BIM/CIM-Integrated Point Cloud As-Built Management: Triple the Accuracy with Easy Implementation!

The use of point cloud data in on-site as-built management is rapidly expanding. Dimension checks that were previously done manually can be dramatically streamlined by 3D "whole-site" data, making it easy for anyone to adopt and offering the potential to greatly improve accuracy (more than threefold compared to traditional methods). In particular, integrating with BIM/CIM makes comparison and verification against design data easier, and a wide range of benefits can be expected, such as labor savings in construction inspections and applications to maintenance management. This article comprehensively explains the basics of using point cloud data in as-built management and the various advantages gained by combining it with BIM/CIM.

Including keywords such as differences from conventional methods, ease of introduction, improvements in accuracy and labor savings, automatic evaluation using heat maps, cloud utilization, digital twin, remote onsite presence, and support for electronic deliverables, this piece provides an overview of as-built management through point cloud × BIM/CIM integration. At the end of the article, we also touch on the recently noted high-precision surveying (LRTK) using smartphones × small GNSS, encouraging the first step toward adopting digital measurement.

What is as-built management? Traditional methods and challenges

First, we outline what as-built management is, its fundamentals, and the challenges of conventional methods. As-built management is a construction management process that verifies and records whether completed structures and formed terrain from construction have the shape and dimensions specified in the design. It is an important task to demonstrate with measurement data that the actual as-built condition conforms to the specification standards (as-built management standards) established by the client. In public works, it is emphasized as a key element of quality assurance because it is a prerequisite for passing inspections and for handover; moreover, in long-term projects, since parts that will later be hidden (such as buried items) need to be reliably recorded, sequential measurement and recording at each stage, including during construction, are required.

Traditional as-built management methods typically involved directly measuring dimensions on site based on control points using measuring tapes, staffs, and levels (surveying instruments), and checking for discrepancies with the design drawings. Surveyors and technicians, often working in teams, would measure each finished location one by one for height, width, thickness, etc., and check whether they were within tolerance. For example, in road construction, they might measure the thickness and height of the roadbed at multiple locations and compile the results in a record sheet. This work required a great deal of manpower and time, and including the task of organizing measurement results into drawings and tables, it placed a heavy burden on site personnel.

In addition, because manual measurements inevitably limit the number of measurement points, it is difficult to comprehensively grasp the as-built condition of the entire site. It is hard to notice any discrepancies or unevenness between measured points, and there was a risk that oversights—such as "key checks passed, but some parts subtly differed from the design"—would be pointed out during later inspections. The larger the structure, the more limited manual measurement becomes, making it easy to miss minor irregularities and variations.

Also, on busy sites, record omissions such as forgetting to take photographs can occur. If photos of hidden elements like rebar that will be buried after completion are forgotten, no evidence remains, and in the worst case this can lead to rework or disputes. In this way, traditional as-built management methods faced issues such as "limited measurable points," "human errors occurring," and "high manpower and time requirements," which imposed a significant burden and stress on site engineers.

Basic Procedures for As-Built Management Using Point Cloud Data

A solution that has attracted attention in recent years is as-built management using 3D point cloud data. Point cloud data (point cloud) are three-dimensional data that record a large number of points composing real-world space with XYZ coordinates, in other words an on-site full-scale 3D copy obtained by scanning the entire space. In recent years, backed by the Ministry of Land, Infrastructure, Transport and Tourism’s *i-Construction* initiative, the introduction of point cloud surveying technologies in the civil engineering and construction industry has accelerated rapidly. High-performance terrestrial laser scanners and drone photogrammetry (photogrammetry) are making methods to measure as-built conditions non-contact and high-density practical, and the use of point clouds for as-built management is now becoming the "new norm".

As-built management using point clouds is, simply put, about enabling you to "measure an entire site in 3D and later extract any cross-section or dimension you want." The general procedure is as follows.

• Preparation for acquiring 3D data: First, create a survey plan and, if necessary, establish reference points or target markers on site that will serve as known points. Select the measurement instruments according to site conditions. Typical instruments include a terrestrial laser scanner mounted on a tripod that projects laser light around the surroundings, and a drone (photogrammetry) that captures aerial photographs and converts them into point clouds. Laser scanners can produce high-density point clouds with millimeter-level accuracy, but the equipment is expensive. Drone photogrammetry can capture large areas in a short time, but adequate image overlap and accuracy control using ground control points (GCPs) are essential. Recently, methods for easily acquiring point clouds using the simple LiDAR built into iPhones and iPads have appeared, but in many cases specialized high-performance equipment is still required to achieve professional-level accuracy and reliability. However, for small sites, convenient measurement is becoming possible, such as a single technician walking around scanning with a smartphone app to acquire point clouds.

