What is Artificial Intelligence and Machine Learning in Construction?
Artificial Intelligence (AI) can be described as an aggregative term that describes how machines can mimic human cognitive abilities, such as solving problems, pattern recognition and learning. Machine Learning is an aspect of AI. It is the subset of artificial intelligence that employs techniques of statistical analysis to provide computers the capability to “learn” from data, without having to be explicitly programmed. A computer becomes more adept in analyzing and providing insight as it is exposed greater amounts of information.
As Trimble engineer in machine learning Bob Banfield put it when we inquired regarding the use of deep learning for construction.
“Machine learning includes many algorithms. Here’s a quick example: if you were looking to find out whether or not you are liable to get some type of disease, one type of learning algorithm might work its way through a tree of questions like, ‘how old are you?’ Then, ‘okay, do you exercise?’ And, so on. If you say yes, you go down one branch, and if you say no, then you go down another. That’s a perfectly valid machine learning algorithm.
In the context of construction the ‘questions’ and algorithms become increasingly complex. For example machines learning programs could track and assess the progress of grading plans to spot potential risks in the schedule early. The algorithms could ask questions regarding cutting and filling volume measurements along with machine uptime and downtime and weather patterns, past initiatives, or any other number of inputs that can be used to create an assessment of risk and decide whether notifications should be sent out.
AI and Machine Learning for Smart Construction
The possibilities for machines learning as well as AI in the field of construction are numerous. Information requests open issues, information, or change orders have become common in the construction industry. Machine learning is an intelligent assistant that is able to analyze this massive amount of information. Then, it alerts project managers to the crucial issues that require their focus. A number of applications have already implemented AI to do this. The benefits of AI range from simple filtering spam emails to more sophisticated security monitoring.
Examples of AI in Construction
1. Prevent cost overruns
The majority of mega projects exceed budget despite having the most competent team of project managers. Artificial neural networks are utilized in projects to anticipate costs overruns based upon factors such as the size of the project as well as the contract type and quality of project managers. The historical data, such as the scheduled dates for start and ending are utilized by predictive models to create realistic timeframes for future projects. AI allows staff to access remotely actual training materials, which help to improve their knowledge and skills quickly. This cuts down the time it takes to integrate new resources into projects. This means that the delivery of projects is speeded up.
2. AI to improve the design of buildings using the generative design process
building information Modeling is an 3D model-based method that gives architects, engineers and construction professionals with the knowledge to efficiently design, plan and build structures and infrastructure. To create and plan the building of a project, 3D models have to be able to take into account the engineering, architecture mechanical, electrical as well as plumbing (MEP) plans as well as the sequence of tasks of the teams involved. The problem is to ensure that the various models of the sub-teams don’t clash with one another.
The industry utilizes machine learning through AI-powered generative design to discover and eliminate conflicts between diverse models developed by various teams in order to avoid the need for rework. There is a software which uses machine learning algorithms to investigate every variation of a model and then generate designs that are more feasible. After a user has set up specifications in the model the software generates 3D models that are optimized for the constraints and then learns every time until it can come up with the most ideal design.
3. Risk mitigation
Each construction project is characterized by a risk, and it can take many types, including quality, safety time, cost, and risk. The bigger construction project is, there’s higher the danger, because there are many subcontractors working on various trades on different construction sites. Today, there are AI and machine-learning solutions currently that general contractors can use to assess and rank risks at the work site, ensuring that the project team can concentrate their time and resources on the most risky elements. AI can be used to determine the priority of problems. Subcontractors are evaluated according to an assessment of risk, which means construction managers are able to work with high-risk teams in order to minimize the risk.
4. Planned project
One company that specializes in construction intelligence was founded in 2017 claiming that its robots and artificial intelligence holds the solution to tackling the issue of over budget and late construction projects. The company makes use of robots that autonomously take 3D images of construction sites. The company then feeds the data into deep neural networks which determines how advanced various sub-projects are. If the project seems to be off track The management team can assist in addressing small issues before they escalate into significant problems. The algorithms of the future will utilize an AI technique referred to “reinforcement learning.” This technique lets algorithms learn from trial and trial and. It can evaluate a variety of options and combinations based upon similar projects. It helps in the planning of projects as it helps to determine the most effective route and then corrects itself as time passes.
