Blood donation operation system and trained model

Abstract

This blood donation management system (1) comprises: a data acquisition unit (11) that acquires population change data for each district demarcated in advance and actual donated blood amount data for each district and each type of blood; and a blood donation amount prediction unit (12) that predicts the blood donation amount for each district and for each type of blood on the basis of the population change data for each district and the actual donated blood amount data for each district and each type of blood acquired by the data acquisition unit (11).

Description

Technical Field

[0001] The present disclosure relates to a blood donation operation system having a function of predicting the blood donation volume for each pre-divided area, and a learned model used in the blood donation operation system. Further, the blood donation operation system can also have a function of optimizing the blood donation location (including the arrangement of blood donation vehicles) according to the predicted blood donation volume for each area.

Background Art

[0002] Blood donation, in which healthy people provide their own blood for patients who need blood transfusion for the treatment of diseases, surgery, etc., is widely known. As places for blood donation, in addition to fixed blood donation centers, mobile blood donation vehicles have also been widely used in recent years.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] For the efficient operation of blood donation, the effective arrangement of blood donation vehicles in various areas is important. For this purpose, it is important to accurately predict the blood donation volume for each area obtained by blood donation for each blood type. Although Patent Document 1 above describes a technique for predicting the demand volume related to the number of medical treatment acts based on various information, a technique related to the prediction of the blood donation volume focused on blood donation is not described, and a technique for accurately predicting the blood donation volume for each target area for each blood type is eagerly awaited.

[0005] The present disclosure has been made to solve the above problems, and an object thereof is to accurately predict the blood donation volume for each target area for each blood type.

Means for Solving the Problems

[0006] There is a technology that uses the structure of a wireless network used by a user's mobile terminal to obtain population trend data representing the population trend for each pre-divided area. Therefore, the applicant has invented a technology for accurately predicting the blood donation volume for each area by blood type as follows, based on the finding that the obtained blood donation volume fluctuates according to the population trend, by making use of the population trend data.

[0007] The blood donation operation system according to the present disclosure includes a data acquisition unit that acquires population trend data for each pre-divided area and blood donation volume performance data for each area and each blood type, and a blood donation volume prediction unit that predicts the blood donation volume for each area by each blood type based on the population trend data for each area and the blood donation volume performance data for each area and each blood type acquired by the data acquisition unit.

[0008] In the above blood donation operation system, the data acquisition unit acquires population trend data for each area and blood donation volume performance data for each area and each blood type, and the blood donation volume prediction unit predicts the blood donation volume for each area by each blood type based on the acquired population trend data for each area and the blood donation volume performance data for each area and each blood type. Here, "blood donation volume prediction" may be, for example, to generate a blood donation volume prediction model for predicting the blood donation volume for each area by machine learning for each blood type, and use the generated blood donation volume prediction model to predict the blood donation volume for each area by each blood type. As described above, it is possible to accurately predict the "blood donation volume for each area" that varies according to the population trend represented by the population trend data for each blood type.

Effect of the Invention

[0009] According to the present disclosure, it is possible to accurately predict the blood donation volume for each target area by each blood type.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Embodiments for Carrying Out the Invention

[0011] Embodiments of the blood donation operation system according to the present disclosure will be described with reference to the accompanying drawings.

[0012] As shown in FIG. 1, the blood donation operation system 1 includes a blood donation amount prediction device 10 that predicts the amount of blood donation, and a blood donation location determination device 20 that determines the blood donation location. Further, as peripheral devices of the blood donation operation system 1, there are a population statistics server 30, a blood donation history management server 40, a blood inventory management server 50, and a blood order management server 60.

[0013] Among the peripheral devices, the population statistics server 30 is a server that acquires and provides population transition data representing the population transition for each pre-divided area using the structure of the wireless network used by the user's mobile terminal, etc. In this embodiment, as the pre-divided area, an area (hereinafter referred to as "mesh") pre-divided in a mesh shape by boundary lines along the east-west and north-south directions is used as an example for explanation. However, it is not essential to use a mesh, and other pre-divided areas such as town blocks of administrative divisions may be used.