• Measurement and Generation of Point Cloud Data: Once preparations are complete, actual measurements are carried out. In the case of a laser scanner, the device is set up on site and emits laser beams in a full 360-degree sweep to acquire many surrounding points at once. If necessary, the device location is changed and multiple scans are performed, and those multiple point clouds are later merged (aligned) to construct the overall 3D data. In the case of drone photogrammetry, numerous photos are taken from various angles from above, and point cloud data are generated by image analysis using dedicated software. When the target area is large or there are blind spots, multiple point cloud datasets are acquired and merged to complete the point cloud model of the entire object. When combining multiple datasets, it is important to align their coordinates to reference points (georeference) to reduce errors. The Geospatial Information Authority of Japan has also warned to “ensure accuracy by sufficiently comparing with existing methods when introducing new technologies,” and it is recommended that point cloud measurements be accuracy-verified through cross-checks with conventional surveying.

• Analysis and verification of as-built condition: Compare the acquired on-site point cloud data with the design data to evaluate the as-built condition. For analysis, point cloud processing software running on a PC and 3D-capable CAD software are used. First, prepare the reference design data, such as design drawings or 3D design models from BIM/CIM, and align the measured point cloud with them in spatial coordinates. There are several comparison methods, but typical ones are comparison with cross-sectional drawings and 3D deviation checks. In cross-sectional comparison, extract cross-sectional profiles from the point cloud at prescribed locations, overlay them with the design cross-sections, and check dimensional differences. In contrast, the 3D deviation check generates a heat map that color-codes the elevation differences between each point in the point cloud and the design surface, allowing you to grasp at a glance how much the as-built deviates from the design. For example, in paving work, traditionally only thickness measurements at individual points were possible, but using point clouds allows evaluation of the irregularities of the entire finished surface as an area, leading to more advanced quality control. In recent years, the Ministry of Land, Infrastructure, Transport and Tourism has also introduced a new "surface management" method that evaluates the as-built condition of the entire surface using areal measurement data such as point clouds, enabling more comprehensive inspections than the conventional point-by-point evaluations. Some analysis software even provides an automatic evaluation function that automatically calculates deviations from the design from point cloud data and performs pass/fail judgments, making it possible to semi-automate as-built inspections.

• Recording, Reporting, and Data Utilization: As-built information analyzed from point clouds is organized and submitted as as-built management charts and reports. The submission format follows the client's procedures; sometimes it is compiled into paper drawings and numerical tables as before, while cases of submitting the 3D data itself or converting analysis results into CAD data for electronic submission have been increasing. Because point cloud data itself can be preserved long-term as high-precision digital evidence, it becomes a more reliable as-built record than paper photo albums. Especially as material for preventing future problems and for maintenance management, keeping as-built records as point clouds has great value. For example, when planning subsequent additional work or renovations at a later date, opening the saved point cloud data allows immediate creation of an accurate 3D model of the current condition and arbitrary cross-sections, saving the effort of re-surveying the site. In this way, point clouds captured at as-built time can be capitalized as a digital twin (a virtual model that is the site's twin) and can also be used for monitoring aging-related changes during the maintenance phase. Also, the construction details captured by point clouds remain as immutable evidence (proof), providing reassurance for quality assurance.

This is the general workflow for as-built management using point clouds. Compared with conventional methods, its greatest feature is that it "can capture the entire site without leaving anything out", and because any missed measurements can be checked and added later from the data, it enables the preservation of "complete, no-omission quality records". The government is also actively promoting the use of 3D data in construction management, and this method is expected to become the standard going forward.

Integration with BIM/CIM Models: Streamlining Design Comparison and Inspections

Point-cloud as-built management becomes even more powerful when linked with BIM/CIM 3D design data. BIM (Building Information Modeling) is a term mainly used in the building sector, while CIM (Construction Information Modeling) is used in the civil infrastructure sector; both refer to three-dimensional digital models created during the design stage. In recent years, cases in which project owners (particularly the Ministry of Land, Infrastructure, Transport and Tourism) provide 3D design data to contractors have increased, and an environment is taking shape on-site where BIM/CIM design models can be utilized for as-built management.

