JP7425663B2 - Disaster victim estimation device, disaster victim estimation method, and disaster victim estimation system - Google Patents

Description

The present invention relates to a disaster victim estimation device, a disaster victim estimation method, and a disaster victim estimation system.

In recent years, various large-scale disasters that were previously unforeseen have occurred, and the way in which support should be provided to disaster victims is being reconsidered. In particular, understanding the number of disaster victims has become an important issue for the national and local governments that support their lives by accepting them into facilities.

Techniques for estimating the number of disaster victims include, for example, a storage means for storing predicted disaster area information indicating the predicted disaster area that is expected to be affected by the disaster, and a storage means for storing predicted disaster area information that is specified based on the predicted disaster area information. means for acquiring the number of mobile communication devices existing in the disaster area; prediction means for predicting the number of people present in the expected disaster area based on the acquired number of mobile communication devices; and the prediction means A system for predicting the number of people in a disaster area (see Patent Document 1) has been proposed.

In this conventional technology, immediately after a disaster occurs, the number of mobile phones is obtained from each mobile phone base station covering the disaster area, and this is divided by the mobile phone penetration rate to calculate the number of people in the disaster area. Estimate.

Japanese Patent Application Publication No. 2007-265115

However, the above-mentioned conventional technology cannot cope with a situation where communication in the mobile phone network is interrupted. In the event of a large-scale disaster, damage may also be caused to base stations of mobile phone networks, and there may be areas where communication with the disaster site is disrupted.

Even if the conventional technology is applied under such circumstances, there is a high possibility that the number of mobile phones cannot be obtained appropriately. In that case, it is also difficult to estimate the number of disaster victims, which changes from moment to moment.

Therefore, an object of the present invention is to provide a technique that makes it possible to estimate the number of disaster victims even if the communication situation at the disaster site is poor.

A disaster victim estimation device of the present invention that solves the above problems includes information on a disaster and the damage situation caused by the disaster, model information that defines correspondence relationships between the disaster and the disaster situation in a plurality of cases, and the number of victims of the disaster. a calculation formula for estimating the scale using the information on the damage situation as a variable; A process of applying the new information to a judgment rule to determine the type of the new information, and determining which model information of the plurality of cases matches the new information according to a predetermined processing method according to the type. Then, as model information to be used later regarding the disaster, a process of identifying the model information found in the determination, estimating the disaster situation based on the identified model information, and applying the information on the disaster situation to the calculation formula. The present invention is characterized in that it includes a process for estimating the scale of disaster victims, and a calculation device that executes the process.

In addition, in the disaster victim estimation method of the present invention, the disaster victim estimation device collects information on a disaster and the damage situation caused by the disaster, model information defining correspondence relationships between the disaster and the disaster damage situation in a plurality of cases, and the disaster victim estimation device. a calculation formula for estimating the number of victims of the disaster using the information on the damage situation as a variable, and when new information is obtained from the information providing device regarding the information on the damage situation of the disaster. , a process of applying the new information to a predetermined judgment rule to determine the type of the new information, and matching the new information among the model information of the plurality of cases according to a predetermined processing method according to the type. The process includes determining the model information found in the determination as model information to be used later regarding the disaster, estimating the disaster situation based on the identified model information, and using the information on the disaster situation in the calculation. The present invention is characterized by executing the following steps: estimating the scale of disaster victims by applying a formula to the formula.

In addition, the disaster victim estimation system of the present invention includes information on a disaster and the damage situation caused by the disaster, model information that defines the correspondence relationship between the disaster and the disaster situation in a plurality of cases, and the scale of the victims of the disaster. a calculation formula for estimating the disaster situation information as a variable; and a storage device that stores a calculation formula for estimating the disaster situation information as a variable; A process of applying the new information to determine the type of the new information, and determining which of the model information of the plurality of cases matches the new information according to a predetermined processing method according to the type; As the model information to be used later regarding the disaster, the process of identifying the model information found in the above judgment, estimating the disaster situation based on the identified model information, and applying the information on the disaster situation to the above calculation formula to identify the disaster victims. The present invention is characterized in that it includes a disaster victim estimation device that includes a process for estimating the scale of a disaster, and an arithmetic device that performs the process.

According to the present invention, it is possible to estimate the number of disaster victims even if the communication situation at the disaster site is poor.

FIG. 1 is a diagram showing an example of the configuration of a disaster victim estimation system according to the present embodiment. FIG. 3 is a diagram showing an example of the data structure of a mobile phone information table according to the present embodiment. It is a figure showing an example of data composition of a seismic intensity information table of this embodiment. It is a figure showing an example of data composition of a building damage information table of this embodiment. It is a figure showing an example of data composition of a water outage rate table of this embodiment. It is a figure showing an example of data composition of a power outage rate table of this embodiment. FIG. 3 is a diagram showing an example of the data structure of a poverty degree table according to the present embodiment. FIG. 3 is a diagram illustrating an example of a data configuration based on a return home rate table according to the present embodiment. It is a figure showing an example of a data composition of a disaster victim estimation table of this embodiment. It is a figure showing an example of a data composition of an estimate table of this embodiment. It is a figure showing an example of a flow of a disaster victim estimation method of this embodiment. It is a figure showing an example of a flow of a disaster victim estimation method of this embodiment. It is a conceptual diagram showing the relationship between unupdated data/updated data in observed/estimated data of this embodiment. It is a figure showing an example of a flow of a disaster victim estimation method of this embodiment. FIG. 2 is a conceptual diagram showing a calculation model setting method in this embodiment. FIG. 2 is a conceptual diagram showing a calculation model setting method in this embodiment. It is a figure which shows the estimation result screen of the required material quantity in this embodiment. It is a figure which shows the estimation result screen of the required material quantity in this embodiment. It is a figure which shows the estimation result screen of the required material quantity in this embodiment.

<<System configuration example>>
EMBODIMENT OF THE INVENTION Hereinafter, the form for implementing this invention is demonstrated using FIG. 1 thru|or FIG. 10C. This example is an example of collecting various information in a disaster area, combining and analyzing that information, estimating the number of disaster victims in the disaster area, and supporting victim support activities when a disaster occurs. shows. However, the present invention is not limited to this embodiment, and the present invention is applicable to any situation in which the number of disaster victims is estimated.

