Incident response assistance system and incident response assistance method

Abstract

This incident response assistance system determines, on the basis of input data from an external system, whether an incident is foreseen, generates, for foreseen incidents, incident amounts and series data pertaining to a prediction error for the incident amounts, decides on a resource amount series on the basis of the incident amount series, decides on an auxiliary resource amount series on the basis of the prediction error series for the incident amounts, and creates a work schedule for an incident response on the basis of the resource amount series, the auxiliary resource amount series, and resource management information.

Description

Incident response support system and incident response support method

[0001] The present invention relates generally to assisting in incident response.

[0002] In the energy business and other businesses that support social activities, strengthening resilience to ensure the continuous provision of services is considered an important issue. For example, electric power companies use a large number of facilities over long periods of time, which means they face the risk of their facilities aging. Furthermore, since much of their facilities are installed outdoors, they are exposed to risks from the natural environment. When an incident actually occurs due to a risk, it is important to have resilience to quickly detect the incident and mitigate its impact.

[0003] Appropriate incident response contributes to improving resilience. Technologies relating to incident response include those disclosed in Patent Literature 1 and 2, for example.

[0004] JP 2007-129547 A JP 2015-42086 A

[0005] Incident response requires resources such as supplies and labor. If resources are depleted, it will be impossible to respond adequately to the incident. Furthermore, if the amount of resources required for incident response is calculated after an incident actually occurs, it will be impossible to respond to the incident quickly. This issue is particularly significant when the risk of an incident occurring is a natural disaster, for example. This is because natural disasters generally cause incidents in multiple regions, and the amount of damage per region (an example of incident volume) varies depending on the severity of the disaster over time and the amount of equipment in the region.

[0006] The techniques disclosed in Patent Documents 1 and 2 cannot solve these problems because Patent Document 1 does not describe how to calculate the amount of resources, such as materials and labor, required to respond to an incident, and the technique disclosed in Patent Document 2 requires that an incident actually occur.

[0007] The incident response support system determines whether an incident can be predicted based on input data from an external system, generates series data of incident volume and prediction error of the incident volume for predicted incidents, determines a series of resource volumes based on the series of incident volumes, determines a series of reserve resource volumes based on the series of prediction error of the incident volumes, and creates a work schedule for responding to the incident based on the series of resource volumes, the series of reserve resource volumes, and resource management information.

[0008] According to the present invention, resources are allocated for appropriate incident response before an incident actually occurs.

[0009] 1 is a diagram showing an example of the configuration of an entire system including an incident response support system according to an embodiment of the present invention. FIG. 2 is a diagram showing an example of the hardware configuration of the incident response support system. FIG. 3 is a diagram showing an example of resistance and resilience. FIG. 4 is a diagram showing an example of data for ensemble prediction of typhoon paths. FIG. 5 is a diagram showing an example of forecast data representing a time series of average rainfall values. FIG. 6 is a diagram showing an example of the concept of the relationship between the amount of damage, the prediction error of the amount of damage, and the amount of resources. FIG. 7 is a graph showing a step-by-step recovery process when a complex disaster occurs. FIG. 8 is a graph modeling the recovery of a system state during the initial response to an incident. FIG. 9 is a graph showing the concept of evaluating the severity of risk.

[0010] In the following description, an "interface apparatus" may be one or more interface devices. The one or more interface devices may be at least one of the following: - One or more I / O (Input / Output) interface devices. The I / O (Input / Output) interface device is an interface device for at least one of an I / O device and a remote display computer. The I / O interface device for the display computer may be a communication interface device. The at least one I / O device may be a user interface device, for example, either an input device such as a keyboard and a pointing device, or an output device such as a display device. - One or more communication interface devices. The one or more communication interface devices may be one or more homogeneous communication interface devices (e.g., one or more NICs (Network Interface Cards)) or two or more heterogeneous communication interface devices (e.g., a NIC and an HBA (Host Bus Adapter)).

[0011] In the following description, "memory" refers to one or more memory devices, typically a primary storage device. At least one of the memory devices may be a volatile memory device or a non-volatile memory device.

[0012] In the following description, a "persistent storage device" refers to one or more persistent storage devices. A persistent storage device is typically a non-volatile storage device (e.g., an auxiliary storage device), and specifically, for example, a hard disk drive (HDD) or a solid state drive (SSD).

[0013] Also, in the following description, a "storage device" may be at least one of a memory and a persistent storage device (for example, a memory).

[0014] In the following description, a "processor" refers to one or more processor devices. The at least one processor device is typically a microprocessor device such as a CPU (Central Processing Unit), but may also be other types of processor devices such as a GPU (Graphics Processing Unit). The at least one processor device may be a single-core or multi-core. The at least one processor device may also be a processor core. The at least one processor device may also be a processor device in a broader sense, such as a hardware circuit (for example, an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit)) that performs part or all of the processing.