[0014] The blood donation history management server 40 is a server that stores and manages, for example, the blood donation history data shown in FIG. 2(a). As the blood donation history data, information regarding one blood donation such as the blood donation date and time, sex, age group, blood type, blood donation location, and blood donation amount is stored as one record. Note that the "blood type" when managing the blood donation amount is a concept that widely includes blood types such as type A and type B, and types such as red blood cells and platelets. In this embodiment, an example using blood types such as type A and type B as the "blood type" will be described. More specifically, an example of managing the blood donation amount for each mesh, for each sex and age group, and for each blood type will be described. Further, the blood donation history management server 40 stores and manages the location information of predetermined blood donation points within each mesh, and provides this information to the blood donation location determination device 20. The above-mentioned "blood donation points within each mesh" refer to the location of the blood donation station for meshes where a fixed blood donation station exists, and for meshes where no fixed blood donation station exists, it is a predetermined point where a blood donation vehicle can park and conduct blood donation.

[0015] The blood inventory management server 50 is a server that stores and manages blood inventory data as shown in, for example, Fig. 2(b). In the blood inventory data, information for each blood product, such as sex, age, blood type, blood volume, expiration date, destination of shipment, shipment time, and status, is stored as one record with the blood product number as the key.

[0016] The blood order management server 60 is a server that stores and manages blood order data as shown in, for example, Fig. 2(c). As the blood order data, information regarding one order, such as the hospital that placed the blood order, the details of the blood product order, the order time, status, delivery deadline, and shipment time, is stored as one record.

[0017] In this embodiment, an example will be described in which the predicted blood donation volume is predicted for each sex-age group and each blood type in mesh units, and further, based on the information including the predicted blood donation volume values for each sex-age group and each blood type in mesh units obtained, a blood donation location including the location of the mobile blood donation vehicle in a time series is determined. Note that the requirement of "for each sex-age group" among the above is not essential, and the prediction of the blood donation volume etc. may be performed regardless of sex and age. However, by adding the requirement of "for each sex-age group", it becomes possible to obtain the "insufficient blood donation volume" etc. for each sex-age group, and there is an advantage that enlightenment activities such as promoting blood donation participation for sex-age groups with a large or increasing tendency of insufficient blood donation volume can be effectively implemented.

[0018] Now, in the blood donation operation system 1 of Fig. 1, the blood donation volume prediction device 10 includes a data acquisition unit 11, a blood donation volume prediction unit 12, and an information transmission unit 13, and the blood donation location determination device 20 includes an information acquisition unit 21, a blood donation location determination unit 22, and an information output unit 23. Hereinafter, the functions of each unit will be described.

[0019] The data acquisition unit 11 is a functional unit that acquires population transition data for each pre-divided mesh from the population statistics server 30 and acquires the blood donation history data illustrated in Fig. 2(a) from the blood donation history management server 40.

[0020] The blood donation volume prediction unit 12 is a functional unit that aggregates the blood donation history data illustrated in Fig. 2(a) for each mesh, for each gender and age group, and for each blood type, to obtain the blood donation volume performance data for each mesh, for each gender and age group, and for each blood type, and predicts the blood donation volume for each mesh for each gender and age group and for each blood type based on the obtained blood donation volume performance data and the population transition data for each mesh. The details of the prediction process by the blood donation volume prediction unit 12 will be described later with reference to Fig. 3 and the like.

[0021] The information transmission unit 13 is a functional unit that transmits the information on the predicted blood donation volume for each mesh, for each gender and age group, and for each blood type obtained by the prediction process by the blood donation volume prediction unit 12 to the blood donation location determination device 20.

[0022] The information acquisition unit 21 is a functional unit that acquires various information necessary for determining the blood donation location. For example, the information acquisition unit 21 acquires the information on the predicted blood donation volume for each mesh, for each gender and age group, and for each blood type from the information transmission unit 13, acquires the blood inventory data illustrated in Fig. 2(b) from the blood inventory management server 50, acquires the blood order data illustrated in Fig. 2(c) from the blood order management server 60, and further acquires the location information of the predetermined blood donation locations within each mesh from the blood donation history management server 40. Note that it is not necessary to acquire the location information of the blood donation locations within each mesh every time, and it may be acquired each time there is a change.

