Construction employs almost 7% of the global workforce and is therefore an important sector of the economy. The general public and businesses spend €8,766 billion a year on construction activities (McKinsey, 2017). While other sectors are using Artificial Intelligence and other technologies to improve their productivity, construction is only slowly incorporating these advances.
Worldwide, growth in the construction sector has been just 1% per annum in recent decades. Meanwhile, the manufacturing industry has grown by 3.6% and the global economy by 2.8%. Productivity - total economic output per employee - has remained constant in the construction sector. In comparison, productivity in retail, manufacturing and agriculture has increased by 1500% since 1945. This is because construction is a sector in which digital tools are not widely used, and new technologies are adopted only slowly (McKinsey, 2017). Adopting the latest technologies can seem daunting to your teams. But machine learning and artificial intelligence make worksites more efficient and save money. AI solutions that have benefited other sectors are starting to emerge in construction.
What is artificial intelligence and machine learning?
Artificial intelligence (AI) is an umbrella term for a machine's ability to mimic human cognitive functions, including problem solving, pattern identification and learning. Machine learning is a sub-field of AI that uses statistical techniques to enable computer systems to "learn" from data, without any special programming: the more data they are fed, the more efficient they are
In the short term, McKinsey predicts limited development of AI in the construction sector(McKinsey, 2018). Nevertheless, a tremor is perceptible. Industry players can no longer afford to let the relevance of AI benefit only others. Engineering and construction must appropriate AI methods and applications. This is the only way to face up to new competitors and stay in the race.
AI and machine learning for intelligent construction
There are many possible applications for machine learning and AI in the construction sector. Requests for information, problems to be dealt with and corrective orders are typical in this sector. Machine learning represents an intelligent assistant capable of examining this mountain of data. It alerts project managers to the main issues to be addressed. Several applications are already using AI in this way. The benefits range from e-mail and spam sorting to advanced security monitoring.
Consult the free guide Why embrace new technologies to discover how the right choice of cutting-edge technology can help a construction company stay ahead of the game.
10 examples of AI use in the construction sector
1. Avoiding extra costs
Most mega-projects come in under budget, even when in the hands of the best teams. Artificial neural networks are used in some projects to predict different cost overruns based on factors such as project size, contract type and skill level of project managers. Historical data such as planned start and finish dates are used by predictive models to realistically define future project deadlines. In addition, AI helps teams to access training materials remotely, and thus to rapidly develop their skills and knowledge. As a result, the time needed to integrate a new resource into a project is reduced. Project delivery is consequently accelerated.
2. AI for better-designed buildings thanks to generative design
Building information modeling is a process that relies on 3D models to provide architecture, engineering and construction professionals with useful information for the efficient planning, design, construction and management of buildings and infrastructures. To plan and design the construction of a building, 3D models need to take into account the architecture, engineering, mechanical, electrical and plumbing plans, as well as the sequence of activities of the various teams.
The challenge is to ensure that the models of each sub-team do not conflict. The industry is currently trying to use machine learning in the form of generative design to identify and limit conflicts between the different models generated by each team during the planning and design phase, and thus avoid rework.
Software is available that exploits machine learning algorithms to explore all variations of a solution and generate alternative designs. Machine learning is used to specifically create 3D models of mechanical, electrical and plumbing systems, while simultaneously ensuring that these do not conflict with the building's architecture. With each iteration, it learns to propose an optimal solution.
3. Risk mitigation
Every construction project involves various forms of risk in terms of quality, safety, deadlines and costs. The larger the project, the greater the risks, as several subcontractors work in parallel on construction sites. AI and machine learning solutions are now being used by prime contractors to track and prioritize risks, so that the team can focus its time and resources on the key risk factors. AI is used to automatically assign priority to problems encountered. Subcontractors are rated according to a risk level so that site managers can ensure that risks are mitigated with the most exposed teams.
