system

Abstract

We provide a system that enables local governments to carry out unified and efficient disaster response, and enables rapid and effective disaster response. [Solution] A system including: a means for collecting disaster risk information, past disaster data, geographical information, and infrastructure information; a means for analyzing disaster risk based on the collected data; a means for creating and proposing risk reports based on the analysis results; a means for automatically generating disaster response plans; a means for constructing training plans based on the generated disaster response plans; a means for monitoring the implementation status and effectiveness of training in real time and making adjustments as necessary; a means for quickly initiating a response based on a previously proposed response plan when a disaster occurs; and a means for collecting data after disaster response, evaluating the effectiveness of measures, and proposing improvements for the next disaster response.

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] When a disaster occurs, local governments are often unable to respond quickly and effectively, and local governments, particularly those facing labor shortages and an increasing number of non-regular employees, face the challenge of responding calmly and in an orderly manner. Furthermore, there are areas where past lessons have not been fully utilized, resulting in disparities in the responses of each local government, creating a need for a unified and efficient disaster response system. Therefore, it is necessary to provide a system that analyzes disaster risks in advance and proposes optimal measures, thereby enabling a fast and effective disaster response. [Means for solving the problem]

[0005] This invention includes a means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information, and a means for analyzing disaster risks based on the collected data. It also includes a means for creating and proposing risk reports based on the analysis results, and a means for automatically generating disaster response plans. It also includes a means for creating training plans based on the generated disaster response plans, and a means for monitoring the implementation status and effectiveness of training in real time and making adjustments as necessary. It also includes a means for quickly initiating a response based on a pre-proposed response plan when a disaster occurs, and a means for collecting post-disaster response data, evaluating the effectiveness of measures, and proposing improvements for the next disaster response. This enables local governments to respond to disasters in an integrated and efficient manner, thereby achieving rapid and effective disaster response.

[0006] "Disaster risk information" refers to information that indicates the possibility of future disasters occurring in a specific area, calculated based on past disaster history and disaster prediction models.

[0007] "Past disaster data" refers to data that records the type, scale, extent of impact, and damage situation of disasters that have occurred in the past.

[0008] "Geographic information" refers to information that indicates the geographical characteristics of a particular region, such as topography, geology, climate, and population distribution.

[0009] "Infrastructure information" refers to information about the location, condition, and operation status of core infrastructure such as electricity, water, gas, and communications.

[0010] A "risk report" is a report that organizes the results of disaster risk analysis and summarizes them in a visually easy-to-understand manner.

[0011] A "disaster preparedness plan" is a specific action plan for disaster prevention and response, including the establishment of evacuation shelters, distribution of relief supplies, and medical response in the event of a disaster.

[0012] A "training plan" is a plan for simulations and practical training conducted to improve the effectiveness of disaster response plans.

[0013] "Monitoring" refers to the activity of monitoring the status of training and the progress of responses in the event of a disaster in real time, in order to detect and respond to problems early.

[0014] "Improvement proposals" are proposals based on data from after a disaster response, to reflect the knowledge gained from that experience in response measures for the next disaster.

[0015] "Collection method" means the method for capturing and organizing specified data.

[0016] "Analytical tools" are means for assessing disaster risk based on collected data and predicting the scope of impact and probability of occurrence.

[0017] "Generation means" refers to the means for creating risk reports based on the analysis results and automatically generating disaster prevention plans.

[0018] "Monitoring measures" refer to the means for monitoring the implementation of training and disaster prevention measures in real time, and for immediately proposing countermeasures if any problems arise.

[0019] "Evaluation methods" are means for analyzing data collected after a disaster response and evaluating the effectiveness of countermeasures. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4]FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0021] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0022] First, the terms used in the following description will be explained.

[0023] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0024] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0025] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0026] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0027] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0031] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0032] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0033] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0034] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0035] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0038] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0039] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0040] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0041] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users (local government officials) as follows:

[0042] Data collection

[0043] First, the server collects necessary disaster risk information, past disaster data, geographic information, and infrastructure information from the databases and APIs of local governments and related organizations, including population distribution data, national geographic information, and infrastructure company operation data.

[0044] Examples:

[0045] The server collects past earthquake occurrence history for each region from the database of local government A and imports the latest geographical information and supply infrastructure information from other related organizations.

[0046] Data Preprocessing

[0047] The server cleanses and integrates the collected data. Cleansing involves filling in missing data and processing outliers. Integration involves storing data in different formats in a unified database.

[0048] Risk Analysis

[0049] The server then inputs the organized data into an AI model to analyze disaster risk, which then predicts the disaster risk, impact area, and damage for each region.

[0050] Examples:

[0051] The server analyzes the probability of an earthquake occurring in Area B, predicts the epicenter and seismic intensity, and combines this data with population distribution data to identify the areas that will be most affected.

[0052] Creating a risk report

[0053] The server then generates a risk report based on the analysis results, which includes a visualization of disaster risk, identification of the most affected areas, and an assessment of their impact. The risk report is available in PDF and online dashboard format.

[0054] Examples:

[0055] The server creates an earthquake risk report for Area B, showing areas likely to be affected on a map. The report is generated as a PDF and sent to local government officials.

[0056] Disaster prevention plan proposal

[0057] The server generates a disaster response plan based on the risk report, including the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities.

[0058] Examples:

[0059] The server generates a list of suitable evacuation sites in Area B, calculates and provides a table of supplies needed for each site, and assesses the capacity of nearby major medical facilities to determine whether additional medical assistance is needed.

[0060] Building a training plan

[0061] A terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan, which includes hypothetical scenarios and a schedule of practical training exercises.

[0062] Examples:

[0063] The local government's terminal creates an evacuation drill plan assuming an earthquake occurs in Area B. The plan includes confirmation of evacuation routes and a simulation of opening evacuation shelters.

[0064] Status Monitoring

[0065] The server monitors in real time the status of training and disaster response measures, detecting progress and problems and making adjustments as necessary.

[0066] Examples:

[0067] The server monitors the progress of evacuation drills being conducted in Area B in real time and immediately notifies local government officials if any problems arise. For example, if a particular evacuation route is congested, the server will use that information to suggest an alternative route.

[0068] Emergency response

[0069] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance. The system will support the smooth implementation of planned responses, such as setting up evacuation shelters, distributing relief supplies, and providing medical care.

[0070] Examples:

[0071] If an earthquake actually occurs in Area B, local government officials will follow the instructions from the server to quickly open evacuation shelters and guide residents there, as well as deliver necessary relief supplies to the appropriate locations and deploy medical teams.

[0072] Post-recovery feedback and improvements

[0073] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. Based on the results, it makes suggestions for improvements to be made in the next disaster response.

[0074] Examples:

[0075] After the disaster has subsided, the server will collect data on shelter operations and relief supply distribution in Area B, analyze which aspects were effective and which areas need improvement, and create a report based on the results to propose even more effective response measures for the next disaster.

[0076] In this way, the server, terminal, and user work together to realize a system that provides effective disaster risk analysis and countermeasures.

[0077] The processing flow will be explained below.

[0078] Step 1: Collect data

[0079] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[0080] Specific behavior:

[0081] Obtain population distribution data for residents from the database of Municipality A.

[0082] Download the latest topographic data from the Geospatial Information Authority of Japan.

[0083] It collects information on power supply status and water supply operation sent from infrastructure companies.

[0084] Step 2: Preprocessing the data

[0085] The server cleanses and consolidates the collected data: first, it completes missing data, then it detects outliers and handles them appropriately.

[0086] Specific behavior:

[0087] Complement missing data with historical statistical data.

[0088] Detect outliers (e.g., extremely high population growth rates) and identify their causes.

[0089] Convert data in different formats into a unified format (e.g., CSV format).

[0090] Step 3: Risk analysis

[0091] The server inputs the preprocessed data into an AI model to analyze the disaster risk in each region.

[0092] Specific behavior:

[0093] Geographical information and past earthquake data are input into the earthquake prediction model to calculate the probability of an earthquake occurring in each region.

[0094] River and precipitation data are input into the flood risk model to identify vulnerable areas.

[0095] Step 4: Create a risk report

[0096] The server creates a visually easy-to-understand risk report based on the results of the risk analysis.

[0097] Specific behavior:

[0098] The probability of an earthquake occurring and predicted seismic intensity are plotted on a map to show the area of ​​impact.

[0099] Areas at high risk of flooding are color-coded.

[0100] Generate reports as PDFs or online dashboards and send them to city officials.

[0101] Step 5: Propose a disaster response plan

[0102] The server automatically generates a disaster recovery plan based on the risk report.

[0103] Specific behavior:

[0104] Create a list of shelter locations and needed supplies.

[0105] Establish a schedule for relief operations and develop a resource allocation plan based on that schedule.

[0106] Assess the capacity of medical facilities and determine the number of medical support teams needed.

[0107] Step 6: Develop a training plan

[0108] The terminal will then implement a training plan based on the proposed disaster response plan.

[0109] Specific behavior:

[0110] Create hypothetical scenarios and plan simulation training.

[0111] Schedule the training and notify participants.

[0112] Schedule on-the-job training and prepare necessary supplies and equipment.

[0113] Step 7: Status monitoring

[0114] The server monitors the progress and implementation status of training in real time.

[0115] Specific behavior:

[0116] Collect data during training and display progress on a dashboard.

[0117] When a problem occurs, we will notify you immediately and give you instructions on how to fix it.

[0118] Evaluate each stage in real time and adjust your plan as needed.

[0119] Step 8: Emergency response

[0120] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance.

[0121] Specific behavior:

[0122] Quickly set up evacuation shelters and guide residents there.

[0123] Check the delivery status of relief supplies and arrange for additional supplies if necessary.

[0124] Medical teams will be deployed to provide emergency response.

[0125] Step 9: Post-recovery feedback and improvement

[0126] The server collects data after disaster response and evaluates the effectiveness of countermeasures. Based on the results, it makes suggestions for improvements to be made in future disaster responses.

[0127] Specific behavior:

[0128] Collect and analyze evacuation center operation data and material distribution status.

[0129] We will extract good points and problems in the response and create an evaluation report.

[0130] A report summarizing areas for improvement will be provided to local government officials as feedback for the next event.

[0131] Example 1

[0132] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0133] In recent years, the frequency and scale of disasters have been increasing, making it extremely important to assess disaster risks and prepare appropriate countermeasures in advance. However, in conventional systems, data collection, risk analysis, and countermeasure proposals are dispersed, preventing efficient and integrated countermeasures. In particular, data preprocessing and real-time situation monitoring are lacking, making it difficult to implement rapid and accurate disaster countermeasures.

[0134] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0135] In this invention, the server includes a means for importing disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs, a means for cleansing and integrating the collected data, completing missing data, and processing outliers, and a means for inputting the preprocessed data into a generative AI model to analyze disaster risks. This makes it possible to assess disaster risks in advance and prepare appropriate countermeasures quickly and accurately.

[0136] "Disaster risk information" refers to data and information used to assess the likelihood of a disaster occurring and its impact. Specifically, this includes the probability of occurrence and damage forecasts for natural disasters such as earthquakes, typhoons, tsunamis, floods, and volcanic eruptions.

[0137] "Past disaster data" refers to detailed records of past disasters, including information on the date, location, scale, damage, and response measures.

[0138] "Geographic information" is data that indicates the geographical characteristics of a specific area, including information on topography, elevation, water systems, land use, infrastructure layout, etc.

[0139] "Infrastructure information" refers to data on infrastructure such as electricity, gas, water, communications, and transportation, including the location, operating status, and supply capacity of each piece of infrastructure.

[0140] A "database" is a system for systematically collecting, storing, and managing specific information. It is used to integrate information from multiple different data sources.

[0141] "API" stands for Application Programming Interface, a set of protocols and tools that allow data to be exchanged between different pieces of software.

[0142] "Cleansing" is the process of correcting or removing missing or outlier values ​​to improve data quality.

[0143] "Integration" is the process of bringing together data from different formats into one standardized format.

[0144] A "generative AI model" is an artificial intelligence model that learns from collected data and performs risk analysis and predictions.

[0145] A "risk report" is a report summarizing the results of a risk analysis, including predictions of the probability of a disaster occurring and the extent of its impact.

[0146] A "disaster response plan" is a specific action plan for minimizing damage in the event of a disaster. It includes the establishment of evacuation shelters, preparation of supplies, medical response, etc.

[0147] A "training plan" is a plan that defines the specific schedule and procedures for training to be conducted based on a disaster response plan.

[0148] "Monitoring" is the process of monitoring the response situation during training and disasters in real time and making adjustments as necessary.

[0149] "Rapid response" refers to immediately taking appropriate countermeasures when a disaster occurs.

[0150] "Evaluation" is the process of analyzing the results of disaster response, measuring effectiveness, and identifying areas for improvement.

[0151] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users as follows:

[0152] Data collection

[0153] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from databases and APIs of local governments and related organizations. At this stage, API calls and database connections are made using a high-performance server machine and Python. For example, the server collects resident population distribution data, national geographic information, and infrastructure company operation data.

[0154] Data Preprocessing

[0155] The server cleanses and integrates the collected data. It uses the Pandas library and NumPy to fill in missing data and process outliers. It also stores data in a unified format in the database. This process produces reliable data.

[0156] Risk Analysis

[0157] The server inputs the preprocessed data into a generative AI model to analyze disaster risk. The AI ​​model used is trained using Tensorflow (registered trademark) or PyTorch, enabling highly accurate predictions. For example, the server analyzes the probability of an earthquake occurring in a specific area and identifies areas likely to be affected.

[0158] Creating a risk report

[0159] The server creates a risk report based on the results of the risk analysis. The risk report includes visualization of disaster risks, identification of the most affected areas, and their assessment. This report is generated in PDF format and converted using tools such as Adobe Acrobat. For example, the server displays the affected areas on a map, generates a PDF report, and sends it to local government officials.

[0160] Disaster prevention plan proposal

[0161] The server automatically generates a disaster response plan based on the risk report. This plan includes the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities. For example, the server generates a list of appropriate evacuation shelters and calculates the amount of supplies needed for each shelter. It also assesses the capacity of nearby medical facilities and determines whether additional medical assistance is needed.

[0162] Building a training plan

[0163] The terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan. The training plan includes hypothetical scenarios and a schedule for practical training. As a specific example, the terminal creates an evacuation training plan in an Excel file and notifies each department.

[0164] Status Monitoring

[0165] The server monitors in real time the status of training and the implementation of countermeasures in the event of a disaster. Zabbix and Nagios are used for this monitoring, detecting progress and problems and making adjustments as necessary. For example, the server monitors the progress of training in real time and immediately notifies local government officials if a problem occurs.

[0166] Emergency response

[0167] When a disaster occurs, users (local government officials) can quickly respond based on the prepared response plan. For example, users can use mobile devices and PCs to set up evacuation shelters, distribute relief supplies, and provide medical care.

[0168] Post-recovery feedback and improvements

[0169] The server collects data after disaster response and evaluates the effectiveness of the response. During this process, it reanalyzes information from the database and proposes future improvements. For example, the server analyzes data on evacuation center operations and relief supply distribution to identify areas for improvement.

[0170] In this way, the system provides effective disaster risk analysis and countermeasures by having the server, terminals, and users work together.

[0171] Example prompt for a generative AI model:

[0172] "Please explain in detail the procedures for collecting and preprocessing geographic information, past disaster data, infrastructure information, and population distribution data as input data for the AI ​​model to analyze disaster risk."

[0173] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0174] Step 1: Data collection

[0175] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from databases and APIs of local governments and related organizations. Specifically, it uses the Python requests library to obtain data from APIs and connects to various databases to import information. Input data includes resident population distribution data, national geographic information, and infrastructure operation status data, and is stored on the server. Output data is the collected raw data in various formats.

[0176] Specific operation: When the server retrieves data from the API, it uses the requests.get() method to retrieve the required data and parses it in JSON format.

[0177] Step 2: Data Preprocessing

[0178] The server cleanses and integrates the collected data. Specifically, it uses the Pandas library and NumPy to impute incomplete data, process outliers, and standardize data in different formats. The input data is the collected raw data, and the output data is the cleansed, integrated, and reliable data.

[0179] Specific operation: The server uses the Pandas library to fill in missing values ​​using the fillna() method, etc., and masks outliers.

[0180] Step 3: Risk analysis

[0181] The server inputs the preprocessed data into a generative AI model to analyze disaster risk. The AI ​​models used include TensorFlow and PyTorch, which utilize advanced machine learning algorithms. The input data is preprocessed, reliable data, and the output data is risk assessment results and prediction data.

[0182] Specific operation: The server uses TensorFlow's predict method to input the preprocessed data into the model and obtain prediction results.

[0183] Step 4: Create a risk report

[0184] The server creates the actual risk report based on the results of the risk analysis. It uses Matplotlib and Seaborn libraries to generate graphs and maps and create a PDF version of the risk report. The input data are the risk assessment results and prediction data, and the output data is the generated risk report.

[0185] What it does: The server uses Matplotlib to plot the risk assessment results and then saves the generated graph to a PDF.

[0186] Step 5: Propose a disaster response plan

[0187] The server automatically generates a disaster response plan based on the risk report. This plan includes the location of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities. The input data is the risk report, and the output data is the disaster response plan.

[0188] Specific operation: The server analyzes the risk report data and calculates the evacuation shelter list and the amount of necessary supplies.

[0189] Step 6: Develop a training plan

[0190] The terminal (e.g., the local government's computer system) creates a training plan based on the proposed disaster response plan. The training schedule is created in an Excel file and notified to each department. The input data is the disaster response plan, and the output data is the training plan.

[0191] Specific operation: The terminal uses the openpyxl library to write and save the training schedule in an Excel file.

[0192] Step 7: Status monitoring

[0193] The server monitors in real time the status of training and the implementation of countermeasures in the event of a disaster. Monitoring is performed using Zabbix API and Nagios, and if an abnormality is detected, it immediately notifies. The input data is training and disaster response data collected in real time, and the output data is the monitoring results and alerts.

[0194] Specific operation: The server uses the Zabbix API to obtain real-time data and issues an alert if an abnormality is detected.

[0195] Step 8: Emergency response

[0196] When a disaster occurs, users (local government officials) quickly begin responding based on a pre-proposed response plan. They use mobile devices and PCs to set up evacuation shelters, distribute relief supplies, and provide medical care. The input data is the disaster response plan and real-time situation data, and the output data is the actual response history.

[0197] Specific actions: The user follows the plan, opens designated evacuation centers, and distributes necessary supplies.

[0198] Step 9: Post-recovery feedback and improvement

[0199] The server collects data after a disaster response and evaluates the effectiveness of the response. Based on this evaluation data, it makes suggestions for improving future disaster responses. The input data is various data from after the disaster response, and the output data is suggestions for improving future disaster responses.

[0200] Specific operation: The server evaluates post-disaster response data, analyzes which parts were effective, and identifies areas for improvement.

[0201] (Application example 1)

[0202] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0203] Conventional disaster risk management systems lack sufficient real-time risk monitoring, immediate response when anomalies are detected, and proposals for advance countermeasure plans. This has led to problems such as reduced efficiency in systems for quickly responding to disasters and evaluating the effectiveness of countermeasures. Furthermore, there is a lack of a way to present disaster risk analysis results in a visually understandable manner.

[0204] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0205] In this invention, the server includes: means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information; means for analyzing disaster risk based on the collected data; means for creating and proposing a risk report based on the analysis results; means for automatically generating a disaster response plan; means for constructing a training plan based on the generated disaster response plan; means for monitoring the implementation status and effectiveness of training in real time and making adjustments as necessary; means for promptly initiating a response based on a previously proposed response plan when a disaster occurs; means for collecting post-disaster response data, evaluating the effectiveness of measures, and proposing improvements for the next disaster response; means for monitoring disaster risk information in real time and implementing emergency measures when an abnormality is detected; means for predicting disaster risk information based on an AI model and proposing a pre-disaster response plan; and means for graphing and displaying disaster risk analysis results and generating a report in PDF format. This enables real-time disaster risk monitoring, rapid emergency response in emergencies, and the provision of visually easy-to-understand risk analysis results.

[0206] "Disaster risk information" is data that indicates possible risks related to disasters, and includes information on natural disasters such as earthquakes, tsunamis, and typhoons.

[0207] "Past disaster data" refers to data that records the history of past disasters, the extent of their impact, and the extent of the damage.

[0208] "Geographic information" refers to data relating to the topography and geographical characteristics of a particular area, including map information and topographical maps.

[0209] "Infrastructure information" refers to data on public services and facilities such as electricity, gas, water, and transportation.

[0210] "Means for analyzing disaster risk" are methods and tools for assessing the likelihood of occurrence and the scope of impact based on collected disaster risk information.

[0211] "Means for preparing and proposing risk reports" refers to a method of preparing the results of disaster risk analysis in report format and proposing countermeasures based on that information.

[0212] "Means for automatically generating disaster prevention plans" refers to a method by which the system automatically plans and proposes optimal countermeasures based on the results of disaster risk analysis.

[0213] The "means for constructing a training plan" refers to a method for setting up practical training or scenario-based training based on the generated disaster response plan.

[0214] "Real-time monitoring and adjustment measures" refer to methods for immediately monitoring the progress of training exercises and actual disaster responses and making necessary adjustments.

[0215] "Measures to initiate a rapid response" are methods for quickly putting into action measures that have been planned in advance when a disaster occurs.

[0216] "Means for collecting data and evaluating effectiveness" refers to methods for collecting data after disaster response and evaluating the effectiveness of the measures taken.

[0217] "Means for monitoring disaster risk information in real time" refers to a method for continuously monitoring disaster risk information in real time.

[0218] The "means of implementing emergency measures when an abnormality is detected" refers to a method for quickly implementing an emergency response when an abnormality is detected in disaster risk information.

[0219] "Means for making predictions based on AI models and proposing advance countermeasure plans" refers to a method for predicting disaster risks using artificial intelligence models and proposing effective countermeasure plans in advance based on the results.

[0220] "Means for displaying graphs based on risk analysis results and generating reports in PDF format" refers to a method for visually displaying the results of disaster risk analysis and creating reports in PDF format.

[0221] MODE FOR CARRYING OUT THE INVENTION

[0222] This invention is a factory disaster risk management system that collects disaster risk information, past disaster data, geographical information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users (local government officials).

