Streamline Disaster Recovery with ICT! Cut Recovery Time by 50% Using Drones

Challenges and Needs in Disaster Recovery

In recent years, natural disasters such as earthquakes, typhoons, and heavy rainfall have occurred frequently across Japan. When roads are severed, river embankments breach, or landslides occur, the swift restoration of infrastructure is directly tied to the safety and security of communities. However, traditional civil engineering disaster recovery often requires large numbers of personnel and long durations, and it can take a long time to remove isolation and restore lifelines in affected areas. The larger the scale of damage, the longer recovery work tends to take, and during that time the regional economy and residents’ lives are significantly affected. From this situation, there is a strong demand for improving the efficiency of recovery works, namely drastically shortening work time and reducing labor.

In addition, the construction and civil engineering industries face a serious shortage of workers, meaning that when large-scale disasters occur, a limited workforce must cover many recovery sites. Attention is therefore focused on the use of drones and ICT (information and communication technology). By leveraging the latest technologies to streamline processes from survey and measurement to design and construction management, it is becoming possible to shorten the overall recovery construction period to half or less of conventional durations. This article explains concrete methods to accelerate and improve accuracy in each process by introducing ICT and drones into disaster recovery sites, achieving recovery time reductions of 50% or more.

Limits of Conventional Methods and Causes of Delay

When a large-scale disaster occurs, the first steps are to assess the site and damage, followed by consideration and design of recovery methods, and then contracting and construction. However, conventional methods have various bottlenecks at each stage, which tend to delay the start of recovery. Major causes of delay include the following points.

• Surveys and measurements take time: Traditionally, engineers went to the site and used total stations and staffs to measure the dimensions and topography of damaged objects one by one. Extensive surveys could take several days to several weeks, during which full-scale recovery design could not begin.

• Work restrictions in hazardous areas: Disaster sites often have unstable ground and risk of secondary disasters. It is difficult to obtain information about places people cannot approach, such as steep slopes or swollen rivers, and work may have to be suspended or postponed to ensure safety.

• Delays in information sharing and decision-making: If damage information and survey data obtained on site are managed as paper drawings or photos, it takes time to share them among stakeholders. Time is spent sending data to responsible departments or partner companies and explaining in meetings, which delays decisions on recovery policy and the start of design.

• Person-dependent tasks and labor shortages: Traditional surveying and design work relies heavily on the experience and intuition of skilled personnel, concentrating burden on a limited number of technicians. When many sites are damaged by a single disaster, there were problems with insufficient manpower to quickly cover all locations.

Because of these factors, traditional disaster recovery often took a long time from immediately after the disaster to project completion. The expected solution to these issues is the reformation of business processes through drone and ICT utilization.

Speed and Accuracy of Situation Assessment by Drone

A drone (unmanned aerial vehicle) aerial survey and photogrammetry are extremely effective for understanding the situation at disaster sites. Flying a drone over the area allows capturing the entire damage range, including dangerous places where people cannot enter, in a short time. By capturing consecutive aerial photos using a drone equipped with a high-resolution camera and analyzing them with dedicated software, detailed orthophotos (top-down real images) and three-dimensional models of the site can be created. Surveys that traditionally took days when done manually can be completed in just a few hours to about one day by combining drone aerial photography and image analysis. For example, at one road recovery site, topographic surveying that had taken four days in total was completed in half a day after drone introduction. In another river project, initial site surveys that had taken over a week were reduced to about two days with a drone. By using drones in this way, the time required for situation assessment can be dramatically compressed, enabling earlier recovery plan formulation.

The advantages of drone surveying are not limited to speed. Data acquired from the air are extremely high-density and high-accuracy, providing detailed topographic information unobtainable by manual surveying. Point cloud data totaling tens of millions of points allow accurate grasping of fine terrain undulations and tilts of damaged structures. This enables detection of small collapses and cracks that were previously overlooked, leading to more appropriate choices of recovery methods. Also, because drones can record damage without exposing workers to danger, they are excellent from a safety improvement perspective. In fact, during Typhoon No. 19 in 2019 (the East Japan Typhoon), drone footage of breached embankments was quickly shared and used to help plan recovery work. Initial inspections that were previously reliant on helicopters or manpower can now be performed quickly and in detail thanks to the mobility of drones.

