Model generation device, model generation method, and program

Abstract

This model generation device (10) is provided with a control unit (13) that: acquires first facility data (21) indicating the characteristics of a first facility attached to a first bridge, first water level data (22) indicating the water level of a river in which the first bridge is provided, and inspection data (24) indicating the presence or absence of a disaster due to flooding in the first facility; and generates a prediction model by using the first facility data (21), the first water level data (22), and the inspection data (24), the prediction model predicting the possibility of disaster due to flooding in a second facility when second facility data (61) indicating the characteristics of a second facility attached to a second bridge and second water level data (62) indicating the water level of the river in which the second bridge is provided are input.

Description

Model generation device, model generation method, and program

[0001] The present disclosure relates to a model generation device, a model generation method, and a program.

[0002] It is known that bridge vulnerability can be quantitatively assessed by analyzing bridge damage caused by floods. For example, Non-Patent Document 1 describes a method for estimating the current risk of damage and future risk of scouring through multivariate analysis based on various parameters related to railway bridges obtained through routine inspection work and past damage data, and for identifying bridge piers that are at risk of damage when rivers flood. Non-Patent Document 2 also describes an index for classifying bridge damage based on a statistical analysis of damage to bridges along the Misasa River in Hiroshima Prefecture, where many bridges were damaged by heavy rains that occurred in July 2018.

[0003] Masahiko Samizo and three others, "Statistical Method for Identifying Bridge Piers at Risk of Damage During Flooding of Railway Bridges," Journal of the Japan Society of Civil Engineers, D3 (Civil Engineering Planning), Vol. 69, No. 13, 2013. Takashi Inoue and two others, "Study on Bridge Damage Risk Assessment Indicators During Heavy Rains - Using Longitudinal Distribution Data of Damage on the Misasa River Bridge in the 2018 Western Japan Heavy Rains," Journal of the Japan Society of Civil Engineers, B1 (Hydraulic Engineering), Vol. 75, No. 2, I_1417-I_1422, 2019.

[0004] As such, while it is known to analyze bridge damage caused by floods, it is not known to predict the possibility of damage to facilities attached to bridges.

[0005] The purpose of this disclosure, made in consideration of the above circumstances, is to make it possible to predict the possibility that equipment attached to a bridge will be damaged by flooding.

[0006] A model generation device according to one embodiment acquires first equipment data indicating the characteristics of a first equipment attached to a first bridge, first water level data indicating the water level of the river on which the first bridge is located, and inspection data indicating whether the first equipment has been damaged by flooding; and includes a control unit that, when second equipment data indicating the characteristics of a second equipment attached to a second bridge and second water level data indicating the water level of the river on which the second bridge is located are input, generates a prediction model that predicts the possibility of flood damage to the second equipment using the first equipment data, the first water level data, and the inspection data.

[0007] A model generation method according to one embodiment is a model generation method executed by a model generation device, and includes acquiring first equipment data indicating characteristics of a first equipment attached to a first bridge, first water level data indicating the water level of a river on which the first bridge is located, and inspection data indicating whether the first equipment has been damaged by flooding; and when second equipment data indicating characteristics of a second equipment attached to a second bridge and second water level data indicating the water level of the river on which the second bridge is located are input, generating a prediction model that predicts the possibility of flood damage to the second equipment using the first equipment data, the first water level data, and the inspection data.

[0008] A program according to one embodiment causes a computer to function as the model generating device.

[0009] According to the present disclosure, it becomes possible to predict the possibility that equipment attached to a bridge will be damaged by flooding.

[0010] 8 is a block diagram showing the configuration of a model generation device according to one embodiment. FIG. 9 is a flowchart showing the operation of the model generation device. FIG. 10 is a schematic diagram showing an example of a first pipe attached to a bridge. FIG. 11 is a diagram showing examples of first equipment data, first water level data, first rainfall data, and inspection data. FIG. 12 is a diagram showing dummy variables for each category of items included in the first equipment data, first water level data, first rainfall data, and inspection data shown in FIG. 4. FIG. 13 is a diagram showing an example of a damage coefficient. FIG. 14 is a block diagram showing the configuration of a prediction device according to one embodiment. FIG. 15 is a flowchart showing the operation of the prediction device. FIG. 16 is a diagram showing an example of the success or failure of prediction by the prediction device shown in FIG.

[0011] An embodiment will be described below with reference to the drawings.

[0012] In each drawing, the same or corresponding parts are denoted by the same reference numerals. In the description of this embodiment, the description of the same or corresponding parts will be omitted or simplified as appropriate.

[0013] The configuration of a model generating device 10 according to this embodiment will be described with reference to FIG.

[0014] The model generating device 10 is a general-purpose computer such as a PC, a server computer such as a cloud server, or a dedicated computer. "PC" is an abbreviation for personal computer.

[0015] The model generating device 10 may be capable of communicating with other devices via a network. The network may include, for example, the Internet, at least one WAN, at least one MAN, or any combination thereof. "WAN" is an abbreviation for wide area network. "MAN" is an abbreviation for metropolitan area network. The network may include at least one wireless network, at least one optical network, or any combination thereof. The wireless network may be, for example, an ad hoc network, a cellular network, a wireless LAN, a satellite communication network, or a terrestrial microwave network. "LAN" is an abbreviation for local area network.

[0016] The model generating device 10 includes an input unit 11 , a storage unit 12 , a control unit 13 , an output unit 14 , and a communication unit 15 .

[0017] The input unit 11 is, for example, a physical key, a capacitance key, a pointing device, a touch screen integrated with a display, a camera, or a microphone. The input unit 11 accepts an operation to input data used in the operation of the model generation device 10. The input unit 11 may be connected to the model generation device 10 as an external input device instead of being provided in the model generation device 10. As a connection interface, an interface compatible with standards such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used. "USB" is an abbreviation for Universal Serial Bus. "HDMI (registered trademark)" is an abbreviation for High-Definition Multimedia Interface.

