to focus on the most futuristic elements of automation, visualising droids laying bricks on-site or AI lead designers generating perfect models.
Rather than automation for the sake of it in the form of cyborg-led construction sites, the goal should be process improvement, specifically through the industrialisation of construction. Pursuing this path over the next decade should lead to improvements that enable us to deliver increasingly sustainable, safe, and efficient construction projects that address the systemic failures of the sort that led to the Grenfell tragedy to create a platform for future advances.
Industrialised construction is the convergence of our industry with the principles of manufacturing. This means applying controlled and refined processes for greater certainty. Modern Methods of Construction (MMC), which involves the use of integrated processes and products (including off-site manufacturing), will play a significant role. And, yes, so will robotics – where appropriate.
Automation will be central to this improvement, whether it is applied to processes at the factory and job site, to the design phase, or the flow and analysis of data from the whole supply chain (with crucial insights offered by AI). High quality data from sensors throughout the process will drive real time improvements for better outcomes. This is already happening, particularly in Europe, with the likes of ABB and Weinmann helping firms implement BIM-driven factory automation.
Process improvements are sorely needed within construction, which is facing widespread labour shortages from traditional skills [1] to digital talent [2], a challenge intensified by low productivity. [3]
Projects are growing in complexity, but predictability remains elusive. There is a high level of waste, a poor safety record and a lack of trust throughout supply chains. With labour shortages projected to worsen, all these issues will become even more acute. [4] Architects can respond by embracing industrialised construction and taking a more active role in process improvement.
Industrialised construction – supercharged by automation – has the potential to overhaul both how we build and the relationships between architects, clients and the whole supply chain. This could create a role for architects that spans the highest creativity and the most strategic buildability – an even more promising prospect than bot bricklayers.
A vision of 2034
By 2034, the act of construction will have been transformed, with automation – and especially robotics – playing a growing role. Many of the technologies are already in use in pockets today. For example, Shimizu in Japan has developed a smart site, [5] complete with autonomous robots like the Robo-Buddy, which handles construction work for ceilings and floors.
Likewise, robots and drones are frequently used for automated site mapping to track progress in minute detail without health and safety risks for workers. During the next decade, we’ll see companies identifying which applications of robotics are the most effective. It might be that whereas they are suitable for assembling precast concrete, they are not suitable for MEP pipework, for instance.
The application of robotics on site is important but the real paradigm shift will be where construction work is completed and the processes used. MMC will become the predominant model for construction. Moving work off-site into the controlled conditions of the factory can immediately provide a higher level of certainty, productivity and safety. This is simply because people can work at the right height with tools close at hand and because the controlled conditions mean it is possible to know exactly when components will be ready.
Companies like Mace are demonstrating the value of this standardised approach, developing a system called High Rise Solutions across multiple developments. On the ongoing Chapter London Bridge project, a 39-storey student housing tower, the team has used 120 precast components preassembled off-site. [6] This has improved quality, reduced the number of on-site workers used and enabled each construction team member to deliver £1.59m of value over a year, compared to the overall company average of £1m per person. Importantly, lessons will be carried forward onto the next development, so it is not one-and-done.
All these workflows depend on data, which at present are not well-utilised across much of design and construction. According to analysis by Deloitte, 80% of global construction businesses today have beginner or emerging data capabilities, meaning limited use of data, strategy and skills. [7]
This is set to change by 2034, when data will be used more frequently and much more effectively throughout the industry, as technology matures and the workforce upskills. With more sophisticated digital tools and greater sensor use, accurate information will flow more readily between stakeholders, from the initial planning phase through to operations.
Artificial intelligence will play a key role in analysing, augmenting and automating this information to ensure that decision-makers can quickly act on key insights and implement long-term improvements.
Generative AI will also be in greater use as the companies currently experimenting with it demonstrate its advantages. For example, the Dutch construction company Van Wijnen, has applied generative design at the urban scale [8] to explore its possibilities for the layout of residential areas based on financial, sociological, or environmental metrics. The programming tool could quickly iterate designs, structuring and managing the complex and sometimes conflicting requirements of urban developments.
