Unlocking the Benefits of AI in the Construction Sector

Some of the most innovative use-cases and the deep-learning algorithms driving them are being developed by Indian companies.

By Vikshut Mundkur

An economy in its own right, the construction industry contributes massively to the Indian economy. Currently, it makes up 9% of national GDP, employs about 51 million people, and attracted 13% of all FDI inflows in FY21. The importance of this sector is expected to get even more in the years to come. As per reliable projections, the Real Estate industry in India is expected to reach $1 Trillion by 2030 and contribute 13% to India’s GDP

Needless to say, this sector is huge and its operations are so complex that issues like lack of coordination, frequent re-works, inefficient monitoring, and flawed estimations are considered commonplace. Adding to the woes is the fact that the construction sector has been slow in adopting technology.

The application of AI in construction is scaling rapidly and could potentially be a $4.5 Billion global market by 2026.

Technology

Firstly, the increasing project complexity and an ever-growing demand to complete projects faster, cheaper, and with fewer reworks is forcing the construction executives to look at better ways of getting things done. These days, the executive team of a construction project, namely the project management consultancies (PMC), and their sub-contractors are proactively seeking technology solutions to streamline their workflow, monitor progress, and reduce rework and waste. Secondly, the leadership team among the real estate developers and their financiers want an accurate and reliable solution to evaluate project feasibility, estimation, and reporting solution.

And lastly, the recent developments in the technology landscape – especially in computer vision, natural language processing (NLP), machine learning, cloud computing, drone-tech, and Internet-of-Things (IoT) have opened up new real-world possibilities. Construction-tech companies are finding innovative ways to unlock use-cases that seemed impossible earlier.

Known challenges in the industry

• Constant revision/updates to the plan
• A diverse set of parties (contractors, developers, banks, unit owners)
• Non-transparent reporting – heavy reliance on handwritten notes, reports, emails, and spreadsheets
• Sub-optimal and generic tools for project planning and monitoring
• Low adoption of technology by ground staff
• Lack of visual evidence and tech-backed insights for decision making

These three factors have in fact created a niche ecosystem of forward-looking construction executives and innovative construction-tech companies. A promising endeavor of this ecosystem is to unlock the benefits of AI. And this niche is fast becoming mainstream.

The value of artificial intelligence (AI) in the construction market is expected to reach $ 4.51 Billion by 2026 according to ‘Reports and Data’.

Real estate and construction companies are steadily adopting visual-intelligence platforms for remote management of their projects. Many of these solutions are built on AI technologies and by Indian companies.

Manifestation of AI in construction technology

An integral part of most AI applications is the concept of machine learning (ML). ML (at a very high level) are computer algorithms that learn and improve through experience. Engineers train the algorithm to make sense of architectural drawings, site plans, building components, quality traits, and Building Information Modelling (BIM) files.

Algorithms are trained to understand construction collaterals as close to a human eye as possible. Then applied to solving problems.

The common applications of AI in the construction sector include,

• Work-site surveying
• Insights for architectural designs
• Visualizations (3D modeling) of the site
• Resource estimation & allocation
• Workflow streamlining/Task scheduling
• Remote construction monitoring
• Detailed inspection of work-site details
• Detecting deviations from the plan
• Early warning of faulty construction
• Quality Checks (QC)
• Snag management
• Safety monitoring

AI for Land Surveying – A deep dive

Before any construction starts the first step is to identify a suitable parcel of land, study it and create a feasibility analysis. Traditionally, a team of surveyors visit the site and carry out inspections. This is key to the collection of accurate data. But, despite best efforts, certain human errors occasionally creep in requiring revisits and rework. With AI these challenges are overcome.

A drone equipped with a high-resolution camera (the “eye”) captures images of a building’s exterior. A 360-degree camera captures every construction detail of the interiors. This is the raw data necessary for the ML, deep-learning engine. AI then weaves its magic!

Computer vision (CV) is a branch of AI that enables computers to derive meaningful information from digital images, videos, and other visual inputs. It enables algorithms to make recommendations and call out anomalies and derive actionable insights.

How drone-assisted land surveying & AI-assisted analysis looks like,

• Drone-based land surveys can be conducted in one-third the time of traditional means
• Costs 1/3rd the price of the traditional approach
• Capture 6+ layers of topographical data in one flight
• Captures the natural and man-made structures in the land parcel
• CV can visualize your land in 2D & 3D
• Avoid the cost and time of commuting

Major developers are trying-out (and depending on) visual-intelligence platforms to monitor their projects spread across geographies from their desks (laptops) or even on mobile apps. They generate automated progress reports, and undertake virtual interior walkthroughs and exterior monitoring, using 360-degree cameras and drones operated by an on-site crew.

These are early days but the future of Construction-Tech is promising.

Mundkur is CEO – HUVIAiR Technologies. On a mission to unlock the power of AI for large scale construction projects

feedbackvnd@cybermedia.co.in