Road disaster response support system and program, road management method

Abstract

There was a lack of understanding of cavity warning signs during normal times, and in the event of a disaster, advance road clearance plans could not be dynamically updated to reflect the actual situation, resulting in delays and personal delays. There was no integrated operation of authority adjustment across multiple administrators, allocation of equipment and personnel, preparation for communication outages, or reflecting knowledge from training and past performance. Solution: The information acquisition department collects data from satellites, optical fiber, vehicles, etc., and the analysis department performs integrated analysis. The road clearance department registers a preliminary plan, and the road clearance implementation plan is dynamically optimized based on the analysis results. The feasibility of achieving 24 / 48 / 72-hour targets is evaluated using reinforcement learning, and consideration is given to determining the transfer of authority, spatial optimization of bases and equipment, redundant communications, heavy equipment operation and robotics collaboration, training and performance learning, and conflict assessment through intellectual property investigations. Multi-hazard risks are quantified and weighted by priority, and a redundant network is formed in conjunction with the evaluation of bases such as roadside stations. Special equipment and sea and air transport are integrated, and abandoned vehicle removal and the automatic assembly of contracted companies are controlled, and emergency restoration plans are automatically generated along with alternative route selection.

Description

[Technical Field]

[0001] The present invention relates to technology related to road maintenance and management and road clearance during disasters, and in particular to a road disaster response support system, related programs, and road management methods that, in normal times, evaluate the risk of hollowing out of road infrastructure and determine its soundness based on information acquired by multiple sensing means (satellite sensing, optical fiber sensing, vehicle travel sensing, etc.), and, in the event of a disaster, utilizes road clearance plans formulated in advance to quickly formulate road clearance implementation plans to ensure the passage of emergency vehicles, etc. Furthermore, the aim is to provide a flexible and advanced support system that can function effectively in both peacetime and disaster situations, by achieving both wide-area, continuous infrastructure monitoring, which has been difficult to achieve in the past, and rapid decision-making and planning support during the initial disaster response. This system will contribute to the creation of a road management system with high disaster resilience by seamlessly supporting the preventive maintenance phase and the disaster response phase. [Background technology]

[0002] In conventional road infrastructure management, emergency response measures are generally taken after accidents such as road surface collapses or subsidence have occurred, but there has been no adequately established system for widely and continuously monitoring the risk of road hollowing or signs of deterioration during peacetime.In addition, in the event of a disaster, rapid road clearance is required to ensure the passage of emergency vehicles such as ambulances, fire departments, and the Self-Defense Forces, but the information acquisition and judgment processing required to formulate an implementation plan tend to be personal, and there has been a lack of efficient and scientific support technology. In recent years, advances in advanced sensing technologies such as satellites, optical fiber, and connected cars have made wide-area road monitoring technically possible, but there have been limited examples of their integrated use in both infrastructure health assessments and disaster road clearance plans. Furthermore, ground-penetrating radar and other detection methods are point-specific and intermittent, making them incapable of continuous, area-wide assessments. Therefore, there are limitations to their use in detecting signs of disaster risk and in making immediate route decisions immediately after a disaster occurs. On the other hand, in the event of a large-scale disaster, an important administrative task is to carry out emergency restoration (road clearance) prior to normal restoration procedures and ensure access to medical institutions, evacuation centers, and disaster bases. Road clearance here refers to the process of removing obstacles and making simple repairs immediately after a disaster to quickly enable the passage of emergency vehicles, and a system is required to compare a road clearance plan formulated in advance with actual information at the time of the disaster and quickly incorporate it into a road clearance implementation plan. However, conventional technology lacks the configuration to dynamically utilize road clearance plans, and does not adequately implement processes such as updating plans to reflect damage conditions and traffic closure information after a disaster, or optimizing implementation timing. The configuration for flexibly processing information interpretation, weighting, route optimization, etc. using AI technology is also insufficient, which has resulted in a burden on on-site decision-making and delays in initial responses. Furthermore, with the recent increase in inbound tourism, the risk of foreign tourists visiting Japan facing natural disasters is also increasing. However, when a disaster occurs, foreign tourists have limited means to quickly and accurately obtain information on whether roads are passable and evacuation routes, and language barriers and differences in information acquisition routes have led to delayed responses and confusion. Therefore, there is a need to develop a system that provides easy-to-understand, immediate disaster-related traffic information to a diverse range of users, including inbound tourists. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2024-014948 [Patent Document 2] Japanese Patent Publication No. 2024-071243 [Patent Document 3] Japanese Patent Publication No. 2023-162007 [Patent Document 4] JP 2024-136347 A Summary of the Invention [Problem to be solved by the invention]

[0004] Road infrastructure is at high risk of collapses and traffic disruptions during disasters due to subsurface cavities and ground deformation. Conventionally, emergency responses have generally been taken after these events have become apparent, and there has been insufficient understanding and management of cavity risks over a wide area and continuously during peacetime. Furthermore, although individual technologies such as optical fiber sensing, satellite remote sensing, and vehicle driving data exist, a system that integrates and analyzes these to support both preventive maintenance and emergency response in the event of a disaster had not yet been established. For example, the following prior art is known. Patent Document 1 discloses technology for detecting road abnormalities using optical fiber sensors, but does not disclose integrated analysis of satellite data and vehicle driving data, or the use of ground survey results to improve AI models. Patent Document 2 describes a technology for detecting road obstacles using satellite images and determining whether or not a road is passable, but does not disclose heterogeneous sensor integration, risk score output using an AI model, or learning improvements. Patent Document 3 uses AI to analyze images from an in-vehicle camera to determine the disaster situation, but does not explicitly state that it will continuously improve through heterogeneous sensor fusion or retraining. Patent Document 4 discloses a configuration for visualizing data from radar-equipped vehicles on a map, but does not mention multi-dimensional integration, cavity risk assessment using AI, or the use of ground survey results. As such, most conventional technologies are limited to detecting local anomalies based on a single sensor, and a comprehensive system that includes both area-wide and continuous assessment of the risk of hollowing out during peacetime and rational and immediate response support processing in the event of a disaster has not yet been realized. [Means for solving the problem]

[0005] In order to solve the above problem, a road disaster response support system according to the present invention may have the following configuration. a road clearance department that registers road clearance plans that have been formulated in advance and formulates or updates road clearance implementation plans based on the road clearance plans when a disaster occurs; An analysis unit that uses multiple sensing methods (such as satellites, optical fibers, and vehicle traffic) during normal times to evaluate and predict the health of road infrastructure; The system is configured to include an improvement unit that learns and improves the analysis results so that they can be used to make decisions about infrastructure maintenance and management or disaster response support. The road clearance unit may store time milestones (e.g., 24 / 48 / 72 hours) according to the time elapsed since the disaster as internal parameters, and may sequentially correct the road clearance implementation plan according to the estimated probability of achievement based on indicators such as the damage overview, road obstructions, and restoration progress. Ground surveys may be conducted for areas with a high risk of cavities predicted by the analysis unit, and the results may be fed back into the AI ​​model to continuously improve prediction accuracy. After a disaster occurs, the AI ​​model and accumulated geospatial data may be used to help determine rational road reopening routes. Furthermore, the system may be configured to identify impassable sections immediately after a disaster using high-frequency information from optical fiber and satellites, determine priority routes for emergency vehicles, and provide a function to visualize risk scores and the reasons for decisions using explainable AI. It may also include a mode of enhancing the formulation or updating of road clearance implementation plans by combining functions such as authority determination, equipment allocation, base selection, multi-hazard assessment, communication / transport / robotics collaboration, and learning / intellectual property analysis. Furthermore, in some embodiments, the system may include an intellectual property research unit that analyzes claim text from public web data and patent databases, normalizes it to absorb variations in claim wording and differences in the order of constituent elements, evaluates the correspondence between the processing procedures or functions included in road clearance plans and road clearance implementation plans and the constituent elements described in third-party patent claims, and estimates the likelihood of applicability (possibility of conflict) based on the correspondence. This may include automatic claim chart generation, term normalization, requirements mapping, semantic matching of equivalent concepts, etc. The output of the intellectual property research unit may be used to determine whether or not to adopt a function, present alternatives, consider licensing policies, notify risks during external collaboration, and create risk review materials. An example of the configuration of the present invention will be described below in accordance with the claims. [1] A road disaster response support system comprising: an information acquisition unit that acquires information regarding road disaster conditions; an analysis unit that analyzes the information acquired by the information acquisition unit to grasp the road disaster conditions; and a road clearance unit that registers a road clearance plan that has been formulated in advance, the road clearance plan including a timeline including time milestones, road clearance bases that function as bases for dispatching work teams to carry out road clearance work and as collection bases for equipment and materials, and road clearance routes from the road clearance bases, and that optimizes the road clearance plan based on the analysis results by the analysis unit and dynamically formulates or updates a road clearance implementation plan. [2] A road disaster response support system as described in [1], wherein the road clearance unit formulates the road clearance implementation plan with the goal of opening wide-area travel routes within 24 hours, access routes within 48 hours, and routes within the disaster area within 72 hours, depending on the time milestones on the timeline after the disaster. [3] A road disaster response support system as described in [2], wherein the road clearance unit aims to open the wide-area travel route within 24 hours, the access route within 48 hours, and the route within the disaster area within 72 hours, depending on the time elapsed since the disaster; and the analysis unit evaluates the possibility of achieving the time milestones using a reinforcement learning model with at least one of traffic volume simulation, abandoned vehicle information, short-term weather forecast, rescue request information, damage status information, recovery status information, or the operating status of the equipment and personnel as input, and dynamically formulates or updates the road clearance implementation plan based on the evaluation results. [4] [1] A road disaster response support system as described in [1], wherein the analysis unit divides disaster response into the phases of initial response, emergency recovery, and full-scale recovery, and determines whether or not to transition to each phase taking into account the achievement status of the time milestones, and the road clearance unit dynamically switches the contents of the road clearance implementation plan based on the determination result, and optimizes the allocation of materials and equipment, the deployment of personnel, and the priority of the road clearance work according to each phase. [5] [1] A road disaster response support system as described in [1], wherein the analysis unit calculates the combined probability of occurrence of earthquakes, tsunamis, floods, landslides, snow damage, volcanic eruptions (including ash fall disasters), and nuclear disasters using a multi-hazard analysis model, and quantifies the risk of disruption that each disaster poses to the road clearance route, and the road clearance unit dynamically weights the priority of the road clearance work for each time milestone based on the quantified risk score, and updates the road clearance implementation plan. [6] [1] A road disaster response support system as described in [1], characterized in that the road clearance unit takes into account the occurrence of multiple disasters, including earthquakes, tsunamis, floods, landslides, snow damage, volcanic eruptions (including ash fall disasters), and nuclear disasters, evaluates the impact of each disaster, reevaluates the road clearance plan, and dynamically formulates or updates the road clearance implementation plan based on the reevaluation. [7] [1] A road disaster response support system as described in [1], wherein the information acquisition unit acquires drone aerial footage, in-vehicle drive recorder footage, or vehicle driving sensor data in real time, the analysis unit analyzes the real-time data using image analysis or a machine learning model to identify the presence or absence of rubble, flooding, fire, infrastructure damage, progress of restoration work, or road obstructions, and classifies the obstructions or estimates the required restoration time, and the road clearance unit dynamically formulates or updates the road clearance implementation plan, including switching the road clearance route, changing the priority of the road clearance work, or rearranging the equipment and materials, based on the analysis results. [8] [1] A road disaster response support system as described in [1], wherein the road clearance unit estimates the quantity of the equipment and materials required for the road clearance implementation plan, and if a shortage is expected compared with the stockpiled amount, dynamically selects a temporary storage site or supply base for the equipment and materials and reflects this in the road clearance implementation plan. [9] [8] A road disaster response support system as described in [8], wherein the road clearance unit predicts the demand for heavy machinery, personnel, fuel and the above-mentioned materials and equipment required for each disaster scenario, and when a shortage is expected compared with the stockpiled amounts, dynamically selects a temporary storage site for the materials and equipment or a fuel supply base based on a space optimization model that inputs geographic information system data and land use data, and reflects the selection results in the road clearance implementation plan.

[10] [1] A road disaster response support system as described in [1], wherein the road clearance unit acquires in real time at least one of the remaining fuel, operating time, failure prediction information, and available operating time for the heavy equipment or vehicles used in the road clearance work, evaluates the possibility of continued operation based on the acquired information, determines the need for replenishment or maintenance, and reflects this in the selection of a temporary storage site or fuel supply base for the equipment and materials, and in updating the road clearance implementation plan.

[11] [1] A road disaster response support system as described in [1], wherein the road clearance unit registers port facilities, temporary landing bases, temporary maritime transport bases, airport facilities, or river boat transport bases, as well as Self-Defense Force air cushion boats, transport ships, and other special equipment as part of the road clearance bases, and the analysis unit comprehensively evaluates the transportation time, fuel consumption, weather conditions, tidal conditions, and safety of maritime transport routes, air transport routes, and land road clearance routes, optimizes the transportation efficiency of relief supplies, medical supplies, or the equipment and materials, and further optimizes the material transportation plan simultaneously with the road clearance implementation plan.

[12] [1] A road disaster response support system as described in [1], comprising a robotics unit that controls disaster response robots or robotic equipment to perform the road clearance work, and the robotics unit collects operation data of the disaster response robots or robotic equipment from training for the road clearance work in peacetime, normal work at construction sites, or operation records in a simulation environment for each disaster category of earthquake, tsunami, flood, landslide, snow damage, volcanic eruption (including ash fall disaster), and nuclear disaster, learns from the data to improve the behavioral model, and continuously improves performance by feeding back the results of activities during actual disasters, and reflects the results of the improvements in the allocation of robot work or auxiliary work plans in the road clearance implementation plan.

[13]

[12] A road disaster response support system as described in

[13]

[12] , wherein the robotics unit controls the disaster response robot or the robotics equipment based on the analysis results by the analysis unit and the road clearance implementation plan to remove rubble, excavate, transport the equipment and materials, or perform temporary construction, and feeds back the results of the execution to the analysis unit to reflect them in improving the next road clearance implementation plan.

[14]

[13] A road disaster response support system as described in

[14]

[13] , wherein the robotics unit cooperatively controls a plurality of the disaster response robots or a plurality of the robotics devices using a common control signal, and coordinates a plurality of the disaster response robots or a plurality of the robotics devices made by different manufacturers or models to carry out continuous construction work of excavation, transportation, dumping, and compaction, and further feeds back the results of the cooperative construction to improve the division of roles or the cooperative algorithm and reflect this in the next road clearance implementation plan.

[15] [1] A road disaster response support system as described in [1], characterized in that the road disaster response support system uses a large-scale language model to collect data on the Internet and learn about disaster prevention measures regardless of the type of language, thereby continuously improving the road clearance plan, the road clearance implementation plan, or the road clearance work.

[16] [1] A road disaster response support system as described in [1], comprising an intellectual property research department, which analyzes patent claim information registered in a patent database using a large-scale language model to normalize constituent elements, absorbs variations in expression and differences in the order of the constituent elements, evaluates the correspondence with the processing procedures or functions included in the road clearance implementation plan, estimates the possibility that the processing procedures or the functions fall under the constituent elements described in the scope of a third-party patent claim, and outputs the estimation result.

[17] [1] A road disaster response support system as described in [1], characterized in that the road disaster response support system includes an intellectual property research department, which collects information on third-party systems, programs, or methods published on the Internet, analyzes patent claim information related to the road disaster response support system using a large-scale language model to normalize constituent elements, absorbs variations in expression and differences in the order of the constituent elements, evaluates the correspondence between the constituent elements and the collected information, estimates the possibility that a third-party implementation will satisfy the constituent elements based on the correspondence, and outputs the estimated result.

