Automation of As-Built Civil Engineering Management for Labor Reduction & Safety Improvement! ICT Technologies That Change the Field

(Main target readers of this article: construction managers, client engineers, municipal staff, surveying personnel)

As-built management in civil engineering is an important construction control process that verifies and records whether completed structures and terrain conform to the designed shapes and dimensions. While indispensable for ensuring quality, traditional as-built management work has required large amounts of labor and time, and because measurement points were limited, labor reduction and prevention of oversights were major challenges. Ensuring worker safety is also a key issue when surveying or conducting on-site inspections in hazardous areas. In recent years, a wave of automation using ICT technologies has swept into the field of “as-built civil engineering management,” producing dramatic labor reduction and safety improvements. This article comprehensively explains the latest technologies and benefits that are transforming as-built management, including GNSS surveying, point-cloud measurement with drones and laser scanners, AI-based automatic acceptance judgment, and cloud-based information sharing. We also cover operational case studies that enable as-built inspections with few personnel and contactless methods, effects such as reduced contact risk and decreased workload from adoption, and future prospects for automated construction and smart infrastructure management. At the end of the article, we introduce the easy high-precision surveying tool “LRTK” that combines a smartphone and a compact GNSS, and propose ways to promote labor reduction, safety, and simplicity by introducing ICT on-site.

Labor savings and improved work safety through the use of GNSS surveying

In the field of civil surveying including as-built management, the use of satellite positioning systems (GNSS) has greatly contributed to labor savings and improved safety. Traditionally, surveying with levels or total stations required multiple people including experienced technicians to set up equipment, place targets, and take readings. However, with RTK-GNSS receivers, one-person rapid point acquisition—one-man surveying—is possible. For example, for post-construction checks of roadbed height or thickness, where teams of a supervisor and assistants once measured dozens of points, a technician carrying a GNSS rover can now carry out successive measurements alone. This greatly shortens working time and reduces the burden of staffing.

The advantages of GNSS surveying are not limited to efficiency. Reducing personnel also reduces the number of people who must enter hazardous areas, contributing to improved safety. For instance, in as-built measurements on busy roadworks, surveying staff previously risked holding rods near traffic lanes, but GNSS allows measurements from a safe distance without relying on heavy equipment operators. Also, with properly established reference points, GNSS allows measurements while moving across wide areas, reducing the need to set up equipment repeatedly on high places or steep slopes, thereby lowering worker burden and accident risk. Under the Ministry of Land, Infrastructure, Transport and Tourism’s *i-Construction*, site surveying in combination with ICT-equipped machinery has been advanced, and there are reports of more than 30% reduction in total working time through efficiency gains with RTK-GNSS and auto-tracking total stations. GNSS surveying is a technology that can deliver major effects in both labor reduction and ensuring safety.

Current status of point-cloud data acquisition with drones, laser scanners, and smartphones, and automatic processing

A new trend attracting attention in as-built management is 3D point-cloud data measurement. A point cloud is a digital 3D copy that captures the surface of a structure or terrain with countless measured points, and it is characterized by the ability to grasp as-built conditions in a surface-wise manner. Various methods and devices for acquiring point clouds have emerged, each with its advantages. Representative ones include:

• Terrestrial laser scanner (TLS): Devices mounted on a tripod emit 360-degree laser beams to acquire high-density point clouds with millimeter accuracy. They excel in precision and detail, capturing fine irregularities of large structures, but the high cost of the equipment requires significant initial investment. When scanning from multiple positions, post-processing to merge the point clouds (alignment) is required.

• Drone photogrammetry: Uses a large number of aerial photos of the site to reconstruct 3D models (SfM analysis) in software and generate point clouds. It can cover wide areas in a short time and is effective for measuring as-built conditions in places inaccessible to people. However, sufficient overlapping shots and control points (ground markers with known coordinates) are essential to ensure accuracy. Wind conditions and flight restrictions must also be considered.

• Smartphones and handheld LiDAR: Recently, simple LiDAR built into iPhones and iPads, and photo-scan apps for dedicated devices have made it easy to capture surrounding 3D point clouds. Methods that allow scanning by simply walking through the site have emerged, and for small-scale as-built measurements even a single novice can handle the task. However, current built-in smartphone sensors have limitations in achieving surveying-grade accuracy and reliability, so for medium- to large-scale inspections it is still recommended to use them in combination with high-performance equipment or to cross-check results.