[0223] Data collection

[0224] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs, including population distribution data, geographic information, and infrastructure company operation data.

[0225] Example: The server collects historical earthquake occurrence history for each region from the local government database and imports up-to-date geographical and supply infrastructure information from other relevant agencies.

[0226] Data Preprocessing

[0227] The server cleanses and integrates the collected data. Cleansing involves filling in missing data and processing outliers. Integration involves storing data in different formats in a unified database.

[0228] Risk Analysis

[0229] The server inputs the organized data into an AI model to analyze disaster risk. This allows for predictions of disaster risk, impact area, and damage for each region. The server performs the analysis using TensorFlow and Scikit-learn.

[0230] Example: The server analyzes the probability of an earthquake occurring in a specific area, predicts the epicenter and intensity, and combines this with population distribution data to identify the areas most affected.

[0231] Creating a risk report

[0232] The server generates a risk report based on the analysis results. This report includes visualization of disaster risks, identification of the most affected areas, and assessment of their impact. The risk report is provided in PDF and online dashboard format. Matplotlib and ReportLab are used for visualization and report generation.

[0233] Example: The server creates an earthquake risk report for a specific area, showing areas likely to be affected on a map. The report is generated as a PDF and sent to local government officials.

[0234] Disaster prevention plan proposal

[0235] The server generates a disaster response plan based on the risk report, including the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities.

[0236] Example: A server generates a list of suitable evacuation sites in a particular area, calculates and provides a list of supplies needed for each site, and assesses the capacity of nearby major medical facilities to determine if additional medical assistance is needed.

[0237] Building a training plan

[0238] A terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan, which includes hypothetical scenarios and a schedule of practical training exercises.

[0239] Example: A local government device creates an evacuation drill plan for a specific area in the event of an earthquake, including confirmation of evacuation routes and a simulation of the establishment of evacuation shelters.

[0240] Status Monitoring

[0241] The server monitors in real time the status of training and disaster response measures, detecting progress and problems and making adjustments as necessary.

[0242] Example: The server monitors the progress of evacuation drills in a specific area in real time and immediately notifies local government officials if a problem occurs. For example, if a specific evacuation route is congested, it will use that information to suggest an alternative route.

[0243] Emergency response

[0244] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance. The system will support the smooth implementation of planned responses, such as setting up evacuation shelters, distributing relief supplies, and providing medical care.

[0245] Example: If an earthquake actually occurs in a specific area, local government officials will follow instructions from the server to quickly set up evacuation shelters and guide residents, as well as deliver necessary relief supplies to the appropriate locations and deploy medical teams.

[0246] Post-recovery feedback and improvements

[0247] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. Based on the results, it makes suggestions for improvements to be made in the next disaster response.

[0248] Example: After a disaster has subsided, the server collects data on the operation of evacuation centers and the distribution of relief supplies in a specific area, analyzes which aspects were effective and which areas need improvement, and creates a report based on the results to propose more effective response measures for the next disaster.

[0249] Prompt Sentence Examples

[0250] "Analyze the latest earthquake risks in the designated area and take safety measures."

[0251] "How can we collect, analyze, and display real-time disaster risk information on a dashboard?"

[0252] In this way, the server, terminal, and user work together to realize a system that provides effective disaster risk analysis and countermeasures.

[0253] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0254] MODE FOR CARRYING OUT THE INVENTION (PROGRAM PROCESSING STEPS)

[0255] Step 1:

[0256] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs. The server imports resident population distribution data, geographic information, and infrastructure company operation data. The input is data from APIs and databases, and the output is a dataset that integrates this information. Specifically, the data is stored in a data frame using Python's pandas library.

[0257] Step 2:

[0258] The server cleanses and integrates the collected data. It uses the collected raw data as input and performs tasks such as filling in missing values ​​and removing outliers. The output is a cleansed dataset stored in a unified format. Specifically, it uses pandas functions to interpolate missing values ​​and remove outliers.

[0259] Step 3:

[0260] The server inputs the organized data into an AI model to analyze disaster risk. The input is a cleansed dataset, and the output is a disaster risk prediction. Specifically, TensorFlow and Scikit-learn are used to input data into the model and make predictions. For example, the probability of an earthquake occurring and the extent of its impact can be predicted.

[0261] Step 4:

[0262] The server creates a risk report based on the analysis results. The input is the prediction results of the AI ​​model, and the output is a risk report (in PDF or online dashboard format). Specifically, Matplotlib is used to create graphs, and ReportLab is used to generate the PDF report. High-risk areas and key countermeasures are visually displayed.

[0263] Step 5:

[0264] The server generates a disaster response plan based on the risk report. The input is the risk report, and the output is a specific disaster response plan. Specifically, it calculates the locations of evacuation shelters and the amount of supplies needed, and compiles them in list form.

[0265] Step 6:

[0266] The terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan. The input is the disaster response plan, and the output is the training plan. Specific operations include setting the date and scenario for the evacuation training.

[0267] Step 7:

[0268] The server monitors the status and effectiveness of training in real time and makes adjustments as necessary. The input is training progress data, and the output is monitoring results and adjustments. Specifically, it collects sensor data and GPS information in real time and displays it on a dashboard.

[0269] Step 8:

[0270] When a disaster occurs, the user (local government official) will quickly begin responding based on a countermeasure plan proposed in advance. The input is the countermeasure plan, and the output is the status of the countermeasures that have been implemented. Specific actions include setting up evacuation shelters and distributing supplies.

[0271] Step 9:

[0272] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The input is disaster response data, and the output is an effectiveness evaluation report. Specifically, the server analyzes the effectiveness of the countermeasures based on the collected data and proposes improvements for the next disaster response.

[0273] Prompt Sentence Examples

[0274] "Analyze the latest earthquake risks in the designated area and take safety measures."

[0275] "How can we collect, analyze, and display real-time disaster risk information on a dashboard?"

[0276] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0277] This invention combines a system that collects disaster risk information, past disaster data, geographical information, and infrastructure information, analyzes disaster risks, and proposes optimal disaster countermeasures with an emotion engine that recognizes the user's emotions. This system operates in cooperation with a server, terminal, and user (local government official) as follows:

[0278] Data collection

[0279] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations, including resident demographic distribution data, up-to-date geographical information, and supply infrastructure information.

[0280] Examples:

[0281] The server collects past disaster occurrence history for each region from the database of X municipality and imports the latest geographical information and lifeline information from other related organizations.

[0282] Data Preprocessing

[0283] The server cleanses and consolidates the collected data, imputes missing data, detects outliers and handles them appropriately, and then converts data from different formats into a unified format.

[0284] Risk Analysis

[0285] The server inputs the preprocessed data into an AI model to analyze the disaster risk for each region, thereby predicting the disaster risk, impact area, and damage for each region.

[0286] Examples:

[0287] The server analyzes the probability of an earthquake occurring in region X, predicts the epicenter and seismic intensity, and combines this with population distribution data to identify the areas that will be most affected.

[0288] Creating a risk report

[0289] The server generates a risk report based on the results of the risk analysis. This report visualizes disaster risks, identifies vulnerable areas, and assesses their impact. The risk report is provided in PDF and online dashboard format.

[0290] Examples:

[0291] The server creates an earthquake risk report for region X, showing areas likely to be affected on a map, and generates a PDF report that is sent to local government officials.

[0292] Introducing the Emotion Engine

[0293] Implement an emotion engine to assess the psychological impact of disaster response plans and help adjust training plans and implementation.

[0294] Examples:

[0295] The server monitors the stress levels and emotional state of users (local government officials) during the training and adjusts the training content as necessary. For example, if many of the training participants are feeling high levels of stress, it will temporarily lower the difficulty of the training.

[0296] Disaster prevention plan proposal

[0297] The server generates disaster response plans based on the risk reports, including the locations of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities, while also taking into account the psychological impact of the plans through an emotion engine.

[0298] Examples:

[0299] The server generates a list of suitable evacuation sites in area X, calculates the amount of supplies needed for each site, and uses an emotion engine to suggest the best psychological placement of the evacuation sites.

[0300] Building a training plan

[0301] The device creates a training plan based on the proposed disaster response plan, creates a virtual scenario and a schedule for practical training, and adjusts the training content based on the user's emotional state during the training using an emotion engine.

[0302] Examples:

[0303] The local government's terminal creates an evacuation drill plan assuming an earthquake occurs in region X. The plan includes confirmation of evacuation routes and a simulation of opening evacuation shelters. An emotion engine is used to detect users' stress and anxiety that arise during the drill in real time and adjust the drill content accordingly.

[0304] Status Monitoring

[0305] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[0306] Examples:

[0307] The server monitors the progress of evacuation drills being conducted in Region X in real time, and immediately notifies local government officials if any problems arise. In addition, if officials feel extremely stressed during the drill, the server temporarily halts the drill and allows them time to refresh.

[0308] Emergency response

[0309] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine takes into account the user's emotional state and helps implement the optimal response.

[0310] Examples:

[0311] If an earthquake actually occurs in region X, local government officials will follow the server's instructions to quickly open evacuation shelters and guide residents. The emotion engine will monitor the officials' emotional state and provide appropriate support if overwork or stress is detected.

[0312] Post-recovery feedback and improvements

[0313] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The emotion engine also takes into account psychological impacts and proposes improvements for the next disaster response.

[0314] Examples:

[0315] After the disaster subsides, the server collects data on shelter operations and relief supply distribution in region X, analyzing which aspects were effective and where there is room for improvement. It also analyzes the psychological data of staff members during the operation and evaluates areas where stress and anxiety became a problem. Based on the results, a report is created proposing optimal psychological measures for the next disaster response.

[0316] In this way, by having the server, terminal, and user work together, it is possible to realize a system that provides effective disaster risk analysis and countermeasures that also take into account the user's emotions.

[0317] The processing flow will be explained below.

[0318] Step 1: Collect data

[0319] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[0320] Specific behavior:

[0321] Obtain the history of past disasters by region from the database of X municipality.

[0322] Download the latest topographic data from the Geospatial Information Authority of Japan.

[0323] It collects information on power supply status and water supply operation sent from infrastructure companies.

[0324] Step 2: Preprocessing the data

[0325] The server cleanses and integrates the collected data, supplements missing data with past statistical data, detects and appropriately processes outliers (e.g., extremely high population growth rates), and converts data in different formats into a unified format (e.g., CSV format).

[0326] Specific behavior:

[0327] Complement missing earthquake occurrence data from the database.

[0328] Check for abnormal values ​​and analyze the cause if necessary.

[0329] Integrate topographical and population data in one unified format.

[0330] Step 3: Risk analysis

[0331] The server inputs the preprocessed data into an AI model to analyze the disaster risk in each region.

[0332] Specific behavior:

[0333] Geographical information and past earthquake data are input into the earthquake prediction model to calculate the probability of an earthquake occurring in each region.

[0334] River and precipitation data are input into the flood risk model to identify vulnerable areas.

[0335] Step 4: Create a risk report

[0336] The server generates a visually easy-to-understand risk report based on the results of the risk analysis, which is available in PDF format and an online dashboard.

[0337] Specific behavior:

[0338] The probability of an earthquake occurring and predicted seismic intensity are plotted on a map to show the area of ​​impact.

[0339] Areas at high risk of flooding are color-coded.

[0340] Generate a report in PDF format and email it to city officials.

[0341] Step 5: Implementing the Emotion Engine

[0342] Using an emotion engine, the psychological impact of disaster response plans is assessed and reflected in training plans and implementation.

[0343] Specific behavior:

[0344] The server monitors the stress level and emotional state of the user (local government official) during the training.

[0345] The server analyzes the emotion engine data in real time and adjusts the training content accordingly, for example, suggesting that the training be made easier if stress levels are high.

[0346] Step 6: Propose a disaster response plan

[0347] The server generates disaster response plans based on the risk reports, including the locations of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities. An emotion engine also takes into account the psychological impact of the plans.

[0348] Specific behavior:

[0349] Create a list of shelter locations and needed supplies.

[0350] Establish a schedule for relief operations and develop a resource allocation plan based on that schedule.

[0351] Assess the capacity of medical facilities and determine the number of medical support teams needed.

[0352] Using an emotion engine, we propose shelter layouts that have minimal psychological impact.

[0353] Step 7: Develop a training plan

[0354] The device then creates a training plan based on the proposed disaster response plan, creating virtual scenarios and practical training schedules, and using data from the emotion engine to monitor the user's emotional state and adjust the training content.

[0355] Specific behavior:

[0356] Create hypothetical scenarios and plan simulation training.

[0357] Schedule the training and notify participants.

[0358] Schedule on-the-job training and prepare necessary supplies and equipment.

[0359] The emotion engine monitors the user's emotional state, detects stress or anxiety during training, and adjusts the training progress.

[0360] Step 8: Status monitoring

[0361] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[0362] Specific behavior:

[0363] Collect data during training and display progress on a dashboard.

[0364] When a problem occurs, we will notify you immediately and give you instructions on how to fix it.

[0365] Evaluate each stage in real time and adjust your plan as needed.

[0366] An emotion engine monitors the user's emotional state and takes appropriate measures if excessive stress is detected.

[0367] Step 9: Emergency response

[0368] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine takes into account the user's emotional state and helps implement the optimal response.

[0369] Specific behavior:

[0370] Quickly set up evacuation shelters and guide residents there.

[0371] Check the delivery status of relief supplies and make additional arrangements as necessary.

[0372] Medical teams will be deployed to provide emergency response.

[0373] An emotion engine monitors the user's emotional state and provides supportive measures if overwork or stress is detected.

[0374] Step 10: Post-recovery feedback and improvement

[0375] The server collects data after disaster response and evaluates the effectiveness of countermeasures. It also takes into account psychological impacts using an emotion engine and proposes improvements for future disaster response.

[0376] Specific behavior:

[0377] Collect and analyze evacuation center operation data and material distribution status.

[0378] We analyze users' emotional data during disaster response and evaluate the areas where stress and anxiety became an issue.

[0379] We will extract good points and problems in the response and create an evaluation report.

[0380] A report summarizing areas for improvement will be provided to local government officials as feedback for the next event.

[0381] Example 2

[0382] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0383] In recent years, the frequency and scale of disasters have increased, creating a need for rapid and accurate disaster prevention measures. However, current systems are limited to collecting and analyzing disaster risk information and generating disaster prevention plans, and do not adequately consider user emotions in their training and responses. Furthermore, cleansing and integrating collected data and real-time monitoring often require manual intervention, resulting in reduced efficiency. In this situation, there is a need for a system that can achieve more effective and efficient disaster prevention measures by taking users' psychological states into account.

[0384] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0385] In this invention, the server includes means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information, means for cleansing and integrating the collected data, means for analyzing disaster risk using preprocessed data, means for creating and proposing risk reports, means for recognizing a user's emotional state using an emotion engine, means for automatically generating a disaster response plan and optimizing it taking into account the user's emotional state, means for building a training plan based on the proposed disaster response plan, means for monitoring the implementation status and effectiveness of the training in real time and making adjustments as necessary, means for promptly initiating a response based on the previously proposed response plan when a disaster occurs and providing support taking into account the user's emotional state, and means for collecting post-disaster response data, evaluating the effectiveness of the response measures, and proposing improvements for the next disaster response. This enables highly accurate analysis of disaster risk, reduces the user's psychological burden, and enables more effective and practical disaster response.

[0386] "Disaster risk information" is information that indicates the probability of a disaster occurring and the extent of its impact, and is data used to assess future disaster risks.

[0387] "Past disaster data" refers to detailed information about disasters that have occurred in the past, including the type of disaster, the location of the disaster, and the extent of the damage.

[0388] "Geographic information" is information that describes the geographical characteristics of a particular area, such as topography, land use, and infrastructure layout.

[0389] "Infrastructure information" refers to information related to the basic infrastructure of daily life, such as transportation networks, water, electricity, and gas, and is data necessary for maintaining urban functions.

[0390] "Data cleansing" is the process of removing errors and inconsistencies from collected data to improve the quality of the data.

[0391] "Data integration" is the process of combining multiple data sets collected from different sources in a consistent format to create uniformly usable data.

[0392] "Disaster risk analysis" is the process of using AI models and algorithms to assess the probability of a disaster occurring and the extent of its impact based on collected data.

[0393] A "risk report" is a report summarizing the results of disaster risk analysis, including visualized data and specific risk assessments.

[0394] The "emotion engine" is a system that analyzes a user's emotional state from voice and text data and adjusts training and countermeasures based on that information.

[0395] A "disaster preparedness plan" is a plan that lays out in advance the specific actions and procedures to be taken in the event of a disaster, including the locations of evacuation shelters and plans for the distribution of relief supplies.

[0396] A "training plan" is a plan that defines the detailed schedule and content of training to be conducted based on the proposed disaster response plan.

[0397] "Real-time monitoring" is the process by which systems constantly monitor the situation during drills and disaster response, instantly acquiring and analyzing data and making adjustments as needed.

[0398] "Starting a rapid response" means implementing necessary measures without delay based on a disaster response plan that was prepared in advance when a disaster occurs.

[0399] "Post-disaster response data" refers to data related to responses at the time of a disaster and afterward, and is used to evaluate the effectiveness of responses and areas for improvement.

[0400] "Evaluating effectiveness and proposing improvements" is the process of analyzing the effectiveness of current measures based on collected data and identifying specific areas for improvement in the next disaster response.

[0401] MODE FOR CARRYING OUT THE INVENTION

[0402] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, and combines AI technology with an emotion engine. This system is designed to monitor the user's emotional state in real time and maximize the effectiveness of disaster countermeasures. Below, we explain how this system's program works.

[0403] 1. Data Collection

[0404] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations, including population distribution data, up-to-date geographical information, and supply infrastructure information.

[0405] Hardware / software used: Data collection using RESTful API, Python libraries (requests, pandas)

[0406] Example: The server sends a request to the local government's database API to obtain the history of past disasters by region. It also obtains the latest geographical and lifeline information from other related organizations.

[0407] 2. Data Preprocessing

[0408] The server cleanses and consolidates the collected data, imputes missing data, detects outliers and handles them appropriately, and then converts data from different formats into a unified format.

[0409] Hardware / software used: Python pandas library, database management system (e.g., MySQL (registered trademark))

[0410] Example: The server reads the acquired data in data frame format, imputes missing values ​​with the mean value, filters outliers, and saves the data in a unified format in the database.

[0411] 3. Risk Analysis

[0412] The server inputs the preprocessed data into a generative AI model to analyze disaster risk for each region.

[0413] Hardware / software used: AI model (TensorFlow or PyTorch), data extraction from database

[0414] Example: The server inputs regional earthquake occurrence data into an AI model to predict the epicenter and intensity, and combines this with population distribution data to identify the areas most affected.

[0415] 4. Creating a risk report

[0416] The server generates a risk report based on the results of the risk analysis, which includes visualization of disaster risks, identification of vulnerable areas, and assessment of impacts.

[0417] Hardware / software used: Data visualization tools (Matplotlib, D3.js), report generation tool (PDFKit)

[0418] Example: The server visualizes the risk level on a map based on the analysis results, generates a risk report in PDF format, and provides it to the user.

[0419] 5. Introducing the Emotion Engine

[0420] The server uses an emotion engine to monitor the user's emotional state during training and execution, with the aim of reducing psychological burden and realizing effective training and disaster response.

[0421] Hardware / software used: Emotion recognition AI model (e.g. OpenCV, NLTK)

[0422] Example: The server collects speech data during training and inputs it into an emotion recognition model to analyze stress and anxiety levels in real time.

[0423] 6. Proposal of disaster prevention plan

[0424] Based on the risk report, the server generates a disaster response plan, which includes the location of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities, and also takes into account psychological impacts using an emotion engine.

[0425] Hardware / software used: Data analysis tools (e.g., SciPy), AI models

[0426] Example: The server proposes psychologically optimal shelter layouts based on the user's emotional state.

[0427] 7. Developing a training plan

[0428] The device then creates a training plan based on the proposed disaster response plan, creates a virtual scenario and a training schedule, and adjusts the training content based on the user's emotional state during the training using an emotion engine.

[0429] Hardware / software used: Scheduling tool (calendar application), real-time monitoring system

[0430] Example: The device creates a training schedule based on hypothetical scenarios and uses an emotion engine to monitor stress levels during training.

[0431] 8. Status Monitoring

[0432] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[0433] Hardware / software used: Real-time data analysis tools, dashboards (Grafana, Kibana)

[0434] Example: The server monitors the progress of training and immediately notifies the user if an abnormality is detected. It also adjusts the training content if it detects excessive stress.

[0435] 9. Emergency Response

[0436] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine will monitor the user's emotional state and provide the necessary support.

[0437] Hardware / software used: Digital instruction system, emotion recognition system

[0438] Example: The user quickly opens a shelter and guides residents. If the emotion engine detects overwork or high stress, it receives support from the server.

[0439] 10. Post-recovery feedback and improvements

[0440] The server evaluates the effectiveness of countermeasures based on post-disaster response data and makes suggestions for improvements for the next disaster response, taking into account the psychological impact.

[0441] Hardware / software used: Data analysis tools, report generation tools

[0442] Example: The server collects data from disaster response and analyzes it together with emotional data. Based on the results, it formulates a response plan for the next disaster and provides feedback to the user.

[0443] Examples of prompt statements

[0444] For example, the following might be an example of a prompt sentence to input to a generative AI model:

[0445] "Based on disaster data from the past 10 years in the region, we analyzed the current disaster risk and created a risk report."

[0446] "Create a plan for your next evacuation drill and monitor the emotional state of participants during the drill."

[0447] "We propose a response plan in the event of a disaster and provide optimal responses taking into account the stress levels of those in charge."

[0448] In this way, the present invention realizes a system that provides effective disaster risk analysis and countermeasures that take into account the user's emotions, through the cooperation of the server, terminal, and user.

[0449] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0450] Step 1:

[0451] Data collection

[0452] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[0453] Specific behavior:

[0454] Input: Database API endpoint, authentication information

[0455] Processing: The server uses a RESTful API to request various information and retrieves data in JSON or XML format.

[0456] Output: Collected raw data (e.g., past disaster history, latest geographic information, supply infrastructure information)

[0457] Hardware / software used: RESTful API, Python requests library

[0458] Step 2:

[0459] Data Preprocessing

[0460] The server cleanses and consolidates the collected data.