[18] A road management method using a computer, wherein the computer acquires information relating to road disaster conditions via a network, analyzes the acquired information to grasp the road disaster conditions, and registers a road clearance plan that has been formulated in advance, the road clearance plan including a timeline including time milestones, road clearance bases that function as bases for dispatching work teams and as collection bases for equipment and materials for carrying out road clearance work, and road clearance routes from the road clearance bases, and optimizes the road clearance plan based on the results of the analysis, and executes a process to dynamically formulate or update a road clearance implementation plan.

[19]

[18] A road management method according to the present invention, characterized in that the computer executes a process to formulate the road reopening implementation plan with the goal of opening wide-area travel routes within 24 hours, access routes within 48 hours, and routes within the disaster area within 72 hours, depending on the time elapsed since the disaster as the time milestones on the timeline.

[20]

[19] A road management method according to the present invention, characterized in that the computer sets the goal of opening the wide-area travel route within 24 hours, the access route within 48 hours, and the route within the disaster area within 72 hours, depending on the time elapsed since the occurrence of a disaster, and further evaluates the possibility of achieving the time milestones using a reinforcement learning model with at least one of traffic volume simulation, abandoned vehicle information, short-term weather forecast, rescue request information, damage status information, recovery status information, or the operating status of the equipment and personnel as input, and dynamically formulates or updates the road opening implementation plan based on the evaluation results.

[21]

[18] A road management method as described in

[18] , characterized in that the computer divides disaster response into the phases of initial response, emergency recovery, and full-scale recovery, determines whether or not to transition to each phase taking into account the achievement status of the time milestones, dynamically switches the contents of the road clearance implementation plan based on the determination result, and executes processing to optimize the allocation of materials and equipment, the deployment of personnel, and the priority of the road clearance work according to each phase.

[22]

[18] A road management method as described in

[18] , characterized in that the computer calculates the combined probability of occurrence of earthquakes, tsunamis, floods, landslides, snow damage, volcanic eruptions (including ash fall disasters), and nuclear disasters using a multi-hazard analysis model, quantifies the risk of disruption that each disaster poses to the road clearance route, dynamically weights the priority of the road clearance work for each time milestone based on the quantified risk score, and executes a process to update the road clearance implementation plan.

[23]

[18] A road management method as described in

[18] , characterized in that the computer quantifies and calculates multiple evaluation indicators such as the speed of rescue operations, the urgency of saving lives, the importance of restoring lifelines, the need to ensure logistics, and the scale of isolated settlements or isolated people, weights them including the isolation status of medical institutions, welfare facilities, or evacuation shelters, evaluates the evaluation indicators in an integrated manner to calculate the priority of the road clearance work, and dynamically formulates or updates the road clearance implementation plan based on the priority.

[24]

[18] A road management method as described in

[18] , wherein the computer simulates scenarios including secondary disasters such as personnel shortages, bridge collapses, other infrastructure damage, renewed snowfall, or aftershocks in addition to communication outages, fuel supply outages, or delays in procuring the materials and equipment, evaluates the feasibility of the road clearance implementation plan under each scenario using a probabilistic model or multi-agent simulation, and based on the evaluation results, selects a redundant alternative route or alternative base from multiple alternative routes or alternative bases, and dynamically formulates or updates the road clearance implementation plan.

[25] A program including a sequence of instructions for causing a computer to execute at least one of the following (A) to (C): (A) the function of the road disaster response support system described in any one of [1] to [2], [4] to [8],

[10] to

[12] , and

[15] to

[17] ; (B) the processing of the road management method described in any one of

[23] to

[24] ; (C) at least one of the functions of the following (1) to (4): (1) the function described in [3] of dynamically formulating or updating a road clearance implementation plan based on the evaluation results; (2) the function described in [9] of reflecting the selection results in the road clearance implementation plan; (3) the function described in

[13] of feeding back the execution results to the analysis unit and reflecting them in improving the next road clearance implementation plan; (4) the function described in

[14] of improving the role allocation or collaboration algorithm based on the results of collaborative construction and reflecting them in the road clearance implementation plan. [Effects of the Invention]

[0006] According to the present invention, the risk of hollowing out of road infrastructure can be grasped area-wide and continuously during normal times, and advanced preventive maintenance that was difficult to achieve with conventional inspection-based management can be realized. In the event of a disaster, a road clearance implementation plan can be quickly formulated based on a road clearance plan that was prepared in advance, and rational route selection that reflects the results of AI analysis of infrastructure conditions and the impact of the disaster will ensure the passage of emergency vehicles and optimize initial responses. In addition, by integrating and analyzing information from different sensors such as optical fiber, satellites, and vehicle driving data, it is possible to grasp cavity risks and fault conditions with high accuracy, and by continuously training this information into an AI model, it is possible to improve prediction accuracy and responsiveness in the long term. Furthermore, by providing visualization functions and feedback mechanisms based on explainable AI, it will support the decisions of on-site users and local government officials, enabling practical and reliable disaster response. In addition, by equipping the system with functions for obtaining correction information using robotics, multilingual support, and notification functions based on traveler attributes, it will be possible to quickly provide information on passability and danger avoidance in the event of a disaster to foreign tourists visiting Japan, contributing to reducing confusion and secondary damage. [Brief explanation of the drawings]

[0007] [Figure 1] 1 is a system configuration diagram of a road disaster response support system according to the present invention. [Figure 2] FIG. 10 is a flowchart illustrating an example of a patrol flow. [Figure 3] FIG. 10 is a flowchart illustrating an example of a flow of a fixed camera. [Figure 4] FIG. 10 is a diagram illustrating an example of a patrol situation. [Figure 5] FIG. 10 is a diagram illustrating an example of an optical fiber inspection situation. [Figure 6] FIG. 10 is a diagram illustrating an example of a satellite survey situation. [Figure 7] FIG. 10 is a diagram illustrating an example of weather conditions. [Figure 8] FIG. 2 is a diagram illustrating an example of a vehicle driving situation. [Figure 9]1 is a flowchart illustrating an example of the flow of a road disaster response support system according to the present invention. [Figure 10] FIG. 10 is a diagram showing an example of a determination criterion in a determination unit of the present invention. [Figure 11] 11 is a diagram showing an example of a hardware configuration of a road disaster response support system according to the present invention. Note that the configuration diagrams and flowcharts shown in FIGS. 1, 9, 10, and 11 are merely examples and do not comprehensively illustrate all elements related to the present invention. Configurations and processes not explicitly shown in the drawings (e.g., road clearance department, robotics department, intellectual property investigation department, authority determination, equipment and material allocation, base selection, multi-hazard evaluation, communication / transport / robotics collaboration, intellectual property investigation, etc.) may also be included in embodiments of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, one embodiment of the road disaster response support system of the present invention will be described with reference to the drawings. Note that the following description does not unduly limit the technical idea of ​​the present invention described in the claims. Not all of the configurations described in this embodiment are essential constituent elements of the present invention, and each individual element constituting a group of features may be an independent invention. In addition to supporting emergency response in the event of a disaster, the system of this invention can also be used to assess the soundness of road infrastructure during normal times and to detect signs of risk of subsurface cavities. This will contribute to preventing accidents and road collapses, and to optimizing long-term maintenance costs through infrastructure maintenance support. The system may also be configured to provide information in multiple languages ​​to foreign users, including inbound tourists, via smartphones, etc. This will enable foreign tourists to instantly grasp the passability of their current location and the vicinity of their destination, risk avoidance routes, road conditions, etc. in the event of a disaster, and support them in making decisions about safe actions. It should be noted that "information relating to road disaster conditions" is a concept that encompasses multiple types of information used to grasp the state of road infrastructure during a disaster, such as disaster information, road information, traffic information, and weather information.

[0009] 1 shows an example of the configuration of a road disaster response support system 600 according to the present invention. The road disaster response support system 600 includes an information acquisition unit 610 that acquires information on patrol status, optical fiber survey status, satellite survey status, weather conditions, and vehicle driving status, an analysis unit 620 that analyzes the various conditions acquired by the information acquisition unit 610, and a decision unit 630 that decides the need for road disaster response based on the analysis results. The road disaster response support system 600 of the embodiment is communicatively connected to a patrol status providing server 100, an optical fiber investigation status providing server 200, a satellite investigation status providing server 300, a weather condition providing server 400, and a vehicle driving status providing server 500 via a network NW. In order to grasp the patrol situation, only one vehicle Vh and terminal device TM are shown in Fig. 1, but multiple vehicles Vh and terminal devices TM may be connected to the network NW. In order to grasp the disaster situation, etc., only one fixed camera CAM is shown in Fig. 1, but multiple fixed camera CAMs may be connected to the network NW. Based on the analysis results, the decision unit 630 may determine whether road infrastructure maintenance and disaster response are necessary, and may instruct response processing such as the formulation and updating of repair plans and road clearance implementation plans as necessary. Processing related to road clearance may be handled by the road clearance unit, which may distribute and update plans based on the estimated results of time milestones and achievement probabilities. Furthermore, the road disaster response support system 600 may include functional blocks such as an infrastructure maintenance department, an improvement department, a robotics department, and an intellectual property research department. If necessary, it may be configured to incorporate information on road service status, output from the prediction department, and multilingual distribution by the information provider. The output from the intellectual property research department may be reflected in the adoption or rejection of functions, the presentation of alternative methods, consideration of contracts and licenses, and risk notification during external collaboration.

[0010] The terminal device TM, fixed camera CAM, patrol status providing server 100, optical fiber investigation status providing server 200, satellite investigation status providing server 300, weather status providing server 400, vehicle driving status providing server 500, and road disaster response support system 600 communicate via a network NW. The network NW includes, for example, some or all of a WAN (Wide Area Network), LAN (Local Area Network), the Internet, a provider device, a wireless base station, a dedicated line, a satellite line, etc. The communication method is not limited to the network NW, but data can also be sent and received via a memory card. Data can also be downloaded and uploaded via the network NW. In addition, in the present invention, the road disaster response support system 600 may be provided with a function for providing information regarding the road service situation, and the information may be acquired from a dedicated server, a terminal device, or the like via the network NW.

[0011] The terminal device TM is used by a user who gets into the vehicle Vh. The terminal device TM is a mobile phone such as a smartphone, a tablet terminal, or the like. The terminal device TM may be a communication-type drive recorder or a stationary in-vehicle device mounted on the vehicle Vh, or may have an image analysis function using AI (artificial intelligence).The vehicle Vh may also have an under-road cavity detection function (technology that irradiates electromagnetic waves from above the road toward below the road and estimates the locations of cavities and buried pipes from the reflected waves), or the vehicle Vh may be an under-road cavity detection vehicle. The terminal device TM has a road patrol application installed therein that cooperates with the patrol status providing server 100 . The terminal device TM has a positioning device such as a GPS (Global Positioning System) receiver, a communication device for connecting to the network NW, an input / output device such as a G sensor (acceleration sensor), a camera, and a touch panel, and a processor such as a CPU (Central Processing Unit).

[0012] 2 is a flowchart showing an example of the flow of patrol. When the terminal device TM presses a patrol start button on the road patrol app (S1), it starts collecting location information, acceleration information, video, etc. (S2). After the patrol is completed, the user presses the patrol end button in the road patrol application (S3), and the position information, acceleration information, video, etc. of the terminal device TM are transmitted to the patrol status providing server 100 (S4). The patrol status providing server 100 determines whether the road surface is uneven or not based on the measurement information transmitted from the terminal device TM, and identifies the location of the road surface that is determined to be uneven. Also, based on the transmitted video and images, it determines the state of damage to the road surface, and identifies the location of the road surface that is determined to be a risk location.

[0013] Fixed camera CAMs are installed on buildings and roadside posts and poles around areas prone to flooding, such as roads (highways, major trunk roads, roads with heavy traffic, major bus routes, roads connecting to schools, public facilities, and emergency hospitals, roads along mountains and in mountainous areas, etc.), underpasses (roads that are dug down at intersections), roads along rivers, and roads along the sea. Fixed camera CAMs include live cameras, web cameras, and network cameras that are capable of communication. The fixed camera CAM may be a communication-type drive recorder or a small unmanned aerial vehicle such as a drone, or it may be equipped with image analysis functions using AI (artificial intelligence). The fixed camera CAM has a built-in camera application that communicates with the patrol status providing server 100 . The fixed camera CAM includes a lens, an image sensor, a positioning device such as a GPS (Global Positioning System) receiver, a communication device for connecting to the network NW, and a processor such as a CPU (Central Processing Unit).

[0014] 3 is a flowchart showing an example of the flow of the fixed camera CAM. The fixed camera CAM periodically or intermittently collects road conditions (S5), and periodically or intermittently automatically transmits images, location information, date and time information, etc. to the patrol status providing server 100 (S6). The patrol status providing server 100 judges the state of damage to the road surface based on the video and images transmitted from the fixed camera CAM, and identifies the locations judged to be risky locations.

[0015] The patrol status providing server 100 provides the patrol status via the network NW to the road disaster response support system 600. The provided patrol status is information for each road, and includes some or all of the disaster status and obstacles to vehicle traffic, such as road damage, road subsidence, roadbed washout, roadway collapse, pavement damage, roadside gutter damage, road surface unevenness, cavities under the road, liquefaction, snow accumulation, snow quality, and ice on the road surface, collapsed buildings, vehicle traffic history, power outages, earthquake damage, tsunami damage, typhoon damage, eruption damage, volcanic activity damage, tornado damage, landslides, landslides, soil runoff, slope collapses, tunnel collapses, collapsed shoulders, fallen trees, falling rocks, road scouring, total bridge damage, river bank collapses, river flooding, flooding, dense fog, avalanches, blizzards, and the presence or absence of accident vehicles and stranded vehicles. FIG. 4 is a diagram showing an example of a patrol situation, in which damaged road areas 110 are displayed in black on the map.

[0016] The optical fiber investigation status providing server 200 utilizes optical fiber sensing technology that uses optical fiber as a sensor, receives backscattered light from communication optical fibers included in cables laid on roads, etc., detects vibration patterns according to the driving conditions of vehicles on the roads, etc. based on the backscattered light, and acquires the driving conditions of vehicles on the roads, etc. and the surrounding road conditions, etc. from the detected vibration patterns and a learning model (information from a camera connected to the optical fiber, etc., may also be used).In addition, by analyzing the intensity and frequency changes of minute vibrations propagating underground and continuous abnormal patterns of waveforms, it is possible to grasp the risk of underground cavities occurring and signs of ground deformation. The optical fiber inspection status providing server 200 provides the optical fiber inspection status to the road disaster response support system 600 via the network NW. The optical fiber inspection status provided is information for each road, and includes some or all of the following disaster conditions (including images of the road) and obstacles to vehicle traffic, such as vehicle traffic history, traffic volume, traffic congestion, sudden vehicle stops, traffic accidents, snow accumulation on the road surface, cavities under the road surface, collapses in tunnels, accidents, power outages, water leaks and water outages, earthquake prediction information, earthquake damage, tsunami damage, typhoon damage, eruption damage, tornado damage, and optical fiber blockages and disconnections. FIG. 5 is a diagram showing an example of an optical fiber investigation situation, in which a collapsed area 210 in the tunnel is displayed in black on the map.