Displacement Analysis and Volume Estimation Using Point Cloud Data

Three-dimensional point cloud data obtained by drones or laser scanners are powerful for quantitatively analyzing topographic changes and displacements of structures caused by disasters. A point cloud is digital data that represents the surfaces of terrain or objects by countless measured points; by comparing pre- and post-disaster topography, you can determine where and how much displacement or movement occurred. For example, in the case of a large landslide, overlaying pre-collapse topography data and a post-collapse point cloud model allows precise calculation of the volume of collapsed soil (earthwork volume). Traditionally, on-site longitudinal and cross-sectional surveys were conducted and approximately estimated, which took time, but using digital point cloud comparison makes it possible to obtain reliable figures in a short time.

Displacement analysis works similarly. On a slope where a landslide has occurred, you can read from point cloud data how many meters (how many ft) the ground surface moved compared to before the disaster, and which parts subsided or heaved. Dedicated analysis software can display the elevation difference between two point cloud models as a color-coded heat map so deformation amounts are immediately understood. This enables quick prioritization of areas with particularly large changes or high risk within the damaged site. In addition, the calculated soil volume directly informs the number of trucks and heavy equipment to arrange and waste disposal planning, making it important data for optimizing the recovery schedule and costs. By using point cloud data, it becomes possible to grasp disaster-induced topographic changes quantitatively and intuitively, allowing recovery policy planning at a previously unattainable speed.

Immediate Use from Photogrammetry to Design and Contracting

Orthophotos and point cloud data obtained from drone photogrammetry can be immediately used in the design phase of recovery works. Traditionally, survey and cross-sectional drawings of the damaged site were created before designers examined construction methods, but using photogrammetry deliverables can greatly eliminate that labor. Specifically, on the 3D model of the current situation obtained by drone, you can directly consider the placement of recovery structures and dimension planning.

For example, when constructing a retaining wall on a collapsed slope, you can extract arbitrary cross-sections from the point cloud data to compare pre- and post-disaster topography and instantly derive required wall heights and foundation positions. Designers can examine recovery proposals on their office PCs while viewing the 3D model and calculate construction quantities (earthwork volumes and required materials) on the spot. What used to be a process of on-site surveying → drawing → design planning → quantity calculation taking several days can, with photogrammetry data, be almost completed as a one-stop process.

Moreover, this digital data is useful at the contracting stage. Contracting authorities such as municipalities or managers can provide 3D models and orthophotos directly as explanatory materials to bidders, allowing bidders to prepare estimates and construction plans with an accurate understanding of the site. In some cases, bidders can obtain necessary information from drone data and submit bids without visiting the site. This is a major advantage under emergency contracting with limited time. For contractors, early sharing of detailed as-built data and design conditions enables efficient preparation before construction. By directly linking photogrammetry-derived data to design and contracting processes, unnecessary time lags can be eliminated and the overall duration of disaster recovery projects can be shortened.

Cloud-based Progress and Stakeholder Sharing

In ICT-enabled disaster recovery, data cloud sharing is key to smooth information transmission and efficient progress management. Traditionally, survey data and photos were taken back on USB drives and stored on internal servers, or paper drawings were distributed to stakeholders, but using the cloud allows direct uploading and immediate sharing from the site. For example, if 3D models and orthophotos generated from drone imagery are saved to a project folder on the cloud, design staff and contracting authorities in remote offices can view and review the data the same day. If design modifications are required, the latest data can be exchanged in real time between the site and the office, enabling rapid adjustment of on-site responses.

Also effective is visualizing progress on cloud services and sharing it among all stakeholders. If schedules, completion status of each phase, and daily site photos and drone imagery are consolidated on a dashboard, contractors, designers, and contracting authorities can discuss while viewing the same up-to-date information. This reduces recognition gaps when problems arise and speeds decision-making on countermeasures. Disaster recovery often involves multiple agencies (national, prefectural, municipal governments, and lifeline operators), so centralizing data and reports on the cloud prevents coordination errors and enables smooth role allocation.

