Table of contents

• Introduction: the importance of differential earthwork volume management and conventional challenges

• What is a point cloud? Its value as on-site survey data

• Basics of earthwork volume calculation and the definition of "differential earthwork volume"

• Workflow and limitations of conventional differential volume calculation methods (TS, level, CAD processing)

• Emergence of smartphone point-cloud surveying: overview and features of LRTK

• Automatic differential volume calculation from point clouds: LRTK Cloud functionality explained

• Accuracy, speed, safety: strengths of smartphone point-cloud surveying seen in comparison

• Practical advantages applicable to as-built management, interim inspections, and client reports

• Case studies: specific examples of shortened schedules and reduced labor costs from adoption

• Low barriers to adoption and ease of in-house rollout

• Conclusion: a proposal to adopt LRTK as the first step toward simplified surveying and smarter site management

• FAQ: point cloud density and accuracy, satellite reception conditions, handling flat and sloped terrain, how to check point-cloud differences, report output, etc.

Introduction: the importance of differential earthwork volume management and conventional challenges

On civil engineering and land development sites, accurately understanding and managing the amount of soil and earth moved (earthwork volume) due to filling or excavation is essential. Whether the prescribed amount of earth has been placed or removed as planned directly affects construction progress management (quantity management), pass/fail of as-built inspections, and even the calculation of construction costs. Accordingly, it is indispensable to survey terrain before and after construction and calculate the differential earthwork volume (the volumetric difference between two ground surfaces) to verify whether embankments were placed according to design and whether the specified excavation quantity was achieved.

However, conventional differential earthwork volume management has faced challenges. Terrain surveying requires manpower and specialized equipment, which is time-consuming, and sparse measurement points on a large site often lead to errors. For example, with coarse spacing over a vast area, small undulations, surplus or shortage of fill, or unexcavated spots can be overlooked, creating risks of later rework or additional work. Surveying in hazardous areas such as cliffs or steep slopes also poses major safety concerns for workers. As earthwork scale increases, traditional manual methods relying on labor become difficult to sustain, prompting demand for more efficient and higher-precision earthwork measurement methods.

Against this backdrop, a new approach has emerged in recent years: differential earthwork calculation using smartphone point-cloud surveying. This article explains the method of differential volume calculation using point-cloud measurements with smartphones and high-precision GNSS technology (LRTK), comparing it with traditional manual work, dedicated surveying instruments, and CAD processing, and describes its advantages in efficiency, labor and effort reduction, safety, accuracy, and cost in an easy-to-understand way.

What is a point cloud? Its value as on-site survey data

Point cloud data (point clouds) are three-dimensional data that represent the surfaces of objects or terrain as a collection of countless points. Each point contains X, Y, and Z coordinate values (and sometimes color or reflectance intensity), and can be acquired by laser scanners or photogrammetry. By analyzing and visualizing the obtained large number of points, complex terrain and structures can be reproduced as detailed and precise 3D models. Surface variations and shapes that are difficult to grasp from planar drawings or photos can be recorded and visualized in situ with point clouds, making them useful across design, construction, and maintenance.

Point clouds have many values as on-site survey data. First, because the current state can be recorded in true three dimensions, it is easy to extract arbitrary sections later or remeasure dimensions—providing high reusability of the data. Once a point cloud is acquired, cross-sections or shape checks for required locations can be created in the office, reducing the need for additional field surveys and improving efficiency. Second, because a point cloud consists of an enormous number of points, it functions as a sort of spatial scan that comprehensively captures terrain undulations and shapes. This enables detection of minute elevation differences and terrain changes that limited point surveys may miss, reducing variability in construction quality. More recently, it has become possible to overlay design data onto acquired point clouds and color-code the difference between existing conditions and design (creating heat maps). This allows intuitive judgment of acceptability in construction management and inspections, positioning point clouds as a promising new visualization tool for sites.

Basics of earthwork volume calculation and the definition of "differential earthwork volume"

Earthwork volume calculation means computing the volume of terrain or the volumetric amount of fill/excavation. Generally, it involves integrating the volume of a solid bounded by a reference surface and the target terrain. Examples include the volume of fill relative to a flat reference plane or the amount of soil moved between original ground and developed ground. Among these, the quantity obtained by calculating the volumetric difference between two different ground surfaces is called the differential earthwork volume, and it is used to evaluate increases or decreases in earth materials (fill or excavation) before and after construction.