[0018] The storage unit 12 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or any combination thereof. The semiconductor memory is, for example, a RAM, a ROM, or a flash memory. "RAM" is an abbreviation for random access memory. "ROM" is an abbreviation for read only memory. The RAM is, for example, an SRAM or a DRAM. "SRAM" is an abbreviation for static random access memory. "DRAM" is an abbreviation for dynamic random access memory. The ROM is, for example, an EEPROM. "EEPROM" is an abbreviation for electrically erasable programmable read only memory. The flash memory is, for example, an SSD. "SSD" is an abbreviation for solid-state drive. The magnetic memory is, for example, an HDD. "HDD" is an abbreviation for hard disk drive. The storage unit 52 functions, for example, as a main storage device, an auxiliary storage device, or a cache memory. The memory unit 12 stores data used in the operation of the model generation device 10, such as first equipment data 21, first water level data 22, first rainfall data 23, and inspection data 24, as well as data obtained by the operation of the model generation device 10.

[0019] The control unit 13 includes at least one processor, at least one programmable circuit, at least one dedicated circuit, or any combination thereof. The processor is a general-purpose processor such as a CPU or GPU, or a dedicated processor specialized for specific processing. "CPU" is an abbreviation for central processing unit. "GPU" is an abbreviation for graphics processing unit. An example of the programmable circuit is an FPGA. "FPGA" is an abbreviation for field-programmable gate array. An example of the dedicated circuit is an ASIC. "ASIC" is an abbreviation for application specific integrated circuit. The control unit 13 executes processing related to the operation of the model generation device 10 while controlling each part of the model generation device 10.

[0020] The output unit 14 is, for example, a display, a printer, or a speaker. The display is, for example, an LCD or an organic EL display. "LCD" is an abbreviation for liquid crystal display. "EL" is an abbreviation for electroluminescent. The output unit 14 outputs data obtained by the operation of the model generation device 10. The output unit 14 may be connected to the model generation device 10 as an external output device instead of being provided in the model generation device 10. The connection interface may be an interface compatible with standards such as USB, HDMI (registered trademark), or Bluetooth (registered trademark).

[0021] The communication unit 15 includes at least one communication module. The communication module is, for example, a module compatible with a wired LAN communication standard such as Ethernet (registered trademark), a wireless LAN communication standard such as IEEE 802.11, or a mobile communication standard such as LTE, 4G, or 5G. "IEEE" is an abbreviation for Institute of Electrical and Electronics Engineers. "LTE" is an abbreviation for Long Term Evolution. "4G" is an abbreviation for 4th generation. "5G" is an abbreviation for 5th generation. The communication unit 15 may communicate with a prediction device 50 as shown in FIG. 7. The communication unit 15 receives data used in the operation of the model generation device 10 and transmits data obtained by the operation of the model generation device 10.

[0022] The functions of the model generation device 10 are realized by executing a model generation program according to this embodiment on a processor serving as the control unit 13. That is, the functions of the model generation device 10 are realized by software. The model generation program causes a computer to execute the operations of the model generation device 10, thereby causing the computer to function as the model generation device 10. That is, the computer functions as the model generation device 10 by executing the operations of the model generation device 10 in accordance with the model generation program.

[0023] A program such as a model generation program can be stored on a non-transitory computer-readable medium. Examples of non-transitory computer-readable media include flash memory, magnetic recording devices, optical disks, magneto-optical recording media, and ROMs. A program can be distributed by selling, transferring, or lending portable media such as SD cards, DVDs, or CD-ROMs that store the program. "SD" is an abbreviation for Secure Digital. "DVD" is an abbreviation for digital versatile disc. "CD-ROM" is an abbreviation for compact disc read only memory. A program can also be distributed by storing it in the storage of a server and transferring it from the server to another computer. A program can also be provided as a program product.

[0024] A computer temporarily stores a program stored on a portable medium or transferred from a server in its main storage device. The computer then reads the program stored in the main storage device with its processor and executes processing in accordance with the read program. The computer may also read the program directly from the portable medium and execute processing in accordance with the program. The computer may also execute processing in accordance with the received program each time a program is transferred from the server to the computer. Processing may also be executed using a so-called ASP-type service that realizes functions simply by issuing execution instructions and obtaining results, without transferring the program from the server to the computer. "ASP" is an abbreviation for application service provider. A program is information used for processing by a computer and includes something equivalent to a program. For example, data that is not a direct instruction to a computer but has properties that define computer processing falls under the category of "something equivalent to a program."

[0025] Some or all of the functions of the model generating device 10 may be realized by a programmable circuit or a dedicated circuit as the control unit 13. In other words, some or all of the functions of the model generating device 10 may be realized by hardware.

[0026] The operation of the model generation device 10 according to this embodiment will be described with reference to Fig. 2. The operation described below corresponds to the model generation method according to this embodiment. That is, the model generation method according to the present disclosure includes, for example, steps S11 to S19 shown in Fig. 2.

[0027] First, in steps S11 to S14, the control unit 13 acquires first equipment data 21 indicating characteristics of the first equipment attached to the first bridge, first water level data 22 indicating the water level of the river on which the first bridge is located, and inspection data 24 indicating whether the first equipment has been damaged by flooding. In this embodiment, the control unit 13 may acquire multiple pieces of first equipment data indicating characteristics of multiple pieces of first equipment attached to multiple first bridges, respectively, first water level data 22 indicating the water level of the river on which the multiple first bridges are located, and multiple pieces of inspection data 24 indicating whether the multiple first equipment has been damaged by flooding. Although not required, in this embodiment, the control unit 13 further acquires first rainfall data 23 indicating the rainfall at the location where the first bridge 40 is located. Flood damage to the first equipment refers to a state in which the first equipment cannot be used continuously unless repair or reinforcement measures are taken due to changes in the water level of the river. Damage caused by flooding to equipment attached to bridges includes scratches, flattening, bending, or joint detachment on the surface of the pipes due to collisions with water currents or drifting debris. Equipment attached to bridges will also be damaged if the bridge girders of the bridge itself are washed away, or if the bridge piers are tilted or overturned. However, since it is difficult to predict damage to the bridge itself without detailed information about the bridge and the river, predictions of damage to attached equipment when the bridge itself is damaged are outside the scope of this technology.