FIG. 1 shows a configuration example of a disaster victim estimation system 10 in this embodiment. The disaster victim estimation system 10 in this embodiment has a configuration in which a disaster victim estimation device 100, a disaster countermeasure headquarters system 160, a disaster information sharing system 170, and a data provision system 180 are communicably connected via communication lines 150A and 150B. It becomes.

The disaster victim estimation device 100, which mainly constitutes the disaster victim estimation system 10, includes a central control device 140, a main storage device 110, an auxiliary storage device 130, a communication device 141, and an input/output device 142. These are interconnected by a bus.

Among these, the auxiliary storage device 130 stores an observation/estimate database 131, a model database 132, and a disaster victim number estimation result database 133.

On the other hand, the main storage device 110 stores the following programs. That is, they are an observed/estimated data acquisition section 111, an observed/estimated data reading section 112, a model reading section 113, a parameter calculation section 114, and a disaster victim number estimation section 115.

Hereinafter, when the subject is written as "○○ part", the central controller 140 reads each program from the auxiliary storage device 130, loads it into the main storage device 110, and then writes the function of each program (details will be described later). This applies to those who have achieved this.

Here, the disaster response headquarters system 160, the disaster information sharing system 170, and the data providing system 180 are general computers.

The disaster response headquarters system 160 is a system that collects and analyzes the situation in disaster-affected areas and provides support for the government, local governments, etc. to take countermeasures. This disaster response headquarters system 160 can be assumed to be operated by a local government in a disaster-stricken area that takes the initiative in taking countermeasures, or operated within a government in the case of a large-scale disaster that crosses regions.

Although the system operates as a separate system for each local government and each government agency in charge, it may also be a system shared by local governments in multiple regions. Alternatively, it may be a system shared by multiple ministries and agencies. Alternatively, the system may be shared between one or more local governments and one or more ministries and agencies.

The disaster information sharing system 170 is a system that provides the disaster victim estimation device 100 with disaster damage information. Disaster damage information includes information such as the state of collapsed buildings and the state of road damage.

Further, the data providing system 180 is a system that provides the disaster victim estimation device 100 with information useful for disaster response among information managed by private companies. Information useful for disaster response may include, for example, information on the number of mobile phone subscribers compiled by region.

On the other hand, the disaster victim estimation device 100 communicates with the disaster control headquarters system 160 and the disaster information sharing system 170 via the communication line 150A. Further, the disaster victim estimation device 100 communicates with a disaster countermeasure headquarters system 160, a disaster information sharing system 170, and a data providing system 180 via a communication line 150B.

These communication lines 150A and 150B include LAN (Local Area Network).
ork), various networks such as a dedicated line, a WAN (Wide Area Network), a power line network, a wireless network, a public line network, a mobile phone network, and a satellite communication line can be employed.

Note that the communication line 150A is assumed to be a communication line within an administrative body such as a government or a local government. Therefore, when using a communication line open to the public outside the government, such as a public line network, it is assumed that the line is pseudo-dedicated using VPN (Virtual Private Network) technology.

Furthermore, the communication line 150B is assumed to be a communication line open to the public outside the government. Therefore, only when the communication contents are to be kept secret, a pseudo-dedicated line is created using the above-mentioned VPN technology or the like.

In addition, in FIG. 1, the disaster victim estimation device 100 performs input and output independently, but it is connected to other terminals via the communication line 150A or 150B, and inputs and outputs information using the input/output device of the terminal. Good as a thing.

Further, the disaster victim estimation device 100 may be a subsystem within the disaster countermeasure headquarters system 160 or the disaster information sharing system 170.

<<Database configuration>>
Here, details of the database held by the disaster victim estimation device 100 will be explained. First, FIGS. 2A to 2C show examples of each table included in the observation/estimation database 131.

The observation/estimation database 131 stores data observed in the disaster area and data estimated based on the data observed in the disaster area.

The observation/estimation database 131 includes, for example, a mobile phone information table 200 (FIG. 2A), a seismic intensity information table 210 (FIG. 2B), and a building damage information table 220 (FIG. 2C).

Among these, the mobile phone information table 200 stores data regarding the number of people staying in the disaster area, which is estimated based on the position information and number of mobile phones observed in the disaster area. The mobile phone information table 200 stores data distributed from the data providing system 180. In this case, the operator of the data providing system 180 is a mobile phone company.

The mobile phone information table 200 includes date and time 201, area ID 202, number of people 203, area resident ratio 204, age 205, and gender 206.

The keys for identifying each record are the date and time 201 and the area ID 202. This date and time 201 indicates the date and time when the mobile phone information was observed. On the other hand, the region ID 202 is an identifier for uniquely identifying the region where the mobile phone information was observed.

Further, among the values associated with the above-mentioned key, the number of people 203 is the number of people staying in the area, estimated based on the location information and number of mobile phones observed with the area ID 202 at the date and time 201.

Further, the local resident ratio 204 indicates the ratio of residents living in the disaster area among the number of people 203 estimated as described above. Whether a person is a resident of the disaster-stricken area is determined based on the address of the mobile phone subscriber. However, if it is unknown, it may be determined from the statistical information of the mobile phone subscriber.

Furthermore, age 205 and gender 206 may be added as keys for identifying each record. The age 205 is one of the attributes related to the estimated number of people, and is an identifier for each age group, such as whether the estimated person is in his or her teens or twenties.

Furthermore, gender 206 is one of the attributes related to the estimated number of people, and is an identifier for identifying whether the estimated person is male or female.

Next, in the seismic intensity information table 210 illustrated in FIG. 2B, data regarding the seismic intensity of an earthquake is stored. This seismic intensity information table 210 stores, for example, data distributed from the disaster information sharing system 170.

The seismic intensity information table 210 includes an occurrence date and time 211, a region ID 212, and an earthquake intensity 213. The keys for identifying each record are the date and time of occurrence 211 and the region ID 212.

Of these, the occurrence date and time 211 indicates the date and time when the earthquake was observed. Moreover, the region ID 212 is an identifier for uniquely identifying the region where seismic intensity information was observed. Seismic intensity 213 indicates seismic intensity information of the observed earthquake.

Further, in the building damage information table 220 illustrated in FIG. 2C, data regarding building damage caused by an earthquake is stored.