Specifically, by overlaying point cloud data acquired during construction onto the design BIM/CIM model and comparing them, you can intuitively visualize the differences between the as-built condition and the design (出来形と設計との差異を直感的に視覚化). Traditionally, inspection work involved comparing plan views and cross-sections, but because it only requires displaying the 3D model and point cloud simultaneously on the screen, the checking process becomes dramatically more efficient. For example, in inspections of tunnels and bridges, comparing a laser-scanner-measured point cloud with the design 3D model makes it immediately clear which parts deviate from the design. Minute construction deviations that manual measurements had missed can be captured with three-dimensional color-difference displays, directly contributing to the advancement of construction management and as-built inspections. Dedicated viewer software and cloud services that can display BIM/CIM models and point clouds integrated in the same coordinate system have also emerged, enabling anyone to easily compare 3D data on a PC. This smooths information sharing with supervisors and inspectors, and there are reports that it leads to simplification of on-site inspections and reduction of the burden of preparing construction documents. In actual sites, cases have been reported where overlaying as-built point clouds obtained by total stations equipped with laser scanners onto the design model for as-built verification achieved substantial work time reduction and streamlining of inspection procedures compared with conventional methods.

Linking BIM/CIM with as-built point cloud data also leads to the use of digital twins in the operations and maintenance phase. By associating the design-phase 3D model with the point cloud model that represents actual construction results, the "virtual and physical" aspects of an asset can be managed in a unified way, bringing us closer to realizing a true digital twin. For example, if an as-built point cloud is available at completion, advanced maintenance—such as analyzing long-term changes by comparing it with point clouds acquired during future periodic inspections—becomes possible. In this way, by combining it with the data foundation of BIM/CIM, the value of as-built management data is realized not only immediately after construction but throughout the entire lifecycle.

Benefits of Point Cloud As-Built Management and Differences from Traditional Methods

So, how does as-built management using point clouds differ from traditional manual surveying? Here, we summarize the main benefits.

• Dramatic improvement in accuracy and coverage: Previously, construction accuracy was evaluated by sampling a limited number of points, but with point clouds you can record the actual site shape completely with countless points. Because it can capture minute irregularities at the millimeter level (mm-level, 0.04 in), differences from the design can be detected down to the smallest details. It can cover areas that human inspection would have missed, and the precision of as-built management is greatly improved. As a result, this leads to early detection and correction of construction errors and overall quality improvement. In particular, because parts that will later become hidden—such as the interior of concrete—can also be recorded as 3D data, verification with a level of coverage and reliability not possible before is enabled.

• Reduced work time and improved inspection efficiency: The introduction of 3D measurement technology has made it possible to acquire a large amount of data in a single measurement even for wide-area as-built conditions. For example, surveys that used to take a team including veterans a full day can, with a laser scanner, be completed in a short time (in some cases a few hours). In fact, according to a survey by the Ministry of Land, Infrastructure, Transport and Tourism, earthwork projects that introduced ICT construction (such as 3D surveying and machine guidance) reported that total work time was reduced by an average of about 30%. Point cloud measurement is non-contact and speedy, so interruptions due to waiting for heavy equipment operation and remeasuring rework are reduced, contributing to an overall shortening of construction schedules. Furthermore, because software assists with automatic calculations when analyzing the acquired data, the labor involved in manual calculations and drawing creation is reduced and the overall productivity of as-built inspection operations is improved.

• Labor-saving and Improved Safety: Point cloud measurement is a technology that is operable by a small team, and in some cases the workload is light enough that even a single newcomer can operate the equipment. Unlike traditional methods that require multiple people to perform layout marking and fixed-point measurements, work can proceed efficiently even on sites with labor shortages. This is also effective for the construction industry's broader challenge of labor saving (addressing workforce shortages). In addition, because laser measurement and drone photography allow measurements to be taken from a distance, surveyors themselves do not need to enter dangerous locations such as high places, steep slopes, or busy roadways. In other words, the use of point clouds greatly contributes to on-site safety and reduced work burden, enabling as-built management to be implemented "with fewer personnel, more safely".