[0028] The first facility may be a first pipeline 30 as shown in Fig. 3. The first pipeline 30 may be a pipeline for transporting water or gas, or for laying cables for electricity, communications, etc. The first bridge 40 may include a deck 41, a bridge girder 42, and a support member 43 that supports the first pipeline 30, and may be, for example, a road bridge spanning a river or the like.

[0029] Specifically, in S11, the control unit 13 acquires the inspection data 24 from the storage unit 12. For example, the inspection data 24 is input via the input unit 11 and stored in advance in the storage unit 12. Alternatively, the inspection data 24 may be transmitted from a device that manages the inspection data 24, received by the communication unit 15, and stored in advance in the storage unit 12.

[0030] The inspection data 24 is data indicating whether or not the first facility has suffered flood damage. The presence or absence of flood damage indicated by the inspection data 24 may be the presence or absence of flood damage in the past.

[0031] In S12, the control unit 13 acquires the first equipment data 21 from the storage unit 12. For example, the first equipment data 21 is input via the input unit 11 and stored in advance in the storage unit 12. Alternatively, the first equipment data 21 may be transmitted from a device that manages the first equipment data 21, received by the communication unit 15, and stored in advance in the storage unit 12.

[0032] As described above, in a configuration in which the first equipment is the first pipeline 30, as shown in Fig. 4, the first equipment data 21 may include data indicating the pipe type of the first pipeline 30. The pipe type is the type of the first pipeline 30 determined according to the material, type or strength of the joint, and diameter (outer diameter and inner diameter) of the first pipeline 30.

[0033] Furthermore, in a configuration in which the first facility is the first pipeline 30, the first facility data 21 may include data indicating an attachment position at which the first pipeline 30 is attached to the first bridge 40. The attachment position is a position at which the first pipeline 30 is attached to a bridge girder 42 of the first bridge 40. The attachment position includes, for example, the upstream or downstream side of the river on which the first bridge 40 is provided, the back surface between two main girders, etc.

[0034] Furthermore, in a configuration in which the first facility is the first conduit 30, the first facility data 21 may include data indicating the length of the section in which the first conduit 30 is attached to the first bridge 40. In a configuration in which the first conduit 30 is attached over the entire length of the first bridge 40, the section length is the bridge length. Furthermore, the bridge length can be considered to be the width of the river in which the first bridge 40 is provided.

[0035] The first equipment data 21 may also include data indicating the mounting method of the first pipeline 30, the spacing of the support members 43, etc. In a configuration in which the first equipment includes a plurality of first pipelines 30, the first equipment data 21 may also include data indicating the total number of first pipelines 30, the total number of first pipelines 30 arranged in each of the horizontal and vertical directions, etc.

[0036] The first facility data 21 may also include data indicating deterioration of the first pipeline 30. Specifically, the first facility data 21 may include data indicating the year in which the first pipeline 30 was attached to the first bridge 40, or the number of years elapsed from the year in which the first pipeline 30 was attached to the bridge to the time when the prediction is made, the distance from the coast to the first bridge 40, the amount of snow-melting agent sprayed on the first pipeline 30, or the hours of sunshine at the location where the first bridge 40 is installed.

[0037] In S13, the control unit 13 acquires the first water level data 22 from the storage unit 12. For example, the first water level data 22 is input via the input unit 11 and stored in advance in the storage unit 12. Alternatively, the first water level data 22 may be transmitted from a device that manages the first water level data 22, received by the communication unit 15, and stored in advance in the storage unit 12. The device that manages the first water level data 22 may be operated by, for example, an administrative agency, a local government, or any other organization.

[0038] The first water level data 22 is data indicating the water level of the river where the first bridge 40 is located. The first water level data 22 may be data indicating the maximum water level from the start to the end of rainfall, or may be water depth converted from the estimated inundation depth in a hazard map published by an administrative agency, local government, or the like.

[0039] For example, the water level of the river may be the water level from a predetermined water level reference at an observation point related to the point where the first bridge 40 is installed. The observation point may be a point where the water level at the observation point is highly correlated with the water level at the point where the first bridge 40 is installed. For example, the observation point may be a point where a measuring instrument capable of measuring the water level is installed, which is closest to the point where the first bridge 40 is installed.

[0040] The first water level data 22 may also include data indicating the water pressure of the river at a location where the first bridge 40, to which the first conduit 30 is attached, is located. The first water level data 22 may also include data indicating drifting debris in the river at a location where the first bridge 40, to which the first conduit 30 is attached, is located. The first water level data 22 may also include data indicating the flow velocity of the river at a location where the first bridge 40, to which the first conduit 30 is attached, is located. The first water level data 22 may also include data indicating the flow rate of the river at a location where the first bridge 40, to which the first conduit 30 is attached, is located.

[0041] In S14, the control unit 13 acquires the first rainfall data 23 from the storage unit 12. For example, the first rainfall data 23 may be input via the input unit 11 and stored in advance in the storage unit 12. Alternatively, the first rainfall data 23 may be transmitted from a device that manages the first rainfall data 23, received by the communication unit 15, and stored in advance in the storage unit 12. The device that manages the first rainfall data 23 may be operated by, for example, an administrative agency, a local government, or any other organization.

[0042] The first rainfall data 23 is data indicating the amount of rainfall at the location where the first bridge 40 is located. The amount of rainfall indicated by the first rainfall data 23 is correlated with the change in water level indicated by the first water level data 22. For example, the amount of rainfall is correlated with the amount of change in water level indicated by the first water level data 22 from the start to the end of rainfall.