To compute differential earthwork volume, survey the pre-construction ground and the post-construction terrain separately, and calculate volume from height differences between the two. Traditionally, the ground before and after works was represented by a grid or multiple survey points along section lines, and volume was commonly calculated by the average end area method using section areas and distances derived from elevation differences at each point. Another approach is to generate TIN (triangulated irregular network) models from pre- and post-construction survey points and use CAD software to automatically compute the volume difference between the two terrain models. In any case, the key is estimating earthwork volume by taking the difference between two reference terrain datasets, and accurate execution of this process ensures fairness and economic efficiency in earthworks.

Workflow and limitations of conventional differential volume calculation methods (TS, level, CAD processing)

Conventional differential earthwork calculation mainly required the use of surveying instruments and CAD software. Let’s look at a typical workflow.

• Pre-construction survey (understanding pre-construction terrain): Before starting work, survey ground elevations on site. Use instruments such as total stations (TS) and auto-levels to measure the elevations of many points in a grid or acquire terrain profiles along representative longitudinal and transverse lines. This requires establishing benchmarks (known reference points) and height references via leveling, and to capture a wide area in detail one must plan measurement point layouts and often reposition instruments several times, which is laborious.

• Post-construction survey (understanding post-construction terrain): After embankment or excavation completion, survey the finished terrain again to obtain post-construction elevation data. Methods are similar to pre-construction surveys, but temporary structures or restricted areas after construction may make acquiring necessary data difficult.

• Volume calculation using CAD: Based on pre- and post-construction measurement points, create cross-sections or surface models in drawings or CAD software and compute volumes. For each longitudinal/transverse line, draw pre- and post-construction terrain sections and calculate section area differences multiplied by distances, summing volumes for each segment; or generate surface models from point sets and let the software automatically compute volume differences between the two models. Both approaches require specialized CAD and earthwork calculation skills, and data organization and consistency checks take time.

• Verification and reporting of results: Compare calculated fill and excavation volumes with construction management standards and confirm they are within specified ranges. If shortages or excesses are found, analyze causes and perform additional filling, excavation, or corrective work as needed. If no issues are found, prepare drawings and tables to compile as-built management documents and reports for submission to the client.

These conventional methods have major inefficiencies due to labor-intensive steps. Surveying typically requires a 2–3 person crew and may take from half a day to several days on large sites. Complex terrain or large elevation differences require more measurement points, and trying to cover the entire site increases effort dramatically. Even so, interpolation between points is inevitable, leaving fine unmeasured undulations as sources of error. The CAD-based process of converting survey results into drawings and calculating volumes is also cumbersome and difficult without specialized skills. Because results take time to produce, immediate on-site as-built judgments are often impossible, delaying recovery when rework is needed. Moreover, surveying sometimes requires entering hazardous areas, complicating arrangements for safety personnel and schedules—overall indicating considerable room for improvement.

Emergence of smartphone point-cloud surveying: overview and features of LRTK

A promising solution to these challenges is smartphone point-cloud surveying. In particular, LRTK (Long Range RTK) is a technology that combines smartphones with high-precision GNSS receivers to enable centimeter-level positioning and 3D scanning by anyone. LRTK is an integrated surveying solution combining a smartphone, RTK-GNSS, and cloud services. By attaching a compact receiver device called the "LRTK Phone" to a smartphone, smartphone positioning—normally with meter-level accuracy—improves dramatically, allowing real-time self-positioning with horizontal accuracy of several cm (several in) and vertical accuracy of several cm (several in). The idea is that by simply attaching a device weighing a few hundred grams to a handheld smartphone, the phone becomes a high-precision surveying instrument.

Starting the LRTK-compatible smartphone app allows you to receive RTK correction information while positioning and recording data. It supports network RTK (Ntrip), which delivers correction data from base stations via the internet, enabling centimeter-level positioning even while moving. It can also maintain high accuracy in areas without cellular coverage by receiving augmentation signals from Japan’s quasi-zenith satellite system (QZSS, "Michibiki") via CLAS, so it can be used even where mobile signals don’t reach.