Furthermore, cloud utilization contributes to more efficient data processing. Computation-intensive tasks such as point cloud analysis and generating 3D models from photos can be executed on high-performance cloud servers to obtain results in a short time. The ability to use the latest technologies regardless of the site PC environment is another benefit. Disaster records accumulated in the cloud are also useful for future disaster prevention planning and maintenance management. For example, past disaster point cloud data can be referenced to analyze terrain changes or compared with cases from other regions. As described above, open information sharing via the cloud is an indispensable element for improving efficiency and smooth collaboration among stakeholders in recovery works.

Examples from Municipalities and Contractors (Concrete Use Cases)

Actual cases where drones and ICT have been used to streamline disaster recovery have been increasing across regions. Among municipal initiatives, using drones in the immediate initial response after a disaster is prominent. One local government used drone aerial photography to complete damage assessment of a large landslide site that had previously taken several days of manual surveying, completing it within the same day and enabling quick recovery planning. In another prefecture, orthophotos from drones were used to record the condition of a forest road collapsed by heavy rain in detail, allowing early emergency repair contracts to be issued. When municipalities take the lead in introducing ICT, the diffusion of digital technology use to local construction firms follows, and a new norm of “check the site with a drone first in disasters” is beginning to form.

Among contractors, small and medium-sized civil engineering firms are setting up in-house ICT teams and conducting training in drone operation and 3D data processing. One construction company in Tokyo quickly responded to the Ministry of Land, Infrastructure, Transport and Tourism’s i-Construction initiative and introduced drone surveying and ICT-enabled construction equipment in-house. As a result, surveying work was reduced by nearly 80% compared to before, and safety dramatically improved because personnel no longer needed to enter dangerous slopes. The company used 3D models of collapsed areas to rapidly develop earthwork plans in recovery projects and achieved shorter construction times and labor savings through stake-free construction with ICT-operated machines. Such efforts also strengthen corporate competitiveness, and the firm has been highly regarded by contracting authorities as a contractor that “completes work safely in a short period.”

Simple Surveying on Site Using LRTK (Control Point Surveying, As-built Confirmation, Position Recording)

In addition to upstream process efficiencies from drones and the cloud, new technologies for simple on-site surveying are proving powerful in actual construction. A representative example is LRTK, an RTK surveying technology combining smartphones with GNSS (Global Navigation Satellite System). RTK (Real Time Kinematic) improves positioning accuracy by applying differential corrections to satellite position information, achieving errors down to a few centimeters (a few in). Traditionally, high-precision GNSS equipment costing millions of yen was required, but recently, by attaching a small RTK receiver to a smartphone or using high-performance GNSS chips built into smartphones, a smartphone itself can be used as a high-precision surveying instrument.

Using LRTK dramatically streamlines on-site control point surveying, as-built (finished shape) confirmation, and position recording. For example, the placement of construction control points that formerly required two people and half a day can be done by one person in a short time with just a smartphone and correction information. The acquired control point coordinates can be shared immediately with stakeholders via the cloud, eliminating downtime waiting for survey results. Smartphone-based RTK surveying is also intuitive to operate with apps, so even on-site staff without special qualifications can achieve a certain level of accuracy. This enables on-site teams to independently perform minimum necessary surveying even when arranging professional surveyors is difficult in emergencies.

LRTK is also useful for as-built confirmation. You can measure heights point by point with a smartphone to confirm whether embankments or slope shaping have been constructed according to design. If necessary, you can compare design data and measured points on the smartphone to check for discrepancies. Because these checks can be done in real time, tasks that previously required post-construction confirmation using batter boards (stakes and string) can be avoided, allowing early correction of rework or material shortages. In addition, because smartphones have built-in cameras, measurement point photos can be saved with location information, useful for recording locations of discovered buried objects or positions of cracks requiring observation. Introducing simple surveying with LRTK enables agile measurements and immediate information sharing on site, greatly enhancing on-site response capability for recovery works.

Implementation Steps and Cost Estimate

Introducing the drone and ICT utilization described above requires phased planning and investment, but when done appropriately it yields significant effects. Typical implementation steps and approximate cost ranges are as follows.