Deploying generative AI can not only save time but can significantly improve outcomes. Because of the bespoke nature of construction today, the role of AI has been limited by a lack of datasets to train the system. Presently, the main applications are in concept design (RIBA Stage 2) and go no further.
However, by 2034, as Mark Greaves shows in his Horizon scan on AI, the technology might have developed to the point where the industry has refined the best use cases of generative design for architects.
The role of architects
From robotic automation to AI-assisted processes, the automation of construction will have significant implications for the role of architects over the next decade. The design process will dramatically improve with robust data use that will inform choices and optimise outcomes.
In industrialised construction, productisation is inherent. The creation of standard modules or components that are reused on multiple projects, including anything from plaster walls to bathroom pods.
The combination of generative AI and productisation can supercharge architects’ roles and the quality of the designs they produce. Architects will be able not only to present multiple design options, but those options will comprise manufacturable products with known geometry, data and performance.
Workflows are already available today which allow manufacturers of construction components, from façades to bathroom modules, to provide digital ‘recipes’ for architects to use – along with the parameters that can be adjusted and configured. Architects get the right level of detail for the building design, while manufacturers can be assured that the design will be configured correctly for their capabilities.
This will have significant benefits for the buildability of projects that use the system. As already outlined, it will improve productivity, predictability and safety. It will support the principle of the golden thread of information established by the Building Safety Act 2022. Productised solutions will also allow architects to use real world data to identify the best response to project needs more easily.
For example, the system could provide a project team developing dozens of design options with the cost and schedule estimates for each based on actual data from previous similar projects in a matter of days, rather than the weeks or months it takes today. This would enable the entire team to choose the right option and make the right decisions as early as possible, increasing the likelihood of the project coming in on time and within budget.
The evolution of generative AI will benefit from productization, making it a more useful tool for detailed design. Generative AI’s current limitations (caused by the lack of consistent, comparable datasets) can be overcome with the ‘kit of parts’ approach offered by productised design information. With components – like bathrooms, staircases and walls – underpinned by data, the AI will be able to learn to create the design detail demanded by RIBA Stages 3 and 4.
Generative AI will augment, not replace, the architect’s responsibilities, enabling them to focus on higher value, creative elements of their projects. Specific objectives like cost targets and sustainable performance will be achieved early on in the design process, providing architects with greater scope to focus on the unique elements that will differentiate their projects.
Meanwhile, the use of automation across construction processes will keep architects and builders in lockstep. With the flow of real time data supported by automation, builders will have the information they need to truly reflect design intent, while architects will be even more closely involved in the construction phase.
As an early indication of the future, London-based design and engineering firm Bryden Wood have been realising the benefits of productisation with their Design for Manufacture and Assembly (DfMA) methodology, [9] which centres on buildability.
For example, their work for Heathrow and Gatwick airports aimed to minimise disruption to travellers by establishing manufacturing facilities near each airport, allowing entire sections of corridor to be built off-site. At Heathrow, the construction timescale was reduced by 38% and the number of personnel on-site by 75% compared to traditional construction. This resulted in an exemplary safety record, with zero RIDDORs, lost time injuries and operational impacts.
Across both airports, the total cost saving from using MMC was £15.5 million, or 36% compared to traditional construction. Data-sharing across the supply chain, using a 3D BIM model, enabled Bryden Wood to spatially coordinate much more successfully.
Although the uses of automation – and especially AI – might create some uncertainty for the regulatory process, the improvements it brings will help firms meet legislation requirements like the Building Safety Act more effectively. With higher quality data and transparency between stakeholders, providing the golden thread of information and a true as-built record of the building will be more straightforward.
It may well be that the principles behind the Building Safety Act drive the adoption of industrialised construction, with benefits for owners and occupants across the board.
Evolving relationships and risk
With the move towards industrialised construction, there is the potential for architects to play an even more pivotal role in projects over the next decade, demanding a new relationship with the supply chain.
Currently, responsibility for the actual build – and the liabilities involved – resides almost exclusively with the builder. But for industrialised construction to work, clients and architects must select the methodology together from the beginning, with design detail and procurement decisions moving to the start of the process. This is because, in embracing design for buildability, architects must design with the method of construction in mind.