This building damage information table 220 includes observation/estimated date and time 221, area ID 222, number of completely destroyed buildings 223, number of partially destroyed buildings 224, and number of other buildings 225.

The keys for identifying each record are the observation/estimated date and time 221 and the area ID 222. The observed/estimated date and time 221 indicates the date and time when building damage information was observed or estimated. The region ID 222 is an identifier for uniquely identifying the region where building damage information was observed or estimated.

Further, the number of completely destroyed houses 223 is a numerical value indicating the number of completely destroyed houses in the area ID 222. The number of partially destroyed houses 224 is a numerical value indicating the number of partially destroyed houses in the area ID 222. The number of other houses 225 is a numerical value indicating the number of houses in the area ID 222 that were not completely or partially destroyed and survived with only minor damage.

Note that the above-mentioned total of the number of completely destroyed houses 223, the number of partially destroyed houses 224, and the number of other houses 225 matches the number of all houses in the area ID.

Next, FIGS. 3A to 3E show configuration examples of each table stored in the model database 132.

In the model database 132, calculation models necessary for estimating the number of disaster victims are stored. The model database 132 includes a water outage rate table 300 (FIG. 3A), a power outage rate table 310 (FIG. 3B), a hardship level table 320 (FIG. 3C), a return home rate table 330 (FIG. 3D), and a disaster victim estimation table. 340 (FIG. 3E).

Among these, a water outage rate table 300 illustrated in FIG. 3A stores a calculation model for estimating the rate of water outage in residences in the disaster area. This is a calculation model defined by the user based on existing knowledge such as past disaster experience and related literature.

The water outage rate table 300 includes a model ID 301, a seismic intensity 302, and a water outage rate 303. The key for identifying each record is the model ID 301. The model ID 301 is an identifier for uniquely identifying the calculation model for estimating the water cutoff rate.

The seismic intensity 302 indicates information regarding the standard seismic intensity to which the calculation model is originally applied. On the other hand, the water outage rate 303 is a calculation model for estimating what value the water outage rate will show after the number of days t has passed since the disaster occurred. In this example, the calculation model is a mathematical formula that is generated for each seismic intensity and shows the time course of the water outage rate.

Further, the power outage rate table 310 illustrated in FIG. 3B stores a calculation model for estimating the rate of power outage in residences in the disaster area. This is a calculation model defined by the user based on existing knowledge such as past disaster experience and related literature.

Such a power outage rate table 310 includes a model ID 311, a seismic intensity 312, and a power outage rate 313. The key for identifying each record is the model ID 311. This model ID 311 is an identifier for uniquely identifying the calculation model for estimating the power outage rate. Furthermore, the seismic intensity 312 indicates information regarding the standard seismic intensity to which the calculation model is originally applied.

On the other hand, the power outage rate 313 is a calculation model for estimating what value the power outage rate will show after the number of days t has passed since the disaster occurred. In this example, the calculation model is a mathematical formula that is generated for each seismic intensity and shows the time course of the power outage rate.

In addition, the living hardship level table 320 illustrated in FIG. 3C is a calculation model for estimating the proportion of residents in the disaster area who will be forced to evacuate to evacuation centers due to water outages and power outages. is memorized. This is a calculation model defined by the user based on existing knowledge such as past disaster experience and related literature.

The poverty level table 320 includes a model ID 321, a type 322, and a poverty level 323. The keys for identifying each record are the model ID 321 and the type 322.

Among these models, model ID321 is a calculation model for estimating the degree of living hardship, which is an indicator of the proportion of residents in disaster-stricken areas who are forced to evacuate to evacuation centers due to water outages and power outages. This is an identifier for uniquely identifying.

Further, the type 322 is an identifier for identifying whether the calculation model is a model based on a water outage or a power outage.

On the other hand, the poverty level 323 is a calculation model for estimating what value the poverty level will show after the number of days t has passed since the disaster occurred. In this example, the calculation model is the above-mentioned type 322, that is, a mathematical expression that is generated for each event such as a water outage or a power outage and shows the degree of living hardship over time.

Next, a returning home rate table 330 illustrated in FIG. 3D stores a calculation model for estimating the rate at which residents who do not live in the disaster area (non-local residents) can return to their homes. . This is a calculation model defined by the user based on existing knowledge such as past disaster experience and related literature.

The returning home rate table 330 is composed of a model ID 331 and a returning home rate 332. The key for identifying each record is the model ID 331.

This model ID 331 is an identifier for uniquely identifying a calculation model for estimating the rate at which residents who do not live in the disaster area (non-local residents) can return to their homes.

On the other hand, the returning home rate 332 is a calculation model for estimating what value the returning home rate will show after the number of days t has passed since the disaster occurred.

Further, the disaster victim estimation table 340 illustrated in FIG. 3E stores a calculation model for estimating the number of disaster victims in the disaster area. This is a calculation model defined by the user based on existing knowledge such as past disaster experience and related literature.

Such a disaster victim estimation table 340 is composed of a model ID 341 and a disaster victim estimation value 342. The key for identifying each record is the model ID 341.

This model ID 341 is an identifier for uniquely identifying a calculation model for estimating the number of disaster victims in a disaster area.

On the other hand, the disaster victim estimate 342 is a calculation model for estimating the number of disaster victims in the disaster area when the number of days t has passed since the disaster occurred.

This calculation model includes the number of non-local residents staying in each region, the number of buildings damaged due to the earthquake, such as completely destroyed or partially destroyed, the rate of water outage and power outage in the affected area, the degree of living hardship in the affected area, and the number of non-local residents living in the affected area. It stores calculation models that estimate the number of disaster victims in disaster-stricken areas based on parameters such as return rate, which indicates how many people are able to return home.

The above-mentioned parameters include a mobile phone information table 200 that shows the number of people staying in each area, which is estimated based on the location information of mobile phones and the number of mobile phones, a seismic intensity information table 210 that shows earthquake seismic intensity information, and building damage information due to the earthquake. A building damage information table 220 showing the power outage, a water outage rate table 300 showing a calculation model for estimating the percentage of residences in the disaster area that will experience water outages, and a calculation for estimating the percentage of residences in the disaster area that will experience power outages. A power outage rate table 310 showing a model, and a hardship level table showing a calculation model for estimating the proportion of residents in the disaster area who will be forced to evacuate to evacuation centers due to water outages and power outages. 320, return home rate table 330 showing a calculation model for estimating the percentage of residents who do not live in the disaster area (non-local residents) who are able to return to their homes. is a parameter.

