About 7 percent of the global labor force works in construction, which as an industry thus occupies an important place in the global economy. $10 trillion is spent annually on construction-related activities (McKinsey, 2017). While other industries have adopted AI and other technologies to increase their productivity, construction is making slow progress.
Worldwide, construction as an industry has grown by only 1 percent per year over the past few decades. Compare that figure to the 3.6 percent growth in the manufacturing industry and the 2.8 percent growth in the overall global economy. In construction, productivity, or total economic output per worker, has remained more or less flat. By comparison, since 1945, productivity in retail trade, manufacturing and agriculture has increased 1,500 percent. What accounts for this difference? Construction is one of the least digitized sectors and is slow to adopt new technologies (McKinsey, 2017).
Implementing the latest technology can be a daunting task. But machine learning and artificial intelligence are making construction sites more efficient and saving money at the same time. AI solutions that have already made a change in other industries are now starting to find their way into construction as well.
What is artificial intelligence and machine learning?
Artificial intelligence (AI) is an umbrella term for describing situations in which a machine mimics cognitive functions of humans, such as problem solving, learning and pattern recognition. Machine learning is a subset of AI. Machine learning is a field within artificial intelligence that uses statistical techniques to allow computer systems to "learn" from data without being explicitly programmed. A machine gets better at understanding and providing insight as more data is available.
McKinsey expects the growth of AI within the construction industry to be modest for the foreseeable future(McKinsey, 2018). But a shift is coming - and AI can no longer be dismissed as something of interest only to other industries. Construction, too, will have to work with AI methods and applications. That is the only way to face new competitors and stay relevant.
Building smart with AI and machine learning
The potential applications of machine learning and AI in construction are many. Information requests, open problems and change requests are daily business in the industry. Machine learning is something like a smart assistant that can sift through this mountain of data. It then sends project managers alerts about the critical issues that need their attention. There are already several applications that use AI in this way. They offer all kinds of benefits, ranging from simply filtering out unwanted e-mail messages to advanced security checks.
10 examples of AI in construction
1. Preventing budget overruns
most megaprojects are more expensive despite the use of the best project teams. Artificial neural networks are deployed on projects to predict budget overruns based on factors such as the size of the project, the type of contract and the competency level of project managers. Predictive models use historical data such as planned start and end dates to plot realistic timelines for future projects. Thanks to AI, remote employees have access to realistic training materials that allow them to quickly increase their knowledge and skills. As a result, less time is needed to train new people on projects and projects can be delivered faster.
2. AI for generative design
Building Information Modeling is a 3D modeling process that architects, engineers and structural engineers use to efficiently plan, design, construct and manage buildings and infrastructure. To plan and design the construction of a building, 3D models must take into account architectural, engineering, mechanical and electrical drawings, planned piping and the sequence of activities of different teams.
The problem lies in preventing the different models of the sub-teams from colliding with each other. The industry is trying to apply machine learning in the form of generative design to detect and resolve conflicts between the models of the different teams as early as the planning and design phase to avoid problems at a later stage.
There is software that uses machine learning algorithms to explore all variants of a solution and generate alternative designs. This uses machine learning to create specific 3D models of mechanical, electrical and piping systems while ensuring that the installation systems do not clash with the architecture of the building. The software learns from each intermediate stage to ultimately arrive at an optimal solution.
3. Risk reduction
Every construction project carries certain risks in terms of quality, safety, time and cost. The risk increases as the project gets larger, due to having several subcontractors from different trades working on the site at the same time. AI solutions and machine learning techniques already exist that allow contractors to monitor and prioritize risks on site, allowing the project team to focus limited time and resources on the biggest risk factors. AI is used to automatically prioritize problems. Subcontractors are rated using a risk score so that construction managers can manage risk in close collaboration with the teams that represent the greatest risk.