[0043] In S15 to S18, when second facility data 61 indicating the characteristics of the second facility attached to the second bridge and second water level data 62 indicating the water level of the river on which the second bridge is located are input, the control unit 13 generates a prediction model that predicts the possibility of flood damage to the second facility using the first facility data 21, the first water level data 22, and the inspection data 24. In this embodiment, when second facility data 61 indicating the characteristics of the second facility attached to each of multiple second bridges and second water level data 62 indicating the water level of the river on which the multiple second bridges are located are input, the control unit 13 generates a prediction model that predicts the possibility of flood damage to each of the multiple second facilities using the first facility data 21, the first water level data 22, and the inspection data 24. Although not required, in this embodiment, when the second facility data 61, the second water level data 62, and the second rainfall data 63 indicating the rainfall at the location where the second bridge is located are input as a prediction model, the control unit 13 generates a model that predicts the possibility of flood damage to the second facility by further using the first rainfall data 23. The second facility may be a facility whose possibility of being damaged is unknown. Damage to the second facility refers to a state in which the second facility cannot be used continuously unless repair or reinforcement measures are taken for the second facility.

[0044] As described above, in a configuration in which the first equipment data 21 includes data indicating the pipe type of the first pipeline 30, the second equipment data 61 includes data indicating the pipe type of the second pipeline. The pipe type of the second pipeline is determined according to the material, type or strength of the joint, and diameter (outer diameter and inner diameter) of the second pipeline.

[0045] In addition, in a configuration in which the first equipment data 21 includes data indicating the attachment position where the first pipeline 30 is attached to the first bridge 40, the second equipment data 61 includes data indicating the attachment position where the second pipeline is attached to the second bridge. The attachment position is the position where the second pipeline is attached to the bridge girder of the second bridge. The attachment position includes, for example, the upstream or downstream side of the river on which the bridge is installed, or the back surface between the two main girders.

[0046] In addition, in a configuration in which the first facility data 21 includes data indicating the length of the section in which the first pipeline 30 is attached to the first bridge 40, the second facility data 61 includes data indicating the length of the section in which the second pipeline is attached to the bridge. In a configuration in which the second pipeline is attached to the entire length of the bridge, the section length is the bridge length. The bridge length can be considered to be the width of the river on which the bridge is located.

[0047] Furthermore, in a configuration in which the first equipment data 21 includes data indicating the installation method of the first pipeline 30 or the spacing of the support members 43, the second equipment data 61 includes data indicating the installation method of the second pipeline or the spacing of the support members 43. Furthermore, in a configuration in which the first equipment data 21 includes data indicating the total number of rows of the first pipeline 30, the second equipment data 61 includes data indicating the total number of rows of the second pipeline. Furthermore, in a configuration in which the first equipment data 21 includes data indicating the total number of rows of the first pipeline 30 arranged in each of the horizontal and vertical directions, the second equipment data 61 includes data indicating the number of second pipelines arranged in each of the vertical and vertical directions.

[0048] Furthermore, in a configuration in which the first equipment data 21 includes data indicating deterioration of the first pipeline 30, the second equipment data 61 includes data indicating deterioration of the second pipeline. Specifically, the second equipment data 61 includes the year in which the second pipeline was installed on the bridge, or the number of years elapsed from the year in which the second pipeline was installed on the bridge to the time when the prediction is made, the distance from the coast to the bridge, the amount of snow-melting agent sprayed on the second pipeline, or the hours of sunshine at the location where the bridge is located.

[0049] In steps S15 to S18, the control unit 13 generates a prediction model using quantification theory type 1. An example in which the control unit 13 generates a prediction model using quantification theory type 1 will be described in detail.

[0050] First, in S15 and S16, the control unit 13 calculates the item a, which is information included in the first equipment data 21 and the first water level data 22. ｉ The value of the item a ｉ The category b that discretizes the possible values ​​of ｉｊDummy variables x indicating whether each applies or not ｉｊ At this time, the control unit 13 determines the item a, which is information included in the first rainfall data 23, in addition to the first facility data 21 and the first water level data 22. ｉ The value of the item a ｉ The category b that discretizes the possible values ​​of ｉｊ Dummy variables x indicating whether each applies or not ｉｊ may be determined.

[0051] Specifically, in S15, the control unit 13 determines the item a, which is information included in the first equipment data 21 and the first water level data 22. ｉ The category b that discretizes the possible values ​​of ｉｊ At this time, the control unit 13 determines the item a, which is information included in the first facility data 21, the first water level data 22, and the first rainfall data 23. ｉ The category b that discretizes the possible values ​​of ｉｊ may be determined.

[0052] In this example, item a １ , a ２ , and a ３ are the "pipe type", "attachment position", and "section length" shown in the data included in the first equipment data 21. ４ is the "excess water level" included in the first water level data 22. ５ is the “cumulative rainfall” indicated in the data included in the first rainfall data 23.

[0053] In the example shown in FIG. １ Category B of "pipe type" １１ is "PVC pipe" and category b １２ is "steel pipe". Also, item a ２ Category B of "attached position" ２１ is the "back side" and category b ２２ is the "side". Also, item a ３ Category b of "section length" ３１ is "0m or more but less than 50m", and category b ３２In the example shown in FIG. 5, item a １ The categories of "pipe type" are "PVC pipe" and "steel pipe", but are not limited to these and may include other items such as "others". ２ The same applies to the "attachment position" as described above. Furthermore, the "section length" is discretized into "0 m or more and less than 50 m" and "50 m or less and less than 100 m", but is not limited to this, and a category of 100 m or more may be determined.

[0054] In S16, the control unit 13 ｉｊ For each item a ｉ The value of category b ｉｊ Dummy variables x indicating whether each applies or not ｉｊ That is, item a ｉ In each of the above, any of the categories b ｉｊ Dummy variable x ｉｊ is 1, and the category b ｉｊ Other than category b ｋｌ Dummy variable x ｋｌ is 0 (k≠i, l≠j). i is an integer from 1 to N, and N is the total number of items. Also, j is an integer from 1 to M. ｉ is an integer, and M ｉ is item a ｉ Category B in ｉｊ In the example shown in FIG. 4, the type of pipe corresponding to the facility name "A line 1 to A line 2" is "PVC pipe", so the control unit 13 selects category b as shown in FIG. １１ The dummy variable x for "PVC pipe" is １１ is "1", and category b １２ The dummy variable x for "steel pipe" is １２ is determined to be "0". In this way, the control unit 13 determines dummy variables for each category of items, which are qualitative variables. In the example shown in FIG. 4, the "section length" corresponding to the facility name "A Line 1 to A Line 2" is 45 m, so the control unit 13 determines that category b ３１ The dummy variable x for "0m or more but less than 50m" is ３１ is "1", and category b ３２The dummy variable x for "50m or more but less than 100m" is ３２ is determined to be "0." In this way, the control unit 13 determines the dummy variables for each item category, which is a quantitative variable.