[0015] Furthermore, in the following description, functions may be described using the expression "yyy unit." However, the functions may be realized by one or more computer programs executed by a processor, by one or more hardware circuits (e.g., FPGAs or ASICs), or by a combination thereof. When a function is realized by a program executed by a processor, the specified processing is performed using a storage device and / or an interface device, etc., as appropriate, and therefore the function may be considered to be at least a part of the processor. Processing described using a function as the subject may be processing performed by a processor or a device having the processor. A program may be installed from a program source. The program source may be, for example, a program distribution computer or a computer-readable recording medium (e.g., a non-transitory recording medium). The description of each function is an example, and multiple functions may be combined into one function, or one function may be divided into multiple functions.

[0016] 1 is a diagram showing an example of the overall configuration of a system including an incident response support system according to an embodiment of the present invention. In the following description, "DB" is an abbreviation for database.

[0017] The incident response support system 10 is a computer system that supports incident response. The incident response support system 10 communicates with an external system 20 via a communication network 39 (e.g., the Internet). The incident response support system 10 may communicate with a work planning system 40 and / or an incident response system 30 via the communication network 39 (or another communication network). The incident response support system 10 may communicate with a UI (User Interface) device 15.

[0018] The external system 20 inputs input data used to predict incidents to the incident response support system 10. In other words, the external system 20 is a source of data used to predict incidents. For example, the external system 20 may be at least one of a weather server 21, a satellite image data server 22, an equipment status monitoring server 23, and a news agency server 25. The weather server 21 provides weather data including weather (e.g., rainfall) forecast data for each of one or more regions. The satellite image data server 22 provides satellite image data including data on satellite images of one or more regions. The equipment status monitoring server 23 provides equipment status monitoring data. The news agency server 25 provides news data compiled by news agencies such as newspapers. In this embodiment, the "risk of an incident occurring" is a "natural disaster" and the "incident" is "damage caused by a natural disaster." Therefore, the external system 20 includes at least one of the weather server 21 and the satellite image data server 22.

[0019] The incident response support system 10 may be a physical computer system (one or more physical computers) and may include an interface device, a storage device, and a processor connected thereto. Specifically, for example, as shown in FIG. 2 , the incident response support system 10 may include a CPU 901, a RAM (Random Access Memory) 902, a ROM (Read Only Memory) 903, a HDD (Hard Disk Drive) 904, a communication I / F (Interface) 905, an input / output I / F 906, and a media I / F 907. The CPU 901 may be an example of a processor. The RAM 902 and ROM 903 may be examples of memory. The HDD 904 may be an example of a persistent storage device. The communication I / F 905, the input / output I / F 906, and the media I / F 907 may be examples of interface devices. The communication I / F 905 may communicate with at least the external system 20 among the external system 20, the work planning system 40, and the incident response system 30. The input / output I / F 906 may communicate with the UI device 15. The UI device 15 may be an input / output device such as a keyboard, a mouse, or a display device, or may be an information processing terminal having an input / output device (if the UI device 15 is an information processing terminal, the communication I / F 905 may communicate with the UI device 15). The media I / F 907 may read and write data from a storage medium. Note that instead of being such a physical computer system, the incident response support system 10 may be a logical computer system based on a physical computer system (e.g., a virtual machine or a system as a cloud computing service).

[0020] The incident response support system 10 stores a series database 13 and a resource management database 4 in a storage device. The series database 13 stores incident series data. The resource management database 4 stores information about resources that can be allocated to incidents.

[0021] For example, when the program is executed by a processor, the incident response support system 10 realizes functions such as an incident prediction unit 11, a series data generation unit 12, a resource quantity calculation unit 14, a resource allocation unit 16, and an output control unit 17.

[0022] The incident prediction unit 11 determines whether an incident can be predicted based on input data from the external system 20. The input data may include forecast data and past observation data for weather or other conditions. The input data may also include regional data that includes data that can identify the slope and amount of equipment for each region.

[0023] When an incident is predicted, the series data generator 12 predicts the incident for each time during a target period, by region, based on input data from the external system 20. Specifically, it predicts the amount of damage at each time during the target period. The series data generator 12 also predicts the prediction error of the predicted amount of damage (the difference between the actual amount of damage and the predicted amount of damage) for each time during the target period. The series data generator 12 generates series data of the incident, specifically, data including the predicted amount of damage and the prediction error for each time during the target period (e.g., time-series data of the amount of damage and the prediction error), and stores the generated series data in the series DB 13. The "target period" may be a period during which an incident response is required. The start time of the "target period" may be the time of the disaster (the time the natural disaster occurred), and the time of the disaster may be a past time (the time the natural disaster actually occurred) or a predicted time (the time when the natural disaster was predicted to occur). The end time of the "target period" may be the time of convergence (the predicted end time of the natural disaster) or a later time (for example, the time after a length of time based on a predetermined policy including the status of recovery work is added to the convergence time), and the convergence time may be a predicted time (the time when the natural disaster is predicted to end). In addition, the target period may be set for each region, or may be common to multiple regions. The disaster occurrence time may be the time when a natural disaster is expected to occur and preparations should begin, or the time when a future natural disaster is predicted to occur.