[0023] The blood donation location determination unit 22 is a functional unit that determines the blood donation location including the placement locations of the blood donation vehicles in a time series. The blood donation location determination process by the blood donation location determination unit 22 includes: (a) In an evaluation function with the total blood donation volume obtained from the shortage blood donation volume obtained based on the blood inventory information and blood order information and the acquired blood donation volume obtained based on the predicted blood donation volume for each mesh as the output value, determining the blood donation location by solving an optimization problem of avoiding the total blood donation volume from becoming a negative value or minimizing the absolute value of the negative value; and (b) generating a blood donation location determination model for determining the blood donation location by machine learning and using the blood donation location determination model to determine the blood donation location. In the present embodiment, the method (a) will be described, but the method (b) will be described in Modification 1.

[0024] The information output unit 23 is a functional unit that outputs information on the blood donation location, including the time-series arrangement locations of the blood donation vehicles, determined by the blood donation location determination unit 22. Here, "output" can adopt various forms, such as display on a display, voice output from a speaker, printing by a printer, data output to an external device, etc.

[0025] Next, the processes executed in the blood donation operation system 1 will be described along the flowchart of FIG. 3. The execution timing of this process is arbitrary, and various patterns can be adopted, such as the timing when a pre-scheduled time arrives, the timing when an operator of the blood donation volume prediction device 10 inputs a start command, etc.

[0026] First, the population statistics server 30 provides data on population transition data for each mesh in the past predetermined period (for example, from the previous prediction implementation date to the current prediction implementation date) to the blood donation volume prediction device 10 (step S1), and the data acquisition unit 11 of the blood donation volume prediction device 10 acquires the provided data and passes it to the blood donation volume prediction unit 12. Also, the blood donation history management server 40 provides data on the blood donation history data in the above-mentioned predetermined period shown in FIG. 2(a) to the blood donation volume prediction device 10 (step S2), and the data acquisition unit 11 of the blood donation volume prediction device 10 acquires the provided data and passes it to the blood donation volume prediction unit 12.

[0027] Next, as shown in FIG. 4, the blood donation volume prediction unit 12 aggregates the blood donation history data in FIG. 2(a) for each mesh, each sex-age group, and each blood type, thereby obtaining blood donation volume performance data for each mesh, each sex-age group, and each blood type (step S3). For example, in FIG. 4, the blood donation volume of 400 ml at 12:13 and the blood donation volume of 300 ml at 12:45 for the sex-age group "female in her 20s" with blood type "A" in mesh A on May 28, 2021 are aggregated, and the blood donation volume of 700 ml for the sex-age group "female in her 20s" with blood type "A" in mesh A in the time period from 12:00 to 13:00 on May 28, 2021 is obtained.

[0028] Then, the blood donation volume prediction unit 12 generates and stores a blood donation volume prediction model 12A for each blood type as follows (step S4). For example, as shown in FIG. 5(a), the blood donation volume prediction unit 12 uses the population transition data for each mesh in the acquired past predetermined period (e.g., from the previous prediction execution date to the current prediction execution date) as an explanatory variable, and the blood donation volume performance data for each sex-age group and blood type aggregated hourly for each mesh in the above-mentioned predetermined period as an objective variable, and performs machine learning for each sex-age group and blood type to generate a blood donation volume prediction model 12A for predicting the blood donation volume hourly for each mesh for each sex-age group and blood type. Note that an example targeting "blood type" as the blood type is described, but the same processing applies to blood types other than blood type (types such as red blood cells and platelets).

[0029] After that, when it becomes the day to implement the blood donation volume prediction and it is the timing to execute the blood donation volume prediction, etc., the processes after step S5 in FIG. 3 are executed. In the actual operation of arranging the blood donation vehicle, since the blood donation location needs to be determined in advance and prior notice to the relevant persons and departments is required, for example, when determining the blood donation location in the afternoon of a certain day (including the arrangement of the blood donation vehicle in time series), the processes after step S5 are executed by around 10:00 am on that day, and when determining the blood donation location in the morning of a certain day (including the arrangement of the blood donation vehicle in time series), the processes after step S5 may be executed by around 3:00 pm on the previous day.