[0055] At this time, the control unit 13 may use the presence or absence of flood damage indicated by the inspection data 24 as a response variable and determine a dummy variable for each. Specifically, when the inspection data 24 indicates that damage has occurred, the control unit 13 may determine the dummy variable for the category "yes" in the item "damaged" to be "1" and the dummy variable for the category "no" to be "0." Furthermore, when the inspection data 24 indicates that no damage has occurred, the control unit 13 may determine the dummy variable for the category "yes" in the item "damaged" to be "0" and the dummy variable for the category "no" to be "1."

[0056] In S17, the control unit 13 determines whether or not the inspection data 24 indicates damage caused by flooding and whether or not the inspection data 24 indicates damage caused by flooding. ｉｊ Each dummy variable x ｉｊ Based on this, the damage coefficient c indicates the degree to which the category contributes to the presence or absence of flood damage. ｉｊ For example, the control unit 13 calculates the damage coefficient c ｉｊ Specifically, the control unit 13 uses a dummy variable for the presence or absence of flood damage indicated by the inspection data 24 as a response variable, and dummy variables for each category of the first equipment data 21 and the first water level data 22 as explanatory variables, and calculates the damage coefficient c using quantification theory type 1. ｉｊ Calculate.

[0057] For example, as shown in FIG. 6, the control unit 13 １１ "PVC pipe", category B １２ "Steel pipes", Category B ２１ "Back side" is category b ２２ "Aspect" that is Category b ３１ Category B: "0m or more and less than 50m" ３２ The damage coefficient c for each of "50m or more but less than 100m" ｉｊAt this time, the control unit 13 calculates the damage coefficient c ｉｊ and calculate the constant C in quantification theory type 1.

[0058] In S18, the control unit 13 uses the damage coefficient c ｉｊ We use this to generate a model that predicts the possibility of damage caused by flooding.

[0059] For example, the control unit 13 may use a dummy variable x ｉｊ Damage coefficient c ｉｊ The damage rate R is calculated by adding a constant C to the sum of the values ​​multiplied by ｅ The control unit 13 generates a model that outputs the above-mentioned damage coefficient c ｉｊ In calculating the constant C, the damage rate R ｅ The damage coefficient c is set to be between 0 and 1. ｉｊ and constant C may be calculated. e =c 11 x 11 +c 12 x 12 +・・・+c ij x ij +C (1)

[0060] In this embodiment, the control unit 13 generates a prediction model using quantification theory type 1, but the control unit 13 may generate a prediction model using the first facility data 21, the first water level data 22, and the inspection data 24 by any method other than quantification theory type 1, such as machine learning. Furthermore, the control unit 13 may generate a prediction model using the first rainfall data 23 by any method other than quantification theory type 1, such as machine learning.

[0061] In S19, the output unit 14 outputs the prediction model to the prediction device 50. Alternatively, the communication unit 15 may transmit the prediction model to the prediction device 50.

[0062] The configuration of a prediction device 50 according to this embodiment will be described with reference to FIG.

[0063] The prediction device 50 is a general-purpose computer such as a PC, a server computer such as a cloud server, or a dedicated computer.

[0064] The prediction device 50 may be capable of communicating with other devices over a network, which may include, for example, the Internet, at least one WAN, at least one MAN, or any combination thereof.

[0065] The prediction device 50 includes an input unit 51 , a storage unit 52 , a control unit 53 , an output unit 54 , and a communication unit 55 .

[0066] The input unit 51 is, for example, a physical key, a capacitance key, a pointing device, a touch screen integrated with a display, a camera, or a microphone. The input unit 51 accepts an operation to input data used for the operation of the prediction device 50. The input unit 51 may be connected to the prediction device 50 as an external input device instead of being provided in the prediction device 50. As a connection interface, an interface compatible with standards such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) may be used.

[0067] The storage unit 52 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or any combination thereof. The storage unit 52 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 52 stores data used in the operation of the prediction device 50, such as second facility data 61, second water level data 62, and second rainfall data 63, as well as data obtained by the operation of the prediction device 50.

[0068] The control unit 53 includes at least one processor, at least one programmable circuit, at least one dedicated circuit, or any combination thereof. The control unit 53 executes processes related to the operation of the prediction device 50 while controlling each unit of the prediction device 50.

[0069] The output unit 54 is, for example, a display, a printer, or a speaker. The display is, for example, an LCD or an organic EL display. The output unit 54 outputs data obtained by the operation of the prediction device 50. The output unit 54 may be connected to the prediction device 50 as an external output device instead of being provided in the prediction device 50. As a connection interface, an interface compatible with standards such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used.

[0070] The communication unit 55 includes at least one communication module. The communication unit 55 may communicate with the model generation device 10. The communication unit 55 receives data used in the operation of the prediction device 50 and transmits data obtained by the operation of the prediction device 50.

[0071] The functions of the prediction device 50 are realized by executing a prediction program according to this embodiment on a processor serving as the control unit 53. That is, the functions of the prediction device 50 are realized by software. The prediction program causes a computer to execute the operations of the prediction device 50, thereby causing the computer to function as the prediction device 50. That is, the computer functions as the prediction device 50 by executing the operations of the prediction device 50 in accordance with the prediction program.

[0072] Some or all of the functions of the prediction device 50 may be implemented by a programmable circuit or a dedicated circuit serving as the control unit 53. In other words, some or all of the functions of the prediction device 50 may be implemented by hardware.