Next, FIG. 4 shows an example of the disaster victim number estimation result database 133. This disaster victim number estimation result database 133 is composed of an estimation table 400. This estimation table 400 stores data regarding the estimation results of the number of disaster victims in the disaster area.

The estimation table 400 includes date 401, area ID 402, number of disaster victims 403, age 404, and gender 405.

The keys for identifying each record are the date 401 and the region ID 402. The date 401 indicates the expected date of occurrence of the estimated data. The region ID 402 is an identifier for uniquely identifying the estimated region of the estimated data.

On the other hand, the number of disaster victims 403 indicates the number of disaster victims expected to occur in the area ID 402 on the date 401. Furthermore, age 404 and gender 405 may be added as keys for identifying each record.

The age 404 is one of the attributes related to the estimated number of people, and is an identifier for each age group, such as whether the estimated person is in their teens or 20s.

Furthermore, gender 405 is one of the attributes related to the estimated number of people, and is an identifier for identifying whether the estimated person is male or female.

The more attributes related to the estimated number of people, the more detailed the estimation becomes, and the wider the uses of the estimation result of the number of disaster victims. For example, by estimating the number of women, it is possible to use it to estimate the number of sanitary napkins, which are necessary items in the event of a disaster.

<<Flow of disaster victim estimation method: At the time of disaster>>
Next, a series of processes corresponding to the disaster victim estimation method in this embodiment will be explained. FIG. 5 shows the process by which the disaster victim estimation device 100 detects the occurrence of a disaster and starts estimating the number of disaster victims, and FIG. 6 shows the process for estimating the number of disaster victims.

In addition, the disaster victim estimation device 100 determines the updating method to be adopted for information and models for estimating the number of disaster victims due to a partial update of the observation/estimation database 131 until the start of estimating the number of disaster victims. The process is shown in FIG.

Note that the process of registering the calculation model in the model database 132 is performed before starting the series of processes (processes S500 to S503, processes S600 to S606, and processes S800 to S808) shown in FIGS. This is done by the user.

FIG. 5 shows a processing flow from when the disaster victim estimation device 100 detects the occurrence of a disaster and starts estimating the number of disaster victims. The disaster illustrated here is assumed to be an earthquake as an example, but is not limited to this.

In process S500, the central control unit 140 of the disaster victim estimation device 100 starts a process flow for disaster detection, and proceeds to process S501. The trigger for this start may be a case specified by the user of the disaster victim estimation device 100 (a case where the seismic intensity is above a threshold value, a case where the amount of precipitation is above a threshold value, a combination of these, etc.), a case where a predetermined time has arrived, etc. , various assumptions can be made (the same applies hereafter).

In process S501, the observed/estimated data acquisition unit 111 collects seismic intensity information at the time of a disaster from the disaster information sharing system 170 via the communication device 141, and proceeds to process S502.

At this time, the observed/estimated data acquisition unit 111 stores the seismic intensity information collected above in the seismic intensity information table 210 of the observed/estimated database 131.

Note that when storing seismic intensity information in the seismic intensity information table 210, the date and time when the seismic intensity information of the earthquake was observed is stored in the occurrence date and time 211, the identifier of the area where the seismic intensity information was observed is stored in the area ID 212, and the observation The seismic intensity information of the earthquake that occurred is stored in the seismic intensity 213.

Additionally, along with this, the observed/estimated data acquisition unit 111 collects building damage information from the disaster information sharing system 170 via the communication device 141, and stores this in the building damage information table 220.

When storing the above-mentioned building damage information in the building damage information table 220, the date and time when the building damage was estimated is stored in the observed/estimated date and time 221, and the identifier of the area where the building damage was estimated is stored in the area ID 222. , the number of completely destroyed houses and the number of partially destroyed houses estimated as building damage are stored in the number of completely destroyed houses 223 and the number of partially destroyed houses 224.

Further, the number of houses obtained by subtracting the values of the number of completely destroyed houses 223 and the number of partially destroyed houses 224 from all the numbers of houses obtained in advance is stored in the number of other houses 225.

Subsequently, in process S502, the observed/estimated data acquisition unit 111 collects mobile phone information immediately before the disaster occurs from the data providing system 180 via the communication device 141, and the process proceeds to process S503.

Note that the observed/estimated data acquisition unit 111 stores the mobile phone information collected here in the mobile phone information table 200 of the observed/estimated database 131.

Note that when storing the above mobile phone information in the mobile phone information table 200, the date and time when the mobile phone information was observed is stored in the date and time 201, and the identifier of the area where the mobile phone information was observed is stored in the area ID 202. .

Additionally, the number of people staying in the area, estimated based on the observed location information and number of mobile phones, is stored in the number of people 203, and the ratio of residents living in the disaster area is stored in the area resident ratio 204. shall be taken as a thing.

Further, at this time, when an identifier for each age group such as whether the person is in their teens or 20s is acquired as one of the attributes regarding the number of people 203, information corresponding to the identifier is stored in the age 205.

Further, when an identifier indicating whether the user is male or female is acquired as one of the other attributes, information corresponding to the identifier is stored in the gender 206.

Here, if it is difficult to obtain mobile phone information immediately before the disaster from the data providing system 180 after the disaster occurs, the disaster victim estimation device 100 periodically collects the mobile phone information from the data providing system 180 before the disaster occurs. It is also possible to acquire this information and store it in the mobile phone information table 200 of the observation/estimate database 131.

In that case, it is preferable for the disaster victim estimation device 100 to use the information in the mobile phone information table 200 that was stored immediately after the disaster occurred. Alternatively, the disaster victim estimation device 100 may acquire and use mobile phone information from the mobile phone information table 200 whose date and time when the mobile phone information was periodically acquired and stored is similar to the date and time at the time of the disaster. good.

Note that the determination of whether the dates and times of the mobile phone information are similar may be made from the viewpoint of whether it is a weekday or a weekend or holiday, whether it is morning, noon, or night, whether the seasons are the same, etc.

Subsequently, in step S503, the central processing unit 140 of the disaster victim estimation device 100 ends the processing flow from detecting the occurrence of a disaster through the above-described procedure to starting estimation of the number of disaster victims.