In recent years, smartphones have also begun to include high-performance LiDAR sensors. For example, the latest phones incorporate infrared LiDAR depth sensors that can measure distances up to about 5 m (16.4 ft) and perform quick 3D scans of surroundings. LRTK combines smartphone LiDAR point-cloud measurement capability with RTK-GNSS positioning accuracy. By simply walking around with a smartphone, you can capture a huge cloud of points, each assigned high-precision coordinates in real time. What once required specialized equipment and skilled operators is becoming feasible for anyone with a smartphone.

Automatic differential volume calculation from point clouds: LRTK Cloud functionality explained

High-precision point clouds captured by smartphone can be uploaded to the cloud immediately for analysis. LRTK’s cloud service includes functions to automatically calculate differential earthwork volumes from uploaded point clouds, and the process is very simple.

First, register a reference terrain dataset in the cloud. This can be a pre-construction point cloud or a design-stage 3D terrain model (BIM/CIM data or LandXML files). Next, upload the comparative dataset for post-construction (or any other time). On the cloud, the two datasets are automatically georeferenced (aligned) and displayed overlaid. Because LRTK’s high-precision positioning already places point clouds within the same coordinate system, in many cases no special adjustment is required for accurate overlay.

Once aligned, the software analyzes height differences between the two terrains and calculates volume differences. Specifically, it compares mesh models generated from each point cloud and accumulates the volumes of differing areas. For example, comparing pre-excavation ground point clouds with post-excavation ground point clouds will automatically compute the removed soil volume. Similarly, for embankment works, differential point clouds before and after construction yield the placed fill volume, and for interim as-built checks you can identify deficient or excess fill areas by comparing current point clouds with the design model.

The analysis results are provided not only as numeric volume values but also as color maps (heat maps) for visual feedback. A color distribution map quickly shows where and by how much fill is excessive or insufficient, or where excavation occurred, enabling intuitive understanding of site conditions. You can also slice arbitrary cross-sections and compare the two terrain profiles on the cloud for detailed verification. Analyses that used to be done back at the office on a PC can now be completed immediately from the site by accessing the cloud with a tablet or PC. Large point-cloud processing runs automatically on the server, freeing users from complex CAD tasks and leaving them to simply wait for results.

Accuracy, speed, safety: strengths of smartphone point-cloud surveying seen in comparison

By adopting smartphone point-cloud surveying (LRTK), you gain significant advantages in accuracy, speed, and safety compared with conventional methods. Let’s examine the differences from each perspective.

• Accuracy: In single-point surveying (the measurement accuracy of individual points), conventional instruments like total stations offer millimeter-level precision. Smartphone LiDAR points typically have accuracy on the order of a few centimeters (a few in). However, point-cloud surveying acquires overwhelmingly more points, so statistical errors tend to cancel out, and the overall surface accuracy can reach a sufficiently high level. Demonstration experiments have reported horizontal errors of about 8 mm (0.31 in) when position correction is applied by RTK-GNSS. Moreover, because the entire space is measured as an area, local undulations are not overlooked and you can capture a smoothed, averaged representation of the surface. In other words, smartphone point-cloud surveying secures necessary and sufficient measurement accuracy while improving reliability through comprehensive data coverage.

• Speed: Smartphone point-cloud surveying dramatically shortens the time required for on-site measurement and analysis compared with conventional methods. For example, some sites report that as-built scans with a smartphone were completed in under 5 minutes of actual work. Where conventional surveying crews took half a day to measure and then several hours in the office for CAD calculations, a smartphone scan can yield immediate results on site. Shortening the cycle from measurement to volume calculation enables faster decision-making on site, leading to schedule compression and quick corrective actions. Also, because only a smartphone and a small antenna are needed, setup time is minimal and measurements can be made whenever necessary—an important practical advantage.

• Safety: Point-cloud scanning is non-contact, so measurements can be taken from a safe distance in hazardous areas. On steep slopes or in areas with operating heavy machinery, a brief surrounding scan from a safe position completes the survey, minimizing workers’ exposure to danger. Conventional methods sometimes required placing points on slopes or positioning staff at heights, but with smartphone point-cloud surveying one person can perform a scan from a safe location. Reducing required personnel itself contributes to safety (fewer people means lower risk of human error and accidents). From a site safety perspective, smartphone point-cloud surveying is therefore a valuable tool.