• Simplification of record-keeping and reporting and information sharing: If point cloud data are acquired, dimensions and cross-sections at any desired location can be extracted later, greatly reducing worries such as "forgetting to measure" or "failing to take photos". Creation of as-built drawings and photo albums can also be automated and streamlined by leveraging point cloud data. In fact, trials of high-speed as-built inspections using 3D point clouds + design data have reported that various procedures can be made more efficient compared to conventional methods, such as remote conduct of on-site inspections and major simplification of as-built photo and drawing creation tasks. Because point clouds are electronic data, they are easy to share among stakeholders, enabling new workflows (remote presence-style inspections) such as inspection personnel checking remotely via the cloud.

Thus, by leveraging point clouds you can achieve "more accurate, faster, safer, and less labor-intensive" as-built management. A major difference from conventional methods is that it can prevent human error while enhancing the ability to demonstrate quality. It can truly be called an innovative approach in which a threefold increase in accuracy is not out of the question. On the other hand, challenges such as the purchase cost of high-performance equipment and the processing burden of large-volume data do exist. However, these are gradually being resolved. The next chapter will explain the lowering of implementation hurdles, and finally introduce concrete methods that are easy to get started with.

Utilization in Maintenance and the Potential of Digital Twins

The high-precision point cloud data obtained through as-built management also proves highly valuable in the post-construction maintenance phase. If the point cloud data of the completed form is stored as an electronic deliverable or an internal asset, it can be immediately referenced as an accurate 3D model of the current condition when planning future repairs, upgrades, or additional construction. For example, even when designing road repairs several years later, opening the as-built point cloud acquired at the time of construction allows a detailed understanding of the road’s shape at that time, eliminating the need to revisit the site to re-measure. Moreover, by comparing the stored point cloud data with point clouds acquired during regular inspections, it is possible to quantitatively detect deterioration or settlement due to aging. Through such use, point cloud data functions as a digital twin that reproduces the completed state in virtual space, aiding long-term monitoring and preventive maintenance of infrastructure assets.

Furthermore, if as-built records using point clouds are retained, they are also effective in preventing construction defects and contractual disputes. Even if later there are claims such as “It differs from the drawings” or “Was it not properly constructed?”, having the objective evidence of the point cloud from the time allows facts to be confirmed quickly. Even in situations where paper documents or photographs lack persuasive power, three-dimensional point cloud data can scientifically demonstrate the construction work to clients and third parties, helping to prevent unnecessary rework and disputes. In this way, point-cloud as-built data can be said to be an asset of great value for both advancing maintenance management and quality assurance.

New Field Operations through Cloud Sharing and Remote Presence

The use of point cloud data is well suited to Internet cloud technologies and remote communication technologies, and is creating new forms of on-site operations. In recent years, cloud platforms capable of handling large-volume digital data such as point clouds have been put in place. Cloud services that allow point cloud data and BIM models to be viewed and measured in a web browser without installing dedicated software have also appeared. As a result, it is now practically possible to upload as-built data acquired on site to the cloud on the spot, and for supervisors or clients in the office or at remote locations to immediately review it. For example, workflows are beginning to spread in which point clouds captured and measured with a tablet or smartphone are transmitted from the site and the results are checked in real time on office PCs. Faster data sharing leads to early detection and correction of rework and faster decision-making, improving overall operational efficiency.

Such cloud use also supports a new inspection method called remote on-site attendance (enkaku rinjō). Remote on-site attendance, as the name suggests, is conducting construction status and as-built inspections remotely without going to the site. It is an attempt to carry out site checks and inspection attendance from the office via camera footage and digital data, and in recent years the Ministry of Land, Infrastructure, Transport and Tourism has been actively piloting supervision and inspection using remote on-site attendance. Point cloud data are precisely the ideal material for remote on-site attendance. If three-dimensional point clouds measured on site and as-built drawings with heat maps are shared online, inspectors can examine that data in detail from the office. For example, when confirming the as-built condition of rebar, inspectors can zoom in on captured point cloud data or models on-screen to check dimensions and pitch, enabling confirmations comparable to in-person inspections. At one site, it was reported that they shared as-built reports automatically generated from point clouds with the client on the cloud and conducted efficient inspections by pinpointing and verifying only the necessary locations on site.

In the future, such remote inspections and electronic deliveries based on electronic data will become even more common. Instead of the traditional practice of handing over documents as paper drawings or on DVD media, it is expected that submitting and reviewing 3D data via the cloud will become the mainstream workflow. In fact, on the client side, preparations for acceptance environments are progressing—such as revising electronic delivery guidelines and introducing viewers that support 3D models/point clouds—and the vision that “in the future, all as-built management documentation will be digitized and inspections will be completed online” has become more realistic. The utilization of point cloud data is not limited to mere on-site efficiency improvements; it has become a driving force for digital transformation (DX) of the entire construction process.