[0073] The operation of the prediction device 50 according to this embodiment will be described with reference to Fig. 8. The operation described below corresponds to the prediction method according to this embodiment. That is, the prediction method according to the present disclosure includes, for example, steps S21 to S26 shown in Fig. 8.

[0074] In S21, the control unit 53 acquires the second equipment data 61 from the storage unit 52. For example, the second equipment data 61 is input via the input unit 51 and stored in advance in the storage unit 52. Alternatively, the second equipment data 61 may be transmitted from a device that manages the second equipment data 61, received by the communication unit 55, and stored in advance in the storage unit 52.

[0075] In S22, the control unit 53 acquires the second water level data 62 from the storage unit 52. For example, the second water level data 62 may be input via the input unit 51 and pre-stored in the storage unit 52. Alternatively, the second water level data 62 may be transmitted from a device that manages the second water level data 62, received by the communication unit 55, and pre-stored in the storage unit 52. Here, the second water level data 62 acquired by the control unit 53 may be data indicating an actual water level or data indicating a predicted water level. For example, when heavy rain is expected to occur, the water level indicated by the second water level data 62 may be the predicted water level of the river on which the second bridge is located. Furthermore, the water level indicated by the second water level data 62 may be the highest water level in a predetermined number of years in the past (e.g., the past 100 years). This allows the possibility of damage to the second facility to be predicted in the event of the highest conceivable water level to be predicted, and can be effectively used to formulate disaster prevention measures for the second facility.

[0076] In S23, the control unit 53 acquires the second rainfall data 63 from the storage unit 52. For example, the second rainfall data 63 may be input via the input unit 51 and pre-stored in the storage unit 52. Alternatively, the second rainfall data 63 may be transmitted from a device that manages the second rainfall data 63, received by the communication unit 55, and pre-stored in the storage unit 52. Here, the second rainfall data 63 acquired by the control unit 53 may indicate a predicted rainfall rather than an actual rainfall. For example, when heavy rain is expected to occur, the rainfall indicated by the second rainfall data 63 may be the rainfall predicted in the area where the second bridge is located. Furthermore, the water level indicated by the second rainfall data 63 may be the maximum rainfall over a predetermined number of years in the past (e.g., the past 100 years). This allows the possibility of damage to the second facility to be predicted in the event of the maximum anticipated rainfall to be predicted, and can be effectively used to formulate disaster prevention measures for the second facility.

[0077] In S24, the control unit 53 determines whether the item a ｉ The value of the item a ｉ The category b that discretizes the possible values ​​of ｉｊ Dummy variables x indicating whether each applies or not ｉｊ At this time, the control unit 53 determines, in addition to the second equipment data 61 and the second water level data 62, the item a ｉ The value of the item a ｉ The category b that discretizes the possible values ​​of ｉｊ Dummy variables x indicating whether each applies or not ｉｊ Here, category b ｉｊ is the category b determined by the model generating device 10 ｉｊ is the same as

[0078] In S25, the control unit 53 predicts the possibility that the second facility will be damaged. For example, the control unit 53 may add the item a to the prediction model generated by the model generation device 10. ｉ Category B in each ｉｊ Dummy variable x for eachｉｊ The damage rate R is calculated as the probability that the second facility will be damaged. ｅ As described above, in the configuration in which the model is represented by the formula (1), the control unit 53 adds a dummy variable x ｉｊ By inputting the above, the damage rate R ｅ Predict.

[0079] In addition, the control unit 53 calculates the damage rate R ｅ The control unit 53 determines whether the damage rate R ｅ is equal to or greater than the predetermined threshold, the control unit 53 predicts that the second equipment will be damaged. ｅ If it is determined that the second facility is less than the predetermined threshold, it is predicted that the second facility will not be affected by the disaster.

[0080] In S26, the output unit 54 displays information for identifying the second facilities in association with the possibility of the second facilities being damaged. As an example, the output unit 54 displays a table showing information for identifying each piece of second facilities in association with the possibility of the second facilities being damaged. The output unit 54 also displays a map showing the possibility of the second facilities being damaged in association with the location where the second facilities are installed. The output unit 54 also displays information for identifying each piece of second facilities in association with the damage rate R of the second facilities. ｅ At this time, the information for identifying the second equipment may include the damage rate R ｅ In this way, the possibility of the second facility being damaged can be visually recognized easily.

[0081] In a configuration in which the model generating device 10 generates a prediction model without using the rainfall data 23, the control unit 53 does not need to execute step S23. In this case, in step S24, the control unit 53 ｉｊ Dummy variable x for ｉｊ In step S25, the control unit 53 determines whether the category b of the second rainfall data 63 is ｉｊ Dummy variable x for ｉｊ are not input into the prediction model.

[0082] In this embodiment, the model generation device 10 performs the operation shown in FIG. 2 , thereby enabling the prediction device 50 to generate a prediction model that enables prediction of damage to the second facility attached to the second bridge without using data indicating the characteristics of the second bridge. Generally, it may be difficult for a business operator managing the second facility attached to the second bridge to obtain data related to the second bridge that is not under the business operator's management. Even in such a case, the business operator managing the second facility can predict the possibility of damage to the second facility by using the second facility data 61 and the second water level data 62.

[0083] This allows the operator to take appropriate measures, such as preparing to repair the second facility, based on the possibility of the second facility being damaged predicted by the prediction device 50, in the event of heavy rain or the like, without having to visually inspect the second facility. ｅ The damage rate R ｅ By stopping the communication, electricity, gas, and water services using the cables installed in the second facility where the damage rate R is predicted to be equal to or greater than the threshold, the business operator can provide the infrastructure services appropriately. ｅ By inspecting the second facilities with a higher priority, an efficient restoration plan can be formulated. In addition, the operator can determine whether the damage rate R ｅ For example, businesses can take measures to protect facilities that are predicted to have a high damage rate R ｅ The second facility can be relocated to a lower position, an alternative facility can be provided, or the second facility can be reinforced.