[0017] The satellite survey status providing server 300 utilizes satellite remote sensing technology, which uses observations by satellites equipped with SAR (Synthetic Aperture Radar), optical sensors, microwave sensors, etc., and acquires road and vehicle conditions, etc., from the differences before and after a disaster using data (scattering intensity values, phase information, polarization information, etc.) and images (optical images, SAR images, etc.) observed by the satellite. In particular, by using time-series interference analysis such as InSAR (Interferometric SAR), it is possible to detect minute displacements such as subsidence, uplift, and tilt of the ground with high accuracy, and to estimate signs of underground cavities and the risk of their formation from deformations that appear on the ground surface. In addition, it may be a device that uses AI (artificial intelligence) analysis to compare and detect road conditions, vehicle conditions, etc. before and after a disaster, or a device that uses AI (artificial intelligence) analysis to detect and detect road conditions, vehicle conditions, etc. during a disaster. The satellite survey status providing server 300 provides the satellite survey status via the network NW to the road disaster response support system 600. The provided satellite survey status is information for each road, and includes some or all of the disaster status and obstacles to vehicle traffic, such as collapsed buildings, landslides, landslides, mudslides, slope collapses, tunnel collapses, road damage, collapsed roadways, collapsed shoulders, cavities under the road surface, fallen trees, falling rocks, total bridge damage, river breaches, river flooding, inundation, avalanches, vehicle traffic history, ground subsidence, power outages, water leaks and water outages, earthquake prediction information, earthquake damage, tsunami damage, typhoon damage, volcanic eruption damage, tornado damage, forest damage, crop damage, and the presence or absence of accidental or stranded vehicles. FIG. 6 is a diagram showing an example of a satellite survey situation, where a landslide location 310 is displayed in black on the map.

[0018] The weather condition providing server 400 provides weather conditions via the network NW to the road disaster response support system 600. The weather conditions provided are information for each region, and include some or all of the following: time, weather (clear, rainy, snowy, etc.), temperature, rainfall, snowfall, snow depth, wind speed, emergency warnings (heavy rain, strong winds, high tides, waves, heavy snow, blizzards), warnings (heavy rain, strong winds, floods, heavy snow, blizzards, etc.), information on record-breaking short-term heavy rain, landslide warning information, earthquake forecast information, earthquake information, tsunami information, eruption information, information on volcanic activity, typhoon information, tornado information, and disaster conditions (including images of roads). FIG. 7 shows an example of weather conditions, where a warning 410 and seismic intensity 420 are displayed in letters and numbers.

[0019] The vehicle driving status providing server 500 uses automobile sensing technology to acquire various data from vehicles such as connected cars, such as vehicle driving status and surrounding road conditions. In particular, it is possible to estimate abnormal behavior when a vehicle approaches a cavity based on information such as sudden braking, ABS activation, abnormal acceleration, tire spin, and bump response, making it an effective source of information for detecting risk areas caused by subsurface cavities. The vehicle travel status providing server 500 provides the vehicle travel status to the road disaster response support system 600 via the network NW. The vehicle driving conditions provided are information for each road obtained from private cars, taxis, buses, trucks, etc. (including electric vehicles), and may include some or all of the following: temperature, areas of sudden braking, skidding, tire spin, tire lock, ABS activated areas, flooded areas, and subsurface cavities from vehicle sensors, etc.; power outages, earthquake damage, tsunami damage, typhoon damage, volcanic eruption damage, tornado damage, river flooding, liquefaction, and obstacles on the road from camera footage and images, etc.; identification of collapsed buildings, landslides, collapsed slopes, collapsed tunnels, total bridge damage, damaged roads, and impassable areas from 3D data, etc.; vehicle traffic history (including standard and large vehicles), traffic volume, traffic congestion, passing speed, average speed, acceleration, and whether or not there is rain or snow from wiper operation status from probe information (including ETC2.0), etc. In addition, data may be acquired using autonomous driving technology (sensing technology for autonomous vehicles), or may be detected and interpreted using AI (artificial intelligence) analysis to detect disaster situations and vehicle obstruction information. FIG. 8 is a diagram showing an example of vehicle travel conditions, and a location 510 where no vehicle has traveled is displayed in black on the map.

[0020] The information acquisition unit 610 operating in the road disaster response support system 600 acquires the patrol status from the patrol status providing server 100, the optical fiber investigation status from the optical fiber investigation status providing server 200, the satellite investigation status from the satellite investigation status providing server 300, the weather conditions from the weather condition providing server 400, and the vehicle driving status from the vehicle driving status providing server 500 via the network NW. The patrol status provided by the patrol status providing server 100 is stored as patrol information 640. The optical fiber inspection status provided by the optical fiber inspection status providing server 200 is stored as optical fiber inspection information 650. The satellite inspection status provided by the satellite inspection status providing server 300 is stored as satellite inspection information 660. The weather conditions provided by the weather condition providing server 400 are stored as meteorological information 670. The vehicle driving status provided by the vehicle driving status providing server 500 is stored as vehicle driving information 680. Furthermore, the information acquisition unit 610 may store information related to road service status. The information related to road service status is information for each road and is mainly managed by road service providers (such as the Japan Automobile Federation) (such as information from a road service management system), and includes some or all of the following: rescue requests (date, time, location, rescue details, etc.), rescue requests due to abnormal weather (date, time, location, rescue details, etc.), rescue requests due to natural disasters (date, time, location, rescue details, etc.), dead battery, locked vehicle, running out of gas, flat tire, wheel off / falling off, flooding / submersion, recovery from snowy roads / mud, accident, slip and fall, disaster / damage status (including photographed images of the road), towing / transportation of vehicle, removal / towing / transportation of abandoned vehicle, removal / towing / transportation of damaged vehicle, removal / towing / transportation of accident vehicle, road conditions (including photographed images of the road), traffic conditions, EV charging availability, vehicle inspection results, etc. These road service statuses are stored as road service information. In addition to the above rescue request reception information, rescue activity information obtained through rescue vehicles, communication-type drive recorders, small unmanned aerial vehicles (drones), portable information devices, etc., and member report information provided by members may also be included. Here, in this invention, "rescue vehicle" refers to a vehicle (which may include an emergency vehicle) owned by the road service provider and used to rescue members or general vehicles, and includes tow trucks, service cars, loaders, and vehicles equipped with vehicle rescue equipment. "Rescue request reception information" refers to information regarding a rescue request sent by a member or third party to the road service provider by phone or app, and includes the requester's contact information, location information, the reason and circumstances for the need for rescue, information about the target vehicle, etc. "Rescue operation acquisition information" refers to information including video data, still image data, audio data, location information, on-site environment information, etc. acquired at the rescue operation site using a rescue vehicle, a communication-type drive recorder, a small unmanned aerial vehicle (drone), a portable information device, etc. The information acquired by the information acquisition unit 610 may be only a part of the patrol status, optical fiber survey status, satellite survey status, weather status, vehicle driving status, and road service status. Based on the results of an integrated analysis of this information, the decision unit 630 can be configured to determine the need for preventive maintenance of road infrastructure and the need for response in the event of a disaster (e.g., securing emergency routes, determining repair priorities, etc.), and to execute the necessary response processing.

[0021] An analysis unit 620 operating in the road disaster response support system 600 analyzes each piece of information obtained from the information acquired by the information acquisition unit 610, such as patrol information 640, optical fiber inspection information 650, satellite inspection information 660, weather information 670, vehicle travel information 680, and road service information, and stores the results as analysis results 690. The stored analysis results 690 may be configured to be viewable in the form of a map display, a list display, a time-series graph display, or the like. Furthermore, the analysis unit 620 may be configured not only to process each piece of information individually, but also to collate and compare these different types of data and make a comprehensive evaluation based on the integrated correlation. For example, it may be possible to comprehensively collate and compare at least two or more types of data from satellite data, optical fiber data, and vehicle driving data, and detect the redundancy and consistency of abnormalities at the same location from multiple perspectives such as ground surface displacement, vibration intensity, and driving abnormalities, and evaluate the risk of subsurface cavities. Furthermore, the analysis unit 620 may include an AI (artificial intelligence) model. The AI ​​(artificial intelligence) model may be configured to use a learning algorithm such as a neural network to integrate multiple pieces of sensing data as input and output a risk score (e.g., a continuous value between 0.0 and 1.0 or a risk classification category) for each location. The output score is used as auxiliary information for making decisions in road infrastructure maintenance and disaster response. It may also include a configuration in which cavity presence / absence information, cavity location information, cavity depth information, cavity shape information, etc. obtained from on-site ground surveys are input into an AI (artificial intelligence) model as learning data or update data, and processing (relearning, model update) is performed to continuously improve prediction accuracy and judgment results. For example, it is possible to use measurement data (date and time, location, seismic intensity, weather data, vibration, vehicle behavior, probe information, three-dimensional topographical data, etc.) contained in the various information (patrol information, optical fiber survey information, satellite survey information, weather information, vehicle driving information, road service information) acquired by the information acquisition unit 610 as learning data to analyze and visualize the road disaster situation and the degree of risk of infrastructure damage. Furthermore, the analysis results 690 may be configured to be utilized in cooperation with the prediction unit or improvement unit as necessary, contributing to future disaster prediction and improvement of judgment accuracy. In addition, the information analyzed by the analysis unit 620 may be configured to analyze only a portion of the patrol information 640, optical fiber inspection information 650, satellite inspection information 660, weather information 670, vehicle driving information 680, and road service information.

[0022] Furthermore, the analysis unit 620 may use a deep learning model that automatically detects abnormal signs from multiple disaster-related information sources. For example, by using LSTM (Long Short-Term Memory), abnormal patterns can be detected from time-series changes in weather information, earthquake waveforms, vehicle behavior, etc. This makes it possible to predict disaster precursors and the risk of secondary damage with high accuracy. Furthermore, a graph neural network (GNN) may be applied to analyze the road network structure and the spatial relationships between affected nodes, thereby extracting priority road clearance route candidates and routes with a high risk of damage spreading.

[0023] The analysis unit 620 may also include a process for assigning a reliability score to each data source. The reliability score is dynamically calculated based on the source (public institution / general user), observation conditions, past accuracy history, etc. The analysis unit 620 excludes or weights data below a predetermined threshold to prevent misjudgments due to inaccurate information. Furthermore, the analysis results 690 may be used to evaluate response priorities, and may be weighted and scored based on factors such as the likelihood of human casualties, the status of functional outages at evacuation centers and hospitals, and the impact of damage to lifelines, and visualized as a heat map on a map.

[0024] Furthermore, the analysis unit 620 may have a reliability score generation function that evaluates the reliability of various information used in the analysis. The reliability score generation function may be configured to score each information source acquired by the information acquisition unit 610 based on the frequency of information acquisition, past accuracy, acquisition method (automatic measurement / manual report), etc., and to supply highly reliable information to the analysis process with priority. For example, for disaster reports and resident notification information collected from social media, etc., a reliability score can be calculated taking into account the consistency of location information, the degree of agreement with the damage situation based on image analysis, past notification history, etc., and information with a score below a threshold can be excluded from analysis or corrected. The reliability score may be referred to in each process in the analysis unit 620, prediction unit, and improvement unit, and may be configured to contribute to improving information weighting and anomaly detection accuracy. Furthermore, the transparency of decision support may be increased by providing an interface (such as a reliability label display) that allows the user to explicitly check the information rating.

[0025] Furthermore, the analysis unit 620 may be provided with a citizen participation learning function that utilizes reports, posts, and feedback information from residents and users to improve the accuracy of AI (artificial intelligence) models and optimize judgment criteria. For example, the report contents (text, photos, videos, etc.) regarding damage, depressions, traffic obstructions, etc. on the road are collected together with location information and time information, and whether or not the information is consistent with the analysis result 690 or the judgment result 695 is evaluated, and accurate reports are used as training data for retraining an AI (artificial intelligence) model. In addition, the system may be configured to encourage residents to improve the quality of their posts through a feedback function (report evaluation and report correction) for false and inappropriate reports, thereby contributing to improving the overall performance of the model. This type of citizen-participation learning structure will link the flow of information from society as a whole with the AI ​​(artificial intelligence) learning platform, and is expected to lead to increased participation in the maintenance and management of public infrastructure and strengthening of local disaster prevention capabilities.

[0026] Furthermore, the analysis unit 620 may have a reliability visualization function that outputs, as accompanying information, data on which the judgment is based and the certainty (reliability) of the judgment for various judgment results by AI (artificial intelligence). In this configuration, the reason for the decision and the score are displayed side by side, for example, "This road section is recommended to be closed: reliability 87% (basis: satellite imagery + vibration anomaly + SNS report)," allowing users or administrators to confirm the transparency of the AI ​​(artificial intelligence) decision. The reliability score can be calculated using the probability output (such as softmax output) within the AI ​​(artificial intelligence) model or the consistency of the evidence (the agreement rate between multiple information sources). Furthermore, if the judgment score falls below a threshold, a note such as "Caution required" or "Further investigation recommended" can be added to help prevent overconfidence in judgment. This type of decision visualization configuration will increase social acceptance of the introduction of AI (artificial intelligence) and also contribute to improving the sense of security and satisfaction in actual operations at disaster response sites.

[0027] Furthermore, the analysis unit 620 may include a reliability evaluation mechanism based on multiple information sources. In the event of a disaster, in addition to official sensing data (satellite images, vibration sensors, vehicle behavior, etc.), near-real-time information such as reports from residents, social media posts, and patrol reports may be collected, but the reliability of this information varies. For this reason, this system may be configured to assign a predefined reliability score to each information source (e.g., Japan Meteorological Agency = high, SNS = medium, unregistered reports = low), and weight the information when inputting it into the AI ​​(artificial intelligence) model. In addition, it is also possible to evaluate the consistency (degree of consistency) and spatiotemporal consistency of content across multiple information sources, and automatically determine the reliability rank of the information (high, medium, low), or to add labels such as "needs confirmation" or "pending" to the judgment result for low-reliability information. This type of information reliability evaluation structure is also effective in preventing misjudgments at disaster sites and improving the social transparency of AI (artificial intelligence) decisions.

[0028] The road disaster response support system 600 may also be configured to cooperate with external mobility services and in-vehicle devices. For example, the system may be configured to mutually cooperate with information from mobile systems such as car navigation systems, smartphones, MaaS (Mobility as a Service) apps, self-driving vehicles, and electric vehicles (EVs) to provide road clearance information, passable routes, danger avoidance instructions, and the like in real time during a disaster. In this configuration, disaster conditions and road clearance route determination results are automatically notified to in-vehicle terminals or smartphones, and adaptive navigation instructions are possible based on the user's current location, direction of travel, driving intentions, etc. Furthermore, for autonomous vehicles, a configuration is possible in which "road avoidance commands" and "stop commands" according to road disaster conditions are directly reflected in the control system. Furthermore, API integration with MaaS service providers will enable processes such as optimizing emergency transportation methods, reconstructing routes, and adjusting traffic demand during disasters, contributing to improving disaster response capabilities across society as a whole.

[0029] The road disaster response support system 600 may also have a condition setting function for the timing and trigger of updating the AI ​​(artificial intelligence) model. The model update process may be configured to be executed automatically or semi-automatically when any of the following conditions is met: (1) When a new disaster occurs and actual damage information on the site is acquired (e.g., cavity detection results, structural damage data, image diagnosis results, etc.) (2) When the system's accuracy falls below a predetermined threshold (e.g., a continuous decline in confidence score, an increase in false positive feedback, etc.) (3) When a government or specialist agency issues a model update order (e.g., earthquake response specification revisions, regulation changes, etc.) By setting and managing such update conditions in advance, it is possible to prevent the judgment model from becoming obsolete, and to achieve continuous system maintenance and optimization of disaster response accuracy.

[0030] Furthermore, the road disaster response support system 600 may be configured to be able to dynamically change the decision rules or evaluation criteria depending on the type, scale, and damage situation of the disaster. For example, one possible configuration is to change the type of data to be targeted, the evaluation items to be prioritized (vibration intensity, flood depth, traffic blockage rate, etc.), judgment thresholds, etc. depending on the type of disaster, such as earthquake, heavy rain, landslide, etc. Furthermore, even within the same disaster, the judgment criteria may be optimized for each region, taking into account the characteristics of the affected area (urban / mountainous area, aging rate, damage to lifelines, etc.). Furthermore, by configuring the system to dynamically switch decision priorities and implementation details from the "lifesaving priority phase" to the "supply support phase" and "life restoration phase" depending on the stage of the disaster, it is possible to increase the flexibility and real-time adaptability of on-site responses.