4. Project planning
An AI startup launched in 2018 promises that its robots and artificial intelligence hold the key to solving the problems of delays and cost overruns encountered by projects. The company uses robots to automatically obtain 3D scans of construction sites, then transmit this data to a deep learning neural network that determines the progress of individual sub-projects. If delays are identified, managers can intervene to resolve problems before they become too serious.
The algorithms of the future will use an AI technique called "reinforcement learning". This allows algorithms to learn from trial and error. It can evaluate an unlimited number of combinations and alternatives from similar projects. This facilitates project planning, as the tool can determine the most optimized path and adjust it over time.
5. AI will make building sites more productive
Some companies are beginning to offer autonomous construction machines to perform repetitive tasks more efficiently than their human counterparts, such as pouring concrete, erecting brick walls, welding and demolition. Excavation and earthmoving are carried out by autonomous or semi-autonomous bulldozers capable of preparing the site of a worksite from specifications defined by a human programmer. This means that employees can concentrate on the construction work themselves, reducing the time needed for the project. Project managers can also monitor progress on site in real time. They use facial recognition, on-site cameras and similar technologies to assess productivity and check compliance with procedures.
6. AI for construction safety
Construction workers suffer five times as many fatal workplace accidents as other workers. According to OSHA, the main causes of death in the private construction sector (excluding highway accidents) are falls, impact with objects, electrocution and crushing.
A Boston-based prime contractor with annual sales of $3 billion is developing an algorithm that analyzes photos of construction sites, identifies hazards (such as workers not wearing protective equipment) and relates the images to reported accidents. The company claims to be able to assess the risk level of a project, so as to warn teams when a significant hazard is detected.
7. AI will solve the labor shortage
Labor shortages and a desire to increase the sector's low productivity are driving construction companies to invest in AI and data science. A report compiled by McKinsey in 2017 indicates that construction companies could increase their productivity by at least 50% through real-time data analysis.
Construction companies are starting to use AI and machine learning to better plan the distribution of labor and machinery between different projects. A robot that constantly assesses a project's progress, as well as the location of workers and equipment, enables project managers to know instantly which sites have enough workers and equipment to be completed on time, and which are behind schedule and would need additional manpower. According to experts, robots in the construction sector are becoming increasingly intelligent and autonomous thanks to AI techniques.
8. Off-site construction
Construction companies are increasingly relying on off-site factories equipped with autonomous robots to assemble building components, which are then used by workers on site. Some structures, such as walls, can be built by autonomous machines more efficiently than by humans, who then devote themselves to more specialized tasks such as plumbing, HVAC or electrical systems once the structure has been assembled.
9. AI and Big Data in construction
With a huge volume of data being created every day, AI systems have an unlimited source of learning and improvement. Every construction site becomes a potential source of data for AI. Data generated from images taken from mobile devices, videos recorded by drones, safety sensors, Building information modeling (BIM), etc. have become a wealth of information. Construction industry professionals and their customers now have the opportunity to analyze the information derived from this data and make the most of it, thanks to AI and machine learning systems.
10. AI for post-construction
Construction managers can use AI long after buildings have been delivered. Building Information Modeling or BIM stores information about the structure of buildings. AI can be used to track ongoing problems and even propose solutions to avoid them.
The future of AI in construction
Robotics, AI and connected objects can reduce construction costs by up to 20%. For example, engineers can equip themselves with virtual reality goggles and send mini-robots into buildings under construction to monitor progress using cameras. AI is used to plan electrical and plumbing networks in new buildings, and to develop safety systems for construction sites. It enables real-time monitoring of the interactions of workers, machines and objects on construction sites, and alerts managers to potential hazards, construction errors or productivity problems.
Despite the significant job losses predicted, AI is unlikely to replace the human workforce. Rather, it will change the business models of the construction sector, limiting costly mistakes, reducing the number of accidents on construction sites and making work more efficient. Construction industry leaders need to prioritize investment in areas where AI can have the greatest impact on the specific needs of their business. Those who get in on the ground floor first will define the direction of the industry and benefit from innovations in both the short and long term.