<<Flow of disaster victim estimation method: Estimating the number of disaster victims>>
Next, we will explain the flow of estimating the number of disaster victims. FIG. 6 is a diagram showing a flow of estimating the number of disaster victims in this embodiment.

First, in process S600, the central control device 140 of the disaster victim estimation system 10 starts a process flow for estimating the number of disaster victims. The process advances to processing S601.

Subsequently, in process S601, the observed/estimated data reading unit 112 reads mobile phone information from the mobile phone information table 200, seismic intensity information from the seismic intensity information table 210, and building damage information from the building damage information table 220. Read and proceed to processing S602.

In addition, in process S602, the model reading unit 113 reads respective calculation models from the water outage rate table 300, the power outage rate table 310, the degree of living hardship table 320, the return home rate table 330, and the disaster victim estimation table 340, and processes them. Proceed to S603.

The calculation models read in the above-mentioned process S602 are a calculation model of the water cutoff rate from the water cutoff rate table 300, a calculation model of the power outage rate from the power outage rate table 310, and a calculation model of the degree of living hardship (disaster victim) from the poverty level table 320. From the return home rate table 330, it is calculated how many local residents will be forced to evacuate to evacuation centers due to water outages and power outages. ) is a calculation model for estimating the percentage of people who can return to their homes, and a calculation model for estimating the number of disaster victims from the disaster victim estimation table 340.

Subsequently, in process S603, the parameter calculation unit 114 calculates parameters necessary for estimating disaster victims using the calculation model obtained in process S602 described above, and the process proceeds to process S604.

In this calculation, for the water outage rate, which is a necessary parameter for estimating disaster victims, among the calculation models for water outage rate, select the calculation model for water outage rate 303, which is indicated by the record of seismic intensity 302, which matches the seismic intensity 213 of the earthquake that occurred. Then, the calculation is performed by applying the number of days t that has passed since the disaster occurred to this calculation model.

Similarly, for the power outage rate, which is a necessary parameter for estimating disaster victims, we selected the power outage rate calculation model of 313, which is indicated by the record of seismic intensity 312, which matches the seismic intensity 213 of the earthquake that occurred, among the power outage rate calculation models. , is calculated by applying the number of days t that has passed since the disaster occurred to this calculation model.

In addition, among the degree of hardship, the degree of hardship related to water outage is calculated by applying the number of days t that has passed since the disaster occurred to the calculation model of the degree of hardship related to water outage. Similarly, the degree of hardship related to a power outage is calculated by applying the number of days t that has passed since the disaster occurred to the calculation model for the degree of hardship related to a power outage.

In addition, the parameter calculation unit 114 calculates the number of local residents calculated by multiplying the number of people 203 in the area by the local resident ratio 204 by the total of the number of completely destroyed houses 223, the number of partially destroyed houses 224, and the number of other houses 225, and calculates the number of local residents per house. Calculate the number of residents.

Further, the parameter calculation unit 114 calculates the number of non-local residents (the number of residents who do not live in the disaster area) by subtracting the value obtained by multiplying the number of people 203 in the area by the local resident ratio 204 from the number of people 203.

Subsequently, in process S604, the disaster victim number estimating unit 115 estimates the number of disaster victims, and the process proceeds to process S605.

In this process, each data of mobile phone information, seismic intensity information, and building damage information obtained in the above-mentioned process S601 is used, as well as the water cutoff rate, degree of living hardship related to water outage, power outage rate, and degree of living hardship related to power outage calculated in process S603. , the number of local residents per house, and the number of non-local residents per house are applied to the calculation model for estimating the number of disaster victims obtained in step S602 to estimate the number of disaster victims, and the process proceeds to step S605.

Subsequently, in step S605, the disaster victim number estimating unit 115 outputs the number of disaster victims estimated in the above-described step S604, and the process proceeds to step S606.

In this case, the disaster victim number estimation unit 115 outputs the disaster victim number obtained in the above-described process S604 to the estimation table of the disaster victim number estimation result database 133. Alternatively, the number of disaster victims may be output to the input/output device 142. Furthermore, the number of disaster victims may be transmitted to the disaster response headquarters system 160 via the communication device 141.

In process S606, the central processing unit 140 ends the process flow for estimating the number of disaster victims.

<<Data update concept after disaster>>
Next, based on FIG. 7, a concept of a process of receiving part of the information of the observation/estimate database 131 after a disaster occurs and gradually increasing the area in which the main observation/estimate database 131 is updated will be explained.

In FIG. 7, the horizontal axis shows the number of days that have passed since the disaster occurred. 700 immediately after the disaster occurs, 710 after m days have elapsed, and 720 after m+n days have elapsed, showing how the situation in the disaster area changes over time.

In a large-scale disaster, communications are disrupted in most of the affected area, and there are cases where the disaster victim estimation system 10 cannot receive observed data or estimated data based on it immediately after the disaster occurs.

Therefore, in most of the disaster-stricken areas 701, the number of victims is estimated using observation data before the disaster or estimated data based on it. Thereafter, after m days have elapsed 710, communications are restored in some areas 712 in the disaster area, and the disaster victim estimation system 10 becomes able to receive observed data or estimated data based thereon in the area.

Therefore, in region 712, the number of disaster victims is estimated using newly received observation data or estimated data based on it. On the other hand, communications remain disrupted in most areas 711.

Therefore, the number of victims is estimated using observed data before the disaster or estimated data based on it. Further later, after m+n days have elapsed 720, communication is restored in many areas 722 in the disaster area, and the disaster victim estimation system 10 can receive observation data or estimated data based on it in the area.

Therefore, in region 722, the number of disaster victims is estimated using newly received observation data or estimated data based on it.

On the other hand, since communications in area 721 remain disrupted, the number of victims will be estimated using observation data before the disaster or estimated data based on it.

In this situation, where the area where observation data or estimated data based on it can be received is gradually expanding, the area where observation and estimation data can be obtained will gradually increase after a disaster occurs. Processing is performed to gradually increase the number of regions to be updated.

<<Flow of disaster victim estimation method: At the time of update>>
FIG. 8 is a diagram showing a processing flow for updating data in the observation/estimation database 131 after a disaster occurs. This flow is a specific processing flow of the processing shown in the conceptual diagram of FIG.