[0031] Furthermore, the analysis unit 620 may be configured to dynamically adjust disaster response priorities according to regional characteristics. For example, in areas where facilities requiring special care, such as elderly care facilities, welfare facilities, hospitals, evacuation shelters, and elementary and junior high schools, are concentrated, the analysis unit 620 may be configured to prioritize road clearances around these facilities. It is also possible to perform a multidimensional weighting evaluation based on factors such as population density, the proportion of people vulnerable to disasters, the concentration of lifelines, and the distribution of medical resources, and automatically determine the optimal response order for each region. This enables flexible and rational decisions based not only on the physical damage situation but also on the social needs of the region.

[0032] Furthermore, the road disaster response support system 600 may be configured to handle damage to communication infrastructure or network interruptions during a disaster. For example, various information acquisition units and terminals may be configured to ensure communication without infrastructure dependency by using communication means such as a local network, mesh network, or LPWA (low-power wide area network). Furthermore, a failover configuration allows on-site terminals to autonomously continue making decisions and formulating road clearance plans even when communication with the central server is unavailable, thereby enhancing operational continuity (robustness) during a disaster.

[0033] The Road Disaster Response Support System 600 may also be equipped with a control mechanism for switching between automatic judgment and manual intervention. For example, while responses are usually handled automatically by AI (artificial intelligence), it is possible to configure the system so that expert staff or administrators can manually intervene in cases of high urgency or where there is a high degree of uncertainty. This creates a safety net against the risk of misjudgment by AI (artificial intelligence), improving reliability and flexibility in disaster response. Records of manual intervention are also stored in the system, allowing for a configuration that can contribute to future model improvements.

[0034] Furthermore, the road disaster response support system 600 may be configured to perform specialized processing according to the type of disaster. For example, it may be configured to change the evaluation criteria and weightings used in analysis processing, decision processing, and road clearance route selection, taking into account the different damage characteristics, progression speed, and impact range for each disaster type, such as earthquakes, heavy rain, volcanic eruptions, and tsunamis. This enables optimal decisions to be made for each disaster, improving response accuracy. Specifically, it may be possible to implement AI (artificial intelligence) model selection and switching processing that reflects disaster characteristics, such as prioritizing shaking and ground movement during earthquakes and flood depth and drainage capacity during heavy rain.

[0035] The road disaster response support system 600 may also be configured to gradually switch judgment criteria and processing methods according to the chronological phases of the disaster (before the disaster, immediately after the disaster, emergency response period, and full-scale recovery period). For example, immediately after the disaster, rule-based processing that emphasizes speed may be prioritized, and when the situation has calmed down, switching to AI (artificial intelligence) analysis that emphasizes accuracy may be performed, thereby achieving optimal judgment according to the time series. The configuration may also include an RNN (Recurrent Neural Network) model using time series data and event trigger judgment based on the disaster progression flow.

[0036] The road disaster response support system 600 may also have a hybrid AI (artificial intelligence) configuration. For example, by combining disaster response decisions based on an AI (artificial intelligence) model with explicit rule-based (IF-THEN) decisions, the explainability of the output from the AI ​​(artificial intelligence) model and the basis for the decision can be clarified. In particular, during large-scale disasters, it is important for residents and commanders to understand "why a particular route was selected," so the system may be implemented to present the decision results with an explanation. It may also include a configuration that utilizes LLM (large-scale language model) to automatically generate explanations in natural language for the reasons for route selection.

[0037] The road disaster response support system 600 may also be configured to use live footage of the scene captured by patrol vehicles, drones, fixed cameras, surveillance cameras, robots, etc., and perform image recognition and anomaly detection processing using AI (artificial intelligence). For example, AI (artificial intelligence) can automatically detect visual abnormalities in the footage, such as debris piles, flooding, and cracks in the ground, and execute control processing to increase the priority response level of the relevant location. This enables real-time understanding of the disaster situation and support for judgment without relying on visual inspection, improving the accuracy of road clearance and support vehicle guidance.

[0038] Furthermore, the road disaster response support system 600 may be configured to be capable of disaster response training and simulations in peacetime. For example, it may be configured to input past disaster data and hypothetical disaster scenarios, perform hypothetical judgments using the prediction unit and analysis unit 620, and then, based on the results, execute simulated road clearance plans, support route generation, and confirmation of material transportation plans. This enables practical training before a disaster occurs, improving the ability to respond quickly and accurately in the event of an actual disaster.

[0039] Furthermore, the Road Disaster Response Support System 600 may be configured to process disaster information in real time using not only cloud processing but also edge AI (artificial intelligence) implemented on local terminals. For example, patrol cars, drones, robots, etc. deployed at disaster sites may perform on-site processing even in environments where cloud communication is difficult, and perform local judgment, notification, and control. This makes it possible to provide a certain level of judgment support, road clearance judgment, and danger avoidance even when communication is interrupted, achieving highly resilient disaster response.

[0040] The analysis unit 620 may also be equipped with a triage processing function that assumes a situation in which a large amount of disaster-related information is concentrated at once and prioritizes it according to urgency and importance. For example, when a large number of disaster reports and sensor data are input simultaneously, the system can be configured to prioritize disaster areas directly connected to life and daily infrastructure, and to appropriately allocate resources and determine whether to clear roads, thereby avoiding delays in decision-making and excessive or insufficient responses. The AI ​​(artificial intelligence) model can calculate a priority score based on factors such as the extent of damage, surrounding conditions, and the reliability of the communication source, and determine the processing order based on that score.

[0041] Furthermore, the road disaster response support system 600 may have a hybrid configuration that flexibly uses both cloud and local processing. For example, in situations where network bandwidth is limited, such as immediately after a disaster, important decision-making processes may be executed on local terminals or base servers, and then synchronized and integrated with analysis on the cloud once the network is restored. This achieves both continuity of processing during a disaster and overall optimization. An architecture configuration may also be adopted in which the basis for decisions and processing results are recorded and shared, contributing to subsequent analysis and improvement processes.

[0042] The decision unit 630 operating in the road disaster response support system 600 determines the need for road disaster response and road infrastructure maintenance in an integrated and comprehensive manner based on some or all of the analysis results 690 of various information (including patrol information, optical fiber survey information, satellite survey information, weather information, vehicle driving information, and road service information) analyzed by the analysis unit 620, and stores the determination result 695. The determination result 695 may include determination criteria that reflect the risk score threshold, correlation patterns between analytical data, consistency with ground surveys, etc. The determination result 695 may be configured to be visualized on a map display, a list display, a time series graph display, or a dashboard. The decision unit 630 may also have a decision support function using an AI (artificial intelligence) model, and may be configured to determine the priority of maintenance, the urgency of disaster response, etc. based on statistical threshold decision, rule-based decision, or a predictive model using machine learning. In this case, the decision execution method may be configured as fully automatic processing, semi-automatic processing requiring confirmation by an operator, or a proposal-based support mode. Furthermore, the determination unit 630 may be configured to function as a trigger for executing disaster response support processing, including proposing repair plans, determining road clearance routes, recommending emergency vehicle routes, determining traffic restrictions, etc., as necessary. This enables automatic or semi-automatic disaster response support linked with the analysis results 690. Additionally, the decision unit 630 may be configured to cooperate with the road clearance unit or the robotics unit to formulate a road clearance implementation plan or give instructions for road clearance work by robots based on the determination result 695. This links determination and execution, realizing a fast and efficient disaster response. Furthermore, the judgment results 695 and the analysis results 690 may be used in the improvement department to retrain the AI ​​(artificial intelligence) model and update the judgment criteria, and may be combined with past history data and future scenario data to be reflected in the prediction department's prediction of future disaster risks and the development of advance plans. Furthermore, the road disaster response support system 600 may include an information providing unit that provides to an external party a part or all of the judgment result 695 and the analysis result 690. This information providing unit has the effect of enabling road disaster response policies and countermeasure information to be quickly disseminated and shared with administrative agencies, related businesses, general residents, etc. In addition, the decision unit 630 may be configured to automatically generate an explanation in natural language using a large-scale language model (LLM) for the judgment made by AI (artificial intelligence), thereby increasing the transparency and explainability of the basis for the judgment. Furthermore, the determination result 695 may be configured to be output in a foreign language (such as English or Chinese) through a multilingual translation configuration, and may be used to notify or explain to foreign users.

[0043] FIG. 9 is a flowchart showing an example of the flow of the road disaster response support system 600. The information acquisition unit 610 periodically acquires (e.g., every few minutes) information on patrol status from the patrol status providing server 100 (S10). The information acquisition unit 610 periodically acquires (e.g., every few minutes) information on optical fiber inspection status from the optical fiber inspection status providing server 200 (S11). The information acquisition unit 610 periodically acquires (e.g., every few minutes) information on satellite inspection status from the satellite inspection status providing server 300 (S12). The information acquisition unit 610 periodically acquires (e.g., every few minutes) information on weather conditions from the weather condition providing server 400 (S13). The information acquisition unit 610 periodically acquires (e.g., every few minutes) information on vehicle driving status from the vehicle driving status providing server 500 (S14). The analysis unit 620 then extracts road damage locations and the like from the patrol information 640 (S15). The analysis unit 620 extracts travel history and the like from the optical fiber inspection information 650 (S16). The analysis unit 620 extracts landslide locations and the like from the satellite inspection information 660 (S17). The analysis unit 620 extracts earthquake and tsunami information and the like from the weather information 670 (S18). The analysis unit 620 extracts impassable areas and the like from the vehicle travel information 680 (S19). Next, based on the above analysis results, the decision unit 630 comprehensively determines the necessity of road disaster response, and makes a decision (S20) to execute response processing (for example, determining repair plans and road clearance routes, etc.) as necessary. Furthermore, the road disaster response support system 600 may be equipped with information from road users and residents using SNS (Social Networking Service) (disaster information, victim information, relief information, recovery information, etc.), information from government (police, fire department, Self-Defense Forces, etc.), infrastructure operators (communications, electricity, gas, water, sewerage, etc.), transportation operators (railways, buses, ferries, airplanes, etc.), construction and civil engineering operators (including construction industry associations), delivery companies, tourism operators (inns, hotels, tourist facilities, roadside stations, etc.), designated public institutions (Disaster Countermeasures Basic Act) (disaster information, victim information, relief information, recovery information, etc.), and information from the government's Emergency Disaster Countermeasures Headquarters and Emergency Disaster Countermeasures Headquarters. It should be noted that FIG. 9 shows an example of a typical processing flow, and the present invention is not limited to this.

[0044] 10 is a diagram showing an example of the determination criteria in the determination unit 630. The determination unit 630 can determine the necessity of road disaster response when triggered by events such as the occurrence of a tsunami (element 1) 631, a slope collapse causing an optical fiber disconnection (element 2) 632, road damage and collapsed buildings resulting in no vehicle traffic (element 3) 633, a heavy rain warning being issued and landslides causing tires to spin, preventing vehicles from moving forward and causing severe traffic congestion (element 4) 634, or a heavy snow warning being issued and snow having accumulated on the road surface causing the ABS to activate, causing severe traffic congestion with many stranded vehicles (element 5) 635. Furthermore, with regard to some or all of the judgment result 695 decided by the decision unit 630, the analysis result 690, and the various information acquired by the information acquisition unit 610 (patrol information, optical fiber survey information, satellite survey information, weather information, vehicle driving information, road service information), etc., through the information provision unit, etc., it is possible to provide a function to notify the road administrator by email, a function to provide data to a road information board system, a function to provide data to a road restoration visualization map (a map on the web or the like that displays the road restoration status, main affected areas and damage status, road traffic restrictions, required time between cities, vehicle speed data, vehicle traffic history, population mesh data, etc., all in one place), a function to provide data to a car navigation system, a function to provide data to an autonomous driving system, a function to provide data to MaaS (Mobility as a Service), a function to provide information to the government (police, fire department, Self-Defense Forces, etc.) and the media, and a function to disclose information to road users and residents (on a website, smartphone app, etc.), and it is not necessarily necessary to go through the information provision unit. An example of the information disclosure function for road users and residents is as follows, but is not limited to this: A road disaster response support system characterized by transmitting one or more pieces of information acquired by the information acquisition unit 610, the results of analysis by the analysis unit 620, the necessity of road disaster response determined by the decision unit 630, the road clearance implementation plan formulated by the road clearance unit, the restoration status by the robotics unit, or the prediction of the necessity of road disaster response by the prediction unit, to a mobile terminal device (mobile phone, smartphone, tablet terminal, laptop computer, game console, etc.) or a fixed terminal device (desktop computer, smart TV, set-top box, digital signage, kiosk terminal, car navigation, car display audio, etc.). It should be noted that FIG. 10 shows an example of a typical determination criterion, and the present invention is not limited to this.

[0045] The road disaster response support system 600 may include an improvement unit that continuously improves the analysis results 690 (including the analysis method) obtained by the analysis unit 620, the judgment results 695 output by the decision unit 630, and some or all of the judgment criteria. The improvement unit aims to improve the accuracy of the analysis and judgment processes and reduce erroneous judgments by utilizing disaster-related data (images, numerical data, simulation results, etc.) obtained from external sources. The improvement unit can also cooperate with the infrastructure maintenance unit and the road clearance unit, and can optimize the improvement targets for each process (cavity risk assessment, repair judgment, road clearance route determination, etc.). An example of the improvement process is a processing configuration in which hypothesis data, verification data, future prediction data, etc. related to a disaster are input, and after referring to the history of past judgment results 695 and analysis results 690, the improvement unit performs re-evaluation and relearning using methods such as statistical analysis and machine learning, and the results are fed back to the existing model to improve the accuracy of judgment. The improvement unit may have an analysis function using AI (artificial intelligence) and perform model re-learning processing for image recognition and risk score calculation. For example, the improvement unit inputs image data generated from patrol information, optical fiber survey information, satellite survey information, weather information, vehicle driving information, and road service information acquired by the information acquisition unit 610, as well as various associated sensing data (position information, vibration, temperature, speed, topographical information, etc.), predicts and outputs disaster situations, and uses the corrected and supplemented results to tune the AI ​​(artificial intelligence) model and judgment criteria. In addition, the improvement department can use information on the presence or absence of cavities, cavity location information, cavity depth information, or cavity shape information obtained from on-site ground surveys as "correct data (teaching data)" and, through re-training of the AI ​​(artificial intelligence) model, improve the predictive accuracy of cavity risk assessments and continuously improve the results of infrastructure maintenance decisions. The improvement process may be performed by the improvement unit alone, or may be configured to operate in cooperation with an infrastructure maintenance unit or a road clearance unit. The improvement unit may also be configured to work with the prediction unit as needed to improve the model or reset the criteria based on the disaster prediction results.

[0046] Furthermore, the improvement unit may be configured to flexibly select and apply different decision logic depending on the type and occurrence of the disaster in cooperation with the analysis unit 620 and the road clearance unit. For example, the improvement unit dynamically adapts to the type of disaster, such as adopting analysis logic that emphasizes the vibration frequency of structures in the case of earthquake disasters, and priority decision logic based on flood predictions and road flooding height in the case of flood disasters. This enables flexible and accurate decisions to be made in accordance with the actual situation of the disaster.

[0047] The improvement unit may also be configured to sequentially collect and learn the results of various disaster responses (road clearance record, passability, restoration speed, etc.), and use feedback learning to improve the accuracy of future decisions and implementation plans. In this case, linking the information reliability evaluation results (score) prevents bias in learning due to erroneous information. Furthermore, the improvement unit may have a function to explicitly store model update conditions (deterioration in accuracy, change in disaster type, occurrence of new data, etc.) and automatically update the model when specified conditions are met.