In turn, this means that clients and architects must work closely with the supply chain from the outset of a project, making decisions based on the kits of parts available.
The most successful designers and builders will build longer term relationships with manufacturers to create partnerships over multiple years and projects to support the viability of industrialised construction.
This ‘left shift’ is already being seen, with global scale clients (such as leading cloud-computing companies and pharmaceutical manufacturers) taking more ownership of the process and the risks involved, defining the kit of parts to use and making early supply chain decisions.
Finally, industrialised construction will require moves towards integrated project delivery, a highly collaborative method that brings the design team, delivery and manufacturing suppliers closer to the client. In contrast to the inherently defensive dynamic that can play out in construction projects when responsibility is pushed down the line, this shared risk and reward model requires true project team collaboration.
Preparing for 2034
The advent of automation and the parallel rise of industrialised construction will have widespread repercussions not only for architects but everyone involved in the industry in the next decade. From greater predictability and safety to higher productivity and a more balanced model of risk and reward, the benefits could be enormous. When it happens, the role of architects will evolve very quickly – and the profession must be ready.
A greater awareness of MMC will be vital. Many university courses already include MMC on their curricula, with new architects entering the field ready to use them. This means that existing professionals must stay up to date with the latest capabilities, too.
Now more than ever, it will be important to have a desire and curiosity to embrace new technology. Architects should play with generative AI, become comfortable with it, and understand what it can (and can’t) do. They should make better use of the project data already being produced today to inform future designs – by ensuring that data is well-structured and can yield useful insights. For a possible long term advantage from a professional perspective, they should learn to code.
As industrialised construction takes off, architects will need to understand more about buildability and what the supply chain can do to deliver better project outcomes. Greater openness between collaborators will benefit everyone, from main contractors to clients. This will elevate the architect’s role and create long term opportunities to build much better.
Professional associations and the government will play a crucial role in convincing clients – and the industry as a whole – to embrace the move towards industrialised construction and automation.
By 2034, we won’t have Star Wars’ C-3PO as the job foreman, leading builds overnight. But we could have consistent, predictable and exceptional construction projects, that deliver meaningful societal outcomes in a leaner, cleaner way. And that is far more exciting.
Author biography
Jim Lynch is Senior Vice President & General Manager of Autodesk Construction Solutions. He leads Autodesk’s efforts to accelerate the construction industry’s transformation from analogue-based processes to digital workflows. Jim manages all aspects of business operations, including product development, marketing, sales, and support.
Jim and his team are focused on delivering innovative, cloud-based solutions to help the global construction industry reduce risk using Autodesk technology. During his 20+ year tenure at Autodesk, Jim was a key contributor in scaling Revit into a market-leading solution and establishing BIM as an industry standard. Jim holds a BS in Computer Science from Fitchburg State College.
RIBA Horizons 2034 sponsored by Autodesk
References
[1] International Trade Union Confederation (30 June 2023). More than two million workers will be needed in the construction sector in Europe by 2030
[2] US Bureau of Labor Statistics - C. McAnaw Gallagher (2024). The Construction Industry: Characteristics of the Employed, 2003–20
[3] McKinsey Global Institute - F. Barbosa et al. (27 February 2017). Reinventing construction through a productivity revolution
[4] Construction Industry Training Board - Construction Skills Network (15 May 2024). CSN Industry Outlook – 2023–2027
[5] BuiltWorlds (22 February 2023). Shimizu Smart Site Brings Together People and Robots to Improve Productivity on Job Sites
[6] Building - T. Lane (29 January 2024). Faster, better and less disruptive: the next chapter in Mace’s manufacturing journey
[7] Autodesk - G. Peccolo (26 October 2023). New Report: The State of Data Capabilities in Construction
[8] Autodesk - K. Walmsey and L. Villagi (2019). Generative Urban Design: A Collaboration Between Autodesk Research and Van Wijnen
[9] Autodesk - M. De Wolf (7 July 2022). With DfMA, the full potential of industrialized construction is revealed