[0084] Furthermore, the second facility attached to the second bridge is generally smaller than the second bridge. Therefore, even if the second bridge is not damaged by heavy rain or the like, the second facility may be damaged. Even in such a case, the prediction device 50 can predict the possibility of the second facility being damaged regardless of whether the second bridge is damaged by flooding.

[0085] As described above, the first facility is the first pipeline 30, and the first facility data 21 includes data indicating the pipe type determined according to the material, type or strength of the joints, and diameter of the first pipeline 30. The second facility is the second pipeline, and the second facility data 61 includes data indicating the pipe type determined according to the material and diameter of the second pipeline. Whether the first pipeline 30 is damaged by flood damage and the possibility that the second pipeline will be damaged depend on the pipe type of each pipeline. Therefore, the model generation device 10 can generate a prediction model that enables the prediction device 50 to predict the possibility of the second pipeline being damaged with higher accuracy.

[0086] As described above, the first facility is the first pipeline 30, the first facility data 21 includes data indicating the location where the first pipeline 30 is attached to the first bridge 40, and the second facility is the second pipeline, and the second facility data 61 includes data indicating the location where the second pipeline is attached to the second bridge. Whether the first pipeline 30 is damaged by flood damage and the possibility that the second pipeline will be damaged depend on the location where each pipeline is attached. Therefore, the model generation device 10 can generate a prediction model that enables the prediction device 50 to predict the possibility of the second pipeline being damaged with higher accuracy.

[0087] As described above, the first facility is the first pipeline 30, the first facility data 21 includes data indicating the section where the first pipeline 30 is attached to the first bridge 40, and the second facility is the second pipeline, and the second facility data 61 includes data indicating the section where the second pipeline is attached to the second bridge. Whether the first pipeline 30 is damaged by flood damage and the possibility of the second pipeline being damaged depend on the section where each pipeline is attached. Therefore, the model generation device 10 can generate a prediction model that enables the prediction device 50 to predict the possibility of the second pipeline being damaged with higher accuracy.

[0088] Furthermore, the model generating device 10 generates the prediction model by further using the first rainfall data 23, which allows the prediction device 50 to predict with higher accuracy the possibility of the second facility being damaged. This is because damage to the second facility attached to the second bridge depends not only on the water level of the river on which the second bridge is located, but also on rainfall.

[0089] Here, an example will be described in which a prediction model is generated by the model generation device 10 of this embodiment, and the prediction device 50 uses the generated prediction model to predict the possibility of equipment being damaged.

[0090] In this example, the model generation device 10 defines some of the approximately 580 facilities installed across 10 prefectures as the first facilities described above, and generates a prediction model using first facility data 21 indicating the characteristics of the first facilities, and the first water level data 22, first rainfall data 23, and inspection data 24 described above for each of the first facilities when heavy rains caused by Typhoon No. 19 (East Japan Typhoon) in 2019 and heavy rains in July 2020 occurred. The heavy rains caused by the East Japan Typhoon occurred over a wide area from Shizuoka to Iwate from October 10 to 13, 2019. The heavy rains in July 2020 occurred in southern and northern Kyushu, the Tokai region, and the Tohoku region from July 3 to 31, 2020, due to the influence of a front that remained stationary near Japan.

[0091] Specifically, the model generation device 10 designated three-quarters of the approximately 580 pieces of equipment described above as first pieces of equipment, and generated a prediction model using first equipment data 21 indicating the characteristics of the first pieces of equipment, first water level data 22, first rainfall data 23, and inspection data 24. Then, the prediction device 50 designated the pieces of equipment not designated as first pieces of equipment among the approximately 580 pieces of equipment described above as second pieces of equipment, and input second equipment data 61 indicating the characteristics of the second pieces of equipment, second water level data 62, and second rainfall data 63 into the prediction model generated by the model generation device 10, and predicted the possibility of the second pieces of equipment being damaged.

[0092] As shown in Fig. 9, among the second facilities that were damaged by the above-mentioned heavy rain, the number of second facilities that were predicted to be damaged by the prediction device 50 was "6". Furthermore, among the second facilities that were not damaged by the above-mentioned heavy rain, the number of second facilities that were predicted to be undamaged by the prediction device 50 was "104". Furthermore, among the second facilities that were damaged by the above-mentioned heavy rain, the number of second facilities that were predicted to be undamaged by the prediction device 50 was "0". Furthermore, among the second facilities that were not damaged by the above-mentioned heavy rain, the number of second facilities that were predicted to be damaged by the prediction device 50 was "35".

[0093] That is, the percentage of successful predictions made by the prediction device 50 using the model to predict that damaged equipment would be damaged was 4.1% of the total. Also, the percentage of successful predictions made by the prediction device 50 using the model to predict that second equipment that was not damaged would not be damaged was 71.7% of the total. In contrast, the percentage of unsuccessful predictions made by the prediction device 50 using the model to predict that second equipment that was damaged would not be damaged was 0% of the total. Also, the percentage of unsuccessful predictions made by the prediction device 50 using the model to predict that second equipment that was not damaged was 24.1% of the total.

[0094] Thus, the success rate of predictions was approximately 76% in total. Furthermore, even though a response was required following a disaster, not a single facility was mistakenly determined to be undamaged, and no damage was overlooked. This demonstrates the usefulness of the model generating device 10 and the prediction device 50 according to this embodiment.

[0095] The following additional notes are provided regarding the above-described embodiments.