Technology for automatic processing of point-cloud data has also advanced. Dedicated software can automatically perform image analysis to generate point clouds from photos, completing the modeling simply by shooting. Increasingly, AI-enabled algorithms assist with automatic calculation for aligning acquired point clouds with each other and with design data, reducing tedious manual work. Even when a wide area is measured in multiple segments, software that merges them into a single 3D model with one click is becoming common.

Of particular importance for applying point clouds to as-built management is the analysis and comparison of as-built dimensions from point clouds. This area has also seen automation through software. For instance, acquired point clouds can be overlaid on design drawings or 3D models, cross-sections at specified positions can be automatically extracted and compared, and height differences between the point cloud and design surface can be visualized as colored heat maps to instantly reveal deviations. The Ministry of Land, Infrastructure, Transport and Tourism has recently established guidelines for “surface as-built management” that evaluate overall as-built conditions using surface-based data like point clouds, enabling more comprehensive inspections than traditional single-point checks. For example, pavement thickness inspections using point clouds allow evaluation of surface irregularities across the entire finished surface, leading to higher-level quality control. Moreover, the latest point-cloud processing software has begun to include functions that automatically compute differences from design values for each point in the cloud and perform pass/fail judgments. Such automatic processing is semi-automating what was previously manual comparison of drawings and numbers, reducing the burden on inspection staff and shortening required inspection times.

Mechanisms and usefulness of AI-based automatic as-built acceptance judgment

In addition to the evolution of point-cloud processing software described above, the use of AI (artificial intelligence) has recently further improved the accuracy of automatic acceptance judgments in as-built management. One mechanism is AI-based image recognition and analysis. For example, in rebar placement checks for reinforced concrete work, systems have been developed in which AI instantly reads the number of rebars and spacing from photos taken by a camera and judges whether they conform. By letting AI automatically determine inspection items that were formerly counted or measured by human eyes, the number of people required to attend inspections can be dramatically reduced and human error prevented. In fact, there are examples where rebar inspections are completed with just two people—the photo taker on site and the client’s inspector—simplifying as-built inspection workflows substantially.

AI is also powerful in the point-cloud domain. Research is progressing on techniques that use AI and machine learning to automatically detect and remove unnecessary noise points within point clouds for data cleansing, and to automatically recognize and classify parts of structures (ground, walls, columns, etc.) from point clouds (segmentation). In the future, there is potential for AI to extract distortions and defects from the massive point-cloud data acquired and generate reports without human checking. In other words, a time may come when simply acquiring point clouds on site will allow AI to automatically determine as-built inspection results.

The usefulness of AI-based automatic acceptance judgment is immense. Above all, it dramatically shortens the time required for inspections, enabling quality checks to proceed smoothly even on sites with labor shortages. AI’s judgment criteria do not fluctuate and are consistent, so they can help prevent variability due to inspector skill level and avoid oversights. By using digital records of measurement results together with AI judgments, as-built management with high objectivity and reproducibility—where anyone inspecting obtains the same result—becomes possible. AI does not make human technicians unnecessary; rather, it is a reliable assistant that takes over simple tasks and tedious calculations. Inspectors can focus on confirming AI-presented results and on important decisions and corrective instructions. Overall, AI-based automatic acceptance judgment is emerging as a powerful tool to support safe, high-quality construction on site.

Centralized information management and remote verification enabled by cloud sharing

Another pillar supporting efficiency and sophistication in as-built management is cloud-based data sharing and remote verification. Traditionally, as-built drawings and inspection results were managed by the site representative on paper or in Excel, and during client inspections it was common to spread out drawings and photo albums on site for explanations. This made information easily fragmented and required time to share updates with stakeholders. The Ministry of Land, Infrastructure, Transport and Tourism is therefore promoting the centralized management of drawings, photos, as-built and quality inspection data related to construction management on the cloud, aiming to create an environment where stakeholders can access them from PCs or smartphones at any time. Using the cloud makes it possible to always share the latest as-built data, allowing general contractors, subcontractors, clients, and supervisors to view the same information. Inspectors can check inspection materials online without going to the site office, reducing travel time.