First, in process S800, the central control unit 140 of the disaster victim estimation system 10 starts a process flow for updating observed and estimated data, and proceeds to process S801.

Subsequently, in process S801, the observed/estimated data acquisition unit 111 receives observed/estimated data from the disaster information sharing system 170 and the data providing system 180, and detects an update of the data. The process advances to processing S802. This observed/estimated data becomes the values of each table configuring the observed/estimated database 131.

Subsequently, in process S802, the observed/estimated data acquisition unit 111 determines the type of data received in process S801 described above. The type of data is determined based on the format of the received data, the communication procedure with the system sending the data, the flag added to the data by the data sending system, etc. It can be assumed. Of course, other methods of determination can also be used, and the present invention is not limited to the above example.

As a result of the above determination, if the update data is mobile phone information (S802: mobile phone), the observed/estimated data acquisition unit 111 proceeds to processing S803.

On the other hand, if the updated data is building damage information (S802: building damage), the observed/estimated data acquisition unit 111 proceeds to processing S804.

Further, if the updated data is the water outage rate (S802: water outage rate), the observed/estimated data acquisition unit 111 proceeds to processing S805.

Further, if the updated data is the power outage rate (S802: power outage rate), the observed/estimated data acquisition unit 111 proceeds to processing S806.

In step S803, the parameter calculation unit 114 aggregates the mobile phone information over a certain period of time and sets the aggregated data as update data for the mobile phone information table 200.

Here, the reason for estimating over a fixed period of time is that it is believed that in disaster-stricken areas, victims tend not to keep their mobile phones on all the time after a disaster occurs, due to their psychology of worrying about battery consumption.

Therefore, by totaling mobile phone information for a certain period of time, such as over one day, the total number of people based on the mobile phone information in the area is corrected. At this time, if there is any overlap in the mobile phone information, the overlap is removed and the process advances to step S807.

Further, in step S804, the parameter calculation unit 114 sets the observed data of building damage information as update data of the building damage information table 220.

Here, the reason why the observed data itself is used as updated data instead of correcting it is that the number of completely or partially destroyed houses is not thought to change significantly in a short period of time after a disaster occurs. Therefore, the observation data itself is set as update data and the process advances to step S807.

Subsequently, in step S805, the parameter calculation unit 114 changes the calculation model for estimating the water outage rate based on the observed data on the water outage rate. A conceptual diagram regarding changes in water cutoff rate is shown in FIG. 9A.

In the graph shown in FIG. 9A, the horizontal axis indicates the number of elapsed days (t) 901, and the vertical axis indicates the water cutoff rate 902. Here, when observation data 903 is received, the number of elapsed days ta0 in this observation data 903 is used to substitute the number of elapsed days ta0 into the calculation models 904 to 907 at each seismic intensity to determine the water outage rate of each calculation model.

After that, among the respective water cutoff rates, the water cutoff rate closest to the water cutoff rate Ra0 of the observed data 903 is searched, a calculation model for calculating the water cutoff rate is specified, and the calculation model is set as the updated model after the change. . For example, the water outage rate Ra5 determined by the calculation model 905 for seismic intensity 5 is closest to the water outage rate Ra0 of the observation data 903, not the water outage rate Ra4 calculated by the calculation model 904 for seismic intensity 4, so after changing the calculation model 905 for seismic intensity 5. is set as the calculation model, and the process advances to step S807.

Further, in step S806, the parameter calculation unit 114 changes the calculation model for estimating the power outage rate based on the observed data of the power outage rate. A conceptual diagram regarding this change in power outage rate is shown in FIG. 9B.

In the graph shown in FIG. 9B, the horizontal axis shows the number of elapsed days (t) 911, and the vertical axis shows the power outage rate 912. Here, when observation data 913 is received, using the number of elapsed days tb0 in this observation data 913, the number of elapsed days tb0 is substituted into the calculation models 914 to 917 for each seismic intensity to determine the power outage rate of each calculation model.

After that, out of each power outage rate, the power outage rate closest to the power outage rate Rb0 of the observed data 913 is searched, a calculation model for calculating the power outage rate is specified, and the calculation model is set as the updated model after the change. .

For example, assume that the power outage rate Rb5 determined by the calculation model 915 for seismic intensity 5 is closest to the power outage rate Rb0 of the observed data 913, rather than the power outage rate Rb4 calculated by the calculation model 914 for seismic intensity 4. Therefore, the calculation model 915 for seismic intensity 5 is set as the changed calculation model, and the process advances to step S807.

In addition, in process S807, the parameter calculation unit 114 updates the mobile phone information table 200 if mobile phone information is set in process S803, and updates the building damage information if building damage information is set in process S804. The table 220 is updated, and if the water outage rate calculation model is set in step S805, the water outage rate table 300 is temporarily updated, and if the outage rate calculation model is set in S806, the outage rate table 310 is temporarily updated. is updated, and the process advances to step S808.

Furthermore, in step S808, the central processing unit 140 ends the processing flow for updating the data in the observation/estimation database 131.

Note that when the parameter calculation unit 114 changes the calculation model based on the observed data 903 in the above-mentioned process S805, it not only searches for the water cutoff rate Ra5 closest to the water cutoff rate Ra0 of the observed data 903 and its calculation model 905. , the second closest Ra4 and its calculation model 904 may also be searched, the calculation model 905 and the calculation model 904 may be averaged, and this may be set as the changed calculation model.

Here, in addition to simply averaging the calculation model 905 and the calculation model 904, the calculation model 905 and the calculation model 904 are weighted according to the length of Ra5 from Ra0 and the length of Ra4 from Ra0. It is also possible to average it after that.

Alternatively, the computation model 905 and the computation model 904 may be subjected to addition, subtraction, and multiplication, respectively, and then averaged. Similarly, in step S806, the calculation model 915 and the calculation model 914 may be averaged and set as the changed calculation model. Alternatively, the calculation model 915 and the calculation model 914 may be weighted and then averaged according to the length of Rb5 from Rb0 and the length of Rb4 from Rb0. Alternatively, the computation model 915 and the computation model 914 may be subjected to addition, subtraction, and multiplication, respectively, and then averaged.