[0024] The resource amount calculation unit 14 calculates a series of resource amounts and reserve resource amounts corresponding to the series of damage amounts (resource amounts and reserve resource amounts at each time point in a target period) based on the series data in the series DB 13 and the resource management DB 4. The resource amounts are based on the damage amounts, and the reserve resource amounts are based on the prediction errors of the damage amounts.

[0025] The resource allocation unit 16 allocates resources at each time during the target period, specifically, for example, creates a work schedule, based on the incident series data (here, the series of predicted damage amounts and prediction errors), the series of calculated resource amounts and reserve resource amounts, and the resource management DB 4.

[0026] The output control unit 17 controls the output of the created work schedule. Specifically, for example, the output control unit 17 may display the work schedule on the UI device 15. Furthermore, for example, the output control unit 17 may transmit data representing the work schedule to a work planning system 40 that creates a work plan based on the data. Furthermore, for example, the output control unit 17 may transmit data based on the work schedule to an incident response system 30 based on the data. The work planning system 40 and the incident response system 30 will be described later.

[0027] One of the objectives of this embodiment is to enhance resilience. As illustrated in FIG. 3 , enhancing resilience requires improving resistance and recovery against threats such as natural disasters. Resistance is the ability to suppress damage caused by threats, and can be improved, for example, by making facilities more robust and redundant. On the other hand, resilience is the ability to restore the state of a damaged object (e.g., a system or facility) to a normal level, and can be improved by the incident response support of this embodiment.

[0028] The incident prediction unit 11, the sequence data generation unit 12, the resource amount calculation unit 14, the resource allocation unit 16, and the output control unit 17 will be described in detail below. (1) Incident prediction unit 11

[0029] The incident prediction unit 11 determines whether an incident can be predicted based on input data from the external system 20. The input data includes meteorological data, which may include, for example, the data illustrated in FIG. 4 , i.e., ensemble forecast data for typhoon tracks. The meteorological data in the input data may also include, for example, the data illustrated in FIG. 5 , i.e., forecast data representing a time series of average rainfall values ​​for each region (e.g., meteorological analysis grids 1, 2, ...). The input data may include satellite image data instead of or in addition to the meteorological data.

[0030] The determination of whether an incident is predicted may be performed for each region. Furthermore, the conditions for determining that an incident is predicted may be set in advance. For example, if N or more typhoon paths (N is a natural number) among multiple typhoon paths represented by the ensemble forecast data for typhoon paths overlap with the same region, it may be determined that an incident is predicted for that region. Furthermore, for example, if there is a region where the cumulative average rainfall value for a target period exceeds a certain value, it may be determined that an incident is predicted for that region. In this way, the conditions for determining that an incident is predicted may be set for at least one type of data (value) obtained from the input data.

[0031] The determination of whether an incident is predicted may be performed repeatedly (e.g., periodically), and the sequence data may be generated each time the determination result proves true (i.e., each time an incident is predicted). Also, if it is determined from the input data that a natural disaster has actually occurred, it may be determined that an incident has been predicted. (2) Sequence Data Generator 12

[0032] The series data generator 12 generates series data of incidents. The series data includes data representing a series of damage amounts for each region during a target period, and data representing a series of prediction errors of the damage amounts for each region during a target period. (2-1) Prediction of Damage Amount

[0033] The sequence data generating unit 12 predicts the amount of damage for each region at each time point during a target period from the input data.

[0034] For example, if the natural disaster is a typhoon or rainfall, the amount of damage caused by the typhoon or rainfall for each region can be calculated as follows.

[0035] That is, the series data generation unit 12 identifies the severity of rainfall at each time during the target period for each region from the input data. For example, the series data generation unit 12 may identify the wind speed at each time for each region and calculate the severity of a typhoon (an example of a natural disaster) at each time based on the identified wind speed at each time (the wind speed itself may represent the severity of the typhoon). Furthermore, for example, the series data generation unit 12 may identify the rainfall (average value) and the accumulated rainfall up to that time (the accumulated value of the average rainfall) for each region at each time, and may identify the slope of each region, and based on the rainfall, accumulated rainfall, and slope, identify the severity of rainfall at each time (the severity of landslides caused by rainfall) for each region. Regional data representing the slope of each region may be included in the input data, and the slope of each region may be identified from the input data.

[0036] After specifying the severity for each region, the sequence data generation unit 12 predicts the amount of damage for each region at each time based on the specified severity for each time and the amount of equipment in the region (e.g., the amount of utility poles, steel towers, power lines, etc.). For each time, the amount of damage may be the product of the severity at that time and the amount of equipment. Furthermore, the amount of equipment for each region may be specified from input data or a predetermined database. For example, the input data or a predetermined database may include data related to the amount of equipment at the municipal or other unit level, and the amount of equipment for each region may be specified based on such data.