[0461] Specific behavior:

[0462] Input: Raw collected data (JSON, XML)

[0463] Processing: The server uses the Python pandas library to read the data in data frame format, perform missing value imputation (e.g., inserting the mean or mode), detect and remove outliers, convert data from different formats into a unified format (e.g., CSV, JSON), and save it in the database.

[0464] Output: Clean consolidated data

[0465] Hardware / software used: Python pandas library, database management system (e.g., MySQL)

[0466] Step 3:

[0467] Risk Analysis

[0468] The server inputs the preprocessed data into a generative AI model to analyze disaster risk for each region.

[0469] Specific behavior:

[0470] Input: Clean consolidated data

[0471] Processing: The server feeds the data into a generative AI model (e.g., an AI model using TensorFlow or PyTorch) to calculate disaster risk for each region, including risk assessments for earthquakes, floods, fires, etc.

[0472] Output: Disaster risk assessment results (e.g., probability of occurrence, impact area, damage forecast)

[0473] Hardware / software used: TensorFlow, PyTorch

[0474] Step 4:

[0475] Creating a risk report

[0476] The server creates a risk report based on the risk analysis results.

[0477] Specific behavior:

[0478] Input: Disaster risk assessment results

[0479] Processing: The server uses a data visualization tool (e.g., Matplotlib, D3.js) to visualize the risk data in the form of graphs and maps, and then uses a report generation tool (e.g., PDFKit) to generate the results as a PDF report.

[0480] Output: Risk report (PDF format)

[0481] Hardware / software used: Matplotlib, D3.js, PDFKit

[0482] Step 5:

[0483] Introducing the Emotion Engine

[0484] The server monitors the user's emotional state during training and execution.

[0485] Specific behavior:

[0486] Input: User voice and text data

[0487] Processing: The server inputs the data into an emotion recognition AI model to analyze the user's stress level and emotional state.

[0488] Output: User's emotional state (e.g., stress level, anxiety level)

[0489] Hardware / software used: OpenCV, NLTK

[0490] Step 6:

[0491] Disaster prevention plan proposal

[0492] The server generates a disaster recovery plan based on the risk report.

[0493] Specific behavior:

[0494] Input: Risk report, user emotional state

[0495] Processing: The server uses AI models to assess shelter locations, relief supply needs, and medical facility capacity, optimizing them based on emotional state.

[0496] Output: Disaster Preparedness Plan

[0497] Hardware / software used: SciPy, AI models

[0498] Step 7:

[0499] Building a training plan

[0500] The terminal will create a training plan based on the proposed disaster response plan.

[0501] Specific behavior:

[0502] Input: Disaster Preparedness Plan

[0503] Processing: The device creates virtual scenarios and practical training schedules, and uses an emotion engine to monitor and adjust the user's emotional state during training.

[0504] Output: Training plan

[0505] Hardware / software used: Scheduling tools, real-time monitoring systems

[0506] Step 8:

[0507] Status Monitoring

[0508] The server monitors in real time the status of countermeasures being implemented during training and when disasters occur.

[0509] Specific behavior:

[0510] Input: Training status, user's emotional state

[0511] Processing: The server uses real-time data analysis tools and dashboards to monitor progress and issues, and if anomalies are detected, it immediately notifies you and makes the necessary adjustments.

[0512] Output: Training progress, anomaly detection results

[0513] Hardware / software used: Grafana, Kibana

[0514] Step 9:

[0515] Emergency response

[0516] The user (local government official) will promptly begin responding based on the countermeasure plan proposed in advance.

[0517] Specific behavior:

[0518] Input: Real-time information and countermeasure plans in the event of a disaster

[0519] Action: The user takes action using the digital instruction system. The emotion engine monitors the user's emotional state and provides necessary assistance.

[0520] Output: Measures taken, emotional state log

[0521] Hardware / software used: Digital instruction system, emotion recognition system

[0522] Step 10:

[0523] Post-recovery feedback and improvements

[0524] The server evaluates the effectiveness of countermeasures based on data collected after the disaster response and makes suggestions for improvements for the next disaster response.

[0525] Specific behavior:

[0526] Input: Post-disaster response data, emotional state log

[0527] Processing: The server uses the data analysis tool to evaluate the effectiveness of the measures, and creates an improvement proposal report using the report generation tool.

[0528] Output: Improvement Suggestion Report

[0529] Hardware / software used: Data analysis tools, report generation tools

[0530] (Application example 2)

[0531] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0532] Conventional disaster response systems focus on analyzing disaster risks and proposing countermeasures, but lack functionality that takes into account the psychological burden on responding staff. They also lack the ability to visually grasp real-time disaster information or intuitively understand appropriate countermeasures tailored to the current situation. This makes it difficult to respond quickly and effectively in the event of a disaster, potentially resulting in the expansion of damage.

[0533] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information, means for analyzing disaster risk based on the collected data, and means for creating and proposing a risk report based on the analysis results. This makes it possible to monitor the emotional state of staff and suggest rest if their stress level is high. Furthermore, displaying disaster risk information and countermeasures using augmented reality enables intuitive and rapid response.

[0534] "Disaster risk information" refers to information on natural disasters such as earthquakes, typhoons, and floods, including past occurrence data and forecast data.

[0535] "Past disaster data" refers to information about past disasters, including the extent and impact of damage and the effectiveness of response measures.

[0536] "Geographic information" refers to information about geographical characteristics such as topography, topography, and geology, including land use and building layout.

[0537] "Infrastructure information" refers to information related to social infrastructure such as roads, bridges, power supplies, and water supplies.

[0538] "Means of collection" refers to the technical means used to incorporate the above information into the system using various databases and APIs.

[0539] "Means for analyzing disaster risk" refers to analytical techniques for assessing and predicting disaster risk based on collected data.

[0540] "Risk Report" means a report summarizing the results of a disaster risk analysis, including the scope of impact and damage forecast.

[0541] A "disaster preparedness plan" is a document that outlines specific measures and procedures to be taken in the event of a disaster.

[0542] "Training plan" refers to the specific schedule and content of training conducted based on the disaster response plan.

[0543] "Means of monitoring" refers to technology that monitors the implementation of training and disaster prevention measures in real time.

[0544] "Staff emotional state" refers to the psychological state of logistics center staff, such as stress and anxiety.

[0545] "Emotion monitoring" refers to technology that measures and monitors the emotional state of staff in real time.

[0546] "Measures to suggest rest when stress levels are high" refers to technological measures that encourage appropriate rest when staff stress levels exceed a certain value.

[0547] "Augmented reality display" refers to technology that overlays virtual information on real-world scenes, making it easier to visually understand disaster risks and countermeasures.

[0548] The present invention is implemented as a system for managing disaster risks and monitoring staff emotions in a logistics center. This system operates in cooperation with a server, terminals, and users (logistics center staff).

[0549] Data collection

[0550] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information through various databases and APIs. The collected data includes natural disaster risk around the logistics center, local population density, and the condition of roads and power infrastructure.

[0551] Data Preprocessing

[0552] The server cleanses and consolidates the collected data, converting it into a suitable format for analysis by imputing missing data, detecting outliers, and handling them appropriately.

[0553] Risk Analysis

[0554] The server inputs the preprocessed data into an AI model to analyze disaster risks around the logistics center, thereby evaluating disaster risks such as earthquakes, typhoons, and floods, and predicting the extent of impact and damage.

[0555] Creating a risk report

[0556] The server generates a risk report based on the results of the risk analysis, which includes a visualization of disaster risk and identification of vulnerable areas, and is provided in PDF format or on an online dashboard.

[0557] Introducing the Emotion Engine

[0558] Implement an emotion engine to assess the psychological impact of training and disaster situations on staff, for example by monitoring staff stress levels during training and adjusting training content as needed.

[0559] Disaster prevention plan proposal

[0560] The server generates a disaster response plan based on the risk report, including the location of evacuation sites, the amount of relief supplies needed, and the location of medical facilities. An emotion engine also takes into account the psychological impact of the plan.

[0561] Building a training plan

[0562] The device then creates a training plan based on the generated disaster response plan, including simulations of evacuation routes and the redeployment of relief supplies. It uses an emotion engine to monitor the emotional state of staff in real time and adjusts the training accordingly.

[0563] Augmented reality display

[0564] The server provides a means to display disaster risk information and countermeasures in augmented reality, making it easier for logistics center staff to intuitively understand risks and countermeasures.

[0565] Emergency response

[0566] When a disaster occurs, users (logistics center staff) will quickly begin responding based on a countermeasure plan proposed in advance. The emotion engine takes into account the emotional state of staff members and supports them in taking the optimal response.

[0567] Post-recovery feedback and improvements

[0568] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The emotion engine takes into account the psychological impact and proposes improvements for the next disaster response. This will ensure that the next disaster response is psychologically optimal.

[0569] This system uses the following hardware and software:

[0570] Hardware: Smart glasses (e.g., Google® Glass®, Microsoft® HoloLens®)

[0571] Software: EmotionEngine, ARDisplay (Augmented Reality Display)

[0572] Examples of concrete examples and prompts

[0573] As a concrete example, we present a scenario in which the system is implemented using smart glasses. Logistics center staff wear the smart glasses, and in the event of a disaster, real-time risk information and evacuation routes are visually displayed. If the staff's emotional state is determined to be high stress, appropriate rest is suggested.

[0574] Example prompt sentence:

[0575] "We will implement a system that analyzes current earthquake risks in real time and visualizes the extent of damage and countermeasures. It will also monitor the emotional state of users (logistics center staff) and suggest rest if they are under high stress."

[0576] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0577] Step 1:

[0578] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs. The collected data includes information on natural disaster risk around the logistics center, local population density, and the condition of roads and power infrastructure. Specifically, it calls the API and stores the acquired data in a database. The input is raw data obtained from the API, and the output is an integrated dataset.

[0579] Step 2:

[0580] The server cleanses and consolidates the collected data. Data cleansing involves imputing missing data and detecting and appropriately handling outliers. It then converts data from different formats into a unified format. The input is the raw data collected in step 1, and the output is a clean dataset for analysis.

[0581] Step 3:

[0582] The server inputs the preprocessed data into the AI ​​model and analyzes the disaster risk around the logistics center. The AI ​​model predicts the probability of natural disasters occurring and estimates the extent of impact and damage. The input is the clean dataset created in step 2, and the output is the disaster risk analysis results. Specifically, the data is input into the AI ​​model and the results are obtained.

[0583] Step 4:

[0584] The server creates a risk report based on the results of the risk analysis. The risk report includes visualization of disaster risk and identification of areas susceptible to impact. The risk report is provided in PDF or online dashboard format. The input is the analysis results from Step 3, and the output is the risk report. Specifically, the analysis results are incorporated into a template and a report is generated.

[0585] Step 5:

[0586] The server uses an emotion engine to monitor the emotional state of staff during training or disasters. The emotion engine evaluates the stress level of staff in real time. The input is the physiological data of staff (e.g., heart rate, electrodermal activity), and the output is the stress level evaluation result. Specifically, it collects sensor data, inputs it into the emotion engine, and obtains the analysis results.

[0587] Step 6:

[0588] The server generates a disaster response plan based on the risk report. The plan includes the establishment of evacuation sites, the amount of relief supplies needed, and the location of medical facilities. The emotion engine also considers whether the plan will have a psychological impact. The input is the risk report from Step 4 and the emotion evaluation results from Step 5, and the output is the disaster response plan.

[0589] Step 7:

[0590] The terminal creates a training plan based on the generated disaster response plan. A training scenario is created, including confirmation of evacuation routes and redeployment of relief supplies. An emotion engine is used to monitor the emotional state of staff during the training in real time and adjust the training content accordingly. The input is the disaster response plan from step 6, and the output is the training plan.

[0591] Step 8:

[0592] The server provides a means to display disaster risk information and countermeasures in augmented reality. This allows logistics center staff to visually grasp the risk situation and countermeasures in real time. The input is the analysis results from step 3 and the disaster countermeasure plan from step 6, and the output is an augmented reality display. Specifically, this data is sent to the ARDisplay module and displayed on the AR device.

[0593] Step 9:

[0594] When a disaster occurs, the user (logistics center staff) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine is used to monitor the emotional state of staff, and appropriate assistance is provided if overwork or stress is detected. The input is the augmented reality display and emotion assessment results from Step 8, and the output is the response action.

[0595] Step 10:

[0596] After responding to a disaster, the server collects data and evaluates the effectiveness of the countermeasures. Next, the emotion engine takes into account the psychological impact and proposes improvements for the next disaster response. The input is the actually collected data and the emotion evaluation results, and the output is improvement proposals and a report. Specifically, the system analyzes the collected data and plans the next response measures.

[0597] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0598] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0599] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0600] [Second embodiment]

[0601] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0602] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0603] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0604] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0605] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0606] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0607] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0608] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0609] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0610] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0611] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0612] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0613] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users (local government officials) as follows:

[0614] Data collection

[0615] First, the server collects necessary disaster risk information, past disaster data, geographic information, and infrastructure information from the databases and APIs of local governments and related organizations, including population distribution data, national geographic information, and infrastructure company operation data.

[0616] Examples:

[0617] The server collects past earthquake occurrence history for each region from the database of local government A and imports the latest geographical information and supply infrastructure information from other related organizations.

[0618] Data Preprocessing

[0619] The server cleanses and integrates the collected data. Cleansing involves filling in missing data and processing outliers. Integration involves storing data in different formats in a unified database.

[0620] Risk Analysis

[0621] The server then inputs the organized data into an AI model to analyze disaster risk, which then predicts the disaster risk, impact area, and damage for each region.

[0622] Examples:

[0623] The server analyzes the probability of an earthquake occurring in Area B, predicts the epicenter and seismic intensity, and combines this data with population distribution data to identify the areas that will be most affected.

[0624] Creating a risk report

[0625] The server then generates a risk report based on the analysis results, which includes a visualization of disaster risk, identification of the most affected areas, and an assessment of their impact. The risk report is available in PDF and online dashboard format.

[0626] Examples:

[0627] The server creates an earthquake risk report for Area B, showing areas likely to be affected on a map. The report is generated as a PDF and sent to local government officials.

[0628] Disaster prevention plan proposal

[0629] The server generates a disaster response plan based on the risk report, including the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities.

[0630] Examples:

[0631] The server generates a list of suitable evacuation sites in Area B, calculates and provides a table of supplies needed for each site, and assesses the capacity of nearby major medical facilities to determine whether additional medical assistance is needed.

[0632] Building a training plan

[0633] A terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan, which includes hypothetical scenarios and a schedule of practical training exercises.

[0634] Examples:

[0635] The local government's terminal creates an evacuation drill plan assuming an earthquake occurs in Area B. The plan includes confirmation of evacuation routes and a simulation of opening evacuation shelters.

[0636] Status Monitoring

[0637] The server monitors in real time the status of training and disaster response measures, detecting progress and problems and making adjustments as necessary.

[0638] Examples:

[0639] The server monitors the progress of evacuation drills being conducted in Area B in real time and immediately notifies local government officials if any problems arise. For example, if a particular evacuation route is congested, the server will use that information to suggest an alternative route.

[0640] Emergency response

[0641] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance. The system will support the smooth implementation of planned responses, such as setting up evacuation shelters, distributing relief supplies, and providing medical care.

[0642] Examples:

[0643] If an earthquake actually occurs in Area B, local government officials will follow the instructions from the server to quickly open evacuation shelters and guide residents there, as well as deliver necessary relief supplies to the appropriate locations and deploy medical teams.

[0644] Post-recovery feedback and improvements

[0645] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. Based on the results, it makes suggestions for improvements to be made in the next disaster response.

[0646] Examples:

[0647] After the disaster has subsided, the server will collect data on shelter operations and relief supply distribution in Area B, analyze which aspects were effective and which areas need improvement, and create a report based on the results to propose even more effective response measures for the next disaster.

[0648] In this way, the server, terminal, and user work together to realize a system that provides effective disaster risk analysis and countermeasures.

[0649] The processing flow will be explained below.

[0650] Step 1: Collect data

[0651] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[0652] Specific behavior:

[0653] Obtain population distribution data for residents from the database of Municipality A.

[0654] Download the latest topographic data from the Geospatial Information Authority of Japan.

[0655] It collects information on power supply status and water supply operation sent from infrastructure companies.

[0656] Step 2: Preprocessing the data

[0657] The server cleanses and consolidates the collected data: first, it completes missing data, then it detects outliers and handles them appropriately.

[0658] Specific behavior:

[0659] Complement missing data with historical statistical data.

[0660] Detect outliers (e.g., extremely high population growth rates) and identify their causes.

[0661] Convert data in different formats into a unified format (e.g., CSV format).

[0662] Step 3: Risk analysis

[0663] The server inputs the preprocessed data into an AI model to analyze the disaster risk in each region.

[0664] Specific behavior:

[0665] Geographical information and past earthquake data are input into the earthquake prediction model to calculate the probability of an earthquake occurring in each region.

[0666] River and precipitation data are input into the flood risk model to identify vulnerable areas.

[0667] Step 4: Create a risk report

[0668] The server creates a visually easy-to-understand risk report based on the results of the risk analysis.

[0669] Specific behavior:

[0670] The probability of an earthquake occurring and predicted seismic intensity are plotted on a map to show the area of ​​impact.

[0671] Areas at high risk of flooding are color-coded.

[0672] Generate reports as PDFs or online dashboards and send them to city officials.

[0673] Step 5: Propose a disaster response plan

[0674] The server automatically generates a disaster recovery plan based on the risk report.

[0675] Specific behavior:

[0676] Create a list of shelter locations and needed supplies.

[0677] Establish a schedule for relief operations and develop a resource allocation plan based on that schedule.

[0678] Assess the capacity of medical facilities and determine the number of medical support teams needed.

[0679] Step 6: Develop a training plan

[0680] The terminal will then implement a training plan based on the proposed disaster response plan.

[0681] Specific behavior:

[0682] Create hypothetical scenarios and plan simulation training.

[0683] Schedule the training and notify participants.

[0684] Schedule on-the-job training and prepare necessary supplies and equipment.

[0685] Step 7: Status monitoring

[0686] The server monitors the progress and implementation status of training in real time.

[0687] Specific behavior:

[0688] Collect data during training and display progress on a dashboard.

[0689] When a problem occurs, we will notify you immediately and give you instructions on how to fix it.

[0690] Evaluate each stage in real time and adjust your plan as needed.

[0691] Step 8: Emergency response

[0692] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance.

[0693] Specific behavior:

[0694] Quickly set up evacuation shelters and guide residents there.

[0695] Check the delivery status of relief supplies and arrange for additional supplies if necessary.

[0696] Medical teams will be deployed to provide emergency response.

[0697] Step 9: Post-recovery feedback and improvement

[0698] The server collects data after disaster response and evaluates the effectiveness of countermeasures. Based on the results, it makes suggestions for improvements to be made in future disaster responses.

[0699] Specific behavior:

[0700] Collect and analyze evacuation center operation data and material distribution status.

[0701] We will extract good points and problems in the response and create an evaluation report.

[0702] A report summarizing areas for improvement will be provided to local government officials as feedback for the next event.

[0703] Example 1

[0704] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0705] In recent years, the frequency and scale of disasters have been increasing, making it extremely important to assess disaster risks and prepare appropriate countermeasures in advance. However, in conventional systems, data collection, risk analysis, and countermeasure proposals are dispersed, preventing efficient and integrated countermeasures. In particular, data preprocessing and real-time situation monitoring are lacking, making it difficult to implement rapid and accurate disaster countermeasures.

[0706] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0707] In this invention, the server includes a means for importing disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs, a means for cleansing and integrating the collected data, completing missing data, and processing outliers, and a means for inputting the preprocessed data into a generative AI model to analyze disaster risks. This makes it possible to assess disaster risks in advance and prepare appropriate countermeasures quickly and accurately.

[0708] "Disaster risk information" refers to data and information used to assess the likelihood of a disaster occurring and its impact. Specifically, this includes the probability of occurrence and damage forecasts for natural disasters such as earthquakes, typhoons, tsunamis, floods, and volcanic eruptions.

[0709] "Past disaster data" refers to detailed records of past disasters, including information on the date, location, scale, damage, and response measures.

[0710] "Geographic information" is data that indicates the geographical characteristics of a specific area, including information on topography, elevation, water systems, land use, infrastructure layout, etc.

[0711] "Infrastructure information" refers to data on infrastructure such as electricity, gas, water, communications, and transportation, including the location, operating status, and supply capacity of each piece of infrastructure.

[0712] A "database" is a system for systematically collecting, storing, and managing specific information. It is used to integrate information from multiple different data sources.

[0713] "API" stands for Application Programming Interface, a set of protocols and tools that allow data to be exchanged between different pieces of software.

[0714] "Cleansing" is the process of correcting or removing missing or outlier values ​​to improve data quality.

[0715] "Integration" is the process of bringing together data from different formats into one standardized format.

[0716] A "generative AI model" is an artificial intelligence model that learns from collected data and performs risk analysis and predictions.

[0717] A "risk report" is a report summarizing the results of a risk analysis, including predictions of the probability of a disaster occurring and the extent of its impact.

[0718] A "disaster response plan" is a specific action plan for minimizing damage in the event of a disaster. It includes the establishment of evacuation shelters, preparation of supplies, medical response, etc.

[0719] A "training plan" is a plan that defines the specific schedule and procedures for training to be conducted based on a disaster response plan.

[0720] "Monitoring" is the process of monitoring the response situation during training and disasters in real time and making adjustments as necessary.

[0721] "Rapid response" refers to immediately taking appropriate countermeasures when a disaster occurs.

[0722] "Evaluation" is the process of analyzing the results of disaster response, measuring effectiveness, and identifying areas for improvement.

[0723] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users as follows:

[0724] Data collection

[0725] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from databases and APIs of local governments and related organizations. At this stage, API calls and database connections are made using a high-performance server machine and Python. For example, the server collects resident population distribution data, national geographic information, and infrastructure company operation data.

[0726] Data Preprocessing

[0727] The server cleanses and integrates the collected data. It uses the Pandas library and NumPy to fill in missing data and process outliers. It also stores data in a unified format in the database. This process produces reliable data.