[0096] (Supplementary Item 1) A model generation device comprising a control unit that acquires first equipment data indicating characteristics of first equipment attached to a first bridge, first water level data indicating the water level of a river on which the first bridge is located, and inspection data indicating whether the first equipment will be damaged by flooding, and when second equipment data indicating characteristics of second equipment attached to a second bridge and second water level data indicating the water level of the river on which the second bridge is located are input, generates a prediction model that predicts the possibility of flood damage to the second equipment using the first equipment data, the first water level data, and the inspection data. (Supplementary Item 2) The model generation device according to Supplementary Item 1, wherein the first equipment is a first pipeline, and the first equipment data includes data indicating a pipe type of the first pipeline, and the second equipment data includes data indicating a pipe type of the second pipeline. (Supplementary Item 3) The model generation device according to Supplementary Item 1 or 2, wherein the first facility is a first pipeline, and the first facility data includes data indicating an attachment position of the first pipeline attached to the first bridge, and the second facility is a second pipeline, and the second facility data includes data indicating an attachment position of the second pipeline attached to the second bridge. (Supplementary Item 4) The model generation device according to any one of Supplementary Item 1 to Supplementary Item 3, wherein the first facility is a first pipeline, and the first facility data includes data indicating a section length of the first pipeline attached to the first bridge, and the second facility is a second pipeline, and the second facility data includes data indicating a section length of the second pipeline attached to the second bridge. (Supplementary Item 5) The control unit further acquires first rainfall data indicating the amount of rainfall at a location where the first bridge is located, and when the second equipment data, the second water level data, and the second rainfall data indicating the amount of rainfall at the location where the second bridge is located are input as the prediction model, generates a model that predicts the possibility of damage due to flooding at the second equipment by further using the first rainfall data. This is the model generation device described in any one of Supplementary Item 1 to Supplementary Item 4.(Supplementary Item 6) The model generation device described in any one of Supplementary Items 1 to 5, wherein the control unit: determines dummy variables indicating whether or not the value of an item, which is information included in the first equipment data and the first water level data, falls into each of the categories into which the values ​​that the item can take are discretized; calculates a damage coefficient indicating the degree to which the category contributes to the presence or absence of flood damage based on the presence or absence of flood damage indicated by the inspection data and the dummy variables for each of the categories; and generates a model that predicts the possibility of flood damage using the damage coefficient as the prediction model. (Supplementary Item 7) A model generation method executed by a model generation device, comprising: acquiring first equipment data indicating characteristics of a first equipment attached to a first bridge, first water level data indicating the water level of a river on which the first bridge is located, and inspection data indicating whether the first equipment has been damaged by flood; and when second equipment data indicating characteristics of a second equipment attached to a second bridge and the second water level data indicating the water level of the river on which the second bridge is located are input, generating a prediction model that predicts the possibility of flood damage to the second equipment using the first equipment data, the first water level data, and the inspection data. (Supplementary Item 8) A program that causes a computer to function as the model generation device described in any one of Supplementary Items 1 to 6.

[0097] The present disclosure is not limited to the above-described embodiments. For example, two or more blocks shown in the block diagram may be integrated, or one block may be divided. Two or more steps shown in the flowchart may be executed in parallel or in a different order, instead of being executed in chronological order as described, depending on the processing capabilities of the device executing each step, or as needed. Other modifications are possible within the scope of the present disclosure.

[0098] REFERENCE SIGNS LIST 10 Model generation device 11 Input unit 12 Memory unit 13 Control unit 14 Output unit 15 Communication unit 21 First equipment data 22 First water level data 23 First rainfall data 24 Inspection data 30 First pipeline 40 First bridge 41 Deck slab 42 Bridge girder 43 Support member 50 Prediction device 51 Input unit 52 Memory unit 53 Control unit 54 Output unit 55 Communication unit 61 Second equipment data 62 Second water level data 63 Second rainfall data

Claims

1. A model generation device having a control unit that acquires first equipment data indicating characteristics of a first equipment attached to a first bridge, first water level data indicating the water level of the river on which the first bridge is located, and inspection data indicating whether the first equipment has been damaged by flood, and when second equipment data indicating characteristics of a second equipment attached to a second bridge and second water level data indicating the water level of the river on which the second bridge is located are input, generates a predictive model that predicts the possibility of flood damage to the second equipment using the first equipment data, the first water level data, and the inspection data.

2. The model generating device of claim 1, wherein the first equipment is a first pipeline, and the first equipment data includes data indicating a pipe type of the first pipeline, and the second equipment is a second pipeline, and the second equipment data includes data indicating a pipe type of the second pipeline.

3. A model generating device as described in claim 1, wherein the first equipment is a first pipeline, and the first equipment data includes data indicating an attachment position of the first pipeline attached to the first bridge, and the second equipment is a second pipeline, and the second equipment data includes data indicating an attachment position of the second pipeline attached to the second bridge.

4. A model generating device as described in claim 1, wherein the first equipment is a first pipeline, and the first equipment data includes data indicating a section length of the first pipeline attached to the first bridge, and the second equipment is a second pipeline, and the second equipment data includes data indicating a section length of the second pipeline attached to the second bridge.

5. The control unit further acquires first rainfall data indicating the amount of rainfall at a point where the first bridge is located, and when the second equipment data, the second water level data, and the second rainfall data indicating the amount of rainfall at the point where the second bridge is located are input as the prediction model, the control unit further uses the first rainfall data to generate a model that predicts the possibility of damage due to flooding at the second equipment. A model generation device as described in any one of claims 1 to 4.

6. The control unit determines dummy variables indicating whether or not an item value, which is information contained in the first equipment data and the first water level data, falls into each of the categories into which the values ​​that the item can take are discretized, calculates a damage coefficient indicating the degree to which the category contributes to the presence or absence of damage due to the flood based on the presence or absence of damage due to the flood indicated by the inspection data and the dummy variables for each of the categories, and generates a model that predicts the possibility of damage due to the flood using the damage coefficient as the predictive model. A model generating device as described in any one of claims 1 to 4.

7. A model generation method executed by a model generation device, comprising: acquiring first equipment data indicating characteristics of a first equipment attached to a first bridge, first water level data indicating the water level of a river on which the first bridge is located, and inspection data indicating whether the first equipment has been damaged by flood; and when second equipment data indicating characteristics of a second equipment attached to a second bridge and second water level data indicating the water level of the river on which the second bridge is located are input, generating a predictive model that predicts the possibility of damage to the second equipment by flood using the first equipment data, the first water level data, and the inspection data.

8. A program for causing a computer to function as the model generating device according to any one of claims 1 to 4.