[0037] Natural disasters that pose a risk of incident occurrence do not need to be limited to typhoons and rainfall. For example, if the natural disaster is a forest fire or a forest fire, the sequence data generation unit 12 may identify the temperature at each time (and the cumulative temperature up to that time) and the humidity at each time (and the cumulative humidity up to that time) for each region from the weather data in the input data, and may identify the severity of the weather at each time for each region based on the identified temperature (and cumulative temperature) and humidity (and cumulative humidity). The sequence data generation unit 12 may predict the amount of damage at each time for each region based on the severity at each time and the amount of equipment in the region (e.g., the area of ​​forest).

[0038] Furthermore, natural disaster risks may include the risk of tree contact with power lines or the risk of spontaneous combustion due to tree withering, such as oak wilt. Tree contact disasters are predicted by predicting tree height and growth, and resources are secured for each region to carry out pre-tree cutting (preventive maintenance) or emergency tree cutting (post-maintenance). Furthermore, because tree growth predictions are subject to prediction errors, backup resources may be secured in preparation for a greater number of contact accidents and post-event responses than expected, depending on the prediction error in the damage caused by the tree growth prediction. Tree height and tree growth predictions may be based on image data captured by quasi-zenith satellites, drones, or tractors.

[0039] The risk of spontaneous combustion due to tree withering is measured by measuring the withered area using photographic data from quasi-zenith satellites, drones, and tractors, and predicting the expansion of the area over time. The necessary resources for felling activities are secured for the withered area. In addition, differences in prediction results due to differences in photographic methods are treated as prediction errors, and spare resources are secured. (2-2) Prediction of prediction errors in the amount of damage

[0040] The sequence data generator 12 predicts the prediction error (prediction error of the amount of damage) for each region at each time point during the target period from the input data. The prediction error is predicted, for example, as follows.

[0041] The input data includes sampling error data representing sampling errors of wind speed and rainfall data for each region in relation to data on ensemble predictions of typhoon tracks. The series data generation unit 12 predicts the prediction error of the amount of damage for each region at each time based on the predicted amount of damage for each time and the sampling error represented by the sampling error data for that region (for example, by applying the sampling error to the amount of damage).

[0042] The method for predicting the prediction error is not limited to this example. For example, the current prediction error may be predicted based on statistics of prediction errors as actual measured values ​​in the past. (3) Resource Amount Calculation Unit 14

[0043] The resource amount calculation unit 14 calculates a series of resource amounts (resource amounts at each time in the target period) corresponding to the series of damage amounts based on the series of damage amounts represented by the series data in the series DB 13 and the resource management DB 4.

[0044] For example, the resource management DB 4 includes information representing multiple resource types (e.g., workers, heavy equipment, etc.). More specifically, for each resource, for example, for each time period, the resource management DB 4 may include information representing the relationship between the resource ID (e.g., the ID of a worker or equipment), the resource type (e.g., whether the resource is a worker or equipment (e.g., heavy equipment)), the skills possessed by the resource, the location of the resource (the waiting location of the resource), and the resource reserve information (information including whether the resource is reserved, etc.). The resource amount calculation unit 14 calculates the resource amount for each time period in the target period for each resource type represented by the resource management DB 4 and for each region.

[0045] Take one resource type as an example.

[0046] If the prediction error of the damage amount is zero for all regions at all times from the time of disaster occurrence (start time of the target period) to the time of convergence (end time of the target period), the resource amount calculation unit 14 may calculate the amount of resources sufficient to repair X(t) (the damage amount X at time t) for each time in the target period. The damage amount X may be converted into the resource amount using a predetermined method.

[0047] If there is a time when the prediction error of the damage amount is not zero, the resource amount may be calculated as follows. The resource amount may be calculated periodically. The time interval (period) for calculating the resource amount may be determined based on the time required for the resource to move from its standby location or the time required for the resource to move between regions. For example, the time interval may be six hours. Therefore, assume that one cycle is (t = 1, 2, 3, 4, 5, 6) and the next cycle is (t = 7, 8, 9, 10, 11, 12). The sum of the prediction errors of the damage amount, ΔS, is calculated for each cycle. In this case, each time in the target period may be any time (e.g., the start time) belonging to the cycle. Assume that there are multiple regions, Regions 1 to 9. The prediction error ΔX at time t for Region n is expressed as "ΔXn(t)." In this case, ΔS for the next cycle (t = 7, 8, 9, 10, 11, 12) is as follows: ΔS=ΔX1(7)+ΔX2(7)+...+ΔX9(7) +ΔX1(8)+ΔX2(8)+...+ΔX9(8) +ΔX1(9)+ΔX2(9)+...+ΔX9(9) +ΔX1(10)+ΔX2(10)+...+ΔX9(10) +ΔX1(11)+ΔX2(11)+...+ΔX9(11) +ΔX1(12)+ΔX2(12)+...+ΔX9(12)

[0048] The resource amount calculation unit 14 calculates the value obtained by dividing ΔS by a predetermined value m as the amount of spare resources to be reserved. This is to reserve the amount of resources that may be needed depending on the total amount of prediction errors. The value of m is between 0 and 1, and a user-defined value may be set as the value of m in a storage device (e.g., RAM 902 or HDD 904). m=0 means that there are no spare resources and the prediction error is considered to be zero. m=1 means that a sufficient amount of resources corresponding to the prediction error is reserved.

