5 DX Case Studies to Streamline Civil Engineering Survey Work: Boosting Productivity with AI and IoT

In recent years, as the construction industry faces labor shortages and the need to adapt to work-style reforms, digital transformation (DX) has been rapidly advancing in various survey tasks within the civil engineering field. By introducing AI, IoT, cloud services, and smart sensors into tasks such as topographic surveys, geological surveys, environmental monitoring, traffic surveys, and subsurface structure detection, improvements in efficiency, labor savings, and safety are being achieved. This article introduces five DX case studies applied to civil engineering survey work. For each case we explain the "Background & Challenges", "Technologies Introduced", "Results & Effects", and "Future Prospects", detailing the aims of adoption, specific technologies used, and the benefits realized. At the end of the article, we also touch on the use cases for simple surveying with LRTK combining a smartphone and a high-precision GNSS receiver, considering further possibilities for improving survey operations.

1. Topographic Survey DX with Drones × 3D Surveying

Background & Challenges

Topographic surveys (surveying), essential for civil engineering works, have traditionally involved surveyors on site using total stations and levels to take numerous point measurements, often requiring teams of people to compile topographic maps. In large sites, surveying could take several weeks, and work in mountain areas or on steep slopes always carried risks. The process required significant manpower and time, and there were limits to measurement point density, forcing reliance on experience-based estimation to understand the entire site. With a shrinking and aging workforce, obtaining accurate topographic information quickly with limited personnel has become a major challenge.

Technologies Introduced

To address these issues, 3D surveying using drones (UAVs) has emerged. This technology equips small unmanned aircraft with high-resolution cameras or LiDAR laser scanners to automatically measure terrain from above. By capturing many aerial photographs with a drone and performing photogrammetry, ortho-rectified images and detailed 3D terrain models can be generated. Drones fitted with laser scanners can acquire high-density point cloud data even on wooded slopes or complex terrain. Acquired data are processed automatically in the cloud, enabling visualization and analysis without being physically present on site. The Ministry of Land, Infrastructure, Transport and Tourism promotes drone surveying as part of "i-Construction," and major construction companies have already actively adopted drone surveys for roadworks and dam construction.

Results & Effects

The impact of DX in topographic surveys using drones is substantial. The time required for surveying has been greatly reduced; broad-area surveys that used to take surveying teams weeks can now be completed in days or even hours in some cases. For example, at a dam construction site, drone surveying was performed without setting ground control points and still obtained high-accuracy surface data in a short time. As a result, information needed for design and construction planning can be consolidated rapidly. In addition to efficiency gains, safety has improved. Drones can safely capture the current situation remotely in hazardous cliffs or disaster sites where human entry is dangerous. Reducing work at heights and eliminating the need for personnel to be stationed in risky areas lowers occupational hazard risks. The intuitive 3D models also facilitate smoother explanations and consensus-building with clients. Photos and point cloud data can be directly used for progress management and as-built (deliverable) reporting, reducing the time spent preparing reports. Because point cloud data allow free cross-sectional cuts and measurements of distances and volumes, the need for re-measurements is reduced and productivity increases dramatically.

Future Prospects

Drone × 3D surveying technology will continue to evolve and become standard for survey tasks. In the future, autonomous flight and automated analysis are expected to become commonplace, enabling a "one-stop surveying" workflow in which a drone autonomously flights a specified area and completes everything from data capture to model creation. If real-time capture and processing allow immediate generation of 3D models on site to quickly confirm terrain changes and construction progress, the boundary between surveying and construction management will further blur. Integrating acquired data with BIM/CIM (3D model–based construction management) and automating comparisons with design drawings can streamline plan revisions and as-built inspections. Advances in smaller, higher-performance sensors and cloud analytics will expand applications to continuous monitoring and infrastructure inspections. Drones will play an increasing role in bridge and tunnel maintenance, contributing to labor savings and the sophistication of inspections. Topographic survey DX will continue to progress as a foundation for future unmanned construction and smart construction technologies.