[0728] Risk Analysis

[0729] The server inputs the preprocessed data into a generative AI model to analyze disaster risk. The AI ​​model used is trained using TensorFlow or PyTorch, enabling highly accurate predictions. For example, the server analyzes the probability of an earthquake occurring in a specific area and identifies areas likely to be affected.

[0730] Creating a risk report

[0731] The server creates a risk report based on the results of the risk analysis. The risk report includes visualization of disaster risks, identification of the most affected areas, and their assessment. This report is generated in PDF format and converted using tools such as Adobe Acrobat. For example, the server displays the affected areas on a map, generates a PDF report, and sends it to local government officials.

[0732] Disaster prevention plan proposal

[0733] The server automatically generates a disaster response plan based on the risk report. This plan includes the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities. For example, the server generates a list of appropriate evacuation shelters and calculates the amount of supplies needed for each shelter. It also assesses the capacity of nearby medical facilities and determines whether additional medical assistance is needed.

[0734] Building a training plan

[0735] The terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan. The training plan includes hypothetical scenarios and a schedule for practical training. As a specific example, the terminal creates an evacuation training plan in an Excel file and notifies each department.

[0736] Status Monitoring

[0737] The server monitors in real time the status of training and the implementation of countermeasures in the event of a disaster. Zabbix and Nagios are used for this monitoring, detecting progress and problems and making adjustments as necessary. For example, the server monitors the progress of training in real time and immediately notifies local government officials if a problem occurs.

[0738] Emergency response

[0739] When a disaster occurs, users (local government officials) can quickly respond based on the prepared response plan. For example, users can use mobile devices and PCs to set up evacuation shelters, distribute relief supplies, and provide medical care.

[0740] Post-recovery feedback and improvements

[0741] The server collects data after disaster response and evaluates the effectiveness of the response. During this process, it reanalyzes information from the database and proposes future improvements. For example, the server analyzes data on evacuation center operations and relief supply distribution to identify areas for improvement.

[0742] In this way, the system provides effective disaster risk analysis and countermeasures by having the server, terminals, and users work together.

[0743] Example prompt for a generative AI model:

[0744] "Please explain in detail the procedures for collecting and preprocessing geographic information, past disaster data, infrastructure information, and population distribution data as input data for the AI ​​model to analyze disaster risk."

[0745] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0746] Step 1: Data collection

[0747] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from databases and APIs of local governments and related organizations. Specifically, it uses the Python requests library to obtain data from APIs and connects to various databases to import information. Input data includes resident population distribution data, national geographic information, and infrastructure operation status data, and is stored on the server. Output data is the collected raw data in various formats.

[0748] Specific operation: When the server retrieves data from the API, it uses the requests.get() method to retrieve the required data and parses it in JSON format.

[0749] Step 2: Data Preprocessing

[0750] The server cleanses and integrates the collected data. Specifically, it uses the Pandas library and NumPy to impute incomplete data, process outliers, and standardize data in different formats. The input data is the collected raw data, and the output data is the cleansed, integrated, and reliable data.

[0751] Specific operation: The server uses the Pandas library to fill in missing values ​​using the fillna() method, etc., and masks outliers.

[0752] Step 3: Risk analysis

[0753] The server inputs the preprocessed data into a generative AI model to analyze disaster risk. The AI ​​models used include TensorFlow and PyTorch, which utilize advanced machine learning algorithms. The input data is preprocessed, reliable data, and the output data is risk assessment results and prediction data.

[0754] Specific operation: The server uses TensorFlow's predict method to input the preprocessed data into the model and obtain prediction results.

[0755] Step 4: Create a risk report

[0756] The server creates the actual risk report based on the results of the risk analysis. It uses Matplotlib and Seaborn libraries to generate graphs and maps and create a PDF version of the risk report. The input data are the risk assessment results and prediction data, and the output data is the generated risk report.

[0757] What it does: The server uses Matplotlib to plot the risk assessment results and then saves the generated graph to a PDF.

[0758] Step 5: Propose a disaster response plan

[0759] The server automatically generates a disaster response plan based on the risk report. This plan includes the location of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities. The input data is the risk report, and the output data is the disaster response plan.

[0760] Specific operation: The server analyzes the risk report data and calculates the evacuation shelter list and the amount of necessary supplies.

[0761] Step 6: Develop a training plan

[0762] The terminal (e.g., the local government's computer system) creates a training plan based on the proposed disaster response plan. The training schedule is created in an Excel file and notified to each department. The input data is the disaster response plan, and the output data is the training plan.

[0763] Specific operation: The terminal uses the openpyxl library to write and save the training schedule in an Excel file.

[0764] Step 7: Status monitoring

[0765] The server monitors in real time the status of training and the implementation of countermeasures in the event of a disaster. Monitoring is performed using Zabbix API and Nagios, and if an abnormality is detected, it immediately notifies. The input data is training and disaster response data collected in real time, and the output data is the monitoring results and alerts.

[0766] Specific operation: The server uses the Zabbix API to obtain real-time data and issues an alert if an abnormality is detected.

[0767] Step 8: Emergency response

[0768] When a disaster occurs, users (local government officials) quickly begin responding based on a pre-proposed response plan. They use mobile devices and PCs to set up evacuation shelters, distribute relief supplies, and provide medical care. The input data is the disaster response plan and real-time situation data, and the output data is the actual response history.

[0769] Specific actions: The user follows the plan, opens designated evacuation centers, and distributes necessary supplies.

[0770] Step 9: Post-recovery feedback and improvement

[0771] The server collects data after a disaster response and evaluates the effectiveness of the response. Based on this evaluation data, it makes suggestions for improving future disaster responses. The input data is various data from after the disaster response, and the output data is suggestions for improving future disaster responses.

[0772] Specific operation: The server evaluates post-disaster response data, analyzes which parts were effective, and identifies areas for improvement.

[0773] (Application example 1)

[0774] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0775] Conventional disaster risk management systems lack sufficient real-time risk monitoring, immediate response when anomalies are detected, and proposals for advance countermeasure plans. This has led to problems such as reduced efficiency in systems for quickly responding to disasters and evaluating the effectiveness of countermeasures. Furthermore, there is a lack of a way to present disaster risk analysis results in a visually understandable manner.

[0776] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0777] In this invention, the server includes: means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information; means for analyzing disaster risk based on the collected data; means for creating and proposing a risk report based on the analysis results; means for automatically generating a disaster response plan; means for constructing a training plan based on the generated disaster response plan; means for monitoring the implementation status and effectiveness of training in real time and making adjustments as necessary; means for promptly initiating a response based on a previously proposed response plan when a disaster occurs; means for collecting post-disaster response data, evaluating the effectiveness of measures, and proposing improvements for the next disaster response; means for monitoring disaster risk information in real time and implementing emergency measures when an abnormality is detected; means for predicting disaster risk information based on an AI model and proposing a pre-disaster response plan; and means for graphing and displaying disaster risk analysis results and generating a report in PDF format. This enables real-time disaster risk monitoring, rapid emergency response in emergencies, and the provision of visually easy-to-understand risk analysis results.

[0778] "Disaster risk information" is data that indicates possible risks related to disasters, and includes information on natural disasters such as earthquakes, tsunamis, and typhoons.

[0779] "Past disaster data" refers to data that records the history of past disasters, the extent of their impact, and the extent of the damage.

[0780] "Geographic information" refers to data relating to the topography and geographical characteristics of a particular area, including map information and topographical maps.

[0781] "Infrastructure information" refers to data on public services and facilities such as electricity, gas, water, and transportation.

[0782] "Means for analyzing disaster risk" are methods and tools for assessing the likelihood of occurrence and the scope of impact based on collected disaster risk information.

[0783] "Means for preparing and proposing risk reports" refers to a method of preparing the results of disaster risk analysis in report format and proposing countermeasures based on that information.

[0784] "Means for automatically generating disaster prevention plans" refers to a method by which the system automatically plans and proposes optimal countermeasures based on the results of disaster risk analysis.

[0785] The "means for constructing a training plan" refers to a method for setting up practical training or scenario-based training based on the generated disaster response plan.

[0786] "Real-time monitoring and adjustment measures" refer to methods for immediately monitoring the progress of training exercises and actual disaster responses and making necessary adjustments.

[0787] "Measures to initiate a rapid response" are methods for quickly putting into action measures that have been planned in advance when a disaster occurs.

[0788] "Means for collecting data and evaluating effectiveness" refers to methods for collecting data after disaster response and evaluating the effectiveness of the measures taken.

[0789] "Means for monitoring disaster risk information in real time" refers to a method for continuously monitoring disaster risk information in real time.

[0790] The "means of implementing emergency measures when an abnormality is detected" refers to a method for quickly implementing an emergency response when an abnormality is detected in disaster risk information.

[0791] "Means for making predictions based on AI models and proposing advance countermeasure plans" refers to a method for predicting disaster risks using artificial intelligence models and proposing effective countermeasure plans in advance based on the results.

[0792] "Means for displaying graphs based on risk analysis results and generating reports in PDF format" refers to a method for visually displaying the results of disaster risk analysis and creating reports in PDF format.

[0793] MODE FOR CARRYING OUT THE INVENTION

[0794] This invention is a factory disaster risk management system that collects disaster risk information, past disaster data, geographical information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users (local government officials).

[0795] Data collection

[0796] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs, including population distribution data, geographic information, and infrastructure company operation data.

[0797] Example: The server collects historical earthquake occurrence history for each region from the local government database and imports up-to-date geographical and supply infrastructure information from other relevant agencies.

[0798] Data Preprocessing

[0799] The server cleanses and integrates the collected data. Cleansing involves filling in missing data and processing outliers. Integration involves storing data in different formats in a unified database.

[0800] Risk Analysis

[0801] The server inputs the organized data into an AI model to analyze disaster risk. This allows for predictions of disaster risk, impact area, and damage for each region. The server performs the analysis using TensorFlow and Scikit-learn.

[0802] Example: The server analyzes the probability of an earthquake occurring in a specific area, predicts the epicenter and intensity, and combines this with population distribution data to identify the areas most affected.

[0803] Creating a risk report

[0804] The server generates a risk report based on the analysis results. This report includes visualization of disaster risks, identification of the most affected areas, and assessment of their impact. The risk report is provided in PDF and online dashboard format. Matplotlib and ReportLab are used for visualization and report generation.

[0805] Example: The server creates an earthquake risk report for a specific area, showing areas likely to be affected on a map. The report is generated as a PDF and sent to local government officials.

[0806] Disaster prevention plan proposal

[0807] The server generates a disaster response plan based on the risk report, including the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities.

[0808] Example: A server generates a list of suitable evacuation sites in a particular area, calculates and provides a list of supplies needed for each site, and assesses the capacity of nearby major medical facilities to determine if additional medical assistance is needed.

[0809] Building a training plan

[0810] A terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan, which includes hypothetical scenarios and a schedule of practical training exercises.

[0811] Example: A local government device creates an evacuation drill plan for a specific area in the event of an earthquake, including confirmation of evacuation routes and a simulation of the establishment of evacuation shelters.

[0812] Status Monitoring

[0813] The server monitors in real time the status of training and disaster response measures, detecting progress and problems and making adjustments as necessary.

[0814] Example: The server monitors the progress of evacuation drills in a specific area in real time and immediately notifies local government officials if a problem occurs. For example, if a specific evacuation route is congested, it will use that information to suggest an alternative route.

[0815] Emergency response

[0816] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance. The system will support the smooth implementation of planned responses, such as setting up evacuation shelters, distributing relief supplies, and providing medical care.

[0817] Example: If an earthquake actually occurs in a specific area, local government officials will follow instructions from the server to quickly set up evacuation shelters and guide residents, as well as deliver necessary relief supplies to the appropriate locations and deploy medical teams.

[0818] Post-recovery feedback and improvements

[0819] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. Based on the results, it makes suggestions for improvements to be made in the next disaster response.

[0820] Example: After a disaster has subsided, the server collects data on the operation of evacuation centers and the distribution of relief supplies in a specific area, analyzes which aspects were effective and which areas need improvement, and creates a report based on the results to propose more effective response measures for the next disaster.

[0821] Prompt Sentence Examples

[0822] "Analyze the latest earthquake risks in the designated area and take safety measures."

[0823] "How can we collect, analyze, and display real-time disaster risk information on a dashboard?"

[0824] In this way, the server, terminal, and user work together to realize a system that provides effective disaster risk analysis and countermeasures.

[0825] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0826] MODE FOR CARRYING OUT THE INVENTION (PROGRAM PROCESSING STEPS)

[0827] Step 1:

[0828] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs. The server imports resident population distribution data, geographic information, and infrastructure company operation data. The input is data from APIs and databases, and the output is a dataset that integrates this information. Specifically, the data is stored in a data frame using Python's pandas library.

[0829] Step 2:

[0830] The server cleanses and integrates the collected data. It uses the collected raw data as input and performs tasks such as filling in missing values ​​and removing outliers. The output is a cleansed dataset stored in a unified format. Specifically, it uses pandas functions to interpolate missing values ​​and remove outliers.

[0831] Step 3:

[0832] The server inputs the organized data into an AI model to analyze disaster risk. The input is a cleansed dataset, and the output is a disaster risk prediction. Specifically, TensorFlow and Scikit-learn are used to input data into the model and make predictions. For example, the probability of an earthquake occurring and the extent of its impact can be predicted.

[0833] Step 4:

[0834] The server creates a risk report based on the analysis results. The input is the prediction results of the AI ​​model, and the output is a risk report (in PDF or online dashboard format). Specifically, Matplotlib is used to create graphs, and ReportLab is used to generate the PDF report. High-risk areas and key countermeasures are visually displayed.

[0835] Step 5:

[0836] The server generates a disaster response plan based on the risk report. The input is the risk report, and the output is a specific disaster response plan. Specifically, it calculates the locations of evacuation shelters and the amount of supplies needed, and compiles them in list form.

[0837] Step 6:

[0838] The terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan. The input is the disaster response plan, and the output is the training plan. Specific operations include setting the date and scenario for the evacuation training.

[0839] Step 7:

[0840] The server monitors the status and effectiveness of training in real time and makes adjustments as necessary. The input is training progress data, and the output is monitoring results and adjustments. Specifically, it collects sensor data and GPS information in real time and displays it on a dashboard.

[0841] Step 8:

[0842] When a disaster occurs, the user (local government official) will quickly begin responding based on a countermeasure plan proposed in advance. The input is the countermeasure plan, and the output is the status of the countermeasures that have been implemented. Specific actions include setting up evacuation shelters and distributing supplies.

[0843] Step 9:

[0844] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The input is disaster response data, and the output is an effectiveness evaluation report. Specifically, the server analyzes the effectiveness of the countermeasures based on the collected data and proposes improvements for the next disaster response.

[0845] Prompt Sentence Examples

[0846] "Analyze the latest earthquake risks in the designated area and take safety measures."

[0847] "How can we collect, analyze, and display real-time disaster risk information on a dashboard?"

[0848] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0849] This invention combines a system that collects disaster risk information, past disaster data, geographical information, and infrastructure information, analyzes disaster risks, and proposes optimal disaster countermeasures with an emotion engine that recognizes the user's emotions. This system operates in cooperation with a server, terminal, and user (local government official) as follows:

[0850] Data collection

[0851] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations, including resident demographic distribution data, up-to-date geographical information, and supply infrastructure information.

[0852] Examples:

[0853] The server collects past disaster occurrence history for each region from the database of X municipality and imports the latest geographical information and lifeline information from other related organizations.

[0854] Data Preprocessing

[0855] The server cleanses and consolidates the collected data, imputes missing data, detects outliers and handles them appropriately, and then converts data from different formats into a unified format.

[0856] Risk Analysis

[0857] The server inputs the preprocessed data into an AI model to analyze the disaster risk for each region, thereby predicting the disaster risk, impact area, and damage for each region.

[0858] Examples:

[0859] The server analyzes the probability of an earthquake occurring in region X, predicts the epicenter and seismic intensity, and combines this with population distribution data to identify the areas that will be most affected.

[0860] Creating a risk report

[0861] The server generates a risk report based on the results of the risk analysis. This report visualizes disaster risks, identifies vulnerable areas, and assesses their impact. The risk report is provided in PDF and online dashboard format.

[0862] Examples:

[0863] The server creates an earthquake risk report for region X, showing areas likely to be affected on a map, and generates a PDF report that is sent to local government officials.

[0864] Introducing the Emotion Engine

[0865] Implement an emotion engine to assess the psychological impact of disaster response plans and help adjust training plans and implementation.

[0866] Examples:

[0867] The server monitors the stress levels and emotional state of users (local government officials) during the training and adjusts the training content as necessary. For example, if many of the training participants are feeling high levels of stress, it will temporarily lower the difficulty of the training.

[0868] Disaster prevention plan proposal

[0869] The server generates disaster response plans based on the risk reports, including the locations of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities, while also taking into account the psychological impact of the plans through an emotion engine.

[0870] Examples:

[0871] The server generates a list of suitable evacuation sites in area X, calculates the amount of supplies needed for each site, and uses an emotion engine to suggest the best psychological placement of the evacuation sites.

[0872] Building a training plan

[0873] The device creates a training plan based on the proposed disaster response plan, creates a virtual scenario and a schedule for practical training, and adjusts the training content based on the user's emotional state during the training using an emotion engine.

[0874] Examples:

[0875] The local government's terminal creates an evacuation drill plan assuming an earthquake occurs in region X. The plan includes confirmation of evacuation routes and a simulation of opening evacuation shelters. An emotion engine is used to detect users' stress and anxiety that arise during the drill in real time and adjust the drill content accordingly.

[0876] Status Monitoring

[0877] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[0878] Examples:

[0879] The server monitors the progress of evacuation drills being conducted in Region X in real time, and immediately notifies local government officials if any problems arise. In addition, if officials feel extremely stressed during the drill, the server temporarily halts the drill and allows them time to refresh.

[0880] Emergency response

[0881] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine takes into account the user's emotional state and helps implement the optimal response.

[0882] Examples:

[0883] If an earthquake actually occurs in region X, local government officials will follow the server's instructions to quickly open evacuation shelters and guide residents. The emotion engine will monitor the officials' emotional state and provide appropriate support if overwork or stress is detected.

[0884] Post-recovery feedback and improvements

[0885] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The emotion engine also takes into account psychological impacts and proposes improvements for the next disaster response.

[0886] Examples:

[0887] After the disaster subsides, the server collects data on shelter operations and relief supply distribution in region X, analyzing which aspects were effective and where there is room for improvement. It also analyzes the psychological data of staff members during the operation and evaluates areas where stress and anxiety became a problem. Based on the results, a report is created proposing optimal psychological measures for the next disaster response.

[0888] In this way, by having the server, terminal, and user work together, it is possible to realize a system that provides effective disaster risk analysis and countermeasures that also take into account the user's emotions.

[0889] The processing flow will be explained below.

[0890] Step 1: Collect data

[0891] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[0892] Specific behavior:

[0893] Obtain the history of past disasters by region from the database of X municipality.

[0894] Download the latest topographic data from the Geospatial Information Authority of Japan.

[0895] It collects information on power supply status and water supply operation sent from infrastructure companies.

[0896] Step 2: Preprocessing the data

[0897] The server cleanses and integrates the collected data, supplements missing data with past statistical data, detects and appropriately processes outliers (e.g., extremely high population growth rates), and converts data in different formats into a unified format (e.g., CSV format).

[0898] Specific behavior:

[0899] Complement missing earthquake occurrence data from the database.

[0900] Check for abnormal values ​​and analyze the cause if necessary.

[0901] Integrate topographical and population data in one unified format.

[0902] Step 3: Risk analysis

[0903] The server inputs the preprocessed data into an AI model to analyze the disaster risk in each region.

[0904] Specific behavior:

[0905] Geographical information and past earthquake data are input into the earthquake prediction model to calculate the probability of an earthquake occurring in each region.

[0906] River and precipitation data are input into the flood risk model to identify vulnerable areas.

[0907] Step 4: Create a risk report

[0908] The server generates a visually easy-to-understand risk report based on the results of the risk analysis, which is available in PDF format and an online dashboard.

[0909] Specific behavior:

[0910] The probability of an earthquake occurring and predicted seismic intensity are plotted on a map to show the area of ​​impact.

[0911] Areas at high risk of flooding are color-coded.

[0912] Generate a report in PDF format and email it to city officials.

[0913] Step 5: Implementing the Emotion Engine

[0914] Using an emotion engine, the psychological impact of disaster response plans is assessed and reflected in training plans and implementation.

[0915] Specific behavior:

[0916] The server monitors the stress level and emotional state of the user (local government official) during the training.

[0917] The server analyzes the emotion engine data in real time and adjusts the training content accordingly, for example, suggesting that the training be made easier if stress levels are high.

[0918] Step 6: Propose a disaster response plan

[0919] The server generates disaster response plans based on the risk reports, including the locations of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities. An emotion engine also takes into account the psychological impact of the plans.

[0920] Specific behavior:

[0921] Create a list of shelter locations and needed supplies.

[0922] Establish a schedule for relief operations and develop a resource allocation plan based on that schedule.

[0923] Assess the capacity of medical facilities and determine the number of medical support teams needed.

[0924] Using an emotion engine, we propose shelter layouts that have minimal psychological impact.

[0925] Step 7: Develop a training plan

[0926] The device then creates a training plan based on the proposed disaster response plan, creating virtual scenarios and practical training schedules, and using data from the emotion engine to monitor the user's emotional state and adjust the training content.

[0927] Specific behavior:

[0928] Create hypothetical scenarios and plan simulation training.

[0929] Schedule the training and notify participants.

[0930] Schedule on-the-job training and prepare necessary supplies and equipment.

[0931] The emotion engine monitors the user's emotional state, detects stress or anxiety during training, and adjusts the training progress.

[0932] Step 8: Status monitoring

[0933] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[0934] Specific behavior:

[0935] Collect data during training and display progress on a dashboard.

[0936] When a problem occurs, we will notify you immediately and give you instructions on how to fix it.

[0937] Evaluate each stage in real time and adjust your plan as needed.

[0938] An emotion engine monitors the user's emotional state and takes appropriate measures if excessive stress is detected.

[0939] Step 9: Emergency response

[0940] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine takes into account the user's emotional state and helps implement the optimal response.

[0941] Specific behavior:

[0942] Quickly set up evacuation shelters and guide residents there.

[0943] Check the delivery status of relief supplies and make additional arrangements as necessary.

[0944] Medical teams will be deployed to provide emergency response.

[0945] An emotion engine monitors the user's emotional state and provides supportive measures if overwork or stress is detected.

[0946] Step 10: Post-recovery feedback and improvement

[0947] The server collects data after disaster response and evaluates the effectiveness of countermeasures. It also takes into account psychological impacts using an emotion engine and proposes improvements for future disaster response.

[0948] Specific behavior:

[0949] Collect and analyze evacuation center operation data and material distribution status.

[0950] We analyze users' emotional data during disaster response and evaluate the areas where stress and anxiety became an issue.

[0951] We will extract good points and problems in the response and create an evaluation report.

[0952] A report summarizing areas for improvement will be provided to local government officials as feedback for the next event.

[0953] Example 2

[0954] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0955] In recent years, the frequency and scale of disasters have increased, creating a need for rapid and accurate disaster prevention measures. However, current systems are limited to collecting and analyzing disaster risk information and generating disaster prevention plans, and do not adequately consider user emotions in their training and responses. Furthermore, cleansing and integrating collected data and real-time monitoring often require manual intervention, resulting in reduced efficiency. In this situation, there is a need for a system that can achieve more effective and efficient disaster prevention measures by taking users' psychological states into account.

[0956] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0957] In this invention, the server includes means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information, means for cleansing and integrating the collected data, means for analyzing disaster risk using preprocessed data, means for creating and proposing risk reports, means for recognizing a user's emotional state using an emotion engine, means for automatically generating a disaster response plan and optimizing it taking into account the user's emotional state, means for building a training plan based on the proposed disaster response plan, means for monitoring the implementation status and effectiveness of the training in real time and making adjustments as necessary, means for promptly initiating a response based on the previously proposed response plan when a disaster occurs and providing support taking into account the user's emotional state, and means for collecting post-disaster response data, evaluating the effectiveness of the response measures, and proposing improvements for the next disaster response. This enables highly accurate analysis of disaster risk, reduces the user's psychological burden, and enables more effective and practical disaster response.

[0958] "Disaster risk information" is information that indicates the probability of a disaster occurring and the extent of its impact, and is data used to assess future disaster risks.

[0959] "Past disaster data" refers to detailed information about disasters that have occurred in the past, including the type of disaster, the location of the disaster, and the extent of the damage.

[0960] "Geographic information" is information that describes the geographical characteristics of a particular area, such as topography, land use, and infrastructure layout.

[0961] "Infrastructure information" refers to information related to the basic infrastructure of daily life, such as transportation networks, water, electricity, and gas, and is data necessary for maintaining urban functions.

[0962] "Data cleansing" is the process of removing errors and inconsistencies from collected data to improve the quality of the data.

[0963] "Data integration" is the process of combining multiple data sets collected from different sources in a consistent format to create uniformly usable data.

[0964] "Disaster risk analysis" is the process of using AI models and algorithms to assess the probability of a disaster occurring and the extent of its impact based on collected data.

[0965] A "risk report" is a report summarizing the results of disaster risk analysis, including visualized data and specific risk assessments.

[0966] The "emotion engine" is a system that analyzes a user's emotional state from voice and text data and adjusts training and countermeasures based on that information.

[0967] A "disaster preparedness plan" is a plan that lays out in advance the specific actions and procedures to be taken in the event of a disaster, including the locations of evacuation shelters and plans for the distribution of relief supplies.

[0968] A "training plan" is a plan that defines the detailed schedule and content of training to be conducted based on the proposed disaster response plan.

[0969] "Real-time monitoring" is the process by which systems constantly monitor the situation during drills and disaster response, instantly acquiring and analyzing data and making adjustments as needed.

[0970] "Starting a rapid response" means implementing necessary measures without delay based on a disaster response plan that was prepared in advance when a disaster occurs.

[0971] "Post-disaster response data" refers to data related to responses at the time of a disaster and afterward, and is used to evaluate the effectiveness of responses and areas for improvement.

[0972] "Evaluating effectiveness and proposing improvements" is the process of analyzing the effectiveness of current measures based on collected data and identifying specific areas for improvement in the next disaster response.

[0973] MODE FOR CARRYING OUT THE INVENTION

[0974] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, and combines AI technology with an emotion engine. This system is designed to monitor the user's emotional state in real time and maximize the effectiveness of disaster countermeasures. Below, we explain how this system's program works.

[0975] 1. Data Collection

[0976] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations, including population distribution data, up-to-date geographical information, and supply infrastructure information.

[0977] Hardware / software used: Data collection using RESTful API, Python libraries (requests, pandas)

[0978] Example: The server sends a request to the local government's database API to obtain the history of past disasters by region. It also obtains the latest geographical and lifeline information from other related organizations.

[0979] 2. Data Preprocessing

[0980] The server cleanses and consolidates the collected data, imputes missing data, detects outliers and handles them appropriately, and then converts data from different formats into a unified format.

[0981] Hardware / software used: Python pandas library, database management system (e.g., MySQL)

[0982] Example: The server reads the acquired data in data frame format, imputes missing values ​​with the mean value, filters outliers, and saves the data in a unified format in the database.

[0983] 3. Risk Analysis

[0984] The server inputs the preprocessed data into a generative AI model to analyze disaster risk for each region.

[0985] Hardware / software used: AI model (TensorFlow or PyTorch), data extraction from database

[0986] Example: The server inputs regional earthquake occurrence data into an AI model to predict the epicenter and intensity, and combines this with population distribution data to identify the areas most affected.

[0987] 4. Creating a risk report

[0988] The server generates a risk report based on the results of the risk analysis, which includes visualization of disaster risks, identification of vulnerable areas, and assessment of impacts.

[0989] Hardware / software used: Data visualization tools (Matplotlib, D3.js), report generation tool (PDFKit)

[0990] Example: The server visualizes the risk level on a map based on the analysis results, generates a risk report in PDF format, and provides it to the user.

[0991] 5. Introducing the Emotion Engine

[0992] The server uses an emotion engine to monitor the user's emotional state during training and execution, with the aim of reducing psychological burden and realizing effective training and disaster response.

[0993] Hardware / software used: Emotion recognition AI model (e.g. OpenCV, NLTK)

[0994] Example: The server collects speech data during training and inputs it into an emotion recognition model to analyze stress and anxiety levels in real time.

[0995] 6. Proposal of disaster prevention plan

[0996] Based on the risk report, the server generates a disaster response plan, which includes the location of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities, and also takes into account psychological impacts using an emotion engine.

[0997] Hardware / software used: Data analysis tools (e.g., SciPy), AI models

[0998] Example: The server proposes psychologically optimal shelter layouts based on the user's emotional state.

[0999] 7. Developing a training plan

[1000] The device then creates a training plan based on the proposed disaster response plan, creates a virtual scenario and a training schedule, and adjusts the training content based on the user's emotional state during the training using an emotion engine.

[1001] Hardware / software used: Scheduling tool (calendar application), real-time monitoring system

[1002] Example: The device creates a training schedule based on hypothetical scenarios and uses an emotion engine to monitor stress levels during training.

[1003] 8. Status Monitoring

[1004] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[1005] Hardware / software used: Real-time data analysis tools, dashboards (Grafana, Kibana)

[1006] Example: The server monitors the progress of training and immediately notifies the user if an abnormality is detected. It also adjusts the training content if it detects excessive stress.

[1007] 9. Emergency Response

[1008] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine will monitor the user's emotional state and provide the necessary support.

[1009] Hardware / software used: Digital instruction system, emotion recognition system

[1010] Example: The user quickly opens a shelter and guides residents. If the emotion engine detects overwork or high stress, it receives support from the server.

[1011] 10. Post-recovery feedback and improvements

[1012] The server evaluates the effectiveness of countermeasures based on post-disaster response data and makes suggestions for improvements for the next disaster response, taking into account the psychological impact.

[1013] Hardware / software used: Data analysis tools, report generation tools

[1014] Example: The server collects data from disaster response and analyzes it together with emotional data. Based on the results, it formulates a response plan for the next disaster and provides feedback to the user.

[1015] Examples of prompt statements

[1016] For example, the following might be an example of a prompt sentence to input to a generative AI model:

[1017] "Based on disaster data from the past 10 years in the region, we analyzed the current disaster risk and created a risk report."

[1018] "Create a plan for your next evacuation drill and monitor the emotional state of participants during the drill."

[1019] "We propose a response plan in the event of a disaster and provide optimal responses taking into account the stress levels of those in charge."

[1020] In this way, the present invention realizes a system that provides effective disaster risk analysis and countermeasures that take into account the user's emotions, through the cooperation of the server, terminal, and user.

[1021] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1022] Step 1:

[1023] Data collection

[1024] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[1025] Specific behavior:

[1026] Input: Database API endpoint, authentication information

[1027] Processing: The server uses a RESTful API to request various information and retrieves data in JSON or XML format.

[1028] Output: Collected raw data (e.g., past disaster history, latest geographic information, supply infrastructure information)

[1029] Hardware / software used: RESTful API, Python requests library

[1030] Step 2:

[1031] Data Preprocessing

[1032] The server cleanses and consolidates the collected data.

[1033] Specific behavior:

[1034] Input: Raw collected data (JSON, XML)

[1035] Processing: The server uses the Python pandas library to read the data in data frame format, perform missing value imputation (e.g., inserting the mean or mode), detect and remove outliers, convert data from different formats into a unified format (e.g., CSV, JSON), and save it in the database.

[1036] Output: Clean consolidated data

[1037] Hardware / software used: Python pandas library, database management system (e.g., MySQL)

[1038] Step 3:

[1039] Risk Analysis

[1040] The server inputs the preprocessed data into a generative AI model to analyze disaster risk for each region.

[1041] Specific behavior:

[1042] Input: Clean consolidated data

[1043] Processing: The server feeds the data into a generative AI model (e.g., an AI model using TensorFlow or PyTorch) to calculate disaster risk for each region, including risk assessments for earthquakes, floods, fires, etc.

[1044] Output: Disaster risk assessment results (e.g., probability of occurrence, impact area, damage forecast)

[1045] Hardware / software used: TensorFlow, PyTorch

[1046] Step 4:

[1047] Creating a risk report

[1048] The server creates a risk report based on the risk analysis results.

[1049] Specific behavior:

[1050] Input: Disaster risk assessment results

[1051] Processing: The server uses a data visualization tool (e.g., Matplotlib, D3.js) to visualize the risk data in the form of graphs and maps, and then uses a report generation tool (e.g., PDFKit) to generate the results as a PDF report.

[1052] Output: Risk report (PDF format)

[1053] Hardware / software used: Matplotlib, D3.js, PDFKit

[1054] Step 5:

[1055] Introducing the Emotion Engine

[1056] The server monitors the user's emotional state during training and execution.

[1057] Specific behavior:

[1058] Input: User voice and text data

[1059] Processing: The server inputs the data into an emotion recognition AI model to analyze the user's stress level and emotional state.

[1060] Output: User's emotional state (e.g., stress level, anxiety level)

[1061] Hardware / software used: OpenCV, NLTK

[1062] Step 6:

[1063] Disaster prevention plan proposal

[1064] The server generates a disaster recovery plan based on the risk report.

[1065] Specific behavior:

[1066] Input: Risk report, user emotional state

[1067] Processing: The server uses AI models to assess shelter locations, relief supply needs, and medical facility capacity, optimizing them based on emotional state.

[1068] Output: Disaster Preparedness Plan

[1069] Hardware / software used: SciPy, AI models

[1070] Step 7:

[1071] Building a training plan

[1072] The terminal will create a training plan based on the proposed disaster response plan.

[1073] Specific behavior:

[1074] Input: Disaster Preparedness Plan

[1075] Processing: The device creates virtual scenarios and practical training schedules, and uses an emotion engine to monitor and adjust the user's emotional state during training.

[1076] Output: Training plan

[1077] Hardware / software used: Scheduling tools, real-time monitoring systems

[1078] Step 8:

[1079] Status Monitoring

[1080] The server monitors in real time the status of countermeasures being implemented during training and when disasters occur.

[1081] Specific behavior:

[1082] Input: Training status, user's emotional state

[1083] Processing: The server uses real-time data analysis tools and dashboards to monitor progress and issues, and if anomalies are detected, it immediately notifies you and makes the necessary adjustments.

[1084] Output: Training progress, anomaly detection results

[1085] Hardware / software used: Grafana, Kibana

[1086] Step 9:

[1087] Emergency response

[1088] The user (local government official) will promptly begin responding based on the countermeasure plan proposed in advance.

[1089] Specific behavior:

[1090] Input: Real-time information and countermeasure plans in the event of a disaster

[1091] Action: The user takes action using the digital instruction system. The emotion engine monitors the user's emotional state and provides necessary assistance.

[1092] Output: Measures taken, emotional state log

[1093] Hardware / software used: Digital instruction system, emotion recognition system

[1094] Step 10:

[1095] Post-recovery feedback and improvements

[1096] The server evaluates the effectiveness of countermeasures based on data collected after the disaster response and makes suggestions for improvements for the next disaster response.

[1097] Specific behavior:

[1098] Input: Post-disaster response data, emotional state log

[1099] Processing: The server uses the data analysis tool to evaluate the effectiveness of the measures, and creates an improvement proposal report using the report generation tool.

[1100] Output: Improvement Suggestion Report

[1101] Hardware / software used: Data analysis tools, report generation tools

[1102] (Application example 2)

[1103] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1104] Conventional disaster response systems focus on analyzing disaster risks and proposing countermeasures, but lack functionality that takes into account the psychological burden on responding staff. They also lack the ability to visually grasp real-time disaster information or intuitively understand appropriate countermeasures tailored to the current situation. This makes it difficult to respond quickly and effectively in the event of a disaster, potentially resulting in the expansion of damage.

[1105] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information, means for analyzing disaster risk based on the collected data, and means for creating and proposing a risk report based on the analysis results. This makes it possible to monitor the emotional state of staff and suggest rest if their stress level is high. Furthermore, displaying disaster risk information and countermeasures using augmented reality enables intuitive and rapid response.

[1106] "Disaster risk information" refers to information on natural disasters such as earthquakes, typhoons, and floods, including past occurrence data and forecast data.

[1107] "Past disaster data" refers to information about past disasters, including the extent and impact of damage and the effectiveness of response measures.

[1108] "Geographic information" refers to information about geographical characteristics such as topography, topography, and geology, including land use and building layout.

[1109] "Infrastructure information" refers to information related to social infrastructure such as roads, bridges, power supplies, and water supplies.

[1110] "Means of collection" refers to the technical means used to incorporate the above information into the system using various databases and APIs.

[1111] "Means for analyzing disaster risk" refers to analytical techniques for assessing and predicting disaster risk based on collected data.

[1112] "Risk Report" means a report summarizing the results of a disaster risk analysis, including the scope of impact and damage forecast.

[1113] A "disaster preparedness plan" is a document that outlines specific measures and procedures to be taken in the event of a disaster.

[1114] "Training plan" refers to the specific schedule and content of training conducted based on the disaster response plan.

[1115] "Means of monitoring" refers to technology that monitors the implementation of training and disaster prevention measures in real time.

[1116] "Staff emotional state" refers to the psychological state of logistics center staff, such as stress and anxiety.

[1117] "Emotion monitoring" refers to technology that measures and monitors the emotional state of staff in real time.

[1118] "Measures to suggest rest when stress levels are high" refers to technological measures that encourage appropriate rest when staff stress levels exceed a certain value.

[1119] "Augmented reality display" refers to technology that overlays virtual information on real-world scenes, making it easier to visually understand disaster risks and countermeasures.

[1120] The present invention is implemented as a system for managing disaster risks and monitoring staff emotions in a logistics center. This system operates in cooperation with a server, terminals, and users (logistics center staff).

[1121] Data collection

[1122] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information through various databases and APIs. The collected data includes natural disaster risk around the logistics center, local population density, and the condition of roads and power infrastructure.

[1123] Data Preprocessing

[1124] The server cleanses and consolidates the collected data, converting it into a suitable format for analysis by imputing missing data, detecting outliers, and handling them appropriately.

[1125] Risk Analysis

[1126] The server inputs the preprocessed data into an AI model to analyze disaster risks around the logistics center, thereby evaluating disaster risks such as earthquakes, typhoons, and floods, and predicting the extent of impact and damage.

[1127] Creating a risk report

[1128] The server generates a risk report based on the results of the risk analysis, which includes a visualization of disaster risk and identification of vulnerable areas, and is provided in PDF format or on an online dashboard.

[1129] Introducing the Emotion Engine

[1130] Implement an emotion engine to assess the psychological impact of training and disaster situations on staff, for example by monitoring staff stress levels during training and adjusting training content as needed.

[1131] Disaster prevention plan proposal

[1132] The server generates a disaster response plan based on the risk report, including the location of evacuation sites, the amount of relief supplies needed, and the location of medical facilities. An emotion engine also takes into account the psychological impact of the plan.

[1133] Building a training plan

[1134] The device then creates a training plan based on the generated disaster response plan, including simulations of evacuation routes and the redeployment of relief supplies. It uses an emotion engine to monitor the emotional state of staff in real time and adjusts the training accordingly.

[1135] Augmented reality display

[1136] The server provides a means to display disaster risk information and countermeasures in augmented reality, making it easier for logistics center staff to intuitively understand risks and countermeasures.

[1137] Emergency response

[1138] When a disaster occurs, users (logistics center staff) will quickly begin responding based on a countermeasure plan proposed in advance. The emotion engine takes into account the emotional state of staff members and supports them in taking the optimal response.

[1139] Post-recovery feedback and improvements

[1140] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The emotion engine takes into account the psychological impact and proposes improvements for the next disaster response. This will ensure that the next disaster response is psychologically optimal.

[1141] This system uses the following hardware and software:

[1142] Hardware: Smart glasses (e.g. Google Glass, Microsoft HoloLens)

[1143] Software: EmotionEngine, ARDisplay (Augmented Reality Display)

[1144] Examples of concrete examples and prompts

[1145] As a concrete example, we present a scenario in which the system is implemented using smart glasses. Logistics center staff wear the smart glasses, and in the event of a disaster, real-time risk information and evacuation routes are visually displayed. If the staff's emotional state is determined to be high stress, appropriate rest is suggested.

[1146] Example prompt sentence:

[1147] "We will implement a system that analyzes current earthquake risks in real time and visualizes the extent of damage and countermeasures. It will also monitor the emotional state of users (logistics center staff) and suggest rest if they are under high stress."

[1148] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1149] Step 1:

[1150] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs. The collected data includes information on natural disaster risk around the logistics center, local population density, and the condition of roads and power infrastructure. Specifically, it calls the API and stores the acquired data in a database. The input is raw data obtained from the API, and the output is an integrated dataset.

[1151] Step 2:

[1152] The server cleanses and consolidates the collected data. Data cleansing involves imputing missing data and detecting and appropriately handling outliers. It then converts data from different formats into a unified format. The input is the raw data collected in step 1, and the output is a clean dataset for analysis.

[1153] Step 3:

[1154] The server inputs the preprocessed data into the AI ​​model and analyzes the disaster risk around the logistics center. The AI ​​model predicts the probability of natural disasters occurring and estimates the extent of impact and damage. The input is the clean dataset created in step 2, and the output is the disaster risk analysis results. Specifically, the data is input into the AI ​​model and the results are obtained.

[1155] Step 4:

[1156] The server creates a risk report based on the results of the risk analysis. The risk report includes visualization of disaster risk and identification of areas susceptible to impact. The risk report is provided in PDF or online dashboard format. The input is the analysis results from Step 3, and the output is the risk report. Specifically, the analysis results are incorporated into a template and a report is generated.

[1157] Step 5:

[1158] The server uses an emotion engine to monitor the emotional state of staff during training or disasters. The emotion engine evaluates the stress level of staff in real time. The input is the physiological data of staff (e.g., heart rate, electrodermal activity), and the output is the stress level evaluation result. Specifically, it collects sensor data, inputs it into the emotion engine, and obtains the analysis results.

[1159] Step 6:

[1160] The server generates a disaster response plan based on the risk report. The plan includes the establishment of evacuation sites, the amount of relief supplies needed, and the location of medical facilities. The emotion engine also considers whether the plan will have a psychological impact. The input is the risk report from Step 4 and the emotion evaluation results from Step 5, and the output is the disaster response plan.

[1161] Step 7:

[1162] The terminal creates a training plan based on the generated disaster response plan. A training scenario is created, including confirmation of evacuation routes and redeployment of relief supplies. An emotion engine is used to monitor the emotional state of staff during the training in real time and adjust the training content accordingly. The input is the disaster response plan from step 6, and the output is the training plan.

[1163] Step 8:

[1164] The server provides a means to display disaster risk information and countermeasures in augmented reality. This allows logistics center staff to visually grasp the risk situation and countermeasures in real time. The input is the analysis results from step 3 and the disaster countermeasure plan from step 6, and the output is an augmented reality display. Specifically, this data is sent to the ARDisplay module and displayed on the AR device.

[1165] Step 9:

[1166] When a disaster occurs, the user (logistics center staff) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine is used to monitor the emotional state of staff, and appropriate assistance is provided if overwork or stress is detected. The input is the augmented reality display and emotion assessment results from Step 8, and the output is the response action.

[1167] Step 10:

[1168] After responding to a disaster, the server collects data and evaluates the effectiveness of the countermeasures. Next, the emotion engine takes into account the psychological impact and proposes improvements for the next disaster response. The input is the actually collected data and the emotion evaluation results, and the output is improvement proposals and a report. Specifically, the system analyzes the collected data and plans the next response measures.

[1169] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1170] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1171] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1172] [Third embodiment]

[1173] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1174] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1175] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1176] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1177] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1178] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1179] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1180] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1181] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1182] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1183] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1184] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1185] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users (local government officials) as follows:

[1186] Data collection

[1187] First, the server collects necessary disaster risk information, past disaster data, geographic information, and infrastructure information from the databases and APIs of local governments and related organizations, including population distribution data, national geographic information, and infrastructure company operation data.

[1188] Examples:

[1189] The server collects past earthquake occurrence history for each region from the database of local government A and imports the latest geographical information and supply infrastructure information from other related organizations.

[1190] Data Preprocessing

[1191] The server cleanses and integrates the collected data. Cleansing involves filling in missing data and processing outliers. Integration involves storing data in different formats in a unified database.

[1192] Risk Analysis

[1193] The server then inputs the organized data into an AI model to analyze disaster risk, which then predicts the disaster risk, impact area, and damage for each region.

[1194] Examples:

[1195] The server analyzes the probability of an earthquake occurring in Area B, predicts the epicenter and seismic intensity, and combines this data with population distribution data to identify the areas that will be most affected.

[1196] Creating a risk report

[1197] The server then generates a risk report based on the analysis results, which includes a visualization of disaster risk, identification of the most affected areas, and an assessment of their impact. The risk report is available in PDF and online dashboard format.

[1198] Examples:

[1199] The server creates an earthquake risk report for Area B, showing areas likely to be affected on a map. The report is generated as a PDF and sent to local government officials.

[1200] Disaster prevention plan proposal

[1201] The server generates a disaster response plan based on the risk report, including the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities.

[1202] Examples:

[1203] The server generates a list of suitable evacuation sites in Area B, calculates and provides a table of supplies needed for each site, and assesses the capacity of nearby major medical facilities to determine whether additional medical assistance is needed.

[1204] Building a training plan

[1205] A terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan, which includes hypothetical scenarios and a schedule of practical training exercises.

[1206] Examples:

[1207] The local government's terminal creates an evacuation drill plan assuming an earthquake occurs in Area B. The plan includes confirmation of evacuation routes and a simulation of opening evacuation shelters.

[1208] Status Monitoring

[1209] The server monitors in real time the status of training and disaster response measures, detecting progress and problems and making adjustments as necessary.

[1210] Examples:

[1211] The server monitors the progress of evacuation drills being conducted in Area B in real time and immediately notifies local government officials if any problems arise. For example, if a particular evacuation route is congested, the server will use that information to suggest an alternative route.

[1212] Emergency response

[1213] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance. The system will support the smooth implementation of planned responses, such as setting up evacuation shelters, distributing relief supplies, and providing medical care.

[1214] Examples:

[1215] If an earthquake actually occurs in Area B, local government officials will follow the instructions from the server to quickly open evacuation shelters and guide residents there, as well as deliver necessary relief supplies to the appropriate locations and deploy medical teams.

[1216] Post-recovery feedback and improvements

[1217] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. Based on the results, it makes suggestions for improvements to be made in the next disaster response.

[1218] Examples:

[1219] After the disaster has subsided, the server will collect data on shelter operations and relief supply distribution in Area B, analyze which aspects were effective and which areas need improvement, and create a report based on the results to propose even more effective response measures for the next disaster.

[1220] In this way, the server, terminal, and user work together to realize a system that provides effective disaster risk analysis and countermeasures.

[1221] The processing flow will be explained below.

[1222] Step 1: Collect data

[1223] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[1224] Specific behavior:

[1225] Obtain population distribution data for residents from the database of Municipality A.

[1226] Download the latest topographic data from the Geospatial Information Authority of Japan.

[1227] It collects information on power supply status and water supply operation sent from infrastructure companies.

[1228] Step 2: Preprocessing the data

[1229] The server cleanses and consolidates the collected data: first, it completes missing data, then it detects outliers and handles them appropriately.

[1230] Specific behavior:

[1231] Complement missing data with historical statistical data.

[1232] Detect outliers (e.g., extremely high population growth rates) and identify their causes.

[1233] Convert data in different formats into a unified format (e.g., CSV format).

[1234] Step 3: Risk analysis

[1235] The server inputs the preprocessed data into an AI model to analyze the disaster risk in each region.

[1236] Specific behavior:

[1237] Geographical information and past earthquake data are input into the earthquake prediction model to calculate the probability of an earthquake occurring in each region.

[1238] River and precipitation data are input into the flood risk model to identify vulnerable areas.

[1239] Step 4: Create a risk report

[1240] The server creates a visually easy-to-understand risk report based on the results of the risk analysis.

[1241] Specific behavior:

[1242] The probability of an earthquake occurring and predicted seismic intensity are plotted on a map to show the area of ​​impact.

[1243] Areas at high risk of flooding are color-coded.

[1244] Generate reports as PDFs or online dashboards and send them to city officials.

[1245] Step 5: Propose a disaster response plan

[1246] The server automatically generates a disaster recovery plan based on the risk report.

[1247] Specific behavior:

[1248] Create a list of shelter locations and needed supplies.

[1249] Establish a schedule for relief operations and develop a resource allocation plan based on that schedule.

[1250] Assess the capacity of medical facilities and determine the number of medical support teams needed.

[1251] Step 6: Develop a training plan

[1252] The terminal will then implement a training plan based on the proposed disaster response plan.

[1253] Specific behavior:

[1254] Create hypothetical scenarios and plan simulation training.

[1255] Schedule the training and notify participants.

[1256] Schedule on-the-job training and prepare necessary supplies and equipment.

[1257] Step 7: Status monitoring

[1258] The server monitors the progress and implementation status of training in real time.

[1259] Specific behavior:

[1260] Collect data during training and display progress on a dashboard.

[1261] When a problem occurs, we will notify you immediately and give you instructions on how to fix it.

[1262] Evaluate each stage in real time and adjust your plan as needed.

[1263] Step 8: Emergency response

[1264] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance.

[1265] Specific behavior:

[1266] Quickly set up evacuation shelters and guide residents there.

[1267] Check the delivery status of relief supplies and arrange for additional supplies if necessary.

[1268] Medical teams will be deployed to provide emergency response.

[1269] Step 9: Post-recovery feedback and improvement

[1270] The server collects data after disaster response and evaluates the effectiveness of countermeasures. Based on the results, it makes suggestions for improvements to be made in future disaster responses.

[1271] Specific behavior:

[1272] Collect and analyze evacuation center operation data and material distribution status.

[1273] We will extract good points and problems in the response and create an evaluation report.

[1274] A report summarizing areas for improvement will be provided to local government officials as feedback for the next event.

[1275] Example 1

[1276] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1277] In recent years, the frequency and scale of disasters have been increasing, making it extremely important to assess disaster risks and prepare appropriate countermeasures in advance. However, in conventional systems, data collection, risk analysis, and countermeasure proposals are dispersed, preventing efficient and integrated countermeasures. In particular, data preprocessing and real-time situation monitoring are lacking, making it difficult to implement rapid and accurate disaster countermeasures.

[1278] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1279] In this invention, the server includes a means for importing disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs, a means for cleansing and integrating the collected data, completing missing data, and processing outliers, and a means for inputting the preprocessed data into a generative AI model to analyze disaster risks. This makes it possible to assess disaster risks in advance and prepare appropriate countermeasures quickly and accurately.

[1280] "Disaster risk information" refers to data and information used to assess the likelihood of a disaster occurring and its impact. Specifically, this includes the probability of occurrence and damage forecasts for natural disasters such as earthquakes, typhoons, tsunamis, floods, and volcanic eruptions.

[1281] "Past disaster data" refers to detailed records of past disasters, including information on the date, location, scale, damage, and response measures.

[1282] "Geographic information" is data that indicates the geographical characteristics of a specific area, including information on topography, elevation, water systems, land use, infrastructure layout, etc.

[1283] "Infrastructure information" refers to data on infrastructure such as electricity, gas, water, communications, and transportation, including the location, operating status, and supply capacity of each piece of infrastructure.

[1284] A "database" is a system for systematically collecting, storing, and managing specific information. It is used to integrate information from multiple different data sources.

[1285] "API" stands for Application Programming Interface, a set of protocols and tools that allow data to be exchanged between different pieces of software.

[1286] "Cleansing" is the process of correcting or removing missing or outlier values ​​to improve data quality.

[1287] "Integration" is the process of bringing together data from different formats into one standardized format.

[1288] A "generative AI model" is an artificial intelligence model that learns from collected data and performs risk analysis and predictions.

[1289] A "risk report" is a report summarizing the results of a risk analysis, including predictions of the probability of a disaster occurring and the extent of its impact.

[1290] A "disaster response plan" is a specific action plan for minimizing damage in the event of a disaster. It includes the establishment of evacuation shelters, preparation of supplies, medical response, etc.

[1291] A "training plan" is a plan that defines the specific schedule and procedures for training to be conducted based on a disaster response plan.

[1292] "Monitoring" is the process of monitoring the response situation during training and disasters in real time and making adjustments as necessary.

[1293] "Rapid response" refers to immediately taking appropriate countermeasures when a disaster occurs.

[1294] "Evaluation" is the process of analyzing the results of disaster response, measuring effectiveness, and identifying areas for improvement.

[1295] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users as follows:

[1296] Data collection

[1297] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from databases and APIs of local governments and related organizations. At this stage, API calls and database connections are made using a high-performance server machine and Python. For example, the server collects resident population distribution data, national geographic information, and infrastructure company operation data.

[1298] Data Preprocessing

[1299] The server cleanses and integrates the collected data. It uses the Pandas library and NumPy to fill in missing data and process outliers. It also stores data in a unified format in the database. This process produces reliable data.

[1300] Risk Analysis

[1301] The server inputs the preprocessed data into a generative AI model to analyze disaster risk. The AI ​​model used is trained using TensorFlow or PyTorch, enabling highly accurate predictions. For example, the server analyzes the probability of an earthquake occurring in a specific area and identifies areas likely to be affected.

[1302] Creating a risk report

[1303] The server creates a risk report based on the results of the risk analysis. The risk report includes visualization of disaster risks, identification of the most affected areas, and their assessment. This report is generated in PDF format and converted using tools such as Adobe Acrobat. For example, the server displays the affected areas on a map, generates a PDF report, and sends it to local government officials.

[1304] Disaster prevention plan proposal

[1305] The server automatically generates a disaster response plan based on the risk report. This plan includes the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities. For example, the server generates a list of appropriate evacuation shelters and calculates the amount of supplies needed for each shelter. It also assesses the capacity of nearby medical facilities and determines whether additional medical assistance is needed.

[1306] Building a training plan

[1307] The terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan. The training plan includes hypothetical scenarios and a schedule for practical training. As a specific example, the terminal creates an evacuation training plan in an Excel file and notifies each department.

[1308] Status Monitoring

[1309] The server monitors in real time the status of training and the implementation of countermeasures in the event of a disaster. Zabbix and Nagios are used for this monitoring, detecting progress and problems and making adjustments as necessary. For example, the server monitors the progress of training in real time and immediately notifies local government officials if a problem occurs.

[1310] Emergency response

[1311] When a disaster occurs, users (local government officials) can quickly respond based on the prepared response plan. For example, users can use mobile devices and PCs to set up evacuation shelters, distribute relief supplies, and provide medical care.

[1312] Post-recovery feedback and improvements

[1313] The server collects data after disaster response and evaluates the effectiveness of the response. During this process, it reanalyzes information from the database and proposes future improvements. For example, the server analyzes data on evacuation center operations and relief supply distribution to identify areas for improvement.

[1314] In this way, the system provides effective disaster risk analysis and countermeasures by having the server, terminals, and users work together.

[1315] Example prompt for a generative AI model:

[1316] "Please explain in detail the procedures for collecting and preprocessing geographic information, past disaster data, infrastructure information, and population distribution data as input data for the AI ​​model to analyze disaster risk."

[1317] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1318] Step 1: Data collection

[1319] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from databases and APIs of local governments and related organizations. Specifically, it uses the Python requests library to obtain data from APIs and connects to various databases to import information. Input data includes resident population distribution data, national geographic information, and infrastructure operation status data, and is stored on the server. Output data is the collected raw data in various formats.

[1320] Specific operation: When the server retrieves data from the API, it uses the requests.get() method to retrieve the required data and parses it in JSON format.

[1321] Step 2: Data Preprocessing

[1322] The server cleanses and integrates the collected data. Specifically, it uses the Pandas library and NumPy to impute incomplete data, process outliers, and standardize data in different formats. The input data is the collected raw data, and the output data is the cleansed, integrated, and reliable data.

[1323] Specific operation: The server uses the Pandas library to fill in missing values ​​using the fillna() method, etc., and masks outliers.

[1324] Step 3: Risk analysis

[1325] The server inputs the preprocessed data into a generative AI model to analyze disaster risk. The AI ​​models used include TensorFlow and PyTorch, which utilize advanced machine learning algorithms. The input data is preprocessed, reliable data, and the output data is risk assessment results and prediction data.

[1326] Specific operation: The server uses TensorFlow's predict method to input the preprocessed data into the model and obtain prediction results.

[1327] Step 4: Create a risk report

[1328] The server creates the actual risk report based on the results of the risk analysis. It uses Matplotlib and Seaborn libraries to generate graphs and maps and create a PDF version of the risk report. The input data are the risk assessment results and prediction data, and the output data is the generated risk report.

[1329] What it does: The server uses Matplotlib to plot the risk assessment results and then saves the generated graph to a PDF.

[1330] Step 5: Propose a disaster response plan

[1331] The server automatically generates a disaster response plan based on the risk report. This plan includes the location of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities. The input data is the risk report, and the output data is the disaster response plan.

[1332] Specific operation: The server analyzes the risk report data and calculates the evacuation shelter list and the amount of necessary supplies.

[1333] Step 6: Develop a training plan

[1334] The terminal (e.g., the local government's computer system) creates a training plan based on the proposed disaster response plan. The training schedule is created in an Excel file and notified to each department. The input data is the disaster response plan, and the output data is the training plan.

[1335] Specific operation: The terminal uses the openpyxl library to write and save the training schedule in an Excel file.

[1336] Step 7: Status monitoring

[1337] The server monitors in real time the status of training and the implementation of countermeasures in the event of a disaster. Monitoring is performed using Zabbix API and Nagios, and if an abnormality is detected, it immediately notifies. The input data is training and disaster response data collected in real time, and the output data is the monitoring results and alerts.

[1338] Specific operation: The server uses the Zabbix API to obtain real-time data and issues an alert if an abnormality is detected.

[1339] Step 8: Emergency response

[1340] When a disaster occurs, users (local government officials) quickly begin responding based on a pre-proposed response plan. They use mobile devices and PCs to set up evacuation shelters, distribute relief supplies, and provide medical care. The input data is the disaster response plan and real-time situation data, and the output data is the actual response history.

[1341] Specific actions: The user follows the plan, opens designated evacuation centers, and distributes necessary supplies.

[1342] Step 9: Post-recovery feedback and improvement

[1343] The server collects data after a disaster response and evaluates the effectiveness of the response. Based on this evaluation data, it makes suggestions for improving future disaster responses. The input data is various data from after the disaster response, and the output data is suggestions for improving future disaster responses.

[1344] Specific operation: The server evaluates post-disaster response data, analyzes which parts were effective, and identifies areas for improvement.

[1345] (Application example 1)

[1346] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1347] Conventional disaster risk management systems lack sufficient real-time risk monitoring, immediate response when anomalies are detected, and proposals for advance countermeasure plans. This has led to problems such as reduced efficiency in systems for quickly responding to disasters and evaluating the effectiveness of countermeasures. Furthermore, there is a lack of a way to present disaster risk analysis results in a visually understandable manner.

[1348] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1349] In this invention, the server includes: means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information; means for analyzing disaster risk based on the collected data; means for creating and proposing a risk report based on the analysis results; means for automatically generating a disaster response plan; means for constructing a training plan based on the generated disaster response plan; means for monitoring the implementation status and effectiveness of training in real time and making adjustments as necessary; means for promptly initiating a response based on a previously proposed response plan when a disaster occurs; means for collecting post-disaster response data, evaluating the effectiveness of measures, and proposing improvements for the next disaster response; means for monitoring disaster risk information in real time and implementing emergency measures when an abnormality is detected; means for predicting disaster risk information based on an AI model and proposing a pre-disaster response plan; and means for graphing and displaying disaster risk analysis results and generating a report in PDF format. This enables real-time disaster risk monitoring, rapid emergency response in emergencies, and the provision of visually easy-to-understand risk analysis results.

[1350] "Disaster risk information" is data that indicates possible risks related to disasters, and includes information on natural disasters such as earthquakes, tsunamis, and typhoons.

[1351] "Past disaster data" refers to data that records the history of past disasters, the extent of their impact, and the extent of the damage.

[1352] "Geographic information" refers to data relating to the topography and geographical characteristics of a particular area, including map information and topographical maps.

[1353] "Infrastructure information" refers to data on public services and facilities such as electricity, gas, water, and transportation.

[1354] "Means for analyzing disaster risk" are methods and tools for assessing the likelihood of occurrence and the scope of impact based on collected disaster risk information.

[1355] "Means for preparing and proposing risk reports" refers to a method of preparing the results of disaster risk analysis in report format and proposing countermeasures based on that information.

[1356] "Means for automatically generating disaster prevention plans" refers to a method by which the system automatically plans and proposes optimal countermeasures based on the results of disaster risk analysis.

[1357] The "means for constructing a training plan" refers to a method for setting up practical training or scenario-based training based on the generated disaster response plan.

[1358] "Real-time monitoring and adjustment measures" refer to methods for immediately monitoring the progress of training exercises and actual disaster responses and making necessary adjustments.

[1359] "Measures to initiate a rapid response" are methods for quickly putting into action measures that have been planned in advance when a disaster occurs.

[1360] "Means for collecting data and evaluating effectiveness" refers to methods for collecting data after disaster response and evaluating the effectiveness of the measures taken.

[1361] "Means for monitoring disaster risk information in real time" refers to a method for continuously monitoring disaster risk information in real time.

[1362] The "means of implementing emergency measures when an abnormality is detected" refers to a method for quickly implementing an emergency response when an abnormality is detected in disaster risk information.

[1363] "Means for making predictions based on AI models and proposing advance countermeasure plans" refers to a method for predicting disaster risks using artificial intelligence models and proposing effective countermeasure plans in advance based on the results.

[1364] "Means for displaying graphs based on risk analysis results and generating reports in PDF format" refers to a method for visually displaying the results of disaster risk analysis and creating reports in PDF format.

[1365] MODE FOR CARRYING OUT THE INVENTION

[1366] This invention is a factory disaster risk management system that collects disaster risk information, past disaster data, geographical information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users (local government officials).

[1367] Data collection

[1368] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs, including population distribution data, geographic information, and infrastructure company operation data.

[1369] Example: The server collects historical earthquake occurrence history for each region from the local government database and imports up-to-date geographical and supply infrastructure information from other relevant agencies.

[1370] Data Preprocessing

[1371] The server cleanses and integrates the collected data. Cleansing involves filling in missing data and processing outliers. Integration involves storing data in different formats in a unified database.

[1372] Risk Analysis

[1373] The server inputs the organized data into an AI model to analyze disaster risk. This allows for predictions of disaster risk, impact area, and damage for each region. The server performs the analysis using TensorFlow and Scikit-learn.

[1374] Example: The server analyzes the probability of an earthquake occurring in a specific area, predicts the epicenter and intensity, and combines this with population distribution data to identify the areas most affected.

[1375] Creating a risk report

[1376] The server generates a risk report based on the analysis results. This report includes visualization of disaster risks, identification of the most affected areas, and assessment of their impact. The risk report is provided in PDF and online dashboard format. Matplotlib and ReportLab are used for visualization and report generation.

[1377] Example: The server creates an earthquake risk report for a specific area, showing areas likely to be affected on a map. The report is generated as a PDF and sent to local government officials.

[1378] Disaster prevention plan proposal

[1379] The server generates a disaster response plan based on the risk report, including the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities.

[1380] Example: A server generates a list of suitable evacuation sites in a particular area, calculates and provides a list of supplies needed for each site, and assesses the capacity of nearby major medical facilities to determine if additional medical assistance is needed.

[1381] Building a training plan

[1382] A terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan, which includes hypothetical scenarios and a schedule of practical training exercises.

[1383] Example: A local government device creates an evacuation drill plan for a specific area in the event of an earthquake, including confirmation of evacuation routes and a simulation of the establishment of evacuation shelters.

[1384] Status Monitoring

[1385] The server monitors in real time the status of training and disaster response measures, detecting progress and problems and making adjustments as necessary.

[1386] Example: The server monitors the progress of evacuation drills in a specific area in real time and immediately notifies local government officials if a problem occurs. For example, if a specific evacuation route is congested, it will use that information to suggest an alternative route.

[1387] Emergency response

[1388] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance. The system will support the smooth implementation of planned responses, such as setting up evacuation shelters, distributing relief supplies, and providing medical care.

[1389] Example: If an earthquake actually occurs in a specific area, local government officials will follow instructions from the server to quickly set up evacuation shelters and guide residents, as well as deliver necessary relief supplies to the appropriate locations and deploy medical teams.

[1390] Post-recovery feedback and improvements

[1391] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. Based on the results, it makes suggestions for improvements to be made in the next disaster response.

[1392] Example: After a disaster has subsided, the server collects data on the operation of evacuation centers and the distribution of relief supplies in a specific area, analyzes which aspects were effective and which areas need improvement, and creates a report based on the results to propose more effective response measures for the next disaster.

[1393] Prompt Sentence Examples

[1394] "Analyze the latest earthquake risks in the designated area and take safety measures."

[1395] "How can we collect, analyze, and display real-time disaster risk information on a dashboard?"

[1396] In this way, the server, terminal, and user work together to realize a system that provides effective disaster risk analysis and countermeasures.

[1397] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1398] MODE FOR CARRYING OUT THE INVENTION (PROGRAM PROCESSING STEPS)

[1399] Step 1:

[1400] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs. The server imports resident population distribution data, geographic information, and infrastructure company operation data. The input is data from APIs and databases, and the output is a dataset that integrates this information. Specifically, the data is stored in a data frame using Python's pandas library.

[1401] Step 2:

[1402] The server cleanses and integrates the collected data. It uses the collected raw data as input and performs tasks such as filling in missing values ​​and removing outliers. The output is a cleansed dataset stored in a unified format. Specifically, it uses pandas functions to interpolate missing values ​​and remove outliers.

[1403] Step 3:

[1404] The server inputs the organized data into an AI model to analyze disaster risk. The input is a cleansed dataset, and the output is a disaster risk prediction. Specifically, TensorFlow and Scikit-learn are used to input data into the model and make predictions. For example, the probability of an earthquake occurring and the extent of its impact can be predicted.

[1405] Step 4:

[1406] The server creates a risk report based on the analysis results. The input is the prediction results of the AI ​​model, and the output is a risk report (in PDF or online dashboard format). Specifically, Matplotlib is used to create graphs, and ReportLab is used to generate the PDF report. High-risk areas and key countermeasures are visually displayed.

[1407] Step 5:

[1408] The server generates a disaster response plan based on the risk report. The input is the risk report, and the output is a specific disaster response plan. Specifically, it calculates the locations of evacuation shelters and the amount of supplies needed, and compiles them in list form.

[1409] Step 6:

[1410] The terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan. The input is the disaster response plan, and the output is the training plan. Specific operations include setting the date and scenario for the evacuation training.

[1411] Step 7:

[1412] The server monitors the status and effectiveness of training in real time and makes adjustments as necessary. The input is training progress data, and the output is monitoring results and adjustments. Specifically, it collects sensor data and GPS information in real time and displays it on a dashboard.

[1413] Step 8:

[1414] When a disaster occurs, the user (local government official) will quickly begin responding based on a countermeasure plan proposed in advance. The input is the countermeasure plan, and the output is the status of the countermeasures that have been implemented. Specific actions include setting up evacuation shelters and distributing supplies.

[1415] Step 9:

[1416] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The input is disaster response data, and the output is an effectiveness evaluation report. Specifically, the server analyzes the effectiveness of the countermeasures based on the collected data and proposes improvements for the next disaster response.

[1417] Prompt Sentence Examples

[1418] "Analyze the latest earthquake risks in the designated area and take safety measures."

[1419] "How can we collect, analyze, and display real-time disaster risk information on a dashboard?"

[1420] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1421] This invention combines a system that collects disaster risk information, past disaster data, geographical information, and infrastructure information, analyzes disaster risks, and proposes optimal disaster countermeasures with an emotion engine that recognizes the user's emotions. This system operates in cooperation with a server, terminal, and user (local government official) as follows:

[1422] Data collection

[1423] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations, including resident demographic distribution data, up-to-date geographical information, and supply infrastructure information.

[1424] Examples:

[1425] The server collects past disaster occurrence history for each region from the database of X municipality and imports the latest geographical information and lifeline information from other related organizations.

[1426] Data Preprocessing

[1427] The server cleanses and consolidates the collected data, imputes missing data, detects outliers and handles them appropriately, and then converts data from different formats into a unified format.

[1428] Risk Analysis

[1429] The server inputs the preprocessed data into an AI model to analyze the disaster risk for each region, thereby predicting the disaster risk, impact area, and damage for each region.

[1430] Examples:

[1431] The server analyzes the probability of an earthquake occurring in region X, predicts the epicenter and seismic intensity, and combines this with population distribution data to identify the areas that will be most affected.

[1432] Creating a risk report

[1433] The server generates a risk report based on the results of the risk analysis. This report visualizes disaster risks, identifies vulnerable areas, and assesses their impact. The risk report is provided in PDF and online dashboard format.

[1434] Examples:

[1435] The server creates an earthquake risk report for region X, showing areas likely to be affected on a map, and generates a PDF report that is sent to local government officials.

[1436] Introducing the Emotion Engine

[1437] Implement an emotion engine to assess the psychological impact of disaster response plans and help adjust training plans and implementation.

[1438] Examples:

[1439] The server monitors the stress levels and emotional state of users (local government officials) during the training and adjusts the training content as necessary. For example, if many of the training participants are feeling high levels of stress, it will temporarily lower the difficulty of the training.

[1440] Disaster prevention plan proposal

[1441] The server generates disaster response plans based on the risk reports, including the locations of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities, while also taking into account the psychological impact of the plans through an emotion engine.

[1442] Examples:

[1443] The server generates a list of suitable evacuation sites in area X, calculates the amount of supplies needed for each site, and uses an emotion engine to suggest the best psychological placement of the evacuation sites.

[1444] Building a training plan

[1445] The device creates a training plan based on the proposed disaster response plan, creates a virtual scenario and a schedule for practical training, and adjusts the training content based on the user's emotional state during the training using an emotion engine.

[1446] Examples:

[1447] The local government's terminal creates an evacuation drill plan assuming an earthquake occurs in region X. The plan includes confirmation of evacuation routes and a simulation of opening evacuation shelters. An emotion engine is used to detect users' stress and anxiety that arise during the drill in real time and adjust the drill content accordingly.

[1448] Status Monitoring

[1449] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[1450] Examples:

[1451] The server monitors the progress of evacuation drills being conducted in Region X in real time, and immediately notifies local government officials if any problems arise. In addition, if officials feel extremely stressed during the drill, the server temporarily halts the drill and allows them time to refresh.

[1452] Emergency response

[1453] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine takes into account the user's emotional state and helps implement the optimal response.

[1454] Examples:

[1455] If an earthquake actually occurs in region X, local government officials will follow the server's instructions to quickly open evacuation shelters and guide residents. The emotion engine will monitor the officials' emotional state and provide appropriate support if overwork or stress is detected.

[1456] Post-recovery feedback and improvements

[1457] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The emotion engine also takes into account psychological impacts and proposes improvements for the next disaster response.

[1458] Examples:

[1459] After the disaster subsides, the server collects data on shelter operations and relief supply distribution in region X, analyzing which aspects were effective and where there is room for improvement. It also analyzes the psychological data of staff members during the operation and evaluates areas where stress and anxiety became a problem. Based on the results, a report is created proposing optimal psychological measures for the next disaster response.

[1460] In this way, by having the server, terminal, and user work together, it is possible to realize a system that provides effective disaster risk analysis and countermeasures that also take into account the user's emotions.

[1461] The processing flow will be explained below.

[1462] Step 1: Collect data

[1463] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[1464] Specific behavior:

[1465] Obtain the history of past disasters by region from the database of X municipality.

[1466] Download the latest topographic data from the Geospatial Information Authority of Japan.

[1467] It collects information on power supply status and water supply operation sent from infrastructure companies.

[1468] Step 2: Preprocessing the data

[1469] The server cleanses and integrates the collected data, supplements missing data with past statistical data, detects and appropriately processes outliers (e.g., extremely high population growth rates), and converts data in different formats into a unified format (e.g., CSV format).

[1470] Specific behavior:

[1471] Complement missing earthquake occurrence data from the database.

[1472] Check for abnormal values ​​and analyze the cause if necessary.

[1473] Integrate topographical and population data in one unified format.

[1474] Step 3: Risk analysis

[1475] The server inputs the preprocessed data into an AI model to analyze the disaster risk in each region.

[1476] Specific behavior:

[1477] Geographical information and past earthquake data are input into the earthquake prediction model to calculate the probability of an earthquake occurring in each region.

[1478] River and precipitation data are input into the flood risk model to identify vulnerable areas.

[1479] Step 4: Create a risk report

[1480] The server generates a visually easy-to-understand risk report based on the results of the risk analysis, which is available in PDF format and an online dashboard.

[1481] Specific behavior:

[1482] The probability of an earthquake occurring and predicted seismic intensity are plotted on a map to show the area of ​​impact.

[1483] Areas at high risk of flooding are color-coded.

[1484] Generate a report in PDF format and email it to city officials.

[1485] Step 5: Implementing the Emotion Engine

[1486] Using an emotion engine, the psychological impact of disaster response plans is assessed and reflected in training plans and implementation.

[1487] Specific behavior:

[1488] The server monitors the stress level and emotional state of the user (local government official) during the training.

[1489] The server analyzes the emotion engine data in real time and adjusts the training content accordingly, for example, suggesting that the training be made easier if stress levels are high.

[1490] Step 6: Propose a disaster response plan

[1491] The server generates disaster response plans based on the risk reports, including the locations of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities. An emotion engine also takes into account the psychological impact of the plans.

[1492] Specific behavior:

[1493] Create a list of shelter locations and needed supplies.

[1494] Establish a schedule for relief operations and develop a resource allocation plan based on that schedule.

[1495] Assess the capacity of medical facilities and determine the number of medical support teams needed.

[1496] Using an emotion engine, we propose shelter layouts that have minimal psychological impact.

[1497] Step 7: Develop a training plan

[1498] The device then creates a training plan based on the proposed disaster response plan, creating virtual scenarios and practical training schedules, and using data from the emotion engine to monitor the user's emotional state and adjust the training content.

[1499] Specific behavior:

[1500] Create hypothetical scenarios and plan simulation training.

[1501] Schedule the training and notify participants.

[1502] Schedule on-the-job training and prepare necessary supplies and equipment.

[1503] The emotion engine monitors the user's emotional state, detects stress or anxiety during training, and adjusts the training progress.

[1504] Step 8: Status monitoring

[1505] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[1506] Specific behavior:

[1507] Collect data during training and display progress on a dashboard.

[1508] When a problem occurs, we will notify you immediately and give you instructions on how to fix it.

[1509] Evaluate each stage in real time and adjust your plan as needed.

[1510] An emotion engine monitors the user's emotional state and takes appropriate measures if excessive stress is detected.

[1511] Step 9: Emergency response

[1512] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine takes into account the user's emotional state and helps implement the optimal response.

[1513] Specific behavior:

[1514] Quickly set up evacuation shelters and guide residents there.

[1515] Check the delivery status of relief supplies and make additional arrangements as necessary.

[1516] Medical teams will be deployed to provide emergency response.

[1517] An emotion engine monitors the user's emotional state and provides supportive measures if overwork or stress is detected.

[1518] Step 10: Post-recovery feedback and improvement

[1519] The server collects data after disaster response and evaluates the effectiveness of countermeasures. It also takes into account psychological impacts using an emotion engine and proposes improvements for future disaster response.

[1520] Specific behavior:

[1521] Collect and analyze evacuation center operation data and material distribution status.

[1522] We analyze users' emotional data during disaster response and evaluate the areas where stress and anxiety became an issue.

[1523] We will extract good points and problems in the response and create an evaluation report.

[1524] A report summarizing areas for improvement will be provided to local government officials as feedback for the next event.

[1525] Example 2

[1526] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1527] In recent years, the frequency and scale of disasters have increased, creating a need for rapid and accurate disaster prevention measures. However, current systems are limited to collecting and analyzing disaster risk information and generating disaster prevention plans, and do not adequately consider user emotions in their training and responses. Furthermore, cleansing and integrating collected data and real-time monitoring often require manual intervention, resulting in reduced efficiency. In this situation, there is a need for a system that can achieve more effective and efficient disaster prevention measures by taking users' psychological states into account.

[1528] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1529] In this invention, the server includes means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information, means for cleansing and integrating the collected data, means for analyzing disaster risk using preprocessed data, means for creating and proposing risk reports, means for recognizing a user's emotional state using an emotion engine, means for automatically generating a disaster response plan and optimizing it taking into account the user's emotional state, means for building a training plan based on the proposed disaster response plan, means for monitoring the implementation status and effectiveness of the training in real time and making adjustments as necessary, means for promptly initiating a response based on the previously proposed response plan when a disaster occurs and providing support taking into account the user's emotional state, and means for collecting post-disaster response data, evaluating the effectiveness of the response measures, and proposing improvements for the next disaster response. This enables highly accurate analysis of disaster risk, reduces the user's psychological burden, and enables more effective and practical disaster response.

[1530] "Disaster risk information" is information that indicates the probability of a disaster occurring and the extent of its impact, and is data used to assess future disaster risks.

[1531] "Past disaster data" refers to detailed information about disasters that have occurred in the past, including the type of disaster, the location of the disaster, and the extent of the damage.

[1532] "Geographic information" is information that describes the geographical characteristics of a particular area, such as topography, land use, and infrastructure layout.

[1533] "Infrastructure information" refers to information related to the basic infrastructure of daily life, such as transportation networks, water, electricity, and gas, and is data necessary for maintaining urban functions.

[1534] "Data cleansing" is the process of removing errors and inconsistencies from collected data to improve the quality of the data.

[1535] "Data integration" is the process of combining multiple data sets collected from different sources in a consistent format to create uniformly usable data.

[1536] "Disaster risk analysis" is the process of using AI models and algorithms to assess the probability of a disaster occurring and the extent of its impact based on collected data.

[1537] A "risk report" is a report summarizing the results of disaster risk analysis, including visualized data and specific risk assessments.

[1538] The "emotion engine" is a system that analyzes a user's emotional state from voice and text data and adjusts training and countermeasures based on that information.

[1539] A "disaster preparedness plan" is a plan that lays out in advance the specific actions and procedures to be taken in the event of a disaster, including the locations of evacuation shelters and plans for the distribution of relief supplies.

[1540] A "training plan" is a plan that defines the detailed schedule and content of training to be conducted based on the proposed disaster response plan.

[1541] "Real-time monitoring" is the process by which systems constantly monitor the situation during drills and disaster response, instantly acquiring and analyzing data and making adjustments as needed.

[1542] "Starting a rapid response" means implementing necessary measures without delay based on a disaster response plan that was prepared in advance when a disaster occurs.

[1543] "Post-disaster response data" refers to data related to responses at the time of a disaster and afterward, and is used to evaluate the effectiveness of responses and areas for improvement.

[1544] "Evaluating effectiveness and proposing improvements" is the process of analyzing the effectiveness of current measures based on collected data and identifying specific areas for improvement in the next disaster response.

[1545] MODE FOR CARRYING OUT THE INVENTION

[1546] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, and combines AI technology with an emotion engine. This system is designed to monitor the user's emotional state in real time and maximize the effectiveness of disaster countermeasures. Below, we explain how this system's program works.

[1547] 1. Data Collection

[1548] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations, including population distribution data, up-to-date geographical information, and supply infrastructure information.

[1549] Hardware / software used: Data collection using RESTful API, Python libraries (requests, pandas)

[1550] Example: The server sends a request to the local government's database API to obtain the history of past disasters by region. It also obtains the latest geographical and lifeline information from other related organizations.

[1551] 2. Data Preprocessing

[1552] The server cleanses and consolidates the collected data, imputes missing data, detects outliers and handles them appropriately, and then converts data from different formats into a unified format.

[1553] Hardware / software used: Python pandas library, database management system (e.g., MySQL)

[1554] Example: The server reads the acquired data in data frame format, imputes missing values ​​with the mean value, filters outliers, and saves the data in a unified format in the database.

[1555] 3. Risk Analysis

[1556] The server inputs the preprocessed data into a generative AI model to analyze disaster risk for each region.

[1557] Hardware / software used: AI model (TensorFlow or PyTorch), data extraction from database

[1558] Example: The server inputs regional earthquake occurrence data into an AI model to predict the epicenter and intensity, and combines this with population distribution data to identify the areas most affected.

[1559] 4. Creating a risk report

[1560] The server generates a risk report based on the results of the risk analysis, which includes visualization of disaster risks, identification of vulnerable areas, and assessment of impacts.

[1561] Hardware / software used: Data visualization tools (Matplotlib, D3.js), report generation tool (PDFKit)

[1562] Example: The server visualizes the risk level on a map based on the analysis results, generates a risk report in PDF format, and provides it to the user.

[1563] 5. Introducing the Emotion Engine

[1564] The server uses an emotion engine to monitor the user's emotional state during training and execution, with the aim of reducing psychological burden and realizing effective training and disaster response.

[1565] Hardware / software used: Emotion recognition AI model (e.g. OpenCV, NLTK)

[1566] Example: The server collects speech data during training and inputs it into an emotion recognition model to analyze stress and anxiety levels in real time.

[1567] 6. Proposal of disaster prevention plan

[1568] Based on the risk report, the server generates a disaster response plan, which includes the location of evacuation shelters, the amount of relief supplies needed, and the capacity assessment of medical facilities, and also takes into account psychological impacts using an emotion engine.

[1569] Hardware / software used: Data analysis tools (e.g., SciPy), AI models

[1570] Example: The server proposes psychologically optimal shelter layouts based on the user's emotional state.

[1571] 7. Developing a training plan

[1572] The device then creates a training plan based on the proposed disaster response plan, creates a virtual scenario and a training schedule, and adjusts the training content based on the user's emotional state during the training using an emotion engine.

[1573] Hardware / software used: Scheduling tool (calendar application), real-time monitoring system

[1574] Example: The device creates a training schedule based on hypothetical scenarios and uses an emotion engine to monitor stress levels during training.

[1575] 8. Status Monitoring

[1576] The server monitors the implementation status of drills and disaster countermeasures in real time, and also monitors the user's emotional state using an emotion engine, detecting progress and problems and making adjustments.

[1577] Hardware / software used: Real-time data analysis tools, dashboards (Grafana, Kibana)

[1578] Example: The server monitors the progress of training and immediately notifies the user if an abnormality is detected. It also adjusts the training content if it detects excessive stress.

[1579] 9. Emergency Response

[1580] When a disaster occurs, the user (local government official) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine will monitor the user's emotional state and provide the necessary support.

[1581] Hardware / software used: Digital instruction system, emotion recognition system

[1582] Example: The user quickly opens a shelter and guides residents. If the emotion engine detects overwork or high stress, it receives support from the server.

[1583] 10. Post-recovery feedback and improvements

[1584] The server evaluates the effectiveness of countermeasures based on post-disaster response data and makes suggestions for improvements for the next disaster response, taking into account the psychological impact.

[1585] Hardware / software used: Data analysis tools, report generation tools

[1586] Example: The server collects data from disaster response and analyzes it together with emotional data. Based on the results, it formulates a response plan for the next disaster and provides feedback to the user.

[1587] Examples of prompt statements

[1588] For example, the following might be an example of a prompt sentence to input to a generative AI model:

[1589] "Based on disaster data from the past 10 years in the region, we analyzed the current disaster risk and created a risk report."

[1590] "Create a plan for your next evacuation drill and monitor the emotional state of participants during the drill."

[1591] "We propose a response plan in the event of a disaster and provide optimal responses taking into account the stress levels of those in charge."

[1592] In this way, the present invention realizes a system that provides effective disaster risk analysis and countermeasures that take into account the user's emotions, through the cooperation of the server, terminal, and user.

[1593] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1594] Step 1:

[1595] Data collection

[1596] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[1597] Specific behavior:

[1598] Input: Database API endpoint, authentication information

[1599] Processing: The server uses a RESTful API to request various information and retrieves data in JSON or XML format.

[1600] Output: Collected raw data (e.g., past disaster history, latest geographic information, supply infrastructure information)

[1601] Hardware / software used: RESTful API, Python requests library

[1602] Step 2:

[1603] Data Preprocessing

[1604] The server cleanses and consolidates the collected data.

[1605] Specific behavior:

[1606] Input: Raw collected data (JSON, XML)

[1607] Processing: The server uses the Python pandas library to read the data in data frame format, perform missing value imputation (e.g., inserting the mean or mode), detect and remove outliers, convert data from different formats into a unified format (e.g., CSV, JSON), and save it in the database.

[1608] Output: Clean consolidated data

[1609] Hardware / software used: Python pandas library, database management system (e.g., MySQL)

[1610] Step 3:

[1611] Risk Analysis

[1612] The server inputs the preprocessed data into a generative AI model to analyze disaster risk for each region.

[1613] Specific behavior:

[1614] Input: Clean consolidated data

[1615] Processing: The server feeds the data into a generative AI model (e.g., an AI model using TensorFlow or PyTorch) to calculate disaster risk for each region, including risk assessments for earthquakes, floods, fires, etc.

[1616] Output: Disaster risk assessment results (e.g., probability of occurrence, impact area, damage forecast)

[1617] Hardware / software used: TensorFlow, PyTorch

[1618] Step 4:

[1619] Creating a risk report

[1620] The server creates a risk report based on the risk analysis results.

[1621] Specific behavior:

[1622] Input: Disaster risk assessment results

[1623] Processing: The server uses a data visualization tool (e.g., Matplotlib, D3.js) to visualize the risk data in the form of graphs and maps, and then uses a report generation tool (e.g., PDFKit) to generate the results as a PDF report.

[1624] Output: Risk report (PDF format)

[1625] Hardware / software used: Matplotlib, D3.js, PDFKit

[1626] Step 5:

[1627] Introducing the Emotion Engine

[1628] The server monitors the user's emotional state during training and execution.

[1629] Specific behavior:

[1630] Input: User voice and text data

[1631] Processing: The server inputs the data into an emotion recognition AI model to analyze the user's stress level and emotional state.

[1632] Output: User's emotional state (e.g., stress level, anxiety level)

[1633] Hardware / software used: OpenCV, NLTK

[1634] Step 6:

[1635] Disaster prevention plan proposal

[1636] The server generates a disaster recovery plan based on the risk report.

[1637] Specific behavior:

[1638] Input: Risk report, user emotional state

[1639] Processing: The server uses AI models to assess shelter locations, relief supply needs, and medical facility capacity, optimizing them based on emotional state.

[1640] Output: Disaster Preparedness Plan

[1641] Hardware / software used: SciPy, AI models

[1642] Step 7:

[1643] Building a training plan

[1644] The terminal will create a training plan based on the proposed disaster response plan.

[1645] Specific behavior:

[1646] Input: Disaster Preparedness Plan

[1647] Processing: The device creates virtual scenarios and practical training schedules, and uses an emotion engine to monitor and adjust the user's emotional state during training.

[1648] Output: Training plan

[1649] Hardware / software used: Scheduling tools, real-time monitoring systems

[1650] Step 8:

[1651] Status Monitoring

[1652] The server monitors in real time the status of countermeasures being implemented during training and when disasters occur.

[1653] Specific behavior:

[1654] Input: Training status, user's emotional state

[1655] Processing: The server uses real-time data analysis tools and dashboards to monitor progress and issues, and if anomalies are detected, it immediately notifies you and makes the necessary adjustments.

[1656] Output: Training progress, anomaly detection results

[1657] Hardware / software used: Grafana, Kibana

[1658] Step 9:

[1659] Emergency response

[1660] The user (local government official) will promptly begin responding based on the countermeasure plan proposed in advance.

[1661] Specific behavior:

[1662] Input: Real-time information and countermeasure plans in the event of a disaster

[1663] Action: The user takes action using the digital instruction system. The emotion engine monitors the user's emotional state and provides necessary assistance.

[1664] Output: Measures taken, emotional state log

[1665] Hardware / software used: Digital instruction system, emotion recognition system

[1666] Step 10:

[1667] Post-recovery feedback and improvements

[1668] The server evaluates the effectiveness of countermeasures based on data collected after the disaster response and makes suggestions for improvements for the next disaster response.

[1669] Specific behavior:

[1670] Input: Post-disaster response data, emotional state log

[1671] Processing: The server uses the data analysis tool to evaluate the effectiveness of the measures, and creates an improvement proposal report using the report generation tool.

[1672] Output: Improvement Suggestion Report

[1673] Hardware / software used: Data analysis tools, report generation tools

[1674] (Application example 2)

[1675] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1676] Conventional disaster response systems focus on analyzing disaster risks and proposing countermeasures, but lack functionality that takes into account the psychological burden on responding staff. They also lack the ability to visually grasp real-time disaster information or intuitively understand appropriate countermeasures tailored to the current situation. This makes it difficult to respond quickly and effectively in the event of a disaster, potentially resulting in the expansion of damage.

[1677] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting disaster risk information, past disaster data, geographic information, and infrastructure information, means for analyzing disaster risk based on the collected data, and means for creating and proposing a risk report based on the analysis results. This makes it possible to monitor the emotional state of staff and suggest rest if their stress level is high. Furthermore, displaying disaster risk information and countermeasures using augmented reality enables intuitive and rapid response.

[1678] "Disaster risk information" refers to information on natural disasters such as earthquakes, typhoons, and floods, including past occurrence data and forecast data.

[1679] "Past disaster data" refers to information about past disasters, including the extent and impact of damage and the effectiveness of response measures.

[1680] "Geographic information" refers to information about geographical characteristics such as topography, topography, and geology, including land use and building layout.

[1681] "Infrastructure information" refers to information related to social infrastructure such as roads, bridges, power supplies, and water supplies.

[1682] "Means of collection" refers to the technical means used to incorporate the above information into the system using various databases and APIs.

[1683] "Means for analyzing disaster risk" refers to analytical techniques for assessing and predicting disaster risk based on collected data.

[1684] "Risk Report" means a report summarizing the results of a disaster risk analysis, including the scope of impact and damage forecast.

[1685] A "disaster preparedness plan" is a document that outlines specific measures and procedures to be taken in the event of a disaster.

[1686] "Training plan" refers to the specific schedule and content of training conducted based on the disaster response plan.

[1687] "Means of monitoring" refers to technology that monitors the implementation of training and disaster prevention measures in real time.

[1688] "Staff emotional state" refers to the psychological state of logistics center staff, such as stress and anxiety.

[1689] "Emotion monitoring" refers to technology that measures and monitors the emotional state of staff in real time.

[1690] "Measures to suggest rest when stress levels are high" refers to technological measures that encourage appropriate rest when staff stress levels exceed a certain value.

[1691] "Augmented reality display" refers to technology that overlays virtual information on real-world scenes, making it easier to visually understand disaster risks and countermeasures.

[1692] The present invention is implemented as a system for managing disaster risks and monitoring staff emotions in a logistics center. This system operates in cooperation with a server, terminals, and users (logistics center staff).

[1693] Data collection

[1694] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information through various databases and APIs. The collected data includes natural disaster risk around the logistics center, local population density, and the condition of roads and power infrastructure.

[1695] Data Preprocessing

[1696] The server cleanses and consolidates the collected data, converting it into a suitable format for analysis by imputing missing data, detecting outliers, and handling them appropriately.

[1697] Risk Analysis

[1698] The server inputs the preprocessed data into an AI model to analyze disaster risks around the logistics center, thereby evaluating disaster risks such as earthquakes, typhoons, and floods, and predicting the extent of impact and damage.

[1699] Creating a risk report

[1700] The server generates a risk report based on the results of the risk analysis, which includes a visualization of disaster risk and identification of vulnerable areas, and is provided in PDF format or on an online dashboard.

[1701] Introducing the Emotion Engine

[1702] Implement an emotion engine to assess the psychological impact of training and disaster situations on staff, for example by monitoring staff stress levels during training and adjusting training content as needed.

[1703] Disaster prevention plan proposal

[1704] The server generates a disaster response plan based on the risk report, including the location of evacuation sites, the amount of relief supplies needed, and the location of medical facilities. An emotion engine also takes into account the psychological impact of the plan.

[1705] Building a training plan

[1706] The device then creates a training plan based on the generated disaster response plan, including simulations of evacuation routes and the redeployment of relief supplies. It uses an emotion engine to monitor the emotional state of staff in real time and adjusts the training accordingly.

[1707] Augmented reality display

[1708] The server provides a means to display disaster risk information and countermeasures in augmented reality, making it easier for logistics center staff to intuitively understand risks and countermeasures.

[1709] Emergency response

[1710] When a disaster occurs, users (logistics center staff) will quickly begin responding based on a countermeasure plan proposed in advance. The emotion engine takes into account the emotional state of staff members and supports them in taking the optimal response.

[1711] Post-recovery feedback and improvements

[1712] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. The emotion engine takes into account the psychological impact and proposes improvements for the next disaster response. This will ensure that the next disaster response is psychologically optimal.

[1713] This system uses the following hardware and software:

[1714] Hardware: Smart glasses (e.g. Google Glass, Microsoft HoloLens)

[1715] Software: EmotionEngine, ARDisplay (Augmented Reality Display)

[1716] Examples of concrete examples and prompts

[1717] As a concrete example, we present a scenario in which the system is implemented using smart glasses. Logistics center staff wear the smart glasses, and in the event of a disaster, real-time risk information and evacuation routes are visually displayed. If the staff's emotional state is determined to be high stress, appropriate rest is suggested.

[1718] Example prompt sentence:

[1719] "We will implement a system that analyzes current earthquake risks in real time and visualizes the extent of damage and countermeasures. It will also monitor the emotional state of users (logistics center staff) and suggest rest if they are under high stress."

[1720] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1721] Step 1:

[1722] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs. The collected data includes information on natural disaster risk around the logistics center, local population density, and the condition of roads and power infrastructure. Specifically, it calls the API and stores the acquired data in a database. The input is raw data obtained from the API, and the output is an integrated dataset.

[1723] Step 2:

[1724] The server cleanses and consolidates the collected data. Data cleansing involves imputing missing data and detecting and appropriately handling outliers. It then converts data from different formats into a unified format. The input is the raw data collected in step 1, and the output is a clean dataset for analysis.

[1725] Step 3:

[1726] The server inputs the preprocessed data into the AI ​​model and analyzes the disaster risk around the logistics center. The AI ​​model predicts the probability of natural disasters occurring and estimates the extent of impact and damage. The input is the clean dataset created in step 2, and the output is the disaster risk analysis results. Specifically, the data is input into the AI ​​model and the results are obtained.

[1727] Step 4:

[1728] The server creates a risk report based on the results of the risk analysis. The risk report includes visualization of disaster risk and identification of areas susceptible to impact. The risk report is provided in PDF or online dashboard format. The input is the analysis results from Step 3, and the output is the risk report. Specifically, the analysis results are incorporated into a template and a report is generated.

[1729] Step 5:

[1730] The server uses an emotion engine to monitor the emotional state of staff during training or disasters. The emotion engine evaluates the stress level of staff in real time. The input is the physiological data of staff (e.g., heart rate, electrodermal activity), and the output is the stress level evaluation result. Specifically, it collects sensor data, inputs it into the emotion engine, and obtains the analysis results.

[1731] Step 6:

[1732] The server generates a disaster response plan based on the risk report. The plan includes the establishment of evacuation sites, the amount of relief supplies needed, and the location of medical facilities. The emotion engine also considers whether the plan will have a psychological impact. The input is the risk report from Step 4 and the emotion evaluation results from Step 5, and the output is the disaster response plan.

[1733] Step 7:

[1734] The terminal creates a training plan based on the generated disaster response plan. A training scenario is created, including confirmation of evacuation routes and redeployment of relief supplies. An emotion engine is used to monitor the emotional state of staff during the training in real time and adjust the training content accordingly. The input is the disaster response plan from step 6, and the output is the training plan.

[1735] Step 8:

[1736] The server provides a means to display disaster risk information and countermeasures in augmented reality. This allows logistics center staff to visually grasp the risk situation and countermeasures in real time. The input is the analysis results from step 3 and the disaster countermeasure plan from step 6, and the output is an augmented reality display. Specifically, this data is sent to the ARDisplay module and displayed on the AR device.

[1737] Step 9:

[1738] When a disaster occurs, the user (logistics center staff) will quickly begin responding based on the countermeasure plan proposed in advance. The emotion engine is used to monitor the emotional state of staff, and appropriate assistance is provided if overwork or stress is detected. The input is the augmented reality display and emotion assessment results from Step 8, and the output is the response action.

[1739] Step 10:

[1740] After responding to a disaster, the server collects data and evaluates the effectiveness of the countermeasures. Next, the emotion engine takes into account the psychological impact and proposes improvements for the next disaster response. The input is the actually collected data and the emotion evaluation results, and the output is improvement proposals and a report. Specifically, the system analyzes the collected data and plans the next response measures.

[1741] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1742] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1743] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1744] [Fourth embodiment]

[1745] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1746] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1747] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1748] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1749] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1750] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1751] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1752] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1753] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1754] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1755] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1756] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1757] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1758] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users (local government officials) as follows:

[1759] Data collection

[1760] First, the server collects necessary disaster risk information, past disaster data, geographic information, and infrastructure information from the databases and APIs of local governments and related organizations, including population distribution data, national geographic information, and infrastructure company operation data.

[1761] Examples:

[1762] The server collects past earthquake occurrence history for each region from the database of local government A and imports the latest geographical information and supply infrastructure information from other related organizations.

[1763] Data Preprocessing

[1764] The server cleanses and integrates the collected data. Cleansing involves filling in missing data and processing outliers. Integration involves storing data in different formats in a unified database.

[1765] Risk Analysis

[1766] The server then inputs the organized data into an AI model to analyze disaster risk, which then predicts the disaster risk, impact area, and damage for each region.

[1767] Examples:

[1768] The server analyzes the probability of an earthquake occurring in Area B, predicts the epicenter and seismic intensity, and combines this data with population distribution data to identify the areas that will be most affected.

[1769] Creating a risk report

[1770] The server then generates a risk report based on the analysis results, which includes a visualization of disaster risk, identification of the most affected areas, and an assessment of their impact. The risk report is available in PDF and online dashboard format.

[1771] Examples:

[1772] The server creates an earthquake risk report for Area B, showing areas likely to be affected on a map. The report is generated as a PDF and sent to local government officials.

[1773] Disaster prevention plan proposal

[1774] The server generates a disaster response plan based on the risk report, including the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities.

[1775] Examples:

[1776] The server generates a list of suitable evacuation sites in Area B, calculates and provides a table of supplies needed for each site, and assesses the capacity of nearby major medical facilities to determine whether additional medical assistance is needed.

[1777] Building a training plan

[1778] A terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan, which includes hypothetical scenarios and a schedule of practical training exercises.

[1779] Examples:

[1780] The local government's terminal creates an evacuation drill plan assuming an earthquake occurs in Area B. The plan includes confirmation of evacuation routes and a simulation of opening evacuation shelters.

[1781] Status Monitoring

[1782] The server monitors in real time the status of training and disaster response measures, detecting progress and problems and making adjustments as necessary.

[1783] Examples:

[1784] The server monitors the progress of evacuation drills being conducted in Area B in real time and immediately notifies local government officials if any problems arise. For example, if a particular evacuation route is congested, the server will use that information to suggest an alternative route.

[1785] Emergency response

[1786] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance. The system will support the smooth implementation of planned responses, such as setting up evacuation shelters, distributing relief supplies, and providing medical care.

[1787] Examples:

[1788] If an earthquake actually occurs in Area B, local government officials will follow the instructions from the server to quickly open evacuation shelters and guide residents there, as well as deliver necessary relief supplies to the appropriate locations and deploy medical teams.

[1789] Post-recovery feedback and improvements

[1790] After a disaster response, the server collects data and evaluates the effectiveness of the countermeasures. Based on the results, it makes suggestions for improvements to be made in the next disaster response.

[1791] Examples:

[1792] After the disaster has subsided, the server will collect data on shelter operations and relief supply distribution in Area B, analyze which aspects were effective and which areas need improvement, and create a report based on the results to propose even more effective response measures for the next disaster.

[1793] In this way, the server, terminal, and user work together to realize a system that provides effective disaster risk analysis and countermeasures.

[1794] The processing flow will be explained below.

[1795] Step 1: Collect data

[1796] The server collects disaster risk information, past disaster data, geographical information, and infrastructure information from databases and APIs of local governments and related organizations.

[1797] Specific behavior:

[1798] Obtain population distribution data for residents from the database of Municipality A.

[1799] Download the latest topographic data from the Geospatial Information Authority of Japan.

[1800] It collects information on power supply status and water supply operation sent from infrastructure companies.

[1801] Step 2: Preprocessing the data

[1802] The server cleanses and consolidates the collected data: first, it completes missing data, then it detects outliers and handles them appropriately.

[1803] Specific behavior:

[1804] Complement missing data with historical statistical data.

[1805] Detect outliers (e.g., extremely high population growth rates) and identify their causes.

[1806] Convert data in different formats into a unified format (e.g., CSV format).

[1807] Step 3: Risk analysis

[1808] The server inputs the preprocessed data into an AI model to analyze the disaster risk in each region.

[1809] Specific behavior:

[1810] Geographical information and past earthquake data are input into the earthquake prediction model to calculate the probability of an earthquake occurring in each region.

[1811] River and precipitation data are input into the flood risk model to identify vulnerable areas.

[1812] Step 4: Create a risk report

[1813] The server creates a visually easy-to-understand risk report based on the results of the risk analysis.

[1814] Specific behavior:

[1815] The probability of an earthquake occurring and predicted seismic intensity are plotted on a map to show the area of ​​impact.

[1816] Areas at high risk of flooding are color-coded.

[1817] Generate reports as PDFs or online dashboards and send them to city officials.

[1818] Step 5: Propose a disaster response plan

[1819] The server automatically generates a disaster recovery plan based on the risk report.

[1820] Specific behavior:

[1821] Create a list of shelter locations and needed supplies.

[1822] Establish a schedule for relief operations and develop a resource allocation plan based on that schedule.

[1823] Assess the capacity of medical facilities and determine the number of medical support teams needed.

[1824] Step 6: Develop a training plan

[1825] The terminal will then implement a training plan based on the proposed disaster response plan.

[1826] Specific behavior:

[1827] Create hypothetical scenarios and plan simulation training.

[1828] Schedule the training and notify participants.

[1829] Schedule on-the-job training and prepare necessary supplies and equipment.

[1830] Step 7: Status monitoring

[1831] The server monitors the progress and implementation status of training in real time.

[1832] Specific behavior:

[1833] Collect data during training and display progress on a dashboard.

[1834] When a problem occurs, we will notify you immediately and give you instructions on how to fix it.

[1835] Evaluate each stage in real time and adjust your plan as needed.

[1836] Step 8: Emergency response

[1837] When a disaster occurs, users (local government officials) will quickly begin responding based on the countermeasure plans proposed in advance.

[1838] Specific behavior:

[1839] Quickly set up evacuation shelters and guide residents there.

[1840] Check the delivery status of relief supplies and arrange for additional supplies if necessary.

[1841] Medical teams will be deployed to provide emergency response.

[1842] Step 9: Post-recovery feedback and improvement

[1843] The server collects data after disaster response and evaluates the effectiveness of countermeasures. Based on the results, it makes suggestions for improvements to be made in future disaster responses.

[1844] Specific behavior:

[1845] Collect and analyze evacuation center operation data and material distribution status.

[1846] We will extract good points and problems in the response and create an evaluation report.

[1847] A report summarizing areas for improvement will be provided to local government officials as feedback for the next event.

[1848] Example 1

[1849] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1850] In recent years, the frequency and scale of disasters have been increasing, making it extremely important to assess disaster risks and prepare appropriate countermeasures in advance. However, in conventional systems, data collection, risk analysis, and countermeasure proposals are dispersed, preventing efficient and integrated countermeasures. In particular, data preprocessing and real-time situation monitoring are lacking, making it difficult to implement rapid and accurate disaster countermeasures.

[1851] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1852] In this invention, the server includes a means for importing disaster risk information, past disaster data, geographic information, and infrastructure information from various databases and APIs, a means for cleansing and integrating the collected data, completing missing data, and processing outliers, and a means for inputting the preprocessed data into a generative AI model to analyze disaster risks. This makes it possible to assess disaster risks in advance and prepare appropriate countermeasures quickly and accurately.

[1853] "Disaster risk information" refers to data and information used to assess the likelihood of a disaster occurring and its impact. Specifically, this includes the probability of occurrence and damage forecasts for natural disasters such as earthquakes, typhoons, tsunamis, floods, and volcanic eruptions.

[1854] "Past disaster data" refers to detailed records of past disasters, including information on the date, location, scale, damage, and response measures.

[1855] "Geographic information" is data that indicates the geographical characteristics of a specific area, including information on topography, elevation, water systems, land use, infrastructure layout, etc.

[1856] "Infrastructure information" refers to data on infrastructure such as electricity, gas, water, communications, and transportation, including the location, operating status, and supply capacity of each piece of infrastructure.

[1857] A "database" is a system for systematically collecting, storing, and managing specific information. It is used to integrate information from multiple different data sources.

[1858] "API" stands for Application Programming Interface, a set of protocols and tools that allow data to be exchanged between different pieces of software.

[1859] "Cleansing" is the process of correcting or removing missing or outlier values ​​to improve data quality.

[1860] "Integration" is the process of bringing together data from different formats into one standardized format.

[1861] A "generative AI model" is an artificial intelligence model that learns from collected data and performs risk analysis and predictions.

[1862] A "risk report" is a report summarizing the results of a risk analysis, including predictions of the probability of a disaster occurring and the extent of its impact.

[1863] A "disaster response plan" is a specific action plan for minimizing damage in the event of a disaster. It includes the establishment of evacuation shelters, preparation of supplies, medical response, etc.

[1864] A "training plan" is a plan that defines the specific schedule and procedures for training to be conducted based on a disaster response plan.

[1865] "Monitoring" is the process of monitoring the response situation during training and disasters in real time and making adjustments as necessary.

[1866] "Rapid response" refers to immediately taking appropriate countermeasures when a disaster occurs.

[1867] "Evaluation" is the process of analyzing the results of disaster response, measuring effectiveness, and identifying areas for improvement.

[1868] This invention is a system that collects disaster risk information, past disaster data, geographic information, and infrastructure information, performs risk analysis based on this information, and proposes optimal disaster countermeasures. This system operates in cooperation with a server, terminals, and users as follows:

[1869] Data collection

[1870] The server collects disaster risk information, past disaster data, geographic information, and infrastructure information from databases and APIs of local governments and related organizations. At this stage, API calls and database connections are made using a high-performance server machine and Python. For example, the server collects resident population distribution data, national geographic information, and infrastructure company operation data.

[1871] Data Preprocessing

[1872] The server cleanses and integrates the collected data. It uses the Pandas library and NumPy to fill in missing data and process outliers. It also stores data in a unified format in the database. This process produces reliable data.

[1873] Risk Analysis

[1874] The server inputs the preprocessed data into a generative AI model to analyze disaster risk. The AI ​​model used is trained using TensorFlow or PyTorch, enabling highly accurate predictions. For example, the server analyzes the probability of an earthquake occurring in a specific area and identifies areas likely to be affected.

[1875] Creating a risk report

[1876] The server creates a risk report based on the results of the risk analysis. The risk report includes visualization of disaster risks, identification of the most affected areas, and their assessment. This report is generated in PDF format and converted using tools such as Adobe Acrobat. For example, the server displays the affected areas on a map, generates a PDF report, and sends it to local government officials.

[1877] Disaster prevention plan proposal

[1878] The server automatically generates a disaster response plan based on the risk report. This plan includes the location of evacuation shelters, the amount of relief supplies needed, and an assessment of the capacity of medical facilities. For example, the server generates a list of appropriate evacuation shelters and calculates the amount of supplies needed for each shelter. It also assesses the capacity of nearby medical facilities and determines whether additional medical assistance is needed.

[1879] Building a training plan

[1880] The terminal (e.g., a local government computer system) creates a training plan based on the proposed disaster response plan. The training plan includes hypothetical scenarios and a schedule for practical training. As a specific example, the terminal creates an evacuation training plan in an Excel file and notifies each department.

[1881] Status Monitoring

[1882] The server monitors in real time the status of training and the implementation of countermeasures in the event of a disaster. Zabbix and Nagios are used for this monitoring, detecting progress and problems and making adjustments as necessary. For example, the server monitors the progress of training in real time and immediately notifies local government officials if a problem occurs.

[1883] Emergency response

[1884] When a disaster occurs, users (local government officials) can quickly respond based on the prepared response plan. For example, users can use mobile devices and PCs to set up evacuation shelters, distribute relief supplies, and provide medical care.

[1885] Post-recovery feedback and improvements

[1886] The server collects data after disaster response and evaluates the effectiveness of the response. During this process, it reanalyzes information from the database and proposes future improvements. For example, the server analyzes data on evacuation center operations and relief supply distribution to identify areas for improvement.

[1887] In this way, the system provides effective disaster risk analysis and countermeasures by having the server, terminals, and users work together.

[1888] Example prompt for a generative AI model:

[1889] "Please explain in detail the procedures for collecting and preprocessing geographic information, past disaster data, infrastructure information, and population distribution data as input data for the AI ​​model to analyze disaster risk."

[1890] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1891] St...

Claims

1. A means of collecting disaster risk information, past disaster data, geographic information, and infrastructure information; A means of analyzing disaster risks based on collected data; A method for creating and proposing risk reports based on the analysis results, a means for automatically generating a disaster response plan; A means for constructing a training plan based on the generated disaster response plan; A means to monitor the implementation and effectiveness of training in real time and make adjustments as needed; When a disaster occurs, a means to quickly initiate a response based on a countermeasure plan proposed in advance, and A means of collecting post-disaster response data, evaluating the effectiveness of measures, and proposing improvements for the next disaster response. A system including:

2. 10. The system of claim 1, wherein disaster risk information is imported from various databases and APIs.

3. The system according to claim 1, wherein the collected data is organized, missing data is supplemented, and outliers are processed.

Images