[0049] For each of the multiple regions, the resource amount converted from the damage amount X at time t may be the resource amount at time t. Furthermore, the sum of prediction errors for the regions for the cycle (t=7, 8, 9, 10, 11, 12) may be calculated as ΔXn(7) + ΔXn(8) + ΔXn(9) + ΔXn(10) + ΔXn(11) + ΔXn(12), and the amount obtained by multiplying this sum by m may be used as the reserve resource amount for all of the multiple regions. Furthermore, for region n, the reserve resource amount at that time may be calculated from the prediction error for each time. (4) Resource Allocation Unit 16

[0050] The resource allocation unit 16 allocates resources for each time period during the target period, specifically, creates a work schedule, based on the incident sequence data (sequence of predicted damage volume and prediction error), the calculated sequence of resource volume and reserve resource volume, and the resource management DB 4. Since the resource volume and reserve resource volume have been calculated, a schedule for these may be created. However, the reserve resource is a resource whose actual use has not yet been determined. Specifically, the region to which the reserve resource will be allocated has not yet been determined, and the resource is reserved until a situation requiring the allocation of the reserve resource is detected. The sequence of predicted damage volume and prediction error is used to optimize the area in which such a reserve resource should be placed on standby.

[0051] For example, the resource management DB 4 includes information representing the resource amount and its breakdown (e.g., the amount of reserved resources by allocation region) for each resource type for each region. The resource allocation unit 16 determines the internal resource amount and external resource amount for each region for each time and by resource type based on the resource management DB 4 and the resource amount and spare resource amount for each region at each time. For each region, the "internal resource amount" refers to the amount (typically the number) of internal resources, and the "internal resources" refers to resources present in the internal region (regions that satisfy a predetermined condition (internal region condition), such as being the same region as the region). For each region, the "external resource amount" refers to the amount (typically the number) of external resources, and the "external resources" refers to resources present in external regions (regions that satisfy a predetermined condition (external region condition), such as being a different region from the region). For each region, internal resources for the region are allocated from the internal region, and external resources for the region are allocated from external regions. The resource management DB 4 may also include information representing resource constraints for each region and / or for each resource type. For each region and / or each resource type, resource constraints may be, for example, whether or not the resource can be moved outside the region or whether or not the resource can be taken out.

[0052] For each region and resource type, the resource allocation unit 16 allocates as much internal resource amount as possible (e.g., unreserved internal resource amount) to the resource amount at that time for each region and resource type, and allocates the external resource amount to the sum of the insufficient internal resource amount and the spare resource amount. Note that even after allocating resources for all regions, if there is surplus internal resource amount for a certain region, the resource allocation unit 16 may replace the external resource amount for that region with all or part of the surplus internal resource amount. In the example shown in FIG. 6 , reference numeral 601 denotes a series of damage amounts, reference numeral 602 denotes a series of prediction errors of the damage amounts, reference numeral 603 denotes the external resource amount at a certain time, and reference numeral 604 denotes the internal resource amount at the same time. Reference numerals 603 and 604 indicate that the insufficient internal resource amount is compensated for by the external resource amount.

[0053] A work schedule may be created as a resource allocation, and the work schedule may include, for each region, information indicating whether an incident is predicted for that region. For regions where an incident is predicted, the work schedule may include, for each time, information representing a set of the time, the amount of internal resources allocated (provided) to that region at that time and their breakdown (internal resource amount for each resource type), the amount of external resources allocated to that region at that time and their breakdown (providing region and external resource amount for each resource type), and the amount of resources remaining (on standby) in that region at that time (resource amount for each resource type). For regions where an incident is not predicted, the work schedule may include, for each time, information representing a set of the time and the amount of resources remaining (on standby) in that region at that time (resource amount for each resource type).