[0030] The demographic server 30 provides the blood donation volume prediction device 10 with the population transition data for each mesh in the prediction target period (for example, from the previous prediction execution date to the current prediction execution date) as the current population transition data (step S5). The data acquisition unit 11 of the blood donation volume prediction device 10 acquires the provided population transition data for each mesh. As shown in FIG. 5(b), the blood donation volume prediction unit 12 inputs the population transition data for each mesh in the prediction target period into the blood donation volume prediction models 12A for each sex / age group and each blood type, thereby predicting the blood donation volume for each mesh for each sex / age group and each blood type (step S6). The information on the predicted blood donation volume obtained by this prediction is transferred from the blood donation volume prediction unit 12 to the information transmission unit 13, and the information transmission unit 13 transmits the information on the predicted blood donation volume to the blood donation location determination device 20 (step S7). Similarly, the blood inventory management server 50 transmits the blood inventory data shown in, for example, FIG. 2(b) to the blood donation location determination device 20 (step S8), the blood order management server 60 transmits the blood order data shown in, for example, FIG. 2(c) to the blood donation location determination device 20 (step S9), and the blood donation history management server 40 transmits the position information of the predetermined blood donation locations within each mesh to the blood donation location determination device 20 (step S10). Note that it is not necessary to transmit the position information of the blood donation locations within each mesh every time, and it may be transmitted each time there is a change.

[0031] The information acquisition unit 21 of the blood donation location determination device 20 acquires the above various information and transfers it to the blood donation location determination unit 22. The blood donation location determination unit 22 determines the blood donation location including the arrangement location of the blood donation vehicle in chronological order as follows (step S11). For example, a conversion coefficient α for converting the "insufficient blood volume" into the "insufficient blood donation volume" is obtained in advance. As shown in FIG. 6, the blood donation location determination unit 22 subtracts the blood order data from the blood inventory data for each sex-age group and each blood type to obtain the "insufficient blood volume" for each sex-age group and each blood type, and multiplies the obtained "insufficient blood volume" by the conversion coefficient α to obtain the "insufficient blood donation volume" for each sex-age group and each blood type. Further, the blood donation location determination unit 22 uses the obtained "insufficient blood donation volume" for each sex-age group and each blood type, the predicted blood donation volume per hour for each mesh obtained for each sex-age group and each blood type, and the position information of the blood donation locations within each mesh as variables for the blood donation location including the arrangement location of the blood donation vehicle. In an evaluation function with the obtained blood donation volume based on the predicted blood donation volume per mesh as the acquisition blood donation volume and the "total blood donation volume for each sex-age group and each blood type" obtained from the above insufficient blood donation volume as the output value, an optimization problem of avoiding the total blood donation volume for each sex-age group and each blood type from becoming a negative value or minimizing the absolute value of the negative value is solved with the position information regarding the blood donation locations within each mesh (the pre-determined blood donation vehicle parking positions and the positions of fixed blood donation centers for each mesh) and the predicted blood donation volume per mesh as constraint conditions to determine the blood donation location including the arrangement location of the blood donation vehicle in chronological order. As a result, for example, as shown in the lower right of FIG. 6, the blood donation vehicle 1 is arranged at location H from 12:00 to 15:00 and at location D from 15:00 to 18:00, the blood donation vehicle 2 is arranged at location G from 12:00 to 15:00 and at location C from 15:00 to 18:00, the blood donation vehicle 3 is arranged at location B from 12:00 to 18:00, and the blood donation center F operates at the fixed location F from 12:00 to 18:00, and the like, and the blood donation locations are determined. It should be noted that it is not essential to set the above conversion coefficient α to a uniform value, and as the conversion coefficient α, for example, values obtained in advance for each blood type, values obtained in advance for each sex-age group, values obtained in advance for each combination of sex-age and blood type, etc. may be adopted.