Furthermore, in the graphs of FIGS. 9A and 9B, if the observed data is located in the middle of the model curve, etc., the water outage rate, power outage rate, etc. may be calculated using the values of the existing model according to the distance from the model curve. . The slope may be calculated when there are two or more observation data points, and the curvature may be calculated when there are three or more observation data points, and the water outage rate, power outage rate, etc. may be estimated accordingly. In addition, instead of using the observed data 913 in FIG. 9B, from the closest calculation model 905 obtained by searching in FIG. 9A and the second closest calculation model 904, the first of the practical seismic intensities of the calculation models is 5. Assuming that the second seismic intensity is 4, the power outage rate may be calculated by assuming that the nearest calculation model is 915 and the second closest calculation model is 914 in FIG. 9B. On the contrary, instead of using the observed data 903 in FIG. 9A, the actual seismic intensity of the calculation model is calculated from the closest calculation model 915 obtained by searching in FIG. 9B and the second closest calculation model 914. Assuming that the closest calculation model is 905 and the second closest calculation model is 904 in FIG. 9A, the water cutoff rate may be calculated.

<<About output examples>>
Next, an example of the estimated value output screen displayed on the input/output device 142 in step S605 of FIG. 6 is shown. Examples of output screens include a screen that outputs the value of the estimation table 400 of the disaster victim number estimation result database 133, a screen that outputs the number of disaster victims 403 in the estimation table 400 after re-aggregating it by prefecture, etc. , and a screen that outputs an estimate of the amount of supplies needed for each prefecture or other unit by multiplying it by the amount of supplies that one person is expected to need per day.

In any of the screen examples, a screen may be used to compare the estimation results before and after the observation/estimated data update process shown in FIG. 8 is performed.

Here, FIGS. 10A to 10C are images of example screens in which the quantities of necessary supplies are estimated for each prefecture.

In the projection screen for the amount of necessary supplies shown in FIG. 10A, the user sets in advance the quantity of supplies that one person needs per day as three meals of food, one blanket, and three sanitary products (for women). By multiplying this by the number of disaster victims, the amount of necessary supplies is estimated, and the result is displayed on this screen.

This screen is displayed on the input/output device 142 of the disaster victim estimation system 10. The required material quantity estimation screen shown in FIG. 10A is composed of an overview 1000 and estimation results 1001.

Of these, summary 1000 shows the summary of the estimation results, and in this example, it shows the "results estimated on September 1st, and the amount of supplies needed in the disaster area on September 15th." .

Estimation result 1001 shows details of the estimation result and is composed of region 1002 and item. In this example, the items are food 1003, blanket 1004, and sanitary products 1005, but other items may be used. Here, the results of estimating how much of each item (food 1003, blanket 1004, sanitary napkin 1005) is required for each region 1002 are shown.

In addition, if the estimation result shown in FIG. 10A has been updated after the observation/estimation data shown in FIG. 8 is updated, the required material quantity estimation screen shown in FIG. This is a screen that displays the estimation results before update processing in order to compare them with the previous estimation results.

The configuration of FIG. 10B is the same as that of FIG. 10A, and is composed of an overview 1010 and an estimation result 1011. The overview 1010 shows an overview of the estimation result, and the estimation result 1011 shows the region 1012 and items (in this example, food 1013, A blanket 1014 and sanitary products 1015).

Further, the required material amount estimation screen shown in FIG. 10C is a screen displayed on one screen to compare the estimation results before and after the observation/estimated data update process shown in FIG. 8 is performed.

The structure of FIG. 10C is also similar to that of FIG. 10A, and is composed of an overview 1020 and an estimation result 1021. , blanket 1024, and sanitary products 1025). Furthermore, in order to distinguish between before and after the update process, the estimation results before the update process are displayed in parentheses, and the estimation results after the update process are displayed without parentheses. Note that other display methods may be used as long as it is possible to distinguish before and after the update process.

Although the best mode for carrying out the present invention has been specifically described above, the present invention is not limited thereto and can be modified in various ways without departing from the gist thereof.

According to this embodiment, it is possible to estimate the number of disaster victims even if the communication situation at the disaster site is poor.

The description of this specification clarifies at least the following. That is, in the disaster victim estimation device of the present embodiment, the storage device defines, as the model information, a correspondence relationship between the disaster and the time course of the disaster situation starting from the time of occurrence of the disaster in a plurality of cases. In the process of estimating the scale of disaster victims, the computing device estimates the disaster situation by applying the estimation target period to the identified model information, and uses the information on the disaster situation in the calculation. It may be applied to the formula to estimate the scale of disaster victims.

According to this, it becomes possible to efficiently estimate the damage situation according to the elapsed period from the occurrence of the disaster. Furthermore, even if the communication situation at the disaster site is poor, it becomes possible to estimate the number of disaster victims more efficiently.

Furthermore, in the disaster victim estimation device of the present embodiment, in the process of specifying the model information, the calculation device calculates the value of the disaster situation at the observation period of the disaster situation indicated by the new information and the value of the disaster situation of the plurality of cases. Calculates the degree of separation between each piece of model information and the estimated value of the disaster situation at the observation period, and determines the one with the smallest degree of separation among the model information of the plurality of cases as matching the new information. It may be said that it is.

According to this, even if there is no model information that exactly matches the observed values of the disaster situation in the disaster-stricken area, it is possible to efficiently extract model information that is highly compatible and use it for subsequent estimation of the scale of disaster victims. Can be done. Furthermore, even if the communication situation at the disaster site is poor, it becomes possible to estimate the number of disaster victims more efficiently.

Further, in the disaster victim estimation device of the present embodiment, when the calculation device determines the type of the new information, the new information is of a type that changes depending on the behavioral tendency of the disaster victim. The information on the disaster situation to be applied to the calculation formula may be generated by aggregating the new information for a certain period of time.

According to this, it becomes possible to appropriately eliminate time-varying factors in the information on the disaster situation and use it as appropriate information for the formula for calculating the scale of disaster victims. Furthermore, even if the communication situation at the disaster site is poor, it will be possible to estimate the number of disaster victims more efficiently and with good accuracy.

Furthermore, in the disaster victim estimation device of the present embodiment, as a result of determining the type of the new information, the calculation device determines that the new information is of a type that is unlikely to change for a certain period of time from the occurrence of the disaster. In this case, the new information may be used as is as information on the disaster situation to be applied to the calculation formula.