Practical advantages applicable to as-built management, interim inspections, and client reports

Differential earthwork calculation via smartphone point clouds is useful not only for earthwork management but also across many construction management scenarios. Here are representative practical advantages.

• Use for as-built management: As mentioned above, comparing point-cloud data with design data can generate heat maps that color-code conformity. Areas exceeding tolerances are immediately identifiable, allowing proactive correction before formal inspections. This improves pass rates at client inspections and reduces rework. Point-cloud data can also be treated as deliverables meeting as-built management guidelines, enabling the creation of equivalent drawings and reports for public works.

• Interim inspections and progress management: Regular scans at project milestones make visualizing progress straightforward. For large-scale earthworks, acquiring point clouds weekly or monthly quantifies earthwork progress, improving accuracy of quantity management and interim payments. Sharing cloud data for remote sites enables supervisors or clients to check the site from the office. This reduces travel time and the number of required on-site participants while maintaining necessary information sharing.

• Streamlining report creation: Point-cloud measurement data can be used directly as deliverables for electronic submission; the cloud can export LandXML or PDF outputs. Automatically generated cross-section drawings and heat-map images drastically reduce the manual drafting of drawings and tables. Rapidly preparing as-built drawings and earthwork summary tables, and compiling reports with photos and measurement results, reduces the documentation burden. Because all stakeholders can view the same 3D data, explanatory materials also become more persuasive.

Case studies: specific examples of shortened schedules and reduced labor costs from adoption

What concrete effects can you expect after introducing smartphone point-cloud surveying (LRTK)? Here is an example.

On a road improvement project, conventional practice required a specialized surveying team to visit several times per month to measure fill and excavation volumes. A three-person survey crew spent half a day measuring, and CAD calculations were completed the next day for reporting. After LRTK adoption, the site agent himself scanned the site with a smartphone in about 5–10 minutes and could immediately grasp the differential earthwork volume. For one embankment location, the smartphone scan instantly revealed that planned fill of 500 cubic meters had an actual volume of 480 cubic meters—about a 20 cubic meter shortage—and an immediate instruction for additional fill was issued. Real-time on-site verification of as-built conditions effectively prevents rework and ensures quality.

Additionally, eliminating the need to arrange or attend external survey crews resulted in schedule reductions and labor cost savings. In the cited project, more than 15 person-days of effort previously spent on recurring surveys were nearly eliminated, saving several hundred thousand yen annually. With no need to wait for survey results, work interruptions were reduced and overall construction schedule shortened by about 10% versus the original plan. Site staff reported feeling reassured by being able to measure immediately when needed and praised the ability to check quickly in alignment with equipment availability, improvements that are steadily smartening everyday site management.

Low barriers to adoption and ease of in-house rollout

When introducing new technologies, lack of expertise and high initial costs are common barriers. However, the barriers to smartphone point-cloud surveying (LRTK) are relatively low. In terms of introduction cost, you typically only add a small GNSS antenna device to existing smartphones, which is far less expensive than purchasing dedicated 3D laser scanners or high-end surveying instruments. Subscription-based usage plans are also available, allowing flexible operation for only the required periods.

Operational learning and in-house rollout are also straightforward. The smartphone app’s intuitive UI makes operation simple and easy to learn, so field personnel can become competent in a short time without lengthy training. Younger and veteran employees alike are usually comfortable with smartphones, and reports indicate that even survey novices can acquire basic proficiency after a few uses. Equipping each field worker with a smartphone plus an LRTK device in a one-device-per-person setup enables workers to measure their assigned areas on demand, eliminating waiting time for surveying. From an organizational perspective, you can start small—try one set—and gradually scale up, accumulating know-how from small sites and transitioning to digital surveying without overcommitment.

Conclusion: a proposal to adopt LRTK as the first step toward simplified surveying and smarter site management

Smartphone point-cloud surveying and automatic differential earthwork calculation have the potential to transform manual earthwork management. With LRTK, anyone can quickly acquire high-precision as-built data over wide areas, dramatically improving site productivity, safety, and quality. In the construction industry advancing ICT and DX, adopting such accessible smart tools is an effective first step toward smarter site management.