[0054] The resource allocation unit 16 may refer to the constraints (e.g., resource constraints) represented by the resource management DB 4 and allocate resources (create a work schedule) that satisfy the constraints. For example, the resource allocation unit 16 may allocate resources by solving a constrained optimization problem that satisfies the target index while satisfying the above-mentioned constraints, using any of the functions exemplified below. For example, the resource allocation unit 16 may allocate resources for each region using a model that takes as input the resource amount (and / or damage amount) for each time, the reserve resource amount (and / or prediction error of the damage amount), and the constraints, and outputs the internal resource amount and the external resource amount. - A latent event estimation function using artificial intelligence (AI). - An ensemble error amount estimation function such as the Monte Carlo method. - A data assimilation function for disaster simulation using reports. - A mathematical optimization function. - A reinforcement learning function. (5) Output Control Unit 17

[0055] The output control unit 17 controls the output of the created work schedule. Specifically, for example, the output control unit 17 performs at least one of the following (5-1) to (5-3): (5-1) Display on the UI device 15

[0056] The output control unit 17 displays the work schedule on the UI device 15. The output control unit 17 may accept approval of the work schedule via the UI device 15, or may accept a change in the work schedule (for example, a change in the allocation of the amount of internal resources and the amount of external resources to be provided to at least one region where an incident is predicted, at a time designated by a user (operator of the UI device 15)) and change the work schedule. (5-2) Provision to the work planning system 40

[0057] The output control unit 17 transmits data representing the work schedule to the work planning system 40 .

[0058] The work planning system 40 may be a function realized by executing a program such as an application program on an information processing terminal of a user such as a worker. The work planning system 40 creates a work plan based on data representing a work schedule, and displays the created work plan on the screen of the information processing terminal having the work planning system 40. On the work planning system 40 side, a site supervisor may manually set and carry out the necessary work (tasks). Data representing the results of calculations by the resource allocation unit 16 (an example of a work schedule) may be input to the work planning system 40, and such calculation results (data) may be used to create the work plan. The work plan may be information such as a schedule that allocates resources allocated by the resource allocation unit 16 to tasks set on the work planning system 40 side. (5-3) Provision to the incident response system 30

[0059] The output control unit 17 transmits instructions based on the work schedule to the incident response system 30 .

[0060] For example, the output control unit 17 transmits a response command indicating the execution of a work schedule to the incident response system 30. Specifically, for example, the output control unit 17 converts the work schedule from the resource allocation unit 16 into a workflow in a predetermined format, and transmits the result of the conversion as a response command to the incident response system 30. The incident response system 30 may be a computer system, and responds to incidents by executing a workflow in a predetermined format.

[0061] Although one embodiment has been described above, this is merely an example for explaining the present invention, and the scope of the present invention is not limited to this embodiment. The present invention can be embodied in various other forms. For example, at least a part of the description with reference to FIGS. 7 to 9 may be adopted.

[0062] FIG. 7 is a graph showing a step-by-step recovery process when a complex disaster occurs.

[0063] A complex disaster is a disaster in which multiple interrelated disasters occur with a time lag, causing a deterioration in system status, such as a gradual deterioration of the installation environment. The horizontal axis of the graph in Figure 7 indicates time, and the vertical axis indicates the system status. The higher the position on the graph, the better the system status. Since one factor in the deterioration of the system status is thought to be the extent of the damage, high / low system status may correspond to small / large amounts of damage.

[0064] At time t1, the first disaster (for example, a landslide) of the complex disaster occurs. At this time t1, the effects of the first disaster have not yet spread, and the system state is almost normal.

[0065] At time t2, the system state is slightly degraded from the normal state due to damage caused by the first disaster. From this time t2, the first restoration work is started by operating resources that were reserved in advance.

[0066] At time t3, the damage caused by the first disaster was temporarily restored through the first restoration work. However, at time t3, a second disaster (e.g., internal and external flooding) occurred in the complex disaster.

[0067] At time t4, the system status is lowered from the provisional recovery state due to damage caused by the second disaster. From time t4, a second recovery work is started by operating additional backup resources.

[0068] At time t5, the damage caused by the second disaster was restored to normal (full restoration state) through the second restoration work.

[0069] The output control unit 17 may present the response plan proposed by the resource allocation unit 16 to a response instructor (e.g., a user of the UI device 15), and may also present the graph of Fig. 7 showing the time series of damage conditions to be recovered by the response plan to the response instructor, for example, via the UI device 15. This allows the response instructor to understand, before the restoration work begins, that the response plan proposed by the resource allocation unit 16 does not require all resources to be deployed in the short term, but is intended to restore the system state in stages in the long term.

[0070] In this way, the incident response support system 10 (resource allocation unit 16) of this embodiment allocates allocable resources registered in the resource management DB 4 in stages, taking into account the occurrence status of incidents of complex disasters.

[0071] In other words, because resources are limited, when a complex disaster occurs, the resource allocation unit 16 finds a compromise state that does not require the system state to be restored to a normal state, and allocates resources. Furthermore, the resource allocation unit 16 sets the date (priority date) when the system will be restored to the compromise state on the time axis for allocating resources, making it possible to carry out recovery in stages.

[0072] The reservation processing of the resource allocation unit 16 takes into consideration the degree of influence (CoF: Consequence of Failure) that a risk, which will be described later with reference to FIG. 9, has on the system state.

[0073] This allows for quicker incident response than a method that involves identifying all incidents in a complex disaster and then formulating a response plan.

[0074] Furthermore, the resource allocation unit 16 formulates a response plan while leaving spare resources available. This makes it possible to avoid a situation where resources run out and become insufficient in the latter half of a complex disaster, compared to a method in which resources are allocated on a first-come, first-served basis when an incident is selected.