Cloud sharing shows its true value in remote as-built verification and inspection. In some projects, measured as-built point clouds and photos are already uploaded to the cloud, and technicians at distant headquarters or branches can view the data in real time and complete inspections via online meetings. For example, even for a mountain-area construction site, drone measurement data can be sent via the cloud to urban technicians who can immediately review the as-built check and request on-the-spot additional measurements to be uploaded. Such remote inspections enable, for client witness inspections, alternatives where the client grants electronic approvals on cloud-based 3D models and photos without visiting the site, greatly reducing effort for both parties. In the future, electronic delivery of inspection results via the cloud is likely to become mainstream, and traditional methods of handing over deliverables on paper drawings or DVDs may disappear.

Using a cloud platform also makes it easier to link as-built management data with other construction information. From the design stage, sharing BIM/CIM 3D models on the cloud and continuously comparing them with as-built point clouds during construction becomes realistic. In fact, cloud services that can compare point clouds against BIM models and display deviations in color have already appeared. This enables consistent use of digital data across design, construction, and inspection processes, facilitating early error detection and shared understanding among stakeholders. If all information is available on the cloud, photos taken on-site with a tablet are immediately viewable at headquarters, enabling real-time site visualization and rapid decision-making and support.

Case studies that achieve as-built inspection with few people and contactless methods

By combining the technical elements described above, some sites have completed as-built inspections with very few people and contactless procedures. Below are examples that help visualize how on-site operations change compared to traditional methods.

Case 1: Remote as-built inspection using drones × cloud In a river construction project with a large work area and scheduling difficulties for witness inspections, the contractor adopted drone-based as-built measurement. The completed embankment was photographed from the air and point-cloud data were shared on the cloud. Client engineers checked the cloud-based 3D model from their office and inspected required cross-sections and dimensions. If there were questions, they discussed them via video conference with the field team, collected additional measurement data on the spot, and uploaded them immediately. Only one drone operator was needed on site, and the client’s inspectors could confirm remotely, allowing the as-built inspection to be completed with zero travel and contactless. In this case, inspections that previously required nearly 10 people to attend were reduced to a few people, and same-day responses became possible, realizing significant efficiency and flexibility gains.

Case 2: Simplifying rebar inspection with AI image analysis In typical rebar as-built inspections, the client and contractor attend on site, place a scale next to the rebar, take photos, and people count and measure the number and spacing by eye. In one project, this was replaced with an AI-powered inspection system. When the site representative took photos of the rebar with a smartphone or tablet, a cloud AI analysis server automatically read the rebar arrangement, checked the required items, and returned the results. The client could approve the inspection report on the cloud without visiting the site, and the on-site work was completed by a single photo taker. No one needed to touch the rebar with a scale, and lengthy face-to-face meetings were unnecessary, a case that also helped reduce infection risk during the COVID-19 pandemic.

With such ICT-driven new operations, “inspections that do not rely on people” are becoming a reality. Sensors can automatically inspect without people entering sites while heavy machinery is operating, and remote supervision is possible, significantly reducing accident risk during work and inspections. Small-team inspections address personnel shortages and minimize construction interruptions due to inspection waits. Contactless data review systems are a major achievement in the digital transformation (DX) of construction sites.

Benefits of adoption: reduced contact risk, lower workload, and changes on site

The ICT technologies described above bring various benefits on site. The most notable is reduced contact risk. Contactless measurement and remote inspection reduce physical contacts between workers and between workers and machinery or structures, lowering the risk of accidents and injuries. Tasks where workers once measured at height directly can be replaced by remote sensor measurements, eliminating fall risks, and not having large groups gather for inspections reduces infection risk. The fact that “people do not need to take measurements” and “people do not need to gather” itself is a major safety management advantage.

Next, reduced workload is another important benefit. Workers are freed from carrying heavy surveying equipment over long distances or keeping meticulous records at each point, enabling as-built management to be performed more efficiently and comfortably. With the aging of surveying professionals, replacing physically demanding tasks with machines and software is important for health management and workstyle reform. Reducing simple repetitive measurement tasks allows technicians to spend time on more creative, higher-value work such as improving construction plans or quality analysis. This contributes to both productivity improvement and human resource development.

Furthermore, the advanced as-built data enabled by ICT leads to improved as-built quality. Data makes it easier to detect small defects or deviations that were previously overlooked by human inspection, allowing corrections during construction. As a result, fewer rework and post-delivery complaints can be expected. In this way, ICT-based as-built management not only makes work “safer, easier, and faster” but also has the potential to raise the baseline quality, profoundly changing the field.