According to this, by using information on the disaster situation as it is as application information for the calculation formula for the number of disaster victims, the number of victims can be estimated more efficiently and accurately even if the communication situation at the disaster site is poor. becomes possible.

Further, in the disaster victim estimation device of the present embodiment, the storage device holds a plurality of pieces of the model information for each type of disaster situation, and includes each of the information on the plurality of types of disaster situation as a variable in the calculation formula. , and the arithmetic device calculates, based on the disaster information indicated by the identified model information, information related to the corresponding disaster among the model information of other types of disaster situations held in the storage device. further execute a process of identifying the model information found in the determination as model information to be used thereafter regarding the other types of damage situations of the disaster, and when estimating the scale of disaster victims, each of the identified models Based on the information, each corresponding type of disaster situation may be estimated, and the information on the disaster situation may be applied to the calculation formula to estimate the scale of disaster victims.

According to this, disaster information (eg, seismic intensity) in one model information is used as identification material for other model information, and model information regarding each type of disaster situation can be efficiently identified. Furthermore, even if the communication situation at the disaster site is poor, it becomes possible to estimate the number of disaster victims more efficiently.

10 Victim Estimation System 100 Victim Estimation Device 110 Main Storage Device 111 Observation/Estimation Data Acquisition Unit 112 Observation/Estimation Data Reading Unit 113 Model Reading Unit 114 Parameter Calculation Unit 115 Number of Disaster Victims Estimation Unit 130 Auxiliary Storage Device 131 Observation/Estimation Database 132 Model database 133 Estimated number of disaster victims database 140 Central control device 141 Communication device 142 Input/output device 150A Communication line 150B Communication line
160 Disaster control headquarters system 170 Disaster information sharing system 180 Data provision system 200 Mobile phone information table 210 Seismic intensity information table 220 Building damage information table 300 Water outage rate table 310 Power outage rate table 320 Living hardship level table 330 Return home rate table 340 Victim estimation table 400 estimation table

Claims (8)

Information on the disaster and the damage situation caused by the disaster, model information that defines the correspondence between the disaster and the damage situation in multiple cases, and calculations for estimating the scale of the victims of the disaster using the information on the damage situation as a variable. a storage device for holding an expression;
When new information is obtained from the information providing device regarding the information on the damage situation of the disaster, a process of applying the new information to a predetermined judgment rule and determining the type of the new information, and a process of determining the type of the new information. determining which model information of the plurality of cases matches the new information according to a predetermined processing method, and identifying the model information found in the determination as model information to be used from now on regarding the disaster; , a calculation device that executes a process of estimating a disaster situation based on the identified model information and applying the information on the disaster situation to the calculation formula to estimate the scale of disaster victims;
A disaster victim estimation device characterized by including: The storage device is
As the model information, hold information that defines a plurality of cases of correspondence between the disaster and the time transition of the disaster situation starting from the time of occurrence of the disaster;
The arithmetic device is
In the process of estimating the scale of disaster victims, the period of estimation is applied to the identified model information to estimate the disaster situation, and the information on the disaster situation is applied to the calculation formula to estimate the scale of disaster victims. It is for estimating
The disaster victim estimation device according to claim 1, characterized in that: The arithmetic device is
In the process of specifying the model information, the distance between the value of the disaster situation at the observation period of the disaster situation indicated by the new information and the estimated value of the disaster situation at the observation period in each of the model information of the plurality of cases; calculating the degree of separation, and determining that among the model information of the plurality of cases, the one with the smallest degree of separation matches the new information,
3. The disaster victim estimation device according to claim 2. The arithmetic device is
As a result of determining the type of the new information, if the new information is of a type that changes depending on the behavioral trends of the disaster victims, the calculation formula can be applied by aggregating the new information for a certain period of time. generating information on the disaster situation to be applied;
The disaster victim estimation device according to claim 1, characterized in that: The arithmetic device is
As a result of determining the type of the new information, if the new information is of a type that is difficult to change for a certain period of time from the occurrence of the disaster, the new information is applied as is to the disaster situation in the calculation formula. The information shall be
The disaster victim estimation device according to claim 1, characterized in that: The storage device is
retaining a plurality of pieces of model information for each type of disaster situation, and retaining information on each of the plurality of types of disaster situations as variables in the calculation formula;
The arithmetic device is
Based on the information on the disaster indicated by the identified model information, determine which model information on the other types of disaster situations held in the storage device is related to the information on the corresponding disaster, and further performing a process of identifying the model information found in the determination as model information to be used later regarding the type of disaster situation;
When estimating the scale of disaster victims, each corresponding type of disaster situation is estimated based on the identified model information, and the information on the disaster situation is applied to the calculation formula to estimate the scale of disaster victims. ,
The disaster victim estimation device according to claim 1, characterized in that: The disaster victim estimation device is
Information on the disaster and the damage situation caused by the disaster, model information that defines the correspondence between the disaster and the damage situation in multiple cases, and calculations for estimating the scale of the victims of the disaster using the information on the damage situation as a variable. and a storage device for retaining the expression;
When new information is obtained from the information providing device regarding the information on the damage situation of the disaster, a process of applying the new information to a predetermined judgment rule and determining the type of the new information, and a process of determining the type of the new information. determining which model information of the plurality of cases matches the new information according to a predetermined processing method, and identifying the model information found in the determination as model information to be used from now on regarding the disaster; , a process of estimating the disaster situation based on the identified model information and applying the information on the disaster situation to the calculation formula to estimate the scale of disaster victims;
A disaster victim estimation method characterized by carrying out the following. Information on the disaster and the damage situation caused by the disaster, model information that defines the correspondence between the disaster and the damage situation in multiple cases, and calculations for estimating the scale of the victims of the disaster using the information on the damage situation as a variable. a storage device for holding an expression;
When new information is obtained from the information providing device regarding the information on the damage situation of the disaster, a process of applying the new information to a predetermined judgment rule and determining the type of the new information, and a process of determining the type of the new information. determining which model information of the plurality of cases matches the new information according to a predetermined processing method, and identifying the model information found in the determination as model information to be used thereafter regarding the disaster; , a calculation device that executes a process of estimating a disaster situation based on the identified model information and applying the information on the disaster situation to the calculation formula to estimate the scale of disaster victims;
A disaster victim estimation system comprising a disaster victim estimation device.

Images