[0075] The effects obtained from the incident response support system 10 of this embodiment will be described in detail below with reference to FIGS. 8 and 9. FIG.

[0076] Risk management has traditionally been considered important. Risk management, which involves controlling risks through risk assessment and risk monitoring, is carried out to prevent incidents from occurring. However, even with risk management, it is difficult to reduce incidents to zero, and it is not possible to completely prevent natural disasters.

[0077] Some incidents, such as a blackout caused by lightning, can spread instantly and an initial response can be completed within seconds using protective relay systems. On the other hand, other incidents, such as typhoons, floods, and heavy snow, can worsen over hours or days and require an initial response, such as emergency maintenance to manually remove foreign objects.

[0078] Therefore, it is important to assume that incidents will occur and, even if an incident does occur, to consider resilience, which is the ability to mitigate damage through initial responses. To increase resilience, it is important to have the robustness to prevent damage from incidents, the ability to recover through initial responses when an incident occurs, and the efficiency of recovery.

[0079] FIG. 8 is a graph modeling the recovery of the system state during the initial response to an incident.

[0080] As in Figure 7, the horizontal axis of the graph in Figure 8 represents time, and the vertical axis represents system status. System status deteriorates when an incident, such as a power outage, spreads. In the graph in Figure 8, the degree of deterioration in system status is indicated by the hatched area.

[0081] At time t1, an incident occurs, and the system state deteriorates due to its influence (arrow DL from system state L4 to L1).

[0082] During the initial response period T2, the system state is recovered to a certain level (arrow RL1 from system state L1 to L3) through initial responses such as automatic control and manual emergency repairs.

[0083] In the restoration period T3, restoration activities are carried out to restore the system state to the pre-disaster level (arrow RL2 from system state L1 to system state L4).

[0084] Furthermore, the following parameters are defined for the graph in Figure 8: - Speed ​​of degradation of functionality due to the spread of the impact, α. - Speed ​​of recovery during the initial response period, β1 and β2. - Speed ​​of recovery during the initial response period, which is carried out within a few seconds using automatic control such as protective control, β1. - Speed ​​of recovery during the period of manual system configuration changes and emergency maintenance, which takes several hours to several days, β2. - Speed ​​of recovery during the period of restoration activities, γ.

[0085] FIG. 9 is a graph showing the concept of evaluating the severity of a risk.

[0086] This graph expresses the severity of risk by combining the CoF (CoF) that the risk has on the system status on the horizontal axis with the Probability of Failure (PoF) on the vertical axis. In other words, as at least one of the CoF and PoF increases, the severity of the risk also increases (in the figure, the severity of the risk increases in the order of low severity → medium severity → high severity).

[0087] Risk management, indicated by arrow RS1, is carried out to assess the impact of risk and reduce the probability of risk occurrence (PoF). On the other hand, measures incorporating a resilience perspective, indicated by arrow RS2, aim to reduce not only PoF but also CoF. CoF can be quantified from the magnitude of the deterioration in system status, the level of gradual recovery, and the time required for recovery.

[0088] Responding to incidents is considered to be an effective measure for reducing CoF, but in order to do so, advance planning and preparation for incidents cannot be neglected.

[0089] Therefore, it is desirable for the resource allocation unit 16 to formulate a plan in advance to prepare for an incident so as to meet at least one of the following indexes as a target: - Mitigating the speed (α) at ​​which the impact of the incident spreads in Figure 8. - Increasing the speed at which recovery from the incident occurs (β1, β2, γ>0). - Suppressing the magnitude of the impact of the incident (shortening the length of the arrow DL).

[0090] The incident response support system 10 of this embodiment described above provides the following effects. The resource allocation unit 16 selects incidents that will occur in the future and formulates response plans for them, thereby enabling the incident response system 30 to prevent shortages of necessary resources and execute appropriate incident responses. The resource allocation unit 16 consolidates multiple series of incidents into a single series and formulates response plans for that series, thereby enabling the incident response system 30 to prevent shortages of necessary resources for any of the multiple anticipated series and support the execution of appropriate incident responses.

[0091] The above description can be summarized as follows: Note that the following summary may include supplementary explanations and explanations of modifications of the above description.

[0092] The incident response support system 10 includes an interface device that receives input data from an external system 20, a storage device that stores resource management information (e.g., a resource management DB 4) that includes information about resources allocated for incident response for each region, and a processor connected to the interface device and the storage device. The processor determines whether an incident is predicted based on the input data from the external system 20, and generates series data of incident volume and its prediction error for predicted incidents. The processor determines a series of resource volumes based on the series of incident volumes, determines a series of reserve resource volumes based on the series of prediction errors of the incident volumes, and creates a work schedule for incident response based on the series of resource volumes, the series of reserve resource volumes, and the resource management information.