Future prospects: progression to automated construction and smart infrastructure management

Although automation technologies for as-built management are only beginning to be put into practical use in some areas, they point toward a future of automated construction and smart infrastructure management. The Ministry of Land, Infrastructure, Transport and Tourism’s *i-Construction* initiative continues to deepen, and while ICT use initially focused on earthwork and pavement, it has recently been expanding into structures such as bridges, tunnels, and dams. From fiscal 2024, application of 3D as-built management guidelines is planned for new work types such as steel pipe pile work and tunnel lining concrete, and ICT as-built management is expected to become standardized across the industry. *i-Construction 2.0* also promotes automation and remote operation of construction, and in the future we may see unmanned construction and unmanned inspection by construction robots, or a single technician remotely supervising multiple distant sites simultaneously.

Private construction machinery manufacturers are accelerating development of smart machinery equipped with ICT. For example, systems are under trial that equip bulldozers and rollers with GNSS and sensors to automatically control ground leveling and compaction while measuring as-built conditions in real time and sending the data to the cloud for immediate quality control feedback. A future in which machines autonomously perform construction and inspection and humans only need to review the data later is not far off. If AI, IoT, and robotics come together in smart construction, as-built management will become increasingly labor-saving and sophisticated.

At the same time, as-built data acquired during construction will be leveraged for smart infrastructure management during post-construction maintenance. High-precision point-cloud data captured at as-built time can be preserved as a digital twin of the structure. Using the as-built condition at project completion as a baseline, periodic inspections can acquire new point clouds and compare them to quantify deterioration and deformation over time. For example, slight crack propagation on tunnel linings or bridge settlement can be visualized by overlaying point clouds in a time series. The government is also focusing on 3D data use in infrastructure maintenance, and trials are underway to link as-built electronic data to future management databases. In time, as-built point clouds delivered at handover may be registered directly into facility management ledgers and used for repair planning and risk assessment. ICT-driven as-built management can catalyze the digitalization of the entire infrastructure lifecycle from construction through maintenance to renewal.

Using smartphone × compact GNSS “LRTK” to achieve labor reduction, safety, and simplicity

Given the DX trends in as-built management introduced above, many construction managers may wonder, “We want to introduce ICT ourselves, but where should we start…?” To help, we introduce a practical ICT surveying tool that is easy to deploy on site: smartphone × compact GNSS receivers called LRTK. LRTK (El-Ar-Tee-Kay) is a modern solution that links a smartphone or tablet with a compact high-precision GNSS unit to enable centimeter-level positioning (cm level accuracy (half-inch accuracy)) that anyone can perform easily. By installing a dedicated LRTK app on a smartphone and connecting a small GNSS sensor via Bluetooth, users can obtain high-precision coordinates without complex operations or specialist knowledge. It is truly a “surveying instrument you can carry,” streamlining and simplifying as-built measurements that previously required total stations or large GNSS equipment.

By using LRTK, even staff with limited surveying experience can intuitively collect as-built data via smartphone operation, contributing to labor reduction on sites facing personnel shortages. For example, while a veteran surveyor handles multiple sites, a junior employee can perform preliminary surveying with LRTK, enabling effective use of human resources. Because it is lighter and easier to carry than traditional surveying gear, measurements at height or on slopes can be done more safely, reducing physical burden and hazards for workers. Real-time display of positioning results on the smartphone screen enables immediate on-site verification and smooth on-the-spot data sharing. Combined with LRTK cloud services, acquired as-built coordinates and photos can be instantly uploaded to the cloud for office staff to check in real time.

Thus, LRTK is a tool that significantly lowers the barriers to ICT adoption in civil as-built management. It fits site needs such as “we want labor reduction but cannot afford an expensive 3D laser scanner” or “we primarily need a simple way for one person to safely survey,” and its ease of use—requiring only a familiar smartphone—facilitates smooth acceptance on site. While cutting-edge technology may sound daunting, LRTK uses the everyday smartphone as its platform, lowering psychological barriers and easing on-site adoption. Why not start ICT transformation of familiar as-built management tasks by leveraging LRTK, which enables high-precision surveying by anyone, anywhere, easily? As a first step toward balancing labor reduction and safety improvement, you will likely feel the effects quickly.