[0048] Furthermore, the improvement unit may be configured to sequentially acquire information such as obstacle removal work performed by the robotic equipment by the road clearance unit, passability information on the road clearance route, and on-site restoration progress logs, and use this information to retrain the AI ​​(artificial intelligence) model used by the analysis unit 620. This realizes optimization of robotics control and improvement of environmental adaptability, enabling continuous improvement of road clearance processing accuracy and initial response capabilities in the next disaster. In addition, the improvement unit may be configured to accumulate on-site environmental sensor information (vibration, collision, temperature, terrain change, etc.) acquired during robotics work as training data and use it for model performance evaluation and tuning.

[0049] The road disaster response support system 600 may include a prediction unit that predicts future disaster occurrences and road damage risks in advance by utilizing the history of previously acquired analysis results 690 and judgment results 695. The prediction unit can input and analyze various time-series data related to disasters (actual data, future scenario data, weather forecast data, etc.) and output the need for future road disaster response quantitatively or probabilistically. This will enable road managers to plan and implement advance preparations, response plans, and training plans in preparation for the risk of disasters during peacetime, thereby speeding up and streamlining initial responses when a disaster actually occurs. One example of prediction processing is a configuration in which hypothesis data, verification data, training data, and future prediction data (such as expected patterns of earthquakes or heavy rain disasters that may occur once every few decades) are input into the prediction section, and this is compared with existing analysis history and judgment results to simulate in advance the scope of impact and necessary responses in the event of a disaster. The prediction unit may be equipped with a prediction function using AI (artificial intelligence), and may be configured to use a prediction model that has learned multiple input factors, such as weather conditions, topography, past disaster records, and road structure, to preliminarily identify areas where it is difficult to secure road clearance routes and risk areas that may hinder the passage of emergency vehicles. Furthermore, the prediction results may be utilized in cooperation with the analysis unit 620 and the improvement unit, and may be fed back to the infrastructure maintenance unit and road clearance unit in formulating advance plans as needed.

[0050] Furthermore, the prediction unit may be configured to be able to evaluate not only future disaster risks but also the risk of secondary damage caused by infrastructure aging and insufficient maintenance. For example, the prediction unit may take into account past repair history and infrastructure deterioration indexes to predict areas where damage may expand and the risk of chain collapses that may occur in the future, and reflect this in advance repair and road clearance plans.

[0051] The prediction unit may also be equipped with a scenario simulation function, which performs multi-condition analysis using multiple disaster occurrence conditions (rainfall intensity, location of the epicenter, time of occurrence, etc.) as variables, and may be able to compare and consider the optimal initial response, traffic route, emergency supply transportation route, etc. for each case. In this case, it may be possible to link with a human-in-the-loop decision-making support function to configure a proposal mechanism that assumes the intervention of a manager's judgment.

[0052] The Road Disaster Response Support System 600 functions effectively by using one or more pieces of information from among patrol status, optical fiber survey status, satellite survey status, weather conditions, vehicle driving status, and road service status, and in particular, by combining two or more pieces of information, mutually complementary analysis becomes possible, realizing more accurate situational understanding and decision support. For example, by understanding wide-area surface movements through satellite surveys, detecting local underground vibrations through optical fiber surveys, and acquiring driving records and abnormal behavior through vehicle driving conditions, it becomes possible to achieve both wide-area monitoring and local detection. This will create a multi-layered decision-making platform, from predictive monitoring in peacetime to emergency response decisions in the event of a disaster. In addition, by conducting an integrated analysis that comprehensively compares and collates these multiple pieces of information, the system goes beyond simply listing information and enables the following advanced decision-making processes: (1) Improved accuracy through matching of image data with non-image data: For example, if road damage is detected through image analysis, it can be compared with the corresponding optical fiber displacement and vehicle vibration history to reduce false positives and improve the reliability of disaster assessment. (2) Enhanced situational awareness based on multi-perspective information: Even in situations where detection is difficult using a single sensor, such as at night or in bad weather, it is possible to understand the damage situation in a timely and spatially complementary manner by using other information sources (satellite, driving, weather, etc.). (3) Improved rationality and responsiveness of decisions: Through integrated matching processing based on diverse sensing information and structural analysis using AI (artificial intelligence), risk assessment of road infrastructure, prioritization of disaster response, and determination of road clearance routes can be carried out quickly and rationally. In this way, this system will contribute to minimizing damage and the early recovery of disaster-stricken areas by achieving both a faster initial response when a disaster occurs and more advanced preventive maintenance during peacetime.

[0053] Another embodiment of the present invention may employ a simple configuration for acquiring and analyzing information in stages. In this embodiment, a preliminary determination is first made of the occurrence and severity of a road disaster based on information about weather conditions and information about patrol status. Then, only if a road disaster is determined to be severe, additional information on at least one of vehicle driving status, optical fiber survey status, satellite survey status, and road service status is acquired, and detailed analysis and disaster response decisions are made. This configuration allows for practical disaster response support while reducing system costs and load. In this configuration, the information acquisition unit 610 first acquires information on weather conditions and patrol situations, and the analysis unit 620 evaluates the occurrence and severity of road disasters (for example, earthquakes of seismic intensity 6 or higher, widespread wind and flood damage, snow damage, landslides, etc.) based on this information. If the evaluation result exceeds a predetermined threshold, the information acquisition unit 610 additionally acquires other sensing information, and the analysis unit 620 reanalyzes this information, enabling more advanced disaster response decisions. This enables efficient system operation through step-by-step information utilization. Furthermore, in the processing of this embodiment, the decision unit 630 may determine whether or not a disaster response is necessary and the response policy, as in the normal configuration, and may perform emergency response, notification processing, etc. as necessary. Furthermore, this tiered configuration can also be applied to specific operation modes and simple implementation forms as an auxiliary configuration for the multiple sensing integrated analysis processing, which is the main configuration.

[0054] The road disaster response support system 600 may be provided with a road clearance unit that registers a road clearance plan that has been formulated in advance and determines a road clearance route based on the road clearance plan and acquired information when a disaster occurs. By providing the road disaster response support system 600 with a road clearance unit, it is possible to obtain the effect of quickly clearing roads in the event of a disaster. In typical disasters, the process is emergency restoration followed by full restoration. However, in large-scale disasters, emergency restoration (road clearance) must be carried out before emergency restoration. Road clearance involves quickly removing a minimum amount of rubble and repairing uneven sections to ensure a rescue route, allowing emergency vehicles to pass through for rescue and rescue operations, emergency supply support, and restoration. Road administrators develop road clearance plans in advance, including road clearance bases (bases for support units, disaster prevention centers such as storage areas for supplies and equipment), road clearance routes (wide-area travel routes, access routes, and routes within the affected area), and specific action plans (timelines). A timeline is an action plan that coordinates relevant organizations in the event of a disaster, organizing and sharing in advance a chronological order of who will do what and when. An example of the road clearance unit in the road disaster response support system 600 is a configuration that registers a road clearance plan that was formulated in advance before or after a disaster (it can be registered before or after the disaster), determines a road clearance route in the event of a disaster based on multiple disaster / disaster / traffic related data acquired by the information acquisition unit 610 or the results of an integrated analysis by the analysis unit 620, and formulates an optimized road clearance implementation plan based on that decision. The road clearance plan can be registered by registering plan information entered in advance by an administrator, or by having the computer automatically register road clearance plans obtained from an external system. This allows for both the flexibility of manual input and the speed of automatic processing. The road clearance unit may also have an analysis function and a route determination function using AI (artificial intelligence).For example, two or more pieces of information among patrol information, optical fiber survey information, satellite survey information, weather information, vehicle driving information, and road service information acquired by the information acquisition unit 610, or image, sensor, and three-dimensional topographical data based on these pieces of information, may be input into a machine-learned model, and the road disaster situation may be output, supplemented, and corrected, and then a process may be performed to dynamically determine a road clearance route. Furthermore, the determination of the road clearance route may be configured to select the optimal route in accordance with the disaster situation from multiple candidate routes that avoid high-risk points that should be avoided. The road clearance route and road clearance implementation plan determined by the road clearance unit may be notified to the road administrator via a management terminal or an external server, and may be automatically reflected in coordination with restoration work and traffic regulation instructions. In addition, the determination process may be configured to complement or correct the route determination process by the road clearance unit by referring to on-site information acquired by the robotics unit. Furthermore, the road clearance unit may be configured to cooperate with the analysis unit 620, improvement unit, prediction unit, or infrastructure maintenance unit as necessary, to improve the accuracy of road clearance routes in the event of a disaster and to contribute to dynamic re-planning processing. In this application, a "timeline including time milestones" refers to an action plan that defines status indicators (such as the rate of wide-area travel route availability, the rate of access route availability, the rate of routes within the disaster area availability, the availability rate of road clearance bases, the arrival rate of equipment and materials, and the remaining time for fuel stockpiling) that should be achieved at each predetermined time point (e.g., 24 / 48 / 72 hours) after a disaster occurs. The analysis unit 620 may estimate the achievement rate or probability of achievement of the status indicators based on patrol information, optical fiber survey information, satellite survey information, weather information, vehicle driving information, and road service information. Based on the estimation results, the road clearance unit may dynamically reoptimize the road clearance implementation plan, including route switching, base replacement, equipment and material relocation, or work order rearrangement. Furthermore, the road clearance unit may register and store multiple versions of road clearance plans, such as administrative versions, council versions, and training versions, and select, combine, or apply differences to plan versions based on the estimated results of the achievement rate or achievement probability. The application history (such as version number, applied differences, and application time) may be stored as an audit log. The operation of the timeline may be linked to switching of the judgment criteria or processing method according to the chronological phase of the disaster occurrence (before the disaster, immediately after the occurrence, emergency response period, full-scale recovery period). These time milestones are set with the aim of maximizing the survival rate within 72 hours, taking into account the 72-hour criticality in lifesaving activities.

[0055] Furthermore, the road clearance unit may have a function to comprehensively evaluate past disaster history, current damage status, and the functional status of evacuation, medical, and logistics infrastructure, and automatically generate a road clearance order based on the priority of rescue activities. For example, it may give top priority to areas with the greatest degree of damage and the highest possibility of saving lives, while also performing scoring that takes into account accessibility to important logistics centers and medical institutions, and determine the priority of each segment to be cleared. The score may be configured to be updated in real time based on dynamic conditions (weather, traffic congestion, aftershock risk, etc.).

[0056] The road clearance unit may also be equipped with a multi-criteria optimization function that automatically generates multiple route candidates and evaluates each candidate based on factors such as cost, time, safety, reachability, medical accessibility, and infrastructure constraints such as weight and height restrictions to determine the optimal road clearance route. During the evaluation, weighting based on reliability scores may be applied. Furthermore, routes may be selected through network flow analysis using a spatial optimization model or preference determination (weighted synthesis, etc.) from a Pareto-optimal solution set. In this case, a human-in-the-loop configuration may be adopted in which route proposals proposed by AI are manually reviewed or updated based on feedback. The determined road clearance implementation plan is notified via a management terminal or external server and used to share information and issue work instructions to relevant organizations. The unit may also be configured to work with the robotics unit, analysis unit 620, and improvement unit to dynamically replan based on on-site information.

[0057] Furthermore, the road disaster response support system 600 may be provided with a security configuration to prepare for communication failures and cyber attacks when a disaster occurs. This configuration employs a "zero trust architecture" that implements multiple layers of user authentication, communication encryption, access control, etc., thereby enhancing the security resilience of the entire system. In addition, encrypted communication and mutual authentication are also implemented between each subsystem, minimizing the risk of unauthorized access and information tampering in the event of a disaster. Furthermore, in preparation for a main server failure or network disconnection, a failover configuration (automatic switching to a redundant system) or a configuration with an alternative processing function on a local terminal may be used. For example, even if instructions from the cloud server cannot be received, the local device can be configured to autonomously present and execute response policies using pre-downloaded road clearance plans and AI (artificial intelligence) models. With this enhanced security and resilience configuration, the Road Disaster Response Support System 600 ensures high availability and safety even in the event of a large-scale disaster, contributing to improved reliability for full-scale adoption by local governments and public institutions.

[0058] Furthermore, the road disaster response support system 600 may be configured to perform clustering processing according to the characteristics of the affected municipality or region, and to perform judgment processing that is individually optimized for each region. For example, by using regional characteristic data such as population density, topography, traffic infrastructure density, and disaster history to cluster multiple similar municipalities and apply different models and priority evaluation criteria to each cluster, it becomes possible to respond to disasters in a flexible manner that is not uniform.

[0059] Furthermore, the road clearance unit may be provided with a cooperation configuration for carrying out road clearance work at disaster sites using robotic equipment (autonomous heavy machinery, remotely operated removal devices, etc.). The road clearance unit cooperates with the analysis unit 620 or the improvement unit to determine the type, size, removal means, etc. of obstacles on the road clearance route, and sends work instruction data according to the determination results to the robotic equipment, thereby automating or semi-automating on-site work. Furthermore, work performance information (processing time, obstacle processing history, on-site image and sensor information, etc.) fed back from the robotics equipment can be sent to the analysis unit 620 or improvement unit and reflected in route decisions, work estimates, and work model selection for the next disaster response. This allows for a link between decisions made by AI (artificial intelligence) and robotics as an execution unit, significantly improving responsiveness and safety in disaster response.