[0093] The input data may include at least one of weather data including weather forecast data for each of a plurality of regions and satellite image data including data representing satellite-captured images of the plurality of regions. The incident may be damage caused by a natural disaster. The incident volume may be the volume of damage caused by a natural disaster.

[0094] The processor may predict the amount of damage for each time in a target period by region based on the input data, and may predict a prediction error, which is the difference between the actual amount of damage and the predicted amount of damage, for each time in the target period. The start time of the target period may be the actual time of disaster occurrence or the predicted time of disaster occurrence. The end time of the target period may be the time of convergence as the predicted end time of the natural disaster, or a time later than the convergence time. For each region, the amount of damage for each time may be based on the severity of the natural disaster in that region and at that time, and the amount of equipment in that region.

[0095] The processor may predict, for each period, the amount of resources in each region for that period based on the amount of damage in that region and at that time for each of a plurality of times that belong to that period, and may predict the amount of reserve resources for all of the plurality of regions for that period based on the prediction error for each region and at that time for each of a plurality of times that belong to that period. The series of resource amounts may include the predicted amount of resources for each region for each period. The series of reserve resource amounts may include the predicted amount of reserve resources for all of the plurality of regions for each period.

[0096] The work schedule may be output to a work planning system 40 that creates a work plan based on the work schedule and displays the work plan to workers. The work schedule may include data representing resources and / or backup resources for each time period. For each task set in the work planning system 40, the work plan may include information representing resources and / or backup resources allocated to the task by the work planning system 40 from the work schedule.

[0097] For example, the resource management information may include, for each resource, information indicating the relationship between the resource ID (e.g., the ID of a worker or equipment), the resource type (e.g., whether the resource is a worker or equipment (e.g., heavy machinery)), the resource's skills, the resource's location (the resource's standby location), and the resource's reserve information (information including whether the resource is reserved, etc.) for each time. In a work schedule, the resource and / or the reserve resource for each time may be associated with at least some of the resource information (e.g., at least some of the resource ID, resource type, the resource's skills, the resource's location, and the resource's reserve information) identified from the resource management information for the time and the resource and / or the reserve resource. In order to create a work plan, for each task (e.g., recovery work) set by a user such as a site supervisor, the work planning system 40 may identify a resource and / or a reserve resource associated with resource information matching the attributes of the task (e.g., the task content, the task time, etc.) and assign the resource and / or the reserve resource to the task.

[0098] 10. Incident response support system

Claims

1. An incident response support system comprising: an interface device that receives input data from an external system; a storage device that stores resource management information including, for each region, information on resources to be allocated for incident response; and a processor connected to the interface device and the storage device, wherein the processor determines whether an incident is predicted based on the input data from the external system, generates series data of incident volume and prediction error of the incident volume for predicted incidents, determines a series of resource volumes based on the series of incident volumes, determines a series of reserve resource volumes based on the series of prediction error of the incident volume, and creates a work schedule for incident response based on the series of resource volumes, the series of reserve resource volumes, and the resource management information.

2. The incident response support system of claim 1, wherein the input data includes at least one of meteorological data including weather forecast data for each of a plurality of regions and satellite image data including data representing satellite images of the plurality of regions, the incidents are damage caused by natural disasters, and the incident volume is the volume of damage caused by the natural disaster.

3. The incident response support system of claim 2, wherein the processor: predicts the amount of damage for each time in a target period, by region, based on the input data; predicts a prediction error, which is the difference between the actual amount of damage and the predicted amount of damage, for each time in the target period; the start time of the target period is the actual time of disaster or the predicted time of disaster; the end time of the target period is the time of convergence as the predicted end time of the natural disaster, or a time later than the convergence time; and the amount of damage for each time, by region, is based on the severity of the natural disaster in that region and at that time, and the amount of facilities in that region.

4. The incident response support system of claim 3, wherein the processor: predicts, for each period, the resource amount for each of the regions in that period based on the amount of damage for that region and at that time for each of a plurality of times that belong to that period; predicts the amount of spare resources for all of the multiple regions in that period based on the prediction error for each of the multiple times that belong to that period; the series of resource amounts includes the predicted resource amount for each period for each of the regions; and the series of spare resource amounts includes the predicted spare resource amount for each period for all of the multiple regions.

5. The incident response support system of claim 1, wherein the work schedule is output to a work planning system which creates a work plan based on the work schedule and displays the work plan to workers, the work schedule includes data representing resources and / or backup resources for each time point, and the work plan includes, for each task set in the work planning system, information representing the resources and / or backup resources allocated by the work planning system from the work schedule for that task.

6. An incident response support method performed by a computer, comprising the steps of: determining whether an incident is predicted based on input data from an external system; generating series data of incident volume and prediction error of the incident volume for predicted incidents; determining a series of resource volume based on the series of incident volume; determining a series of reserve resource volume based on the series of incident volume prediction error; and creating a work schedule for incident response based on the series of resource volume, the series of reserve resource volume, and resource management information including, for each region, information on resources allocated for incident response.

Images