The barriers to adopting point cloud technology are decreasing

The advantages of as-built management using point clouds are diverse, but some readers may worry, “Aren't advanced 3D technologies only for large companies or specialized sites?” However, rest assured. The barriers to adoption have dropped significantly in recent years. First, on the hardware side, the emergence of compact, inexpensive measurement devices is notable. 3D laser scanners, which used to be expensive, are now commonly rented, and in some cases small construction firms can obtain point cloud data by, for example, outsourcing measurements to external surveying companies. Photogrammetry services using drones have also become widespread, opening a path to utilize point clouds through outsourcing even without specialized staff. Recently, simple point-cloud surveying apps for smartphones and handheld 3D scanners that can be purchased for around several hundred thousand yen have begun to appear, and the barrier of initial investment costs has definitely started to fall. In fact, there are cases where a small company processes point clouds acquired with an iPhone's LiDAR using free software and uses them for as-built checks.

On the software side, low-cost and free software equipped with point cloud processing and 3D viewer functions is increasing. There are also general-purpose 3D applications and cloud services that can display composite views with BIM/CIM models, so basic analysis and visualization are possible even without expensive dedicated systems. In addition, national and local governments are expanding support measures for ICT adoption by small and medium-sized enterprises, promoting the spread of digital technologies through workshops, guidebook distribution, subsidy programs, and the like. From a practical standpoint, point cloud measurement mainly involves the task of "setting up the equipment and pressing a button," so it does not require extensive skill and is considered easy for new technicians to learn. In short, the attitude of "first, let's start with what is easy to tackle" is important. The government has also called on industry to "start ICT construction on-site with whatever technology is suitable," and point cloud utilization is becoming a solution to challenges across the entire industry, not just for large companies.

The first step in adoption is high-precision surveying with smartphones × GNSS (LRTK)

Although the usefulness of point cloud as-built management can be understood, some people may still wonder, "Where should I start?" In such cases, we recommend beginning with high-precision surveying that combines a smartphone and a compact GNSS receiver. Recently, solutions have appeared that allow easy centimeter-level positioning using small positioning devices that can be attached to a smartphone and an app. This is called LRTK, an innovative method that fuses the portability of smartphones with the accuracy of RTK positioning technology.

Specifically, a small GNSS device attached to a smartphone performs high-precision satellite positioning and applies correction information in real time, reducing the smartphone GPS error that used to be on the order of several meters (several ft) down to a few centimeters (a few in). If you record a position in the dedicated app, the acquired point will be assigned accurate latitude, longitude, and elevation in the World Geodetic System. Using this, even without a veteran surveyor, one person can quickly perform control point surveys and essential as-built measurements. Surveying tasks that used to be carried out by two-person teams setting up a total station can be replaced with just a smartphone and a slender pole. For example, if you want to measure several points of elevation difference on slopes or roads on site, with LRTK you can obtain the three-dimensional coordinates of points one after another just by tapping the positioning button.

Surveying with smartphones and GNSS is low-cost and simple yet can achieve surprisingly high accuracy, making it an ideal entry point for beginners who want to start using point clouds and BIM/CIM. First, experience improved on-site surveying accuracy and increased work efficiency with LRTK, and then, along that progression, if you try smartphone LiDAR scanning features, you can gain hands-on experience acquiring and using point cloud data in a simple way. By gradually becoming accustomed to digital measurement in this way, you should be able to promote your company's on-site DX smoothly without having to assemble large-scale equipment. In fact, smartphone surveying tools like the LRTK series are attracting attention as solutions compatible with *i-Construction*, and they have already begun to be used for rapid terrain surveying at disaster sites and for simplifying as-built management.

The key is to take one step forward from what you can already do. The benefits of point-cloud as-built management and BIM/CIM integration are not only a threefold increase in accuracy but also a major stepping stone toward future operational innovation. While experiencing the benefits of digital technology through easy, high-precision surveying with smartphones × GNSS, gradually expand the scope of 3D point cloud and BIM/CIM utilization. Make the latest technologies your ally and take the first step to evolve future construction management 「efficiently, with high quality, and safely」.