[0060] The road disaster response support system 600 may include a robotics unit that uses AI (artificial intelligence) and robotics technology to use robots (mainly disaster response robots) to perform road clearance work. The robots use AI (artificial intelligence) to determine which roads should be prioritized for clearance, specify the road clearance route, and perform road clearance work using robotics technology. When performing road clearance work, the robots use AI (artificial intelligence) to perform optimal route analysis based on a pre-established road clearance plan registered in the road clearance unit or a road clearance implementation plan established by the road clearance unit, and information acquired by the robotics unit (disaster and damage situation, weather conditions, road conditions, traffic conditions, impassable conditions, rescue situation, recovery situation, obstacle information, topographical information, road clearance progress information, latest on-site information, etc.) (The robots do not necessarily have to be autonomous robots). The Robotics Department will achieve the following benefits: By utilizing autonomous robots, work can be carried out quickly without relying on human labor. Furthermore, by using robots to carry out autonomous work, the dispatch of workers to dangerous areas can be minimized. Furthermore, the coordination of large heavy machinery, small robots, drones, etc. will enable efficient obstacle removal, etc. An example of an embodiment of the robotics section in the road disaster response support system 600 is as follows, but is not limited to this (each robot is connected to a network NW). The collection and learning of this operation data may be organized by disaster category, such as earthquake, tsunami, flood, landslide, snow damage, volcanic eruption (including ash fall disaster), and nuclear disaster. A road disaster response support system (including road management methods) characterized in that the Robotics Department uses artificial intelligence analysis and robotics technology to carry out road clearance work using disaster response robots (large heavy machinery, autonomous heavy machinery, small robots, humanoid robots, four-legged robots, snake-type robots, multi-legged robots, earthworm-type robots, transforming robots, multi-joint robots, crawler robots, autonomous excavation robots, small unmanned flying robots, underwater search robots, rescue robots, etc.) based on a pre-established road clearance plan (a road clearance plan registered with the Road Clearance Department) or a road clearance implementation plan. Examples of robotics technology (including disaster response robots) include, but are not limited to: Remotely or autonomously controlling autonomous heavy machinery (such as bulldozers and excavators) to remove obstacles and repair uneven surfaces; Working in conjunction with radio-controlled debris removal robots to remove small amounts of debris; Utilizing four-legged robots or drones to assist in reconnaissance of disaster areas and light-duty obstacle removal; and analyzing the progress of road clearance in real time using AI (artificial intelligence) to automatically adjust optimal work instructions for robots. Additionally, integrated control of multiple different robots (such as autonomous heavy machinery, small robots, and drones) to optimally allocate tasks. Examples of AI (artificial intelligence) include, but are not limited to: Route optimization AI for calculating road clearance routes and determining priorities (Dijkstra algorithm, reinforcement learning, multi-agent, etc.); Image recognition AI for identifying obstacles using drone and robot sensors and generating 3D maps (convolutional neural networks, PointNet, etc.); Robotics control AI for controlling autonomous heavy machinery, collaborative work by small robots, automatic obstacle avoidance and path planning (imitation learning, reinforcement learning, deep reinforcement learning, multi-agent, simultaneous localization and mapping, etc.); Dynamic route update AI (machine learning, long short-term memory, etc.), and work monitoring by AI (convolutional neural networks, long short-term memory, Transformer, etc.). Examples of input and output data in robotics technology and AI (artificial intelligence) analysis include, but are not limited to, the following: Route optimization AI (artificial intelligence): Input data (road network data, obstacle data, real-time traffic data, priority route information, weather and parcel number data, etc.) → Output data (optimal road clearance routes, emergency routes, work instruction lists, etc.). Image recognition AI (artificial intelligence): Input data (drone footage, LiDAR point cloud data, past disaster data, etc.) → Output data (obstacle maps, obstacle type determination, work priority maps, etc.). Robotics control AI (artificial intelligence): Input data (work area maps, obstacle information, robot status data, terrain data, etc.) → Output data (robot work plans, movement route instructions, obstacle removal operations, etc.). Work monitoring AI (artificial intelligence): Input data (work video data, robot work logs, weather information, etc.) → Output data (progress reports, anomaly detection alerts, work optimization instructions, etc.). By utilizing large-scale language models, road clearance plans, road clearance implementation plans, and road clearance operations can be continuously improved regardless of the type of language. This is particularly effective for understanding road clearance plans formulated in advance, learning disaster countermeasures using vast amounts of data on the Internet, and making real-time decisions in response to disasters. Examples of the use of large-scale language models include, but are not limited to, the following: complementing and optimizing road clearance plans and road clearance implementation plans, learning from global disaster countermeasure data on the Internet, real-time support for robots, and real-time use of disaster data. The robotics unit may operate in cooperation with the analysis unit 620, improvement unit, prediction unit, and road clearance unit, and may include a configuration that dynamically updates the target areas and priorities for road clearance work based on disaster risk information and judgment results provided by each unit. Furthermore, the robotics unit may cooperate with the infrastructure maintenance unit as needed to coordinate with actual restoration work carried out immediately after disaster response and to provide support for regular infrastructure repairs. The robot types listed above are merely examples and are not limited to these. Other types of robots may be used as appropriate depending on the purpose of disaster response and the environment in which they are used. In addition, in the event of a nuclear disaster, dose measurement data, radiation control area information, downwind prediction information, etc. may be included in the input, and the road clearance work may be planned and executed based on remote and non-contact operation.

[0061] The road disaster response support system 600 may include an infrastructure maintenance department. In this specification, "information related to infrastructure integrity" means information related to maintaining the functional or physical integrity of social infrastructure, such as structural abnormalities, cavity risk, subsidence tendency, cracks, abnormal vibrations, signs of deterioration, and other information. The Infrastructure Maintenance Department is responsible for evaluating the health of road infrastructure and making maintenance decisions during peacetime, and in the event of a disaster, it is equipped with a system that links the results of these decisions with disaster response processing, thereby contributing to the compatibility of maintenance and initial response. The infrastructure maintenance unit may include a configuration for calculating a subsurface cavity risk score using an AI (artificial intelligence) model based on at least two or more pieces of information among satellite data, optical fiber data, and vehicle driving data acquired by the information acquisition unit 610. This makes it possible to quantitatively extract locations where there is a concern about the formation of subsurface cavities, and to support the prioritization of infrastructure inspections and repairs. Additionally, for locations that are judged to have a high risk of cavities, on-site ground surveys (for example, underground radar surveys, vibration measurements, camera photography, etc.) are conducted, and the results (presence, location, depth, shape, etc. of cavities) are re-input as learning data or update data for the AI ​​(artificial intelligence) model, thereby enabling continuous improvement in the prediction accuracy of the AI ​​(artificial intelligence) model or infrastructure maintenance decisions. Furthermore, the system may be equipped with XAI (Explainable AI) technology that visualizes the output risk score and the basis for anomaly assessment, and may be configured to prioritize correct data acquisition through an active learning strategy or to use augmentation processing of learning data using GAN (Generative Inverse Network), etc. This will enable improvements in model accuracy and learning efficiency. This configuration will enable road managers to make accurate and rational repair decisions based on infrastructure assessment results based on AI (artificial intelligence) and on-site survey information, contributing to optimizing maintenance costs and preventing accidents. The infrastructure maintenance unit may operate in cooperation with the analysis unit 620 or the improvement unit, and may be configured to update and optimize cavity risk assessment based on analysis results and judgment results. Furthermore, the infrastructure maintenance unit may be configured to cooperate with the prediction unit, road clearance unit, and robotics unit as necessary to support actual restoration work after disaster response and ongoing infrastructure maintenance.

[0062] Furthermore, the road disaster response support system 600 may have a function for controlling cooperation with multiple robotics devices (unmanned vehicles, unmanned heavy machinery, drones, etc.) deployed at the disaster site. In this configuration, it is possible to switch between remote control mode and autonomous operation mode for each robotic device. For example, at the initial stage of a disaster, the device can be deployed within a safe range by remote control, and once stable operation has been confirmed, it can be switched to autonomous operation mode depending on the situation on site. In addition, in cooperation with the analysis unit 620 or the road clearance unit, missions based on the disaster situation (e.g., removing obstacles, taking images, securing access routes) may be automatically assigned to robotic devices, allowing multiple machines to work in parallel and cooperatively. Furthermore, by integrating and analyzing sensing information (images, three-dimensional terrain, vibrations, obstacle detection, etc.) obtained from each device in real time and providing feedback to the behavior of other units, a cooperative control mechanism can be installed, enabling efficient and safe disaster response operations. This type of robotics collaborative control configuration contributes to reducing human risks, improving work efficiency, and expanding the area that can be covered at disaster sites.

[0063] Furthermore, the road disaster response support system 600 may also have an emergency supply transportation support function. For example, it may be configured to link roads to be cleared with the logistics network (medical supplies, food, water, fuel, etc.) and prioritize roads necessary for emergency vehicle traffic for restoration. It may also be possible to link with information on relief supply collection and distribution centers and perform road selection processing that maximizes logistics efficiency. It may also be implemented with processing that optimizes the logistics network during a disaster using AI (artificial intelligence) judgment that takes into account transport schedules, traffic history, road damage levels, etc.

[0064] The road disaster response support system 600 may also have a user interface configuration that visually presents output information such as the analysis results 690 and the judgment results 695 in a variety of output formats. Specifically, information such as response priorities, road reopening routes, and the extent of the disaster impact may be output in the form of a map display using a geographic information system (GIS), augmented reality (AR) navigation, or a list format. This allows for optimal information presentation according to the situation to a variety of users, such as field workers, local government officials, and command headquarters, improving the effectiveness of decision support.

[0065] The road disaster response support system 600 according to the present invention may include, in addition to the integrated analysis configuration using the AI ​​(artificial intelligence) model described above, analysis processing using non-AI methods such as statistical analysis, threshold comparison, rule-based inference, etc., without being limited to AI (artificial intelligence) models. This enables flexible configuration selection according to the operating environment and optimization from the perspectives of real-time performance and processing load. The system may also be configured to switch between integrated analysis using these non-AI methods and analytical processing using an AI (artificial intelligence) model depending on the application, system configuration, and operational conditions. Examples include, but are not limited to, the following: An information acquisition unit 610 that acquires at least two of information on satellite survey status, information on optical fiber survey status, and information on vehicle driving status, an analysis unit 620 that performs an integrated analysis of the information by statistical analysis, threshold comparison, or rule-based inference, and a decision unit 630 that determines whether road infrastructure maintenance or disaster response is required based on the analysis results.

[0066] Furthermore, the road disaster response support system 600 according to the present invention may include a configuration for switching the operation policy between a normal operation mode and a disaster operation mode. In normal times, the main purpose is to identify predictive abnormalities in road infrastructure, assess cavity risks, and make decisions about regular maintenance. However, in the event of a disaster, the system will switch to an operational mode where the main purpose is to secure emergency routes, determine priority for road clearance, and support rescue efforts based on sensing information and integrated analysis results. Such switching is controlled by software based on the operational policy of the entire system, and it is not necessary to provide a dedicated "operation switching unit." For example, in normal mode, the analysis unit 620 can apply processing parameters that focus on cavity risk assessment and abnormality detection, and in disaster mode, the decision unit 630 can perform processing that focuses on extracting roadblocks with high disaster priority and selecting corresponding routes.

[0067] Furthermore, the normal operation mode and the disaster operation mode may include a configuration for switching the risk determination criteria (threshold setting) in the integrated analysis and the prioritization logic for the on-site investigation target points. For example, in normal times, to prioritize wide-area and comprehensive sign detection, the threshold for anomaly detection is set relatively lenient, making it easier to detect potential risks, while in the event of a disaster, the threshold for determining anomaly scores is set stricter, enabling operations to prioritize the extraction and notification of high-risk locations that require rapid response. The road disaster response support system may be configured to dynamically change a risk score calculation, a judgment threshold for anomaly detection, or a prioritization logic for on-site investigation target points depending on whether the system is in a normal operation mode or a disaster operation mode, as shown in the following examples, but is not limited to these.

[0068] The road disaster response support system 600 of the present invention may be configured to continuously acquire and analyze sensing information on disasters, damage, and recovery status, etc., and dynamically reevaluate and reconfigure the analysis results and response plans based on the latest information as needed. This enables flexible responses that respond quickly to changes in the disaster situation.

[0069] To accommodate a variety of users, including foreign tourists, the road disaster response support system 600 may execute notification control processing in the analysis unit 620, the determination unit 630, or the information provision unit according to the user's attribute information and language used. For example, the system may be configured to utilize language setting information and GPS information of the terminal device to automatically notify the user of disaster information and travel route information in the user's language if a dangerous area is present within the foreign tourist's range of movement. Disaster response information may also be provided through a traveler application in cooperation with local governments and tourist facilities. Furthermore, the road disaster response support system 600 may be configured to support foreign languages, and may be configured to output and notify analysis results or road clearance implementation plans translated into multiple languages ​​such as English, Chinese, and Korean via information terminals for foreigners, smartphones, web portals, etc. This makes it possible to deliver accurate and immediate disaster response information to users whose native language is a foreign language. In this case, translation processing and multilingual support can employ an automatic translation configuration that utilizes a large-scale language model (LLM). For example, a pre-trained multilingual translation model can be used to convert specialized disaster terminology and road management terminology into accurate and natural expressions. Furthermore, the system can be configured to dynamically switch the optimal translation model or output format based on the user's device language settings, location information, past usage history, etc., enabling real-time, individually optimized multilingual support. This minimizes delays and misunderstandings in information transmission due to language barriers, enabling safe and effective disaster response support for all users, including foreigners.

[0070] In the following paragraphs, we will describe in order the implementation of the road clearance implementation plan in this embodiment, including the design of evaluation indicators and the optimization framework, collaboration between related agencies and authority judgment, allocation of materials and equipment and selection of base locations, multi-hazard evaluation, collaboration between communications, transportation, and robotics, and learning and intellectual property analysis. The road clearance unit according to this embodiment may retain, as internal parameters, three time milestones on a timeline targeting the opening of wide-area travel routes (t≦24), access routes (t≦48), and routes within the disaster area (t≦72), depending on the time t (unit: hours) that has elapsed since the disaster occurred. The analysis unit 620 may update indicators of the damage overview, road obstructions, and restoration progress in chronological order, and the road clearance unit may evaluate the deviation between the indicators and the time milestones and update the prior probability of whether or not the road can be opened using a statistical method (e.g., sequential Bayesian estimation), thereby correcting the road clearance implementation plan in real time. Furthermore, the analysis unit 620 may associate these evaluation results with time milestones on the timeline (opening of wide-area travel routes, access routes, and routes within disaster areas) and use them to estimate the degree of achievement or probability of achievement of the milestones.

[0071] The time goal may be determined using a weighting function W(t) based on the survival rate curve R(t). R(t) assumes a decrease in survival rate within 72 hours based on general knowledge in disaster medicine, and routes for road clearance may be selected to maximize an evaluation value obtained by dividing the amount of medical resources available for each candidate route by the required time and multiplying the result by W(t). This enables planning to prioritize life-saving activities.

[0072] The analysis unit 620 may aggregate traffic volume simulations (demand forecasts and oncoming traffic congestion estimates), abandoned vehicle information (reports, roadside devices, and image recognition results), short-term weather forecasts (rainfall, wind, and snowfall for the next 1 to 6 hours), rescue request information (fire department, medical, and local government request queues), damage and restoration status (road registers, construction section reports, and daily work reports), and equipment and personnel utilization status (deployment, fuel, and labor hours) as a state vector s. The aggregated s may be input into a reinforcement learning model (e.g., a distributed actor-critic) to estimate the achievement probability P(achievement|s) at each time point of a time milestone (24 / 48 / 72 hours). The estimation results may be reflected in dynamic updates of the road clearance implementation plan (e.g., route switching, priority changes, and equipment reallocation) by the road clearance department.

[0073] When P(achievement|s) falls below a predetermined threshold θ, the road clearance unit may simultaneously optimize the generation of alternative routes (avoiding blocked sections, altitude and gradient constraints, weighting of bridge soundness) and the reallocation of materials and equipment (fuel supply sequence, work crew rotation). Optimization updates the policy based on a search that takes into account multiple objectives such as required time, safety, and fuel consumption on the road network graph, and distributes it with a time tag attached to each task in the road clearance implementation plan.

[0074] When working across management divisions managed by multiple road administrators or related organizations, the road clearance unit may refer to the results (priority section, dangerous section, alternative route) of the analysis unit 620 and determine whether authority for that section needs to be transferred or substituted. The determination is made by combining the authority table (mapping of clauses in the Road Act, the Basic Act on Disaster Management, and individual agreements) associated with the management boundary and a score based on the availability, arrival time, and suitability of the equipment owned by the executing entity.

[0075] The authority determination result is reflected as a task attribute of the road clearance implementation plan (approval required, proxy required, notification only), and may be automatically notified according to the chain of command. The notification may be attached with the reference ID of the basis article or agreement clause, and linked to the approval workflow.

[0076] The Road Clearance Department may analyze relevant laws and regulations and prior agreement documents using natural language processing (article segmentation, legal terminology normalization, and modality extraction of obligations, permissions, and prohibitions) to determine whether or not authority delegation is necessary. Using past agreement implementation records and case summaries as training data, weakly supervised learning can improve the accuracy of the delegation. Furthermore, the equipment availability of multiple administrators (number of heavy equipment owned, operational rate, and transportation method) is assigned to each agent, and the implementation body is selected through a multi-agent system of consensus-based decision-making, with the results then connected to the approval process. Example: In an incident in which heavy rain simultaneously blocked a portion of a national highway and a municipal road, each agent, including the national government, prefectures, municipalities, and construction companies, submitted their legal delegation status and the availability of equipment, materials, and personnel. The selection algorithm determined a delegation and supplementary system between local governments that satisfied criteria such as shortest route, avoidance of bridge constraints, and ease of refueling, and automatically routed approval. Note that this example is merely an example, and the condition settings and combinations of evaluation indicators are not limited to these.

[0077] The Road Clearance Division evaluates the work capabilities (amount of soil processed per hour, cutting and lifting capacity), equipment characteristics (bucket capacity, turning radius, trekking ability), and operating hours (unit regulations, labor, night-time restrictions) of the Self-Defense Forces, fire department, police, private construction companies, and transport companies, and comprehensively optimizes the division of roles among multiple units (advance reconnaissance, heavy equipment unit, subsequent transportation, repair team) and shifts (shifts, rest, supplies).