2. Geological Survey DX Evolving with Sensors and AI

Background & Challenges

For civil planning and disaster prevention, geological surveys must determine ground composition and stability. Traditionally, data were collected fragmentarily through borehole surveys and sampling, and specialists estimated stratigraphy. Point-source investigations, however, cannot fully capture widespread subsurface structures and risk overlooking unknown discontinuities or weak zones. Additionally, identifying potential landslide and debris-flow sites triggered by earthquakes or heavy rainfall often required labor-intensive field inspections in mountainous areas. Heavy reliance on veteran engineers’ experience led to knowledge concentration and inefficiencies.

Technologies Introduced

In this field, DX combining IoT sensors for data collection and AI analysis has advanced. One example is using AI with microtremor exploration that infers subsurface structure by measuring tiny ambient ground vibrations. Multiple vibration sensors are installed to collect ground motion data, and machine learning analyzes this data to automatically generate 3D subsurface models that previously only experts could interpret. A survey company has used microtremor equipment and proprietary AI to visualize underground geological structures, producing 3D ground models that non-experts can understand. More recently, disaster-prevention IoT sensors for real-time ground monitoring have been introduced. Systems place inclinometers (digital tilt sensors) on slopes to detect minute displacements and wirelessly transmit data to the cloud. By detecting early signs of slope movement during heavy rain and issuing alerts, such systems can signal anomalies without requiring personnel to be stationed on site. The massive data from these sensors are centrally managed in the cloud, where AI learns abnormal patterns to help extract hazardous areas and predict failures.

Results & Effects

DX in geological surveys has dramatically improved the accuracy and efficiency of ground information acquisition. AI-generated 3D subsurface models reveal underground conditions that are not apparent from traditional borehole logs or plan views. This makes it intuitive to analyze, for example, how weak layers are distributed within a slope or which areas are high-risk when compared with past disaster histories. Extracting dangerous ground locations from data rather than relying solely on expert judgment is a major achievement. In one study, satellite-derived terrain data and historical disaster records were combined with AI-analyzed subsurface models to automatically identify landslide-prone locations. Moreover, adopting disaster-prevention IoT sensors has enabled continuous monitoring, significantly reducing the patrol burden on municipal staff. Even during the night or heavy rain, remote detection of slope changes supports rapid evacuation decisions and preemptive road closures. In field trials, multiple inclinometer sensors on slopes detected tiny landslides during heavy rain, and subsequent site inspections confirmed actual collapses — capturing early deformations that human patrols might have missed. These results contribute to early warnings and damage mitigation. Overall, sensor + AI geological survey DX enables data-driven, objective ground assessments, reducing labor and improving reliability.

Future Prospects

Future prospects for geological survey DX include further data integration and improved predictive accuracy. Platforms that synthesize geological sensors, drone-derived topography, satellite remote sensing, and weather data will likely emerge to capture comprehensive ground-state changes. For example, AI could simultaneously analyze surface changes (satellite imagery) and subsurface changes (sensor data) to predict landslide occurrence probabilities in real time. Wider adoption of low-cost sensors and LPWA communications (low-power wide-area) will extend smart monitoring to remote and small-to-medium-sized sites that were previously hard to measure. Municipalities could deploy sensor networks around landslide-prone areas and aging infrastructure to realize preventive maintenance through continuous monitoring × AI analysis. Additionally, leveraging vast amounts of cloud-stored ground and disaster data could lead to higher-precision ground risk assessment services and the provision of 3D geological maps. Geological survey DX will continue to develop as an essential technology for disaster prevention, mitigation, and infrastructure maintenance.

3. Hydrological & Meteorological Observation DX: Strengthening Disaster Response with IoT Multi-Point Monitoring

Background & Challenges

Observing river water levels, rainfall, and weather data is vital for disaster prevention and infrastructure operation. Traditionally, hydrological observation stations for small and medium rivers were limited, and many municipalities relied on staff patrolling sites for visual checks. During heavy rain especially, officials sometimes risked visiting rivers to gather information for decision-making. Such manual patrols lack real-time responsiveness and spatial coverage, making it difficult to keep up during simultaneous, widespread heavy rainfall events, particularly at night. With the increasing frequency of typhoons and localized torrential rains, the workload on municipal staff has grown and insufficient rapid evacuation information has become a serious problem. Similarly, construction sites often depended on analogue methods like manual rain checks by workers, complicating objective work-stop decisions and efficient schedule replanning.

