What does it take to maintain Singapore’s glimmering skyline? Traditional building inspection methods involve workers dangling precariously from ropes or gondolas – shining flashlights into crevices to take photos of cracks and corrosion.

These methods are prone to human error, tedious, and unsafe. No longer will this be the case – the nation is trialling the use of AI-powered drones to inspect the facades of buildings.

GovInsider spoke to Wong Wei Loong, JTC’s Director of Building Projects Division, to find out how the agency is trialling drones, AI cameras, and sensors to revolutionise building management in Singapore.

Inspection drones take flight

Drones are deployed to inspect building facades, and are linked to an AI system that automatically identifies defects like cracks and corrosion from what the drones see.

The pilots have helped to make building inspections faster, safer, and cheaper. Using drones have reduced the inspection time of the 31-storey, 128m-tall, JTC Summit building from around 4 weeks to 4 days in a pilot run.

JTC has also halved its manpower costs in trials since 2016, Channel News Asia reported. Drones have taken over the jobs that require heavy-lifting, and freed up manpower for higher-value jobs, says Wong.

Apart from drones, the agency also uses photogrammetry and lidar, a method to measure distance using laser and sensors, to scan buildings while they are being built. The captured information can be overlaid onto the building information model – allowing the team to detect discrepancies and resolve it remotely. “This process also enables us to look into the progress of the project remotely and plan for the next phases of construction works,” says Wong.

Safe-distancing tech

Tech is being piloted at three of JTC’s worksites to facilitate compliance with Covid-19 safe-distancing measures. Bluetooth wearables, for instance, have been distributed to workers to ensure they stand at least 1 metre apart, Wong says. The devices beep when they do not.

This is supplemented by AI cameras that “alert contractors when safe distances are breached”, he adds. Contractors can use data from the cameras to quickly identify high traffic zones and adjust manpower allocations accordingly.

Facial recognition thermal scanners have also been installed at entry and exit points to validate workers’ health records before they enter the worksite, he says.

These technologies are rolled out with the help of construction companies and the Building Construction Authority (BCA), Wong adds. They “make it easier for construction firms to automate and comply with BCA’s safe management measures, such as health checks, contact tracing, and safe distancing.”

Data generated from these devices are plugged into BCA’s BuildSG-COVIDSafe Platform, Wong says. Employers can use the platform to access information about their workers, such as their work permit details.

This platform will be integrated with other government databases and be a single source of information for all construction Covid-19 related matters, according to a BCA factsheet.

Building a smart digital district

The 50-hectare Punggol Digital District in the north-eastern region of Singapore is another project that JTC is excited about. The agency has developed the Open Digital Platform, which functions as a unified operating system for the district.

The platform is connected to a network of sensors and systems that collect building and environmental data, Wong says. This data is then fed into an integrated ops centre that can detect anomalies and optimise the operations of applications like the power grid or electric vehicles.

It also uses a common software framework to allow different digital subsystems – such as building management and energy optimisation – to “speak the same language”, according to the GovTech agency of Singapore.

The repository of real-time district data collected by the platform can be used to create a virtual twin, Wong says. Businesses and students can perform “simulations and rapid prototyping in a risk-free and lifelike environment, reducing the cost of experimentation and time-to-market. “

The virtual twin can be used to train autonomous vehicles, for instance. “You can train your driverless car without having to send the real car out onto the street,” said Mr James Tan, Director of JTC’s Smart District Division, in a press release.

“Companies, students and the public are welcome to use the data to test out new concepts of living, working and delivering services,” Wong says. JTC does not share the data collected with other agencies at the moment.

The possibilities are endless with tech. With the help of drones, AI, and sensors, Singapore is well on its way to become a smart city and nation.

Images by JTC