[0078] The optimization can be formulated as a time-expanded graph that combines a three-layer network of equipment, personnel, and routes, with the goal of minimizing delay penalties to demand points (medical institutions, bases, and isolated settlements). The resulting allocation results are visualized as an operational timeline and distributed to each unit's terminal.

[0079] The Road Clearance Department calculates the quantity of materials and equipment required for the road clearance implementation plan based on the processing coefficient for the type of obstacle (rubble, fallen trees, flooding, road surface damage, etc.) and the estimated length, width, and depth of the target section, and compares this with the stockpile ledger.If a shortage is expected, candidate temporary storage sites and supply bases are dynamically selected by scoring them based on transportation time, disaster resistance, and accessibility, and this is reflected in the plan.

[0080] Demand for heavy machinery, personnel, fuel, and materials and equipment can be predicted using a Bayesian demand model for each disaster scenario (earthquake, tsunami, flood, landslide, snow damage, nuclear power, volcanic ash). If a stockpile shortage is predicted, temporary storage sites and fuel supply bases can be selected through facility layout optimization using geographic information systems (road hierarchy, bridges, elevation, site attributes) and land use data (vacant land, warehouses, parking areas / roadside stations, and ports). [Example] Tsunami inundation predicted areas were set as exclusion constraints, a logistics warehouse site on high ground was provisionally designated as a temporary storage site, and a group of gas stations near the port and an inland oil depot were duplicated to ensure redundancy. Note that this example is merely an example, and the condition settings and combination of evaluation indicators can be changed as appropriate depending on the type of disaster, geographic conditions, and availability of materials and equipment.

[0081] Heavy machinery and vehicle operation monitoring can be performed by acquiring information in real time, such as remaining fuel (CAN / OBD), operating hours (engine hour meter), failure predictions (vibration, temperature, anomaly detection), and available operating hours (operator labor, safety regulations). Based on the acquired information, the possibility of continued operation can be probabilistically evaluated to determine the need for replenishment and maintenance. The results of this assessment can be directly linked to the selection of temporary storage sites and fuel bases and plan updates.

[0082] During the evaluation, the distribution of remaining operational time for each vehicle is estimated and overlaid with the distribution of task durations to determine the timing of vehicle dispatch and resupply that minimizes the risk of mission failure.

[0083] The analysis unit 620 uses inputs of probability and intensity indicators for each hazard (earthquake, tsunami, flood, landslide, snow damage, nuclear power, and volcanic ash), vulnerability indicators of road links to the hazard, and exposure indicators such as traffic volume and substitutability to calculate obstruction risk indicators for each section, and then weights and aggregates the indicators to output an integrated score for the route to quantify the obstruction risk of the road clearance route using a multi-hazard analysis model. The model may be, but is not limited to, a probability simulation (Monte Carlo), a Bayesian network, a scenario tree, or a surrogate model using machine learning. The quantification may be reevaluated using the latest data at time milestones (e.g., 24 / 48 / 72 hours). The road clearance unit dynamically redistributes road clearance priorities at each time milestone using the integrated score as a weight and updates the road clearance implementation plan.

[0084] Priority redistribution may be achieved by using a method that maximizes the utility of combining the importance of demand points (hospitals, evacuation centers, and bases) and the degree of hazard interference. The results are visualized on a dashboard.

[0085] In addition to road clearance bases, the Road Clearance Department will evaluate whether local bases such as roadside stations, logistics bases, school gymnasiums, and industrial parks can be converted into equipment collection bases or relief operation bases, and will incorporate bases that are deemed usable into the road clearance implementation plan.

[0086] For regional bases, parking capacity, fuel facilities, warehouse functions, communications facilities, disaster preparedness stockpiles, power supply facilities, etc. are scored and dynamically updated according to changes in the disaster situation. From the perspective of maximizing redundancy, the network reliability of multiple bases is included in the objective function, and a configuration that is not dependent on a single point of failure is selected. [Example] A coastal roadside station and an inland industrial park are paired, and dual-base operation is planned in preparation for the isolation of either one. Note that this example is just one example, and the evaluation index, objective function, and base combination can be changed as appropriate depending on the disaster situation and regional characteristics.

[0087] As alternative communications in the event of a communications outage, satellite communications (satellite telephones and satellite IP communications) or mobile base station vehicles can be evaluated as the primary means, with dedicated wireless networks and optical fiber redundant lines evaluated as supplementary means, and a redundant communications network can be formed between the disconnection points. The placement and operational priority of backup links will be defined in advance from the perspective of link reserve rates and terrain obstructions.

[0088] In addition to port facilities, temporary landing sites, temporary marine transport, airports, and river boat transport, Self-Defense Force air cushion craft and transport ships may also be registered as part of the clearance bases. The analysis unit 620 performs an integrated evaluation of the required time, fuel consumption, weather, tides, and safety for each marine, air, and land transport route to optimize the transport efficiency of relief supplies, medical supplies, and equipment. [Example] In a situation where roads were cut off, a split transport method combining air and water transport was adopted from the airport to a river boat transport base near the disaster area, and the installation plan for a temporary pier was simultaneously optimized. Note that this example is just one example, and the combination of transport modes, condition settings, and evaluation indicators can be changed as appropriate depending on the situation.

[0089] The Road Clearance Department may incorporate data (scenarios, deployment, required time, issues, etc.) obtained from road clearance training and conferences as learning data to improve prediction accuracy in the event of a disaster and reflect the results in the road clearance implementation plan. Indicators (arrival time, work efficiency, supply delays) before and after the training are compared, and the degree of improvement is fed back to the model.

[0090] Training performance data, reports, scenarios, participant feedback, council minutes, video recordings, etc. are normalized using natural language processing, issues are extracted (causes of delays, bottlenecks), improvement proposals are generated, and the data is used as learning data.The road clearance implementation plan is dynamically updated based on the improved model, and a continuous improvement cycle is applied to the next training and actual disaster.

[0091] Using training data, actual disaster data, simulations, and post-mortem verification materials as inputs, the differences between human decision-making and AI proposals can be systematized through extraction, clustering, and reasoning (explainable AI). [Example] The difference between human judgment, which emphasizes avoiding unknown bridge risks, and AI, which emphasizes the shortest route, was visualized using the uncertainty of bridge health as an explanatory variable, and the uncertainty penalty was increased in the next model. Note that this example is just one example, and the feature selection, visualization method, learning settings, etc. can be changed as appropriate depending on the data characteristics and operational requirements.

[0092] This system may use large-scale language models to collect, summarize, and normalize disaster prevention-related information from public web data, and continuously learn knowledge (new technologies, equipment, and operational procedures) that contribute to the advancement of road clearance plans, road clearance implementation plans, and work procedures. Multilingual data may be integrated using language-independent embedding.

[0093] The Intellectual Property Research Department analyzes the claim text in the patent database for the processing procedures or functions included in the road clearance implementation plan using a large-scale language model, performs term normalization and normalization to absorb variations in expression and differences in the order of constituent elements, and evaluates the correspondence between the processing procedures or functions and the constituent elements described in the claims of third parties. Based on the correspondence, it may estimate (score) the possibility of relevance, and may also include automatic claim chart generation and requirement mapping.

[0094] The intellectual property research department may evaluate the correspondence by semantic matching (using a combination of syntax trees and semantic role expressions) that absorbs variations in claim expressions and differences in order, and estimate the possibility of infringement. [Example] The expression differences between "creating an alternative communication means" and "establishing a redundant communication route" were bundled as equivalent concepts, and an infringement score was calculated. Note that this example is just one example, and the method of defining equivalent concepts, matching method, score calculation logic, etc. can be changed as appropriate depending on the situation.

[0095] The Intellectual Property Research Department will collect publicly available information on third-party systems, programs, and methods, normalize the claim elements of this system, and evaluate the likelihood that the third-party implementation will satisfy those elements. The results may be exported as materials for the risk review meeting.

[0096] The Road Clearance Department considers the combined occurrence of earthquakes, tsunamis, floods, landslides, snow damage, nuclear power plants, and volcanic ash, assesses the impact of each disaster, and reevaluates the road clearance implementation plan. In multiple disaster scenarios, the road clearance implementation plan is dynamically updated, including the probability of simultaneous blockages and secondary disasters occurring.

[0097] The information acquisition unit 610 acquires drone aerial footage, in-vehicle drive recorder footage, and vehicle driving sensor data in real time, and the analysis unit 620 uses image analysis and machine learning to identify the presence or absence of rubble, flooding, fire, infrastructure damage, restoration progress, and road obstructions, and classifies the obstructions and estimates the restoration time as necessary. The road clearance unit immediately reflects route switching, priority changes, and equipment redeployment based on the above analysis results.

[0098] The Robotics Department may learn and improve the behavioral models of disaster response robots and robotic equipment based on the operational performance of normal training, normal construction site work, and simulations in each disaster category. The results of activities in actual disasters may be reflected through online learning.

[0099] Based on the analysis results and the road clearance implementation plan, the robotics department controls the robots to remove rubble, excavate, transport materials and equipment, and perform temporary construction, and may feed back the results of the execution (work progress, fuel consumption, equipment load) to the analysis department 620 and use them to improve the next plan.

[0100] The robotics department can coordinate and control multiple robots of different manufacturers and models using a common control signal, automatically organizing the continuous work of excavation, transportation, soil dumping, and compaction as a workflow. [Example] The four-stage process of excavation robot → automatic dump truck transport → bulldozer soil dumping → vibrating roller compaction was synchronized in takt time, and control was applied to automatically increase the number of dump trucks when work delays occurred. Note that this example is just one example, and the process configuration, synchronization method, number adjustment rules, control algorithm, etc. can be changed as appropriate depending on the situation.

[0101] The Road Clearance Department quantifies the speed of rescue operations, the urgency of saving lives, the importance of restoring lifelines, the need to ensure logistics, and the extent of isolation (including the isolation of medical, welfare, and evacuation shelters) as multiple evaluation indicators, weights and integrates them to calculate a priority score, and dynamically formulates and updates road clearance implementation plans based on this score.

[0102] To ensure the explainability of the scores, the contribution of each indicator is calculated (for example, a method such as the Shapley value may be used) and visualized as a basis for decision-making.

[0103] Scenarios for secondary disasters such as communication outages, fuel supply disruptions, delays in procurement of equipment and materials, personnel shortages, bridge collapses, other infrastructure damage, re-snowfall, and aftershocks may be created, and the feasibility of road clearance implementation plans may be evaluated using probabilistic models or multi-agent simulations. [Example] The overall success rate was improved by setting the probability of fuel supply disruptions and selecting redundant routes that include backup tankers and inland bases. Note that this example is merely an example, and the target events, probability setting methods, redundancy means, evaluation indicators, etc. can be changed as appropriate depending on operational requirements. Here, multi-agent simulation refers to a calculation method in which multiple entities involved in road clearance are represented as agents, and interactions on the road network are simulated to calculate indicators such as the feasibility of the plan and the required time.

[0104] The analysis unit 620 classifies responses into initial response, emergency restoration, and full restoration, and determines whether a phase transition is necessary (threshold examples: damage stability, medical demand tightness, material supply stability). At this time, the determination may be made taking into consideration the achievement status of time milestones (e.g., 24 / 48 / 72 hours). Based on the determination results, the road clearance unit dynamically switches the allocation of materials and equipment, personnel deployment, and work priority, and updates the road clearance implementation plan to one optimized for each phase.

[0105] The Road Clearance Department will determine whether abandoned vehicles need to be removed based on the Basic Act on Disaster Countermeasures, etc., and will be able to carry out forced removal procedures through API linkages with towing business associations and contracted companies. The removal results will be reflected in the form of removing obstacles from the road network graph, and the passability determination and route optimization will be updated.

[0106] The system determines whether private construction and transportation companies can operate based on agreement information, disaster risk, and traffic conditions, and evaluates the need for automatic assembly immediately after a disaster occurs. Based on the results of the evaluation, it dynamically creates a road clearance implementation plan, including an automatic assembly notice or standby order, and distributes it to contractor terminals.

[0107] By referencing and analyzing the Basic Disaster Prevention Plan, Regional Disaster Prevention Plan, and Disaster Prevention Operation Plan, priority areas and countermeasures are extracted, linked to the road clearance implementation plan, and coordinated with higher-level plans to create an integrated disaster response plan that is consistent with higher-level plans. Extraction can be done using a hybrid of keyword dictionaries and rule-based / large-scale language models.

[0108] Based on the results of the road clearance implementation plan (arrival time, volume of soil processed, provisional repair level), at least one of the materials and equipment, processes, and personnel allocation required for the emergency restoration plan may be automatically selected and output as an emergency restoration plan. [Example] The amount of mix required for the next day, paving crew shifts, and compactor allocation were automatically calculated based on the provisional restoration thickness and traffic demand, and sent to the relevant office. Note that this example is just one example, and the input indicators, selection targets, output format, distribution destination, etc. can be changed as appropriate depending on operational requirements.

[0109] (Example of evaluation index for time objectives) For a set of candidate routes at time t, the road clearance section may calculate an evaluation value W(t)·M / T based on the survival rate weight W(t), the amount of medical resources available M (for each route), and the required time T (for each route), and select the route that maximizes this evaluation value while considering the achievement probability (24 / 48 / 72 hours) as a constraint. W(t) can be set parametrically as a continuous function that monotonically decreases within 72 hours.

[0110] (Example of reward design in reinforcement learning) Reward R may be given as a linear sum of (i) achievement rewards for achieving each milestone of 24 / 48 / 72 hours, (ii) waiting time penalties at demand points such as hospitals and evacuation centers, (iii) negative rewards for the risk of passing through dangerous sections, and (iv) penalties for excessive fuel and personnel consumption. The policy is updated using a distributed actor-critic, and state transition probabilities are re-estimated online.

[0111] (Example of knowledge representation for authority determination) The knowledge base may store normalized clause IDs, modalities of obligation / permission / prohibition, subjects (nation, prefecture, city, town, village, road administrator, agreement partner), actions (proxy, instruction, approval), etc. Natural language processing extracts <subject, modality, action, condition> from the clause and associates it with the administrative boundary.

[0112] (Example of consensus building for selecting an implementing entity) Multi-agent consensus may be determined using methods such as weighted maximization, weighted voting, and maximum weight agreement based on the utility of each entity (arrival time, available equipment, legal compliance, etc.).

[0113] (Unit / shift optimization: Example) It is also possible to combine objective functions such as reducing total processing time, minimizing demand point delay penalties, and risk avoidance, and solve the problem as mixed integer optimization with labor regulations, equipment operation, and traffic regulations as constraints.

[0114] (Example of equipment demand and facility location) The location of temporary storage sites for equipment and fuel depots can be formulated as an optimization problem that simultaneously minimizes the total transportation burden based on the distance between the demand point and candidate depots, and satisfies constraints on the number of depots p to be opened and exclusion constraints such as areas expected to be flooded. As an example, the p-median problem framework can be used. p is dynamically optimized according to the situation, and the exclusion constraints are updated based on a hazard map.

[0115] (Example of evaluation of the possibility of continued operation) The probability distribution of remaining operational time for each heavy equipment or vehicle can be estimated based on sensor information, and compared with the distribution of task required times. If the probability of failure exceeds a threshold, supply and maintenance can be prioritized. Sensor features can include remaining fuel, temperature, vibration, error codes, continuous operator operation time, etc.

[0116] (Multi-hazard integration: example) The probability of occurrence of each disaster and the degree of disruption impact on each route are combined using policy weights to calculate the overall risk, and correlated weather factors can be corrected using copulas, etc. For road obstruction sections, the route score is recalculated using the overall risk as a weight, and the priority is updated.