Technologies Introduced

To address these challenges, IoT-based multi-point automatic hydrological and meteorological observation has been adopted widely. Specifically, sensors such as river water level gauges, rain gauges, and anemometers are installed at numerous hazard points or across wide areas, and data are aggregated to the cloud via wireless communication. Using LPWA networks (e.g., Sigfox or LoRaWAN), which offer low power consumption and long-range communication, enables affordable sensor networks in mountainous or small-river areas lacking power or communications infrastructure. In one municipal trial, 13 compact ultrasonic IoT water level gauges were installed at five small rivers with a history of flooding, constructing a system that transmits water level data to the cloud at five-minute intervals. Officials can check water levels in real time on a web app from office PCs or smartphones, and automated alerts trigger when thresholds are exceeded so anomalies can be detected even at night. Similarly, private companies now offer IoT weather observation services. Rain and wind sensors installed at construction sites send data via LTE to centralized cloud platforms for unified management. These systems are used for site safety, allowing office-based monitoring of multiple sites and sending email alerts to stakeholders when thresholds are exceeded. For river monitoring, remote cameras with AI are also being combined; network cameras placed on bridges can have AI analyze footage for signs of rising water or overflow, enabling risk assessment even where no water level gauge exists.

Results & Effects

Hydrological and meteorological DX via IoT multi-point observation has enabled real-time, spatially comprehensive disaster information. Sensor networks have made 24/7 data collection possible even on small rivers where gauges were scarce, allowing continuous tracking during heavy rain. As a result, municipal staff no longer need to rush around at night and can perform safe indoor river monitoring, enabling earlier evacuation advisories in communities with vulnerable residents. One town reported that after installing IoT water level gauges, the need for on-site patrols decreased, reducing staff burden and speeding decision-making. Data are stored in the cloud, making time-series analysis easy; analysts can determine which locations begin to rise first and when peaks occur, informing next-flood countermeasures and improving hazard map accuracy. There are also initiatives to share real-time observation data with residents and other agencies. Publishing water level and rainfall data online allows residents to make evacuation decisions and enables upstream and downstream municipalities to coordinate responses. In construction, IoT weather observation allows work-stop and restart decisions to be based on objective data. For example, if a rule states that sustained rainfall above a certain level for one hour triggers an automatic alert and work suspension, safety management can be enforced without relying on human judgement. This leads to fewer near-miss incidents and optimized scheduling. Overall, IoT-based hydrological and meteorological observation DX enhances disaster response capabilities and operational efficiency, giving stakeholders greater peace of mind and capacity.

Future Prospects

Future developments will emphasize data utilization and automated control. For instance, AI may analyze multi-point water level data to forecast river flooding risk and warn municipal disaster-management staff in advance. IoT observation systems could be integrated with gates and drainage pumps to enable automatic water-gate operation and pumping based on sensor inputs, making smart water management technically feasible. Some platforms already trial the use of water level sensor information via interfaces to support remote water-gate operation decisions. In meteorology, private weather companies and research institutes are expected to build mesh sensor networks, deploying many low-cost temperature, humidity, and rain sensors in urban areas to generate high-resolution weather data for monitoring localized heavy rain or urban heat island effects. Large volumes of measured data can be processed as big data and used by AI to detect signs of abnormal weather early and improve weather modeling. Providing observation information to the public is also important: real-time disaster data from IoT observations distributed via smartphone apps would enable anyone to grasp risk on the spot, promoting voluntary evacuation. Hydrological and meteorological observation DX will move beyond mere data accumulation to intelligent utilization, becoming a powerful tool for disaster prevention, mitigation, and reliable infrastructure operation.