[0032] Then, the information on the blood donation locations, including the time-series locations of the blood donation vehicles, is transferred from the blood donation location determination unit 22 to the information output unit 23, and the information output unit 23 outputs the information on the blood donation locations (step S12). For example, it is displayed and output to an operator terminal (not shown) of the blood donation location determination device 20. As a result, the operator can recognize the information on the blood donation locations, including the time-series locations of the blood donation vehicles as illustrated in FIG. 8, and based on this information, can realize an efficient blood donation operation at an optimized blood donation location.

[0033] Here, with reference to FIGS. 7 and 8, an example of determining a blood donation location will be outlined. In FIGS. 7 and 8, to avoid complexity in the figures, an example of targeting a specific gender and age group is described, so it does not represent the values for each gender and age group. In this way, when targeting a specific gender and age group, the blood donation location determination unit 22 determines a blood donation location that avoids the total blood donation volume for each blood type from becoming a negative value or minimizes the absolute value of the negative value. However, in the example of the blood donation location shown in FIG. 7, the total blood donation volume of type B and type AB has become a negative value, which is a non-optimized blood donation location. On the other hand, in the example of the blood donation location shown in FIG. 8, the total blood donation volume of all blood types is a positive value, and the shortages of type B and type AB are eliminated, which is an optimized blood donation location.

[0034] (Effects of this Embodiment) In the blood donation operation system 1 in this embodiment, it is possible to accurately predict the "blood donation volume per mesh" that varies according to the population trend represented by the population transition data for each gender and age group and each blood type.

[0035] Also, along with the above, based on the predicted values of the blood donation volume per mesh, per gender and age group, and per blood type that can be accurately predicted, by solving an optimization problem to avoid the total blood donation volume for each gender and age group from becoming a negative value or minimizing the absolute value of the negative value, among the two examples shown in FIGS. 7 to 8, it is possible to determine an optimized blood donation location that avoids the total blood donation volume from becoming a negative value as shown in FIG. 8.

[0036] In addition, a conversion coefficient for converting the shortage blood volume, which is the difference value obtained by subtracting the blood order information from the blood inventory information, into the shortage blood donation volume is obtained in advance. The blood donation location determination unit multiplies the shortage blood volume by the conversion coefficient to obtain the shortage blood donation volume, and obtains the "total blood donation volume", which is an index for determining the appropriateness of the blood donation location, from the obtained shortage blood donation volume and the predicted blood donation volume (acquired blood donation volume) for each mesh that could be accurately predicted. Therefore, based on the appropriate total blood donation volume, the appropriateness of the blood donation location can be appropriately determined.

[0037] Furthermore, in the present embodiment, an example of predicting the blood donation volume in mesh units "for each gender and age group" and "for each blood type" and then determining a blood donation location including the placement location of the blood donation vehicle over time based on the information including the predicted blood donation volume values for each mesh for each gender and age group and for each blood type has been described. Although the requirement of "for each gender and age group" here is not essential, adding the requirement of "for each gender and age group" enables the calculation of the "shortage blood donation volume" and the like for each gender and age group, and has the advantage that enlightenment activities such as promoting blood donation participation for gender and age groups with a large shortage blood donation volume or an increasing trend can be effectively implemented.

[0038] (Modification Example 1) Modification Example 1 is an example in which when the blood donation location determination unit 22 determines the blood donation location, the above-described method (b) "generating a blood donation location determination model for determining the blood donation location by machine learning and using the blood donation location determination model to determine the blood donation location including the placement location of the blood donation vehicle over time" is adopted.

[0039] As shown in FIG. 9, in the blood donation operation system 1 according to Modification 1, the blood donation location determination unit 22 generates and stores a blood donation location determination model 22A for determining a blood donation location as follows. More specifically, as shown in FIGS. 10(a) and 11, the blood donation location determination unit 22 uses, as explanatory variables, the location information of blood donation locations (blood donation vehicle parking positions and fixed blood donation sites) within each mesh at the time of past blood donation location determination, blood inventory information for each sex / age group and each blood type, blood order information for each sex / age group and each blood type, and the predicted blood donation volume per hour per mesh for each sex / age group and each blood type