5. AI makes jobsites more productive
There are companies beginning to offer self-driving construction machinery that can perform routine tasks more efficiently than human workers, for example, pouring concrete bricklaying, welding, or demolition. The excavation and preparation work is carried out by semi-autonomous bulldozers or autonomous bulldozers that can set up a site with the assistance of human programmers to precisely follow specifications. This frees humans to work on the actual construction and decreases the amount of time needed to finish the job. Project managers are also able to track the work of their workers in real-time. They employ facial recognition, cameras on site as well as similar technologies to evaluate worker performance and their compliance to the processes.
6. AI for safety in construction
Construction workers die at work five times more frequently than other workers. According to OSHA, the most common causes of deaths in the private sector (excluding collisions with highways) for the sector of construction included accidents which were followed by being hitting by objects electrocution, a fall, or being caught in/between. A Boston-based company that develops construction technology has created an algorithm that analyses photographs from its work sites, and scans them for safety hazards, such as workers who are not wearing safety equipment and ties the images to its records of accidents. The company claims that it is able to compute risk ratings for projects so that safety training sessions can be conducted whenever an increase in risk is discovered. It also began rating and releasing safety scores for every U.S. state with respect to COVID-19 compliance by 2020.
7. AI will help address the problem of low-quality labor.
A lack of workers and a desire to improve productivity of the industry have forced construction companies to make investments in AI as well as data-science. The 2017 McKinsey report suggests that construction companies could increase productivity by up to 50% through the an analysis in real time of the data. Construction firms are beginning to utilize AI or machine learning in order to improve their plans for the distribution of workers and equipment across different jobs.
A robot that is constantly monitoring the progress of work as well as the position of the equipment and workers can inform project managers immediately which locations have enough personnel and equipment to finish the project in time and which may be in a position where workers could be needed.
A robot that is powered by AI, like Spot the Dog can autonomously check a site every evening to track progress – this allows a large contractor such as Mortenson to complete more work completed in remote regions where skilled workers are scarce.
8. Construction off-site
Construction firms are increasingly relying on factories that are off-site and operated by robots autonomous which join the elements of a structure, that are then put together by workers on-site. Walls can be constructed in an assembly-line fashion with the help of autonomous machinery, which is more efficient than human workers and human workers are left to do the intricate work like HVAC, plumbing, and electrical systems once the structure is put together.
9. Artificial Intelligence and Big Data in the construction industry
At a time where an immense quantity of new data generated every single daily, AI technology is exposed a vast amount of data that they can learn from and improve upon every day. Every job site can be an opportunity to collect data for AI. Data derived from images taken on mobile devices, drone videos, security cameras, BIM, building information modeling (BIM) and more are now a source of data. This opens up the possibility to professionals from the construction industry as well as clients to analyse and benefit of the information gleaned from these data sources with the aid by AI as well as machine-learning systems.
10. AI for post-construction
Building managers can make use of AI for a long time after the construction has been completed. By collecting data about a structure via drones, sensors as well as other wireless technology advanced analysis and AI-powered algorithms provide invaluable insights into the performance and operation of bridges, buildings roads, or everything else in the built environment. This implies that AI can be utilized to detect the onset of problems and determine when preventative maintenance should to be done or even control the behavior of humans for maximum security and security.
The Future of AI in Construction
Robotics AI, robots along with AI, robotics and the Internet of Things can reduce the cost of building by up to 20 percent. Engineers can put on VR glasses and create mini robots to enter buildings in construction. They use cameras to follow the construction as it goes on. AI is utilized to determine the route of plumbing and electrical systems in modern structures. Businesses are utilizing AI to create safety measures for work sites. AI is used to monitor the live interactions between machinery, workers and other objects at the worksite and notify managers of safety hazards or construction mistakes, as well as problems with productivity.
Despite the rumors of huge job losses AI is not likely to replace human labor. Instead, it will change structures of business within the construction industry. It will lower costly mistakes, decrease workplace injuries, and help increase efficiency in building construction.
The leaders of construction firms should prioritize investment in areas in which AI could have the biggest impact on their business’s specific requirements. The early adopters will establish the course for the construction industry and reap benefits both in the short and long term.