4. Project Planning
An AI startup promised when it was founded in 2018 that its robots and artificial intelligence would provide a solution to late-completed and overpriced construction projects. The company uses robots that autonomously take 3D scans of construction sites and feeds that data into a deep neural network that shows how far different subprojects have progressed. If something does not go according to plan, the management team can intervene and prevent major problems at an early stage.
Future algorithms will use an AI technique called "reinforcement learning." This technique allows algorithms to learn from mistakes by evaluating infinite combinations and alternatives based on similar projects. This helps with project planning by allowing the algorithm to choose the best route and correct itself over time.
5. AI for a productive construction site
There are companies now marketing self-propelled construction machines that perform repetitive tasks such as concrete pouring, bricklaying, welding and demolition more efficiently than humans. Excavation and preparation work is performed by autonomous or semiautonomous bulldozers, which can prepare a construction site to exact specifications with the help of a human programmer. This frees up workers for the construction work itself and reduces the total time required to complete a project. Project managers can also monitor construction site work in real time. To do so, they use facial recognition, on-site cameras and similar technologies to assess construction worker productivity and compliance with procedures.
6. AI for safety in construction
Construction workers are five times more likely to have a fatal accident on the job than other occupations. According to OSHA (the U.S. Occupational Safety and Health Administration), the leading causes of fatal accidents at private construction firms (apart from traffic accidents) are falls, being hit by an object, electrocution and entrapment.
A Boston contractor company with $3 billion in annual sales is developing an algorithm that analyzes photos of its own construction sites and scans them for safety hazards, such as construction workers not wearing protective gear. The algorithm then correlates the images with data on accidents at the company to calculate potential risk scores for projects. If a high risk is detected, special safety instructions can then be issued.
7. AI against personnel shortage
Staff shortages and the desire to address low productivity in the industry are forcing construction companies to invest in AI and data science. According to a 2017 McKinsey report, construction companies can increase productivity by as much as 50 percent using real-time data analytics.
Construction companies are beginning to apply AI and machine learning to better plan the deployment of people and machines on different projects. Thanks to a robot that constantly evaluates a project's progress and the location of people and machines, project managers instantly know which construction sites have enough people and machines deployed to complete the project on time, and which ones need additional people to avoid delays. Experts expect construction robots to become more intelligent and autonomous with AI techniques.
8. Construction activities at another location
Construction companies are increasingly using factories where independent robots fabricate the parts of a building, which are then assembled by workers on the construction site. Parts such as walls can be made more efficiently on a kind of assembly line by autonomous machines than by humans. After the building is assembled, humans do the more detailed work, such as piping, air handling and electrical installations.
9. AI and big data in construction
Huge amounts of data are created every day. As a result, every day AI systems have an endless amount of data to learn from and improve themselves. Every construction site becomes a potential data source for AI. Data from images from mobile devices, drone videos, safety sensors, building information models (BIM) and other systems have become a major source of information. Suppliers and customers in the construction industry can use AI and machine learning systems to analyze that data. The insights they gain that way can then be applied in practice.
10. AI for management and maintenance
Building managers can use AI when the construction of a building has long been completed. A building information model (BIM) stores information about the structure of a building. AI can be used to track any problems and even offers solutions to prevent problems.
The future of AI in construction
Robotics, AI and Internet of Things can reduce construction costs by up to 20 percent. Engineers can put on VR goggles and send mini robots into buildings that are still under construction. These robots use cameras to track the progress of work. AI is being used to plan piping and electrical installations in modern buildings. Companies are using AI to develop safety systems for construction sites. AI is used to track the interactions of construction workers, machinery and objects in real time and to alert superintendents to potential safety issues, construction defects and productivity problems.
Despite predictions of major job losses, AI is unlikely to replace the human workforce. Instead, it will change business models in construction, reduce costly errors, reduce workplace injuries and make construction operations more efficient.
Managers in construction companies should prioritize investments in areas where AI can have the greatest impact on their company's specific needs. Those who get there early will set the course for the industry and benefit in both the short and long term.
Want to know what other innovations installers are using to stay ahead? Then download the free Constructible Magazine!