The automatic differential volume calculation and cloud sharing functions introduced here are only part of the benefits that the latest technologies can bring to the field. It is not necessary to replace everything at once, but even partial adoption of digital surveying can help address labor shortages and improve efficiency. Tasks that once required multiple personnel—surveying and as-built verification—can now be completed by a single person with real-time results, and the smartphone surveying experience is changing site norms. Please consider experiencing this ease and accuracy at your site.

FAQ: point cloud density and accuracy, satellite reception conditions, handling flat and sloped terrain, how to check point-cloud differences, report output, etc.

Q: What point-cloud density and accuracy are required for differential earthwork calculation? Is smartphone-acquired point cloud data sufficient? A: For public-works as-built management, a point density of several dozen points per 1 m² (10.8 ft²) is generally recommended for ground point clouds. Smartphone LiDAR is not as dense as industrial laser scanners, but by walking slowly while scanning you can obtain reasonably fine point clouds. With RTK position correction, individual point accuracy typically falls within a few cm (a few in), which is sufficient for routine earthwork calculations. Except in cases demanding millimeter-level accuracy, differential earthwork volumes can generally be computed satisfactorily from smartphone point clouds. If higher accuracy is needed for specific spots, you can supplement with conventional surveying instruments for cross-checks.

Q: Can LRTK be used in sites with poor GNSS satellite reception? What happens in areas without cellular coverage? A: LRTK can use multiple satellite positioning systems such as GPS, GLONASS, and QZSS (Michibiki), so it provides high-precision positioning if satellite visibility is available. In mountainous areas or environments surrounded by tall buildings, satellite signal reception requires attention, but you can mitigate this by starting scans from open sightlines or using Michibiki augmentation via CLAS. If cellular coverage is absent, CLAS support enables real-time correction in some cases. In locations where satellites cannot be received at all—such as inside tunnels—real-time positioning is difficult; in such cases, obtain a reference coordinate near the tunnel entrance and later merge internal scans with exterior points via post-processing to align them.

Q: Are there measurement method differences to note when surveying flat surfaces (flat ground) versus sloped terrain (cut/fill slopes)? A: The basic measurement procedure is the same, but pay attention to scanning positions and angles on slopes. Smartphone LiDAR’s effective range is on the order of a few meters, so scanning an entire high slope at once may be difficult; measure in segments from top to bottom by approaching closer as needed. If necessary, scan both above and below the slope and merge data to cover the full face. Using smartphone photogrammetry mode can sometimes capture wider areas more comprehensively. Flat ground offers better lines of sight and can be scanned broadly at once; however, for extremely large areas consider placing known reference points (target markers) for reliable georeferencing. In all cases, plan scan routes with overlap to avoid missing data—redundancy is key to accuracy.

Q: How can I check differences between acquired point clouds? I’m concerned whether they truly align. A: LRTK Cloud can overlay multiple uploaded point clouds and automatically compute differences. Difference results are displayed as colored 3D models (heat maps), for example showing areas higher than the design in red and lower areas in blue, making disparities immediately apparent. You can also cut arbitrary cross-sections and compare section lines from the two point clouds. This enables quantitative and visual verification of “where” and “how much” difference exists. Because RTK places each point cloud in the correct coordinate system, in principle no extra alignment is necessary for high-precision difference analysis. If you remain uncertain, compare heights of immovable features (e.g., tops of existing structures) between datasets to confirm matching reference elevations.

Q: Can survey deliverables like reports and drawings be output? Does it support electronic submission standards? A: Yes. LRTK systems can output various data products based on acquired point clouds and calculation results. After computing differential volumes, you can export summary reports as PDFs and save automatically generated cross-sections in DXF format. Exporting a surface model in LandXML enables direct use in other civil-design CADs or machine guidance systems. These output formats are designed to conform to the Ministry of Land, Infrastructure, Transport and Tourism’s electronic submission guidelines (draft), facilitating smooth integration with existing deliverable checks. In short, data from smartphone point-cloud surveying are structured to be submitted as official deliverables, simplifying post-adoption documentation workflows.