[0117] In this specification, "spatial optimization model" refers to a general term for a method of mathematical optimization that uses inputs such as geographic information system (GIS) data, land use data, and transportation network data to perform mathematical optimization for the purposes of base placement, route selection, material and equipment allocation, and transportation efficiency improvement. Specifically, it encompasses location-allocation models, cost-distance analysis, network flow analysis, and multi-agent simulation, and these can be used alone or in combination to optimize base selection, route planning, material and equipment allocation, and inter-modal collaboration during disaster response. In multi-agent simulation, multiple agents on the network make decisions based on given constraints (e.g., road closures, supply base capacity, work time slots) and objectives (e.g., achieving time milestones, reaching high-priority bases), and the resulting spatiotemporal behavior can be evaluated. Furthermore, a plan proposal (base locations, routes, etc.) obtained by a spatial optimization model can be used as an initial solution to be verified in a multi-agent simulation, and the results can be fed back into the optimization process again (an iterative process called coupled optimization simulation). This allows for a more robust plan that takes into account the effects of stochastic events and interactions.

[0118] In this invention, "specialized equipment" refers to equipment that can be used to assist in the transportation of supplies or personnel in disaster environments where ordinary port facilities, airport facilities, or road transport hubs are difficult to use. Specifically, this includes air cushion craft (LCAC), transport ships, floating piers, temporary bridge construction equipment, large hovercraft, heavy machinery-carrying barges, landing craft, temporary runway facilities, and large airlift helicopters (such as CH-47s) owned by the Self-Defense Forces and other organizations. These specialized equipment enable temporary landing, loading, unloading, and transportation even when ports and airports are unusable, enabling the rapid delivery of relief supplies, medical supplies, and recovery equipment.

[0119] Specialized equipment can be used not only as a means of transportation but also for emergency recovery work at disaster sites. For example, floating docks can be used as temporary berthing bases, and hovercrafts can be used to transport materials and equipment to flooded areas. This allows for the formulation of a material transportation plan that integrates land, sea, and air transportation in coordination with road clearance implementation plans, and makes it possible to apply a spatial optimization model based on multiple evaluation indicators such as transportation time, fuel consumption, and safety.

[0120] The "intellectual property research department" in this invention refers to a functional block that collects, summarizes, and normalizes relevant information from public web data and patent databases regarding the processing procedures, functions, data processing flows, etc. included in road clearance plans and road clearance implementation plans, analyzes claim text using a large-scale language model to normalize constituent elements, absorbs variations in claim wording and differences in the order of constituent elements, evaluates the correspondence between the processing procedures or functions and the constituent elements, and estimates the likelihood that a third-party implementation will satisfy the constituent elements based on the correspondence. The intellectual property research department may also score the likelihood of infringement based on automatic claim chart generation, term normalization, requirements mapping, and semantic matching (e.g., a combination of syntax trees and semantic role representations) that absorbs variations in claim wording and differences in order. The evaluation results can be used to present alternative functional solutions, recommend procedure modifications, consider licenses, export risk review meeting materials, etc. Note that this configuration is an example, and the data source, matching method, and score calculation logic can be modified as appropriate depending on operational requirements.

[0121] <Hardware configuration> FIG. 11 illustrates an example of the hardware configuration of a terminal device TM, a fixed camera CAM, a patrol status providing server 100, an optical fiber investigation status providing server 200, a satellite investigation status providing server 300, a weather status providing server 400, a vehicle driving status providing server 500, and a road disaster response support system 600. This diagram illustrates an example in which the terminal device TM is a mobile phone such as a smartphone. The terminal device TM includes, for example, a CPU 701, a RAM 702, a ROM 703, a secondary storage device 704 such as a flash memory, a touch panel 705, and a wireless communication module 706, all interconnected via an internal bus or a dedicated communication line. Application programs such as a road patrol app are downloaded via a network NW and stored in the secondary storage device 704. The wireless communication module 706 may support satellite communications (satellite phone and satellite IP) in addition to cellular, Wi-Fi, and LPWA. The terminal device TM may also include an edge inference accelerator such as an NPU, allowing for simple object detection and roadway obstacle recognition on the terminal side. The fixed camera CAM is configured such that, for example, a CPU 901, RAM 902, ROM 903, secondary storage device 904 such as flash memory, lens / image sensor 905, and communication device 906 are interconnected via an internal bus or a dedicated communication line. Application programs such as camera apps are downloaded via a network NW and stored in the secondary storage device 904. The fixed camera CAM may be equipped with an edge inference accelerator such as an NPU / GPU, and may transmit the results of a primary determination of rubble, flooding, fire, etc. from images, attached as metadata. Each server includes, for example, a NIC 801, a CPU 802, a RAM 803, a ROM 804, a secondary storage device 805 such as a flash memory or a hard disk drive (HDD), and a drive device 806, all interconnected via an internal bus or a dedicated communication line. A portable storage medium such as an optical disk is attached to the drive device 806. A program stored in the secondary storage device 805 or the portable storage medium attached to the drive device 806 is loaded into the RAM 803 by a DMA controller (not shown) or the like, and executed by the CPU 802, thereby realizing the functional units of each server. Patrol information 640, optical fiber inspection information 650, satellite inspection information 660, weather information 670, vehicle driving information 680, analysis results 690, and judgment results 695 are stored in the secondary storage device 805. Note that each server may be implemented using cloud computing. In addition to these, the secondary storage device 805 may store input and output data of the spatial optimization model (location candidates, network graphs, optimization results, etc.), data to be analyzed by the intellectual property research department (claim text, public information), and intellectual property evaluation results (claim charts, requirements mapping, infringement scores), as well as audit logs and operation history. Furthermore, the road disaster response support system 600 may be configured to be able to communicate with each information providing server in order to acquire and process information on the road service status, etc. From the viewpoint of redundant communication, a dedicated wireless network and a satellite communication link may be configured as backup lines, and automatic failover may be performed in the event of a disruption. Furthermore, the road disaster response support system 600 is configured with a computing environment (cloud or on-premise) that has the memory, processors and storage areas necessary for the processing of each component, such as an improvement unit, a prediction unit, an information provision unit, a road clearance unit, an infrastructure maintenance unit, a robotics unit, an intellectual property research unit, etc. The intellectual property research unit may have a connector for securely linking (read-only) with an external patent database or public information source. Furthermore, to execute the AI ​​(artificial intelligence) models used in each component, the system may be configured with computing resources including a GPU (Graphics Processing Unit), a TPU (Tensor Processing Unit), or an AI accelerator.A hybrid inference configuration may also be used, where light-weight inference is performed on the edge side (terminal device TM, fixed camera CAM) and batch learning and high-load inference are performed on the center side (each server, cloud). Note that this diagram is an example of the hardware configuration shown in Fig. 11, and other configurations (edge ​​device configuration, IoT node configuration, distributed processing environment, etc.) may be used depending on the embodiment. To ensure continuity during offline operation, the terminal device TM and fixed camera CAM may be configured to have queuing and local cache mechanisms and to delay the transmission of data after the line is restored.

[0122] Although the present invention has been described above with reference to the drawings, it is not limited to these embodiments or the illustrated configurations. For example, the technical scope of the present invention also includes configurations not shown but described herein, such as processing functions involved in generating analysis results 690 and judgment results 695, the formulation of road clearance implementation plans based on registered road clearance plans, the integrated analysis of various sensing information (optical fiber survey information, satellite survey information, vehicle driving information, etc.), the learning and improvement processing of AI (artificial intelligence) models, the implementation of visualization and feedback functions, the application of spatial optimization models, the use of specialized equipment, the evaluation of rights infringement and risk notification by the intellectual property research department, failover in the event of a disruption using redundant communication links (satellite or dedicated wireless), and the provision of multilingual information. Therefore, the present invention is susceptible to various modifications, alterations, and substitutions without departing from the spirit and scope of the present invention. In addition, functional blocks not explicitly shown in the drawings (road clearance section, authority determination, equipment and material allocation, base selection, multi-hazard assessment, communication / transport / robotics collaboration, intellectual property research section, information provision section, etc.) can be implemented by a person skilled in the art using an appropriate hardware / software configuration. [Explanation of symbols]

[0123] 100: Patrol status server 200: Optical fiber inspection status server 300: Satellite survey status server 400: Weather information server 500: Vehicle driving status server 600: Road Disaster Response Support System 610: Information acquisition department 620: Analysis Department 630: Decision Section 640: Patrol Information 650: Optical fiber survey information 660: Satellite Survey Information 670: Weather information 680: Vehicle driving information 690:Analysis results 695: Judgment result

Claims

1. an information acquisition unit that acquires information about road disaster conditions; an analysis unit that analyzes the information acquired by the information acquisition unit to grasp the road disaster situation; a road clearance unit that registers a road clearance plan (including at least road clearance bases, road clearance routes, and time milestones) that has been formulated in advance; The analysis unit calculates the probability of two or more of a combined occurrence of earthquakes, tsunamis, floods, landslides, snow damage, volcanic eruptions, and nuclear disasters using a multi-hazard analysis model, and quantifies the risk of disruption that each disaster poses to the road opening route; The road clearance unit dynamically weights the priority of road clearance work for each time milestone based on the quantified risk score, and formulates or updates a road clearance implementation plan.

2. A road disaster response support system as described in claim 1, the analysis unit uses a machine learning model as the multi-hazard analysis model, A road disaster response support system characterized by estimating the composite occurrence probability or the disruption risk of each disaster.

3. A road disaster response support system according to claim 1 or claim 2, In addition to the disruption risk of each disaster calculated by the multi-hazard analysis model, the analysis unit The evaluation will be made comprehensively using at least one of the following multiple evaluation indicators: the speed of rescue operations, the urgency of saving lives, the importance of restoring lifelines, the need to ensure logistics, and the scale of isolated settlements or people. The road disaster response support system is characterized in that the road clearance unit calculates the priority of the road clearance implementation plan based on the evaluation results.

4. A road disaster response support system according to claim 1 or claim 2, The analysis unit may also detect disruptions to communications, fuel supplies, or delays in the procurement of materials and equipment. At least one of the following secondary disasters was simulated as a scenario: personnel shortage, bridge collapse, other infrastructure damage, re-snowfall, or aftershocks. assessing the feasibility of the road clearance implementation plan under each scenario using a probabilistic model or multi-agent simulation; A road disaster response support system that ensures redundancy by selecting multiple alternative routes or alternative bases based on the evaluation results.

5. The road disaster response support system according to claim 1, The road clearance department registers at least one of a port facility, a temporary landing base, a temporary marine transport base, an airport facility, or a river boat transport base as part of the road clearance base, If necessary, the Self-Defense Forces' air cushion boats, transport ships, or other special equipment may be registered as the road clearance bases. The analysis unit, for at least one of a marine transport route, an air transport route, and a land road clearance route as each transport route, Evaluate at least one of the following: transit time, fuel consumption, weather conditions, tidal conditions, and safety; The road clearance department optimizes the transportation efficiency of relief supplies, medical supplies, or equipment and materials based on the evaluation results and the spatial optimization model; A road disaster response support system characterized by formulating or updating the road clearance implementation plan based on the optimization results.

6. A road disaster response support system according to claim 1, The road disaster response support system further includes a prediction unit, the prediction unit predicts locations where it is difficult to secure a road clearance route or risk areas that may hinder emergency vehicle passage based on at least one of past disaster record data, road structure data, topographical information, and weather forecast data; The road disaster response support system is characterized by having a function of simulating advance preparation plans or training plans.

7. The road disaster response support system according to claim 1, The road disaster response support system is characterized by continuously improving the road clearance plan, the road clearance implementation plan, or the road clearance work by using a large-scale language model to collect data on the Internet and learn about disaster prevention measures regardless of the type of language.

8. A program including a sequence of instructions for causing a computer to execute at least one of the following (A) or (B): (A) a function of the road disaster response support system according to any one of claims 1 to 2 and claims 5 to 7; (B) At least one of the following functions (1) or (2): (1) A function for comprehensively evaluating, as multiple evaluation indicators, at least one of the following in addition to the risk of disruption from each disaster calculated by the analysis unit as described in claim 3: the speed of rescue operations, the urgency of saving lives, the importance of restoring lifelines, the need to ensure logistics, and the scale of occurrence of isolated settlements or isolated people, and calculating the priority of road clearance implementation plans based on the evaluation results; (2) A function to simulate as scenarios at least one of secondary disasters such as personnel shortages, bridge collapses, other infrastructure damage, renewed snowfall, or aftershocks in addition to communication outages, fuel supply outages, or delays in procuring materials and equipment as described in claim 4, evaluate the feasibility of the road clearance implementation plan under each scenario using a probabilistic model or multi-agent simulation, and ensure redundancy by selecting multiple alternative routes or alternative bases based on the evaluation results.

9. A road management method using a computer, comprising: The computer communicates with the network via Obtain information on road disaster situations, Analyzing the acquired information to grasp the road disaster situation; Register a road clearance plan (including at least road clearance bases, road clearance routes, and time milestones) that has been formulated in advance; The probability of two or more of the following combined events occurring is calculated using a multi-hazard analysis model: earthquake, tsunami, flood, landslide, snow damage, volcanic eruption, and nuclear disaster; and the risk of disruption that each disaster poses to the road closure route is quantified. A road management method characterized by dynamically weighting the priority of road clearance work for each time milestone based on the quantified risk score and executing a process to formulate or update a road clearance implementation plan.

10. A road management method according to claim 9, the computer uses a machine learning model as the multi-hazard analysis model, A road management method characterized by carrying out a process of estimating the composite occurrence probability or the obstruction risk of each disaster.

11. 11. The road management method according to claim 9 or 10, In addition to the disruption risk of each disaster calculated by the multi-hazard analysis model, the computer The evaluation will be made comprehensively using at least one of the following multiple evaluation indicators: the speed of rescue operations, the urgency of saving lives, the importance of restoring lifelines, the need to ensure logistics, and the scale of isolated settlements or people. A road management method characterized by executing a process of calculating the priority of the road clearance implementation plan based on the evaluation results.

12. 11. The road management method according to claim 9 or 10, The computer can detect disruptions to communications, fuel supplies, or delays in procuring materials and equipment, as well as: At least one of the following secondary disasters was simulated as a scenario: personnel shortage, bridge collapse, other infrastructure damage, re-snowfall, or aftershocks. assessing the feasibility of the road clearance implementation plan under each scenario using a probabilistic model or multi-agent simulation; A road management method characterized by selecting multiple alternative routes or alternative bases based on the evaluation results and performing a process to ensure redundancy.

13. A road management method according to claim 9, comprising: The computer registers at least one of a port facility, a temporary landing base, a temporary marine transport base, an airport facility, and a river boat transport base as part of the road clearance base, If necessary, the Self-Defense Forces' air cushion boats, transport ships, or other special equipment may be registered as the road clearance bases. For each transportation route, at least one of the maritime transportation route, the air transportation route, and the land road opening route, Evaluate at least one of the following: transit time, fuel consumption, weather conditions, tidal conditions, and safety; The results of this evaluation and the space optimization model will be used to optimize the efficiency of transporting relief supplies, medical supplies, or equipment. A road management method characterized by executing a process of formulating or updating the road clearance implementation plan based on the optimization results.

14. A road management method according to claim 9, comprising: The computer predicts locations where it is difficult to secure a road clearance route or risk areas that may hinder emergency vehicle passage based on at least one of past disaster record data, road structure data, topographical information, and weather forecast data, A road management method comprising: executing a process for simulating a preparation plan or a training plan.

15. A road management method according to claim 9, comprising: A road management method characterized in that the computer uses a large-scale language model to collect data on the Internet and learn about disaster prevention measures regardless of the type of language, thereby performing a process to continuously improve the road clearance plan, the road clearance implementation plan, or the road clearance work.

Images