4. Traffic Survey DX: 24/7 Automatic Counting with AI Cameras

Background & Challenges

Traffic surveys for roads and city streets provide essential baseline data for urban planning, commercial facility siting, and traffic-safety measures. Traditionally, these surveys relied on surveyors stationed at locations who manually counted passing vehicles and pedestrians. Because personnel were placed at limited time windows such as morning and evening rush hours, the resulting data were temporally and spatially limited. Visual counting by people introduced opportunities for mistakes and subjective bias, undermining accuracy. In addition, substantial costs associated with labor and traffic guidance made frequent surveys infeasible. Consequently, detailed data such as the long-term trend of pedestrian counts in front of a particular store or year-round traffic volumes accounting for seasonal variation were often unavailable, hindering fine-grained urban understanding and long-term analysis.

Technologies Introduced

Recently, DX using AI-equipped cameras and IoT communication has advanced for traffic surveys. Smart street cameras capture video and AI analyzes it in real time to automatically count people and vehicles. One startup developed a traffic survey service using an AI image-recognition algorithm called "IDEA." This uses IoT cameras with edge-processing capabilities (a small computer and communication module built into the camera) mounted on building eaves, where the camera performs on-device analysis of pedestrian and vehicle counts and attributes (such as estimated age groups or gender). Only the result data are sent to the cloud, minimizing bandwidth and protecting privacy. The cloud platform stores and visualizes collected data on a dashboard so users can see real-time traffic graphs and heat maps in a browser. With such a solution, a single camera installation can collect 24/7, year-round traffic data. Local governments have already begun implementing these systems: Mie Prefecture, for example, installed AI cameras at 10 major roads in fiscal 2021 to monitor traffic constantly and publish the data on the prefectural website. Advances in AI now allow higher-level analyses beyond just people and vehicles, including vehicle-type classification (passenger car vs. large vehicle, bicycle vs. pedestrian), speed estimation, and congestion-length (queue length) measurement. In future, integration with road-embedded sensors and communication beacons could evolve these systems into real-time traffic monitoring platforms.

Results & Effects

Traffic survey DX using AI cameras has enabled objective data collection independent of human labor. A major benefit is the dramatic improvement in coverage and continuity of data. Previously only snapshot information at specific times and places was available, but AI cameras can collect data every day, day and night. This accumulates insights on day-of-week and seasonal fluctuations and detailed time-of-day patterns that were previously unattainable. For example, a building owner who installed an AI count camera in front of a commercial building discovered that "weekday daytime foot traffic is high but weekends are surprisingly low" and that "pedestrian counts drop sharply in rain while vehicle traffic rises," yielding new findings. AI cameras also make continuous measurement at pinpoint locations feasible—where earlier only major intersections were surveyed, a camera can be placed at any desired spot to measure long-term foot traffic in front of a vacant storefront for retail catchment analysis. Data accuracy is stable because AI applies the same logic consistently, eliminating human variability and oversight. The 24/7/365 data judged by identical criteria are unique and valuable. Consequently, personnel cost reductions and simplified survey workflows were achieved. Previously several staff were required for setup and aggregation for each survey, but now device installation enables automatic aggregation and graphing in the cloud, leaving staff to simply review the data—resulting in substantial labor savings. Moreover, analyzing accumulated data enables new policy measures. Combining data from multiple AI cameras can map areawide pedestrian and vehicle flows to inform event attendance forecasting or optimize public transport placement, enabling data-driven urban planning.

Future Prospects

Traffic survey DX is expected to evolve into a core technology for smart cities and traffic management. Data from AI cameras can be used in real time to adjust signal control and ease congestion. Some municipalities are trialing systems where AI automatically adjusts signal cycles based on detected traffic volumes as a form of "smart signal control," aiming to reduce peak congestion and CO₂ emissions. In the era of autonomous vehicles, roadside AI camera information will be an important data source for vehicle-to-vehicle and vehicle-to-infrastructure communication, providing real-time traffic flow data to vehicles for route optimization and driving assistance. Analysis of human flows can also support crime prevention and urban vitality initiatives. A trial at a tourist site used AI cameras to capture pedestrian counts and attributes for marketing applications. With privacy protection and anonymization, pedestrian count data can measure event impact or evaluate shopping street vitality. Technically, further AI image-recognition improvements and fusion with acoustic sensors and radar will enable all-weather operation. Stable performance at night and in adverse conditions will build a more reliable data foundation. Traffic survey DX will not only improve survey efficiency but also become essential for building digital twins of cities, transforming urban planning and mobility management.

5. Subsurface Structure Detection DX: Visualizing Buried Objects with GPR and GPS

Background & Challenges

Under roads and underground spaces, various buried objects (subsurface structures) such as water pipes, gas pipes, and communication cables exist. Conducting safe civil engineering and excavation requires pre-construction detection of subsurface conditions, known as buried-object surveys. Traditionally, this work depended on skilled technicians performing manual operations with ground-penetrating radar (GPR) and metal detectors. Even when scanning surfaces with GPR units, interpreting raw waveform data to determine pipe positions and depths required advanced experience, leading to variability in results. Old as-built drawings are often inaccurate, and excavation damage from missed pipes is a frequent cause of accidents. Survey results were often recorded only as rough paper sketches and not effectively used as a future asset, leading to increased project risk and cost due to insufficient subsurface documentation.

Technologies Introduced

DX-based subsurface detection solutions have emerged to tackle these issues. Modern GPR equipment is available in user-friendly formats like compact cart systems and is increasingly integrated with GPS/GNSS receivers and IMUs (inertial measurement units). For example, a manufacturer's "smart GPR" combines a high-sensitivity radar and high-precision GNSS to automatically capture underground utility position information simply by scanning the surface. The operation is simple: a single pass scan displays mapped underground pipe locations on the device screen. Acquired data can be uploaded to the cloud for storage and analysis, enabling a digital 3D buried-utility map to be shared and utilized. AI image-recognition is also being introduced: radar survey data are analyzed by AI in real time to indicate probabilities such as "this reflection is likely a water pipe," allowing non-experts to infer subsurface features. Some integrated construction consultancy firms now offer services that 3D-model underground gas, water, and sewer pipes for customers. By combining high-precision survey equipment, existing drawings, and terrestrial laser scanners, they measure and convert subsurface structures into 3D data—service operation began in some areas in 2021. These DX technologies are rapidly advancing visualization of the underground.

Results & Effects

The main benefits of subsurface detection DX include significant improvements in construction safety and efficiency. Using smart GPR devices, tasks that previously took specialists half a day can now be completed quickly by a single operator. GNSS integration ensures that all survey results’ coordinates are accurately recorded on maps, greatly reducing oversights and recording errors. This prevents accidental damage to pipes and cables during excavation and cuts unnecessary rework and recovery costs. On sites that adopted detection DX, operators reported that "knowing buried-utility locations beforehand allowed accurate adjustment of pouring positions during construction" and "we could proceed with excavation work more confidently than before." Another major advantage is data assetization. Survey results that used to exist only in personal notebooks or paper drawings can now be shared digitally within and outside organizations. Created 3D buried-utility maps can be used for future repairs and maintenance, eliminating repeated surveys of the same location. Building a database with records of new installations and relocations could eventually enable an urban-wide subsurface infrastructure ledger. If data are shared among stakeholders including administrative bodies, buried-utility information can be immediately referenced during roadwork procurement, improving design and cost-estimation accuracy. For example, providers of 3D model services reported more accurate bids and reduced estimation errors in excavation volumes and schedules. Overall, subsurface detection DX reduces safety and economic risks and improves efficiency and quality across the lifecycle from planning through maintenance.

Future Prospects

Subsurface visualization will become increasingly sophisticated and an essential element of infrastructure management. One trend is sensor advancement: beyond electromagnetic radar, future ultra-sensitive technologies such as quantum magnetic sensors may detect deeper buried objects and smaller-diameter pipes. Research will also focus on combining multiple sensing methods to correct for geological effects and boost accuracy. Another trend is data integration: integrating 3D data for various buried utilities in GIS to overlay with surface terrain and building information will enable a combined aboveground-and-underground digital twin, offering a complete view of urban underground spaces for planning new facilities and evaluating the impact of disasters on subsurface structures. AR (augmented reality) integration is also anticipated: when a tablet or smart glasses are pointed at a site, the locations of buried pipes and structures could be displayed on the view. If workers can intuitively confirm hidden hazards using AR at a construction site, safety and work efficiency will improve immensely. From an administrative perspective, pre-excavation surveys and submission of data may become part of standard procurement procedures. Overseas examples already mandate subsurface surveys before excavation, and in Japan, regulatory frameworks may evolve alongside DX adoption. Subsurface detection DX, though understated, supports urban infrastructure foundations and is expected to expand and mature further.

Closing: Emerging Technologies Supporting On-Site DX and the Potential of Simple Surveying with LRTK

Above, we introduced five DX case studies that contribute to improved efficiency in civil engineering survey work. In each case, applying digital technologies such as AI, IoT, and cloud services to field operations has produced labor savings and accuracy improvements that were not possible with conventional methods. While these DX efforts are implemented in specific areas, the overarching point is the importance of accurately and easily acquiring on-site conditions as data and sharing and utilizing that data. To conclude, we highlight LRTK (smartphone × high-precision GNSS receiver) simple surveying as a noteworthy technology that embodies this concept.

LRTK is a recent solution that brings real-time kinematic (RTK) GNSS positioning capabilities to smartphones. By attaching a pocket-sized high-precision GNSS receiver to a smartphone or tablet and launching a dedicated app, on-site positioning and 3D scanning become easy for anyone. Specifically, the smartphone camera (and in some cases built-in LiDAR) captures photos and point clouds while GNSS provides real-time positioning data. This enables on-the-spot generation of georeferenced current-condition data with centimeter-class accuracy. For example, for a small site survey that previously required a surveyor to observe many points and draft drawings, simply walking around the site with an LRTK device and sweeping the phone can capture surrounding terrain and structures as point cloud data. The point clouds are tagged with coordinates in the global geodetic reference, so later overlaying with GIS or CAD yields minimal misalignment and allows precise field verification. Because you can measure distances and heights on the point cloud without using tape measures on site, revisits and re-measurements can be avoided, directly reducing labor.

The advantage of LRTK is that this high-precision surveying can be easily performed by one person. Traditional RTK-GNSS surveying required setting up base stations and specialized knowledge, but LRTK apps automate complex settings so users can simply walk with a smartphone to complete the task. The devices are lightweight, enabling field technicians to carry them routinely and use them quickly when needed. LRTK is useful for field verification of data obtained by drone surveys or subsurface exploration. Survey results that are difficult to interpret on paper become clear when the point cloud model is displayed on a smartphone and compared on site. If additional measurements are required, LRTK can immediately capture coordinate values and upload them to the cloud for sharing. With accuracy comparable to public-survey control points—around 1–2 cm—LRTK is practical for construction management and as-built verification support. Moreover, LRTK devices often include cloud-upload features that automatically generate and analyze 3D models, reducing office work time. Overall, LRTK is a groundbreaking tool that enables anyone, anywhere, immediately to obtain high-precision surveying and position data, significantly streamlining coordinate acquisition and field verification in civil engineering surveys.

While DX in civil engineering surveys is progressing across many fronts, it is crucial to connect these technologies for overall optimization rather than treating them as isolated measures. By combining drones, IoT sensors, AI analytics, and handheld surveying technologies like LRTK, the entire process—from survey planning to data acquisition, analysis, and field feedback—can be seamlessly linked. Repetitive and tedious tasks traditionally handled by humans can be delegated to digital technologies, allowing engineers to focus on higher-value decision-making and creative work. DX-driven efficiency and labor reduction in survey tasks will be increasingly important for infrastructure maintenance and disaster response. Please consider the case studies and technologies presented in this article as inspiration for DX initiatives within your organization. By adopting advanced technologies well, the civil surveying field can continue to achieve both productivity gains and quality improvements.