• Clarify needs and develop a plan: First, identify which parts of your organization’s work processes you want to digitize and improve. Organize needs such as damage recording, design data sharing, and construction management, and formulate an ICT introduction plan.

• Select technologies and prepare equipment: Decide on the ICT technologies and equipment to be used. Specifically: drone platforms (airframes with aerial cameras and, if needed, LiDAR-equipped units), photogrammetry software (or cloud services), 3D design CAD or CIM tools, cloud sharing platforms, and LRTK devices (smartphone GNSS receivers). Also prepare in advance for regulatory procedures such as drone flight permits and radio use applications.

• Develop personnel and build a system: Train operators to handle equipment and personnel to perform data analysis. Drone operation is now subject to national certification, so secure certified staff internally or collaborate with external professionals. Establish a dedicated team or person to promote ICT utilization and serve as a bridge between site and headquarters/design departments.

• Pilot introduction (pilot project): Rather than using it immediately on an actual disaster site, trial the operation on small sites or past data. Test the entire flow from drone shooting to data processing and cloud sharing, identify issues and operational challenges, and conduct internal training and demonstrations to deepen site staff understanding of new technologies.

• Full-scale introduction and standardization: Based on lessons from pilots, integrate ICT into actual disaster recovery operations. Standardize operational rules internally and externally, such as using drone initial surveys and requiring 3D data submission in tender conditions. Verify effects (time savings, improved safety, etc.) through real projects and refine operational flows with feedback and improvements as needed.

Regarding costs, there is a range depending on the equipment and services introduced. As a reference, an industrial drone package including airframe, camera, batteries, and peripheral devices typically costs hundreds of thousands to several million yen. Photogrammetry software for surveying can be bought outright from several hundred thousand yen, and cloud service usage can start from tens of thousands of yen per month. Smartphone-connected GNSS receivers for LRTK are also available on the market for around one hundred thousand yen, making them a lower-cost alternative to expensive dedicated equipment. Although initial investment is required, savings from reduced construction time and labor costs and reduced social losses due to earlier recovery often justify the investment. Also, national and local government programs and subsidies promoting ICT adoption may be available, so it is advisable to proceed step by step while utilizing such support measures.

Future Outlook (AI Integration, Remote Support, Integrated Dashboards)

Disaster recovery support using ICT and digital technologies is advancing rapidly, and more advanced applications are expected in the future. Finally, here are some anticipated prospects.

• Automated damage assessment with AI: Going forward, AI (artificial intelligence) is expected to analyze images and point cloud data obtained by drones to automatically recognize damage areas and assess danger levels. Research-stage efforts are already underway to have AI detect slope movements from multi-temporal point cloud data and classify building damage levels via image recognition. If AI can instantly generate draft damage reports and priority maps for recovery, initial response speed will dramatically improve.

• Remote support and unmanned construction: Use of virtual sites constructed from drone data and live video for remote expert support and instruction will expand. For example, a technician at headquarters could view 3D site imagery through VR goggles and provide precise advice to on-site staff. Advances in communication infrastructure and robotics may also make remote operation of heavy equipment and automated construction common at disaster sites. If remote-controlled machines can remove soil and perform emergency works at dangerous collapse sites without people entering, both safety and speed of operations will further increase.

• Integrated dashboards and disaster prevention platforms: The use of integrated dashboards that manage multiple information sources in one place is also expected to grow. Disaster response involves diverse information: terrain data from drones, site work progress, allocation of materials and personnel, as well as weather information and evacuation status. In the future, a disaster information platform that integrates these data will allow stakeholders to share data and make decisions in real time. Overlaying a 3D map of the affected area with a construction dashboard on a single screen will enable everyone to have the same situational awareness and coordinate and direct recovery work more efficiently than ever.

In this way, the efficiency and sophistication of civil engineering work through ICT are steadily progressing in the disaster recovery field. With technological advances, it is becoming realistic to carry out recovery with speeds and accuracy previously unattainable by humans. Actively adopting advanced technologies such as drones and AI to achieve faster recovery and reconstruction in affected areas is likely to become the standard in civil engineering and disaster prevention going forward.