Car nakadomari evacuation support system, car nakadomari evacuation support method, and program

Abstract

To efficiently collect and unitarily manage information of vehicle Nakadomari evacuees for each vehicle Nakadomari area, and to perform situation grasping and support prediction with time for each vehicle Nakadomari area.SOLUTION: A car Nakadomari and evacuation support system 1 includes a car storage unit (car Nakadomari) that stores car information including attribution information of a car for car Nakadomari, a car Nakadomari area ID for identifying a car Nakadomari area in which the car is parked, and an arrival date and time at the car DB131 area, an evacuee storage unit (evacuee Nakadomari) that stores evacuee information including attribution information of an evacuee on board the car for the car Nakadomari, a statistical information generation unit 122 that generates real-time and chronological statistical information for each car DB132 area based on the car information and the evacuee information, and a quantity prediction learning model that has learned training data including a time for which the car evacuee stays in the car area, a required quantity of each assistance supply, and the evacuee information. Nakadomari Nakadomari Nakadomari, and an aid supply prediction unit 123 that predicts a required quantity of each of the aid supplies from a vehicle Nakadomari area stay time calculated on the basis of the Nakadomari date and time stored in the vehicle storage unit and the evacuee information stored in the evacuee storage unit for a vehicle Nakadomari evacuee, aggregates the required quantity of each of the aid supplies predicted for each of the vehicle Nakadomari areas, and predicts a provision quantity of each of the aid supplies to the vehicle arrival area.SELECTED DRAWING: None

Description

[Technical Field]

[0001] The present invention relates to a vehicle camping evacuation support system, a vehicle camping evacuation support method, and a program for use in a vehicle camping evacuation support system during a disaster. [Background technology]

[0002] In the event of a disaster such as an earthquake or typhoon, in addition to those who have taken refuge in evacuation shelters (hereinafter referred to as evacuees at evacuation shelters), there are also victims who have taken refuge in their own cars or other vehicles (hereinafter referred to as evacuees camping in their cars) rather than using evacuation shelters designated by local governments, etc., due to privacy concerns, concerns about infectious diseases, and evacuation shelter rules such as not allowing pets to accompany people (hereinafter referred to as evacuees camping in their cars). In order to allow evacuees to receive appropriate support from local governments and other organizations, whether they are staying in evacuation shelters or sleeping in their cars, apps are available that allow information to be shared with the city and at evacuation shelters and other organizations (see, for example, Non-Patent Document 1).

[0003] In particular, for those who are evacuating in their cars, methods have been disclosed for identifying those who need assistance when a disaster occurs and improving the technology for providing assistance to those who are evacuating in their cars (see Patent Document 1), as well as a device that accumulates purchase information indicating that users have purchased specified items, and based on the accumulated purchase information, determines the degree to which specified requirements for evacuating in their cars using a vehicle as an evacuation site are met, and estimates which users need information regarding evacuating in their cars, thereby providing information to users who may be evacuating in their cars or are already evacuating in their cars (see Patent Document 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7476784 [Patent Document 2] Patent No. 7468336 [Non-patent literature]

[0005] [Non-Patent Document 1] Tsunagaru+ [URL:https: / / www.city.fukuoka.lg.jp / shimin / t_bousai / tsunagaru_plus.html Searched on June 17, 2024] Summary of the Invention [Problem to be solved by the invention]

[0006] In past disasters, people have evacuated to their cars in various locations, and even designated evacuation centers have not been able to identify those who have evacuated, making it difficult to grasp the situation and provide material support. To solve this problem, it is being considered to have people evacuate to their cars in pre-designated car camping areas, and to centrally manage car camping evacuees by car camping area. However, while the above-mentioned app and Patent Documents 1 and 2 can identify each car camping evacuee who needs assistance, they cannot centrally manage car camping evacuees by car camping area.

[0007] Furthermore, as time passes since the occurrence of a disaster, the situation in the car camping area changes, and the support required by car camping evacuees also changes. However, the above-mentioned app and Patent Documents 1 and 2 do not take into account the time that has passed since the vehicle evacuation began, and therefore have the problem of being unable to respond to changes that occur over time.

[0008] Therefore, the present invention has been made in consideration of the above circumstances, and aims to provide a car camping evacuation support system, a car camping evacuation support method, and a program that efficiently collects and centrally manages information on car camping evacuees for each car camping area, and grasps the situation over time and predicts support for each car camping area. [Means for solving the problem]

[0009] The present invention provides a car camping evacuation support system that includes a vehicle memory unit that stores vehicle information including attribute information of the vehicle being camped in, a car camping area ID that identifies the car camping area in which the vehicle is parked, and the date and time of arrival at the car camping area, an evacuee memory unit that stores evacuee information including attribute information of the car camping evacuees on board the vehicle being camped in, and a statistical information generation unit that generates real-time and time-series statistical information for each car camping area based on the vehicle information and the evacuee information.

[0010] The present invention also provides a car camping evacuation support management system that includes a support material prediction unit that predicts the required quantity of each support material for each car camping evacuee based on the stay time of the car camping evacuees in the car camping area, the required quantity of each support material, and the evacuee information stored in the evacuee memory unit, calculated based on the arrival date and time stored in the vehicle memory unit, based on a quantity prediction learning model that has been learned using teacher data including the stay time of the car camping evacuees in the car camping area, the required quantity of each support material, and the evacuee information, and tallying the predicted required quantity of each support material for each car camping area and predicting the quantity of each support material to be provided to that car camping area.

[0011] The present invention also provides a car camping evacuation support management system that includes a distribution map generation unit that generates a distribution map that displays the quantity of each relief supply to be provided to each car camping area predicted by the relief supply prediction unit at a position on the map that corresponds to the location information of each car camping area.

[0012] The present invention also provides a car camping evacuation support system that includes a health damage risk prediction unit that predicts the health damage risk for each car camping evacuee based on the amount of time the car camping evacuees spend in the car camping area, the health damage risks that may occur to the car camping evacuees, and teacher data including the evacuee information, calculated based on the arrival date and time stored in the vehicle memory unit and the evacuee information stored in the evacuee memory unit, based on a risk learning model that has been learned using training data including the amount of time the car camping evacuees spend in the car camping area, the health damage risks that may occur to the car camping evacuees, and the evacuee information.

[0013] The present invention also provides a car camping evacuation support system that includes an advice memory unit that stores the health damage risks in association with advice related to the health damage risks, and an advice providing unit that refers to the advice memory unit to obtain advice on the health damage risks of each car camping evacuee predicted by the health damage risk prediction unit, and sends the obtained advice to the terminal of each car camping evacuee.

[0014] The present invention also provides a car camping evacuation support system in which the health damage risk prediction unit aggregates the predicted health damage risks for each car camping evacuee for each car camping area and predicts the health damage risk for that car camping area.

[0015] The present invention also provides a method executed by a car camping evacuation support system, which includes the steps of storing vehicle information in a vehicle memory unit, including attribute information of the vehicle being camped in, a car camping area ID that identifies the car camping area in which the vehicle is parked, and the date and time of arrival at the car camping area; storing evacuee information in an evacuee memory unit, including attribute information of the car camping evacuees on board the vehicle being camped in; and generating real-time and time-course statistical information for each car camping area based on the vehicle information and the evacuee information.

[0016] The present invention also provides a program that causes a car camping evacuation support system to function as a vehicle memory unit that stores vehicle information including attribute information of the vehicle being camped in, a car camping area ID that identifies the car camping area in which the vehicle is parked, and the date and time of arrival at the car camping area, an evacuee memory unit that stores evacuee information including attribute information of the car camping evacuees on board the vehicle being camped in, and a statistical information generation unit that generates real-time and time-course statistical information for each car camping area based on the vehicle information and the evacuee information. [Effects of the Invention]

[0017] According to the present invention, information on evacuees camping in their cars can be efficiently collected and centrally managed for each camping area, allowing for a time-dependent understanding of the situation and support prediction for each camping area. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a schematic diagram of a vehicle camping evacuee support system according to an embodiment of the present invention; [Figure 2] 1 is a diagram showing the functional configuration of a vehicle camping evacuee support system according to an embodiment of the present invention; [Figure 3] FIG. 1 is a flowchart showing a process of supporting evacuation during overnight stays in a vehicle by the vehicle overnight stay evacuation support system according to the present embodiment. [Figure 4] 10 is a schematic diagram showing a vehicle camping area DB135 according to the present embodiment. [Figure 5] FIG. 2 is a diagram schematically illustrating a vehicle DB according to the present embodiment. [Figure 6] FIG. 2 is a diagram schematically illustrating an evacuee DB according to the present embodiment. [Figure 7] 2 is a diagram illustrating an advice DB according to the present embodiment. [Figure 8] FIG. 2 is a diagram showing an example of a distribution map according to the embodiment; DETAILED DESCRIPTION OF THE INVENTION

[0019] Hereinafter, embodiments of the present invention (hereinafter referred to as "embodiments") will be described in detail with reference to the accompanying drawings. In the following drawings, the same elements are designated by the same numbers or symbols throughout the description of the embodiments.

[0020] [Basic concept] FIG. 1 is a schematic diagram of a vehicle camping evacuation support system 1 according to an embodiment of the present invention. The car camping evacuation support system 1 is a system that collects vehicle information about vehicles parked in car camping areas and used for car camping during a disaster, as well as evacuee information about evacuees camping in the vehicles, and manages this information for each car camping area.Based on the vehicle information and evacuee information for each car camping area, it generates statistical information over time, predicts the quantity of relief supplies to be provided, and predicts the risk of health damage.

[0021] 1, the vehicle camping evacuation support system 1 includes a vehicle camping evacuation support device 100 and a vehicle camping evacuee terminal (hereinafter referred to as the evacuee terminal) 200. The vehicle camping evacuation support device 100 and the evacuee terminal 200 are connected via a communication line such as the Internet, but in the event of a disaster when normal communication lines such as the Internet cannot be used, the vehicle camping evacuation support device 100 and the evacuee terminal 200 are connected via a local network that is installed in advance in the vehicle camping area.

[0022] The vehicle camping evacuation support device 100 is managed by an administrator of the vehicle camping evacuation support system 1, and may be a cloud server or may be on-premise. The number of vehicle camping evacuation support devices 100 is not limited to one, and multiple devices may be installed in each vehicle camping area.

[0023] The evacuee terminal 200 is a terminal of a vehicle camping evacuee (hereinafter referred to as an evacuee) staying in a vehicle camping area, and is a terminal to be used in the event of a disaster, so it is preferably a mobile terminal such as a smartphone or tablet terminal, but may also be a computer (PC). Although only one evacuee terminal 200 is shown in Fig. 1, there may be two or more evacuee terminals 200, and in an actual system, there may be many evacuee terminals 200.

[0024] An overview of the vehicle camping evacuation support system 1 will be described with reference to FIG. First, after a disaster occurs, an evacuee parks his / her vehicle in a car camping area, and then registers his / her vehicle and vehicle evacuee from the evacuee terminal 200 to the car camping evacuation support device 100 to receive admission (S1). Here, a car camping area is a place for car camping that has been designated in advance by a local government, etc., such as a ground adjacent to a designated evacuation shelter or a parking lot of a large commercial facility. Note that a car camping area is not limited to one ground or parking lot, but may consist of multiple nearby grounds or parking lots.

[0025] In detail, first, after parking a vehicle in the car camping area, a representative of the evacuees in the vehicle (e.g., the driver) saves the vehicle information from the evacuee terminal 200 to the car camping evacuation support device 100 and accepts the vehicle into the car camping area.

[0026] For example, the representative downloads the app from the vehicle camping evacuation support device 100 to the evacuee terminal 200. Next, the driver inputs the vehicle attribute information and the vehicle arrival date and time into the app, and causes a specified terminal to read a QR code generated by the app from the input vehicle attribute information and arrival date and time. Here, the vehicle attribute information includes, for example, the vehicle license plate number, the vehicle owner, the vehicle model, etc. Furthermore, if the representative does not input the arrival date and time, the date and time when the app is launched or the date and time when the QR code is generated may be obtained as the arrival date and time.

[0027] The predetermined terminal transmits vehicle information including the vehicle attribute information and arrival date and time acquired from the QR code, and its own location information, to the vehicle camping evacuation support device 100. The predetermined terminal is connected to the vehicle camping evacuation support device 100 by wire or wirelessly (via the Internet or a local network). Then, when the vehicle camping evacuation support device 100 stores the received vehicle information, the vehicle reception is completed.

[0028] In addition, vehicle information including the vehicle attribute information and arrival date and time entered into the app of the evacuee terminal 200, and the location information of the evacuee terminal 200 (longitude, latitude and address) may be sent directly from the evacuee terminal 200 to the vehicle camping evacuation support device 100.

[0029] Next, the evacuees in the vehicles save the evacuee information, including their own attribute information, from the evacuee terminal 200 to the vehicle camping evacuation support device 100, and are accepted into the vehicle camping area. Here, the attribute information of the evacuees includes, for example, name, age, sex, the vehicle license plate number of the vehicle they are riding in, information about people requiring special care, contact information, height, weight, etc. In addition to the attribute information, the evacuee information also includes information about the evacuees, such as medical history, past medical history, and health condition information such as chronic illnesses.

[0030] For example, an evacuee in a vehicle downloads the app from the vehicle camping evacuation support device 100 to the evacuee terminal 200. Note that if the app is the same as the app used for vehicle reception and has already been downloaded to the evacuee terminal 200, no further download is required. Next, the evacuee enters their own evacuee information into the app and has a designated terminal read the QR code generated by the app from the entered evacuee information. The evacuee may also enter attribute information of other evacuees riding in the same vehicle.

[0031] The predetermined terminal transmits the evacuee information acquired from the read QR code to the vehicle camping evacuation support device 100. Then, when the vehicle camping evacuation support device 100 stores the received evacuee information, the evacuee reception is completed.

[0032] Returning to FIG. 1, next, the vehicle camping evacuation support device 100 saves and manages the vehicle information and evacuee information received from the evacuee terminal 200 (S2).

[0033] Next, the car camping evacuation support device 100 generates real-time and time-varying statistical information for each car camping area based on the vehicle information and evacuee information saved in S2 (S3). The real-time statistical information is statistical information that represents the current status of the car camping area, such as the current number of parked vehicles and the population distribution by generation. The time-varying statistical information includes statistical information from several points in the past and represents past changes in the status of the car camping area, such as the number of parked vehicles two days ago and the number of parked vehicles one day ago, and can represent changes in the number of parked vehicles. The generated real-time and time-series statistical information will be provided to local governments and other organizations providing disaster relief, and will be used by the local governments and other organizations to provide support to evacuees in areas where people are sleeping in their cars.

[0034] Next, the car camping evacuation support device 100 uses machine learning to predict the amount of relief supplies to be provided to the car camping area based on the evacuees' stay time in the car camping area and the stored evacuee information (S4). Specifically, the car camping evacuation support device 100 calculates the duration of evacuees' stay in the car camping area based on the arrival time in the vehicle information. Next, the car camping evacuation support device 100 predicts the required quantity of each preset relief supply for each evacuee using machine learning based on the stored evacuee information. Then, the car camping evacuation support device 100 aggregates the predicted required quantities for each evacuee for each car camping area and generates relief supply information indicating the quantity of each relief supply to be provided to each car camping area. The generated information on relief supplies will be provided to local governments and other organizations and used to provide relief supplies to areas where people are camping in their cars.

[0035] The car camping evacuation support device 100 uses machine learning to predict the health damage risks of evacuees and car camping areas based at least on the time the evacuees stay in the car camping area and the stored evacuee information (S5). In detail, the vehicle camping evacuation support device 100 calculates the duration of time that evacuees will stay in the vehicle camping area based on the arrival time of the vehicle information.

[0036] Next, the vehicle camping evacuation support device 100 predicts the health hazard risk for each evacuee by machine learning based on at least the calculated stay time and the evacuee information. Here, the health hazard risk is the risk of health hazard that may occur to the evacuee, such as economy class syndrome.

[0037] The car camping evacuation support device 100 then aggregates the predicted health hazard risks for each evacuee for each car camping area and generates the health hazard risks for each car camping area as area health hazard risk information. The area health hazard risk information is generated as, for example, "Kumamoto Chuo 1 Area, Area Economy Syndrome Occurrence Risk Level: High: 35 people, 10 people out of 35, Low: 100 people." The generated area health damage risk information will be provided to local governments and other organizations and used to help prevent health damage in areas where people sleep in their cars.

[0038] Next, the vehicle camping evacuation support device 100 acquires advice for the predicted health damage risk for each evacuee from the advice for each health damage risk stored in advance, and transmits the predicted health damage risk and the acquired advice to the evacuee terminal 200 of each evacuee, and displays them on the evacuee terminal 200 (S6). This will allow us to provide each evacuee directly with information about health risks and advice on how to deal with them.

[0039] With this type of car camping evacuation support system, information on each vehicle parked in a car camping area (vehicle information) and information on the evacuees aboard the vehicles (evacuee information) is efficiently collected and centrally managed for each car camping area, making it possible to generate time-series statistical information from this information for each car camping area that can be used to grasp the situation over time for each car camping area.

[0040] Furthermore, for each car camping area, the system can predict the amount of relief supplies to be provided based on the amount of time evacuees will stay in the car camping area and evacuee information using machine learning. This allows evacuees to receive the necessary amount of relief supplies without having to request them. Furthermore, for each evacuee and each vehicle camping area, health hazard risk prediction can be performed using machine learning based on at least vehicle information and evacuee information, thereby helping to prevent health hazards for evacuees and the entire vehicle camping area.

[0041] [Function Configuration] FIG. 2 is a diagram showing the functional configuration of the vehicle camping evacuation support system 1 according to this embodiment. As shown in the figure, the vehicle camping evacuation support system 1 includes a vehicle camping evacuation support device 100 and an evacuee terminal 200.

[0042] The vehicle camping evacuation support device 100 includes a communication unit 110, a control unit 120, and a storage unit . The communication unit 110 is connected to a network wirelessly or via a wired connection, and communicates with the evacuee terminal 200 .

[0043] The control unit 120 is realized by, for example, a computing unit such as a processor or a CPU, and functions in cooperation with various types of memory, a hard disk, and other storage units to execute various programs. The control unit 120 includes a registration unit 121, a statistical information generation unit 122, a relief supply prediction unit 123, a health damage risk prediction unit 124, an advice provision unit 125, and a distribution map generation unit 126. Details of each functional unit included in the control unit 120 will be explained together with the car camping evacuation support flow shown in FIG. 3.

[0044] The registration unit 121 acquires vehicle information from the evacuee terminal 200 via the communication unit 110 and stores the information in the vehicle DB 131. The registration unit 121 also acquires evacuee information from the evacuee terminal 200 via the communication unit 110 and stores the information in the evacuee DB 132. The statistical information generating unit 122 generates time-varying statistical information for each vehicle camping area based on the vehicle DB 131 and the evacuee DB 132.

[0045] The relief supplies prediction unit 123 predicts the quantity of relief supplies (relief supplies information) to be provided for each car camping area using a quantity prediction learning model stored in the learning model 133 based on the vehicle DB 131 and the evacuee DB 132.

[0046] The health damage risk prediction unit 124 predicts the health damage risk of each evacuee using a risk prediction learning model stored in the learning model 133, based on the vehicle DB 131 and the evacuee DB 132. The health damage risk prediction unit 124 also predicts the health damage risk for each vehicle camping area, based on the predicted health damage risk of each evacuee.

[0047] The advice providing unit 125 obtains advice associated with the health damage risk of each evacuee predicted by the health damage risk prediction unit 124 from the advice DB 134, and transmits the obtained advice to the evacuee terminal 200 of each evacuee and displays it.

[0048] The distribution map generation unit 126 generates a distribution map that displays at least one of the time-course statistical information for each car camping area generated by the statistical information generation unit 122, the relief supply information for each car camping area generated by the relief supply prediction unit 123, and the health damage risk for each car camping area predicted by the health damage risk prediction unit 124 at a position on the map that corresponds to the location information of the car camping area.

[0049] The storage unit 130 includes a vehicle DB 131 that stores vehicle information about vehicles parked in car camping areas, a car camping evacuee DB 132 that stores evacuee information about evacuees aboard the vehicles, a learning model DB 133 that stores learning models used by the relief supply prediction unit 123 and the health damage risk prediction unit 124, an advice DB 134 that stores advice for each health damage risk, and a car camping area DB 135 that stores information about car camping areas. Details of each DB will be explained together with the car camping evacuation support flow shown in Figure 3 below.

[0050] The evacuee terminal 200 is equipped with a communication unit (not shown) that connects to a network wirelessly or via a wired connection to communicate with the vehicle camping evacuation support device 100, an input unit (not shown) such as a keyboard that accepts operational input from evacuees, and a display unit (not shown) such as an LCD display that displays various information such as an app screen for registering information and information regarding predicted health damage risks.

[0051] [Processing flow] Next, the vehicle camping evacuation support process performed by the vehicle camping evacuation support system 1 according to this embodiment will be described together with details of each functional configuration of the control unit 120. FIG. 3 is a diagram showing a flowchart of the car camping evacuation support process by the car camping evacuation support system 1 according to this embodiment.

[0052] This process begins when an app for vehicle registration and evacuee registration is downloaded from the car camping evacuation support device 100 to the evacuee terminal 200 of an evacuee in a vehicle parked in a car camping area.

[0053] When the registration unit 121 receives vehicle information from the evacuee terminal 200, including vehicle attribute information entered by the evacuee into the app on the evacuee terminal 200, location information of the car camping area where the vehicle is parked, and the vehicle's arrival date and time, the registration unit 121 stores the information in the vehicle DB 131 and accepts the vehicle into the car camping area (S101).

[0054] In detail, the registration unit 121 refers to the vehicle camping area DB 135 based on the location information of the vehicle camping area received from the evacuee terminal 200, and acquires the vehicle camping area ID of the vehicle camping area.

[0055] Fig. 4 shows a schematic diagram of the car camping area DB 135 according to this embodiment. As shown in Fig. 4, the car camping area information stored in the car camping area DB 135 includes a car camping area ID for identifying the car camping area, a car camping area name, location information (latitude, longitude), location information (address), etc. In this embodiment, the location information is latitude, longitude, and address, but is not limited to this as long as the car camping area can be identified on a map.

[0056] The registration unit 121 then stores the acquired vehicle camping area ID, and the vehicle information including the vehicle attribute information and the vehicle arrival date and time received from the evacuee terminal 200 in the vehicle DB 131.

[0057] Fig. 5 schematically shows the vehicle DB 131 according to this embodiment. As shown in Fig. 5, the vehicle information stored in the vehicle DB 131 includes the overnight stay area ID, the vehicle number which is the vehicle registration number printed on the license plate, the vehicle model, the name of the vehicle owner, the arrival date and time, etc.

[0058] In addition, the registration unit 121 receives evacuee information entered by the evacuee into the app of the evacuee terminal 200 from the evacuee terminal 200, assigns an evacuee ID that identifies the evacuee in the received evacuee information, and stores the information in the evacuee DB 132 (S102).

[0059] Fig. 6 is a schematic diagram of the evacuee DB 132 according to this embodiment. As shown in Fig. 6, the evacuee information stored in the evacuee DB 132 includes an evacuee ID for identifying the evacuee, a vehicle number, a name, a sex, an age, a height, a weight, information on persons requiring special consideration, such as pregnant women, nursing mothers, and persons with disabilities, and contact information. When accepting evacuees into the car camping area, it is not necessary to enter all of the evacuee information items shown in Figure 6; it is sufficient to enter pre-set required items, such as the vehicle number and name. Items that are not entered can be added later.

[0060] Returning to FIG. 3, the statistical information generating unit 122 generates real-time and time-series statistical information based on the vehicle DB 131 and the evacuee DB 132 (S103).

[0061] Specifically, the statistical information generating unit 122 generates real-time statistical information based on all of the vehicle information in the vehicle DB 131 and the evacuee information in the evacuee DB 132. Statistical information includes, for example, the number of parked cars, the number of people sleeping in their cars, the distribution of people by generation (infants, children, adults, elderly people, etc.), the distribution of people by gender, the distribution of people by the length of time they stay in the car camping area, and the distribution of cars by type of car, and indicates the situation in the car camping area.

[0062] In addition, the statistical information generation unit 122 extracts vehicle information and evacuee information at a predetermined time in the past from the vehicle information in the vehicle DB 131 and the evacuee information in the evacuee DB 132, and generates statistical information at a predetermined time in the past based on the extracted vehicle information and evacuee information. The extraction of vehicle information and evacuee information at a predetermined time in the past is performed based on the arrival date and time included in the vehicle information, i.e., by extracting information whose arrival date and time is older than the predetermined time in the past and that was already saved at the predetermined time in the past. Note that there may be multiple predetermined time in the past, and in this case, the statistical information generation unit 122 generates statistical information for each of the predetermined time in the past. The statistical information generating unit 122 then generates time-series statistical information that includes the generated statistical information at a predetermined time point in the past and that represents past changes in the situation of the vehicle camping area.

[0063] Then, the statistical information generating unit 122 transmits the generated real-time and time-course statistical information to a predetermined local government or the like that is providing disaster support. Note that if the statistical information generating unit 122 cannot communicate with the local government or the like, the generated real-time and time-course statistical information may be output to paper or a storage medium and provided to the recipient. Based on the real-time and historical statistical information provided, local governments and other entities can grasp the current status of car camping areas and trends in the situation in car camping areas. As a result, local governments and other entities can provide the support currently needed by evacuees in car camping areas and prepare for support that will be needed in the future.

[0064] Returning to FIG. 3, the relief supplies prediction unit 123 predicts the amount of relief supplies to be provided for each vehicle camping area based on the vehicle DB 131 and the evacuee DB 132 using machine learning (S104).

[0065] In detail, first, the relief supplies prediction unit 123 calculates the stay time of each evacuee in the vehicle camping area based on the arrival date and time in the vehicle DB 131. The stay time may be the time from the arrival date and time to the present time, or may be the time until a predetermined future date and time (also called a designated date and time). By using the time until the designated date and time, it is possible to make a prediction based on the date and time when relief supplies will be provided.

[0066] The relief supplies prediction unit 123 predicts the required quantity of each relief supply for each evacuee using a quantity prediction learning model that inputs the calculated stay time and evacuee information, particularly evacuee attribute information, and outputs a predetermined required quantity of each relief supply. The predicted required quantity of each relief supply for each evacuee is the required quantity of each relief supply appropriate for the evacuee at the current time in the case of the stay time up to the current time, and is the required quantity of each relief supply appropriate for the evacuee at the specified time in the case of the stay time up to a specified time and date.

[0067] In the above case, the quantity prediction learning model is trained using a plurality of training data sets, each of which is composed of a set of the duration of evacuees' stay, the required quantity of each relief supply, and evacuee information (attribute information of the evacuees), as training data. As an example, the learning model is trained by machine learning using a known method such as deep learning. The preset relief supplies are, for example, relief supplies determined based on past disaster experiences, and include various types such as food, daily necessities, medicines, and baby products.

[0068] Then, the relief supplies prediction unit 123 generates relief supplies information indicating the quantity of each relief supply to be provided to each car camping area by aggregating the required quantity of each relief supply predicted for each evacuee for each car camping area. Note that by inputting different amounts of relief supplies only for different durations of stay into the relief supply prediction unit 123, it is possible to predict the required quantities of each relief supply over time for the same evacuee, tally up the required quantities of each relief supply over time, and generate the quantity of each relief supply to the car camping area over time as relief supply information. For example, the relief supply prediction unit 123 can predict the quantity of each relief supply to be provided to the evacuee and the car camping area at the present time, 24 hours later, and 48 hours later.

[0069] Then, the relief supplies prediction unit 123 transmits the generated relief supplies information to a predetermined local government or the like that is providing disaster relief. Note that if the relief supplies prediction unit 123 cannot communicate with the local government or the like, the generated relief supplies information may be output to paper or a storage medium and provided to the recipient. Based on the information on relief supplies provided, local governments and other organizations can grasp the relief supplies needed in the car camping area at the current time or at a specified date and time. As a result, they can provide the appropriate amount of relief supplies needed by evacuees even if they do not request it themselves.

[0070] Returning to FIG. 3, the health damage risk prediction unit 124 predicts the health damage risk for each vehicle camping area based on the vehicle DB 131 and the evacuee DB 132 using machine learning (S105). Here, the health hazard risk refers to the risk of health hazards that may occur to evacuees, such as economy class syndrome, heat stroke, hypothermia, and carbon monoxide poisoning, expressed in three levels, for example, low, medium, and high. The number of levels for risk classification may be arbitrary, such as two levels, "possible" and "not possible," or five levels, with probability of occurrence of 0-20%, 40%, 60%, 80%, and 100%.

[0071] In detail, first, similar to S104, the health damage risk prediction unit 124 calculates the stay time of each evacuee in the vehicle camping area based on the arrival date and time in the vehicle DB 131. If the stay time is defined as the time from the arrival time to a specified future time, the future health damage risk can be predicted.

[0072] The health hazard risk prediction unit 124 predicts the health hazard risk of each evacuee using a risk prediction learning model that inputs the calculated stay time and evacuee information and outputs the health hazard risk. The evacuee information preferably includes not only attribute information but also health status information related to the evacuee's health condition, such as medical history, past medical history, and chronic illnesses. The predicted health hazard risk of each evacuee is the current health hazard risk in the case of the stay time up to the present time, and the health risk at the specified date and time in the case of the stay time up to a specified date and time. For example, the health hazard risk prediction unit 124 predicts that a pregnant woman in her 30s who stays for one hour has a low risk of economy class syndrome, and that a pregnant woman in her 30s who stays for 24 hours has a high risk of economy class syndrome.

[0073] In the above case, the risk prediction learning model is trained using multiple training data sets, each consisting of a set of evacuees' stay time, health hazard risk, and evacuee information, as training data. As an example, the learning model is trained by machine learning using a known technique such as deep learning.

[0074] Then, the health damage risk prediction unit 124 aggregates the health damage risks predicted for each evacuee for each car camping area, and generates the health damage risks for each car camping area as area health damage risk information. It is also possible to predict the health hazard risk over time for the same evacuee by changing only the length of stay and inputting it into the health hazard risk prediction unit 124, and to aggregate the health hazard risks and generate the health hazard risk over time for the car camping area as area health hazard risk information. For example, the health hazard risk prediction unit 124 can predict the health hazard risk at the present time and 24 hours from now for the evacuee and the car camping area.

[0075] Then, the health damage risk prediction unit 124 transmits the generated area health damage risk information to a pre-set local government or the like that is providing disaster support. Note that if the health damage risk prediction unit 124 cannot communicate with the local government or the like, the generated area health damage risk information may be output to paper or a storage medium and provided to the recipient.

[0076] Based on the area health risk information provided, local governments and other organizations can understand the health risk of car camping areas at the current time and at a specified date and time. As a result, they can prevent health damage by taking appropriate measures at the appropriate time. Response measures include, for example, dispatching public health nurses or DMATs to car camping areas, and providing information and warnings about health damage prevention measures.

[0077] Returning to FIG. 3, the advice providing unit 125 transmits advice corresponding to the health damage risk of each evacuee predicted by the health damage risk prediction unit 124 to the evacuee terminal 200 and provides it to the evacuee (S106). In detail, first, the advice providing unit 125 refers to the advice DB 134 and acquires advice corresponding to the health damage risk of each evacuee predicted by the health damage risk prediction unit 124 .

[0078] Fig. 7 schematically shows the advice DB 134 according to this embodiment. As shown in Fig. 7, the advice information stored in the advice DB 134 includes health hazard risks, advice, etc. Note that the advice may be set by generation or gender.

[0079] Next, the advice providing unit 125 refers to the evacuee DB 132 to obtain the contact information of the evacuee who will receive the obtained advice, sends the obtained contact information to the evacuee terminal 200, and displays it on the evacuee terminal 200. Note that if the contact information of the evacuee who will receive the advice is not registered in the evacuee DB 132, the advice may be sent instead to an evacuee who is riding in the same vehicle and whose contact information is registered in the evacuee DB 132.

[0080] Returning to Figure 3, the distribution map generation unit 126 generates a distribution map that displays at least one of the time-series statistical information generated in S103, the relief supply information generated in S104, and the area health damage risk predicted in S105 at a position on the map that corresponds to the location information of the car camping area (S107).

[0081] In detail, first, the distribution map generating unit 126 acquires the location information of the car camping area (also referred to as the displayed car camping area) that displays at least one of the time-series statistical information, the relief supply information, and the area health hazard risk, by referring to the car camping area DB 125. The displayed car camping area may be selected arbitrarily by the administrator, or may be selected according to a predetermined condition, for example, a car camping area in which 100 or more evacuees have a high risk of developing economy class syndrome in the area health hazard risk.

[0082] The distribution map generation unit 126 acquires map information including the displayed car camping area, and displays at least one of the time-series statistical information, relief supply information, and area health hazard risk, for example, in a speech bubble, at a position on the map corresponding to the location information of the car camping area.On the map, only an icon indicating the car camping area may be displayed, and when the icon is selected, the time-series statistical information, relief supply information, and area health hazard risk may be displayed in a speech bubble or the like.

[0083] FIG. 8 is a diagram showing an example of a distribution map according to this embodiment. The distribution map displays three icons indicating the location of the car camping area. Selecting an icon displays a speech bubble, which displays, for example, statistical information over time. Selecting the icon again switches to information on relief supplies and area health risks. It is also possible to display speech bubbles on multiple icons, allowing you to compare the status of the displayed car camping area.

[0084] As described above, according to the car camping evacuation support system of this embodiment, information on vehicles parked in car camping areas (vehicle information) and information on evacuees aboard the vehicles (evacuee information) is efficiently collected and centrally managed for each car camping area, and from this information, time-series statistical information can be generated for each car camping area that allows for an understanding of the situation over time for each car camping area.

[0085] Furthermore, for each car camping area, machine learning can be used to predict the amount of relief supplies to be provided based on the amount of time evacuees will stay in the car camping area and evacuee information. This allows evacuees to receive the necessary amount of relief supplies they need at the time, even if they do not request it. The amount to be provided can be predicted not only for the current time, but also for a specified date and time in the future, so the supporters can take future relief supplies into consideration.

[0086] Furthermore, health hazard risk predictions can be performed for each evacuee and each car camping area using machine learning based at least on vehicle information and evacuee information. By predicting health hazard risks in car camping areas, local governments and other organizations can support the prevention of health hazards throughout the entire car camping area by issuing warnings to car camping areas and supporting health hazard prevention activities. Furthermore, by directly providing evacuees with health hazard risks and advice, they can also support the prevention of health hazards for each evacuee.

[0087] (Variation 1) In this embodiment, the relief supplies prediction unit 123 predicts the required quantities of preset relief supplies, but may also predict the types of relief supplies. In this case, the relief supplies prediction unit 123 uses a relief supplies prediction learning model that inputs the calculated stay time and evacuee information, particularly the evacuee's attribute information, and outputs the relief supplies and the required quantity of those relief supplies together.

[0088] (Variation 2) Furthermore, in this embodiment, the health damage risk prediction unit 124 predicted the health damage risk of each evacuee using a risk prediction learning model that outputs health damage risk by inputting the calculated stay time and evacuee information, but it may also predict the health damage risk of each evacuee using a risk prediction learning model that outputs health damage risk by inputting environmental information such as the current season, current time, and weather information. Adding environmental information can improve the accuracy of predicting health risks, particularly the risks of heatstroke, hypothermia, and carbon monoxide poisoning.

[0089] Although the present invention has been described above using embodiments, it goes without saying that the technical scope of the present invention is not limited to the scope described in the above embodiments. It is clear to those skilled in the art that various modifications or improvements can be made to the above embodiments. Furthermore, it is clear from the claims that forms incorporating such modifications or improvements can also be included in the technical scope of the present invention. In the above embodiments, the present invention has been described as an invention of a vehicle camping evacuation support system, but it can also be considered as an invention of a method executed by the vehicle camping evacuation support system or a program that causes the vehicle camping evacuation support system to function as various means. [Explanation of symbols]

[0090] 1. Car camping evacuation support system 100 Car camping evacuation support device 110 Communications Department 120 control section 121 Registration Department 122 Statistical information generation section 123 Relief Supplies Forecasting Department 124 Health Risk Prediction Department 125 Advice Department 126 Distribution Map Generation Unit 130 Storage section 131 Vehicle DB 132 Evacuee DB 133 Learning Model DB 134 Advice DB 135 Car Camping Area DB 200 Evacuee terminals

Claims

1. a vehicle storage unit that stores vehicle information including attribute information of the vehicle that is sleeping in the car, a car camping area ID that identifies the car camping area where the vehicle is parked, and the date and time of arrival at the car camping area; a refugee storage unit that stores refugee information including attribute information of the refugees who are sleeping in the vehicle; a statistical information generating unit that generates real-time and time-series statistical information for each vehicle camping area based on the vehicle information and the evacuee information; A car camping evacuation support system.

2. Based on a quantitative prediction learning model that has been trained with training data including the duration of stay in the car camping area of ​​evacuees camping in their cars, the required quantity of each relief item, and the evacuee information, For each vehicle camping evacuee, the required quantity of each relief item is predicted based on the vehicle camping area stay time calculated based on the arrival date and time stored in the vehicle storage unit and the evacuee information stored in the evacuee storage unit; The car camping evacuation support system of claim 1, further comprising a support material prediction unit that aggregates the predicted required quantities of each support material for each car camping evacuee for each car camping area and predicts the quantity of each support material to be provided to the car camping area.

3. The car camping evacuation support system of claim 2, further comprising a distribution map generation unit that generates a distribution map in which the quantity of each relief supply to be provided to each car camping area predicted by the relief supply prediction unit is displayed at a position on the map corresponding to the location information of each car camping area.

4. Based on a risk learning model that has been trained on training data including the duration of stay in the car camping area of ​​car camping evacuees, the health hazard risks that may occur to car camping evacuees, and the evacuee information, The car camping evacuation support system described in claim 1 is equipped with a health damage risk prediction unit that predicts the health damage risk for each car camping evacuee based on the time spent in the car camping area calculated based on the arrival date and time stored in the vehicle memory unit and the evacuee information stored in the evacuee memory unit.

5. an advice storage unit that stores the health hazard risk and advice regarding the health hazard risk in association with each other; an advice providing unit that refers to the advice storage unit, acquires advice on the health damage risk of each vehicle camping evacuee predicted by the health damage risk prediction unit, and transmits the acquired advice to a terminal of each vehicle camping evacuee; The vehicle camping evacuation support system according to claim 4, comprising:

6. The car camping evacuation support system of claim 4, wherein the health damage risk prediction unit aggregates the predicted health damage risks for each car camping evacuee for each car camping area and predicts the health damage risk for that car camping area.

7. A method executed by a car camping evacuation support system, Storing vehicle information in a vehicle storage unit, the vehicle information including attribute information of the vehicle being parked, a car camping area ID that identifies the car camping area where the vehicle is parked, and the arrival date and time of the vehicle at the car camping area; Storing evacuee information including attribute information of the evacuees sleeping in the vehicle in an evacuee storage unit; generating real-time and time-varying statistical information for each vehicle camping area based on the vehicle information and the evacuee information; A method having the following.

8. Car camping evacuation support system a vehicle storage unit that stores vehicle information including attribute information of the vehicle being parked overnight, a vehicle camping area ID that identifies the vehicle camping area where the vehicle is parked, and the date and time of arrival at the vehicle camping area; a refugee storage unit that stores refugee information including attribute information of the refugees sleeping in the vehicle who are sleeping in the vehicle; a statistical information generating unit that generates real-time and time-series statistical information for each vehicle camping area based on the vehicle information and the evacuee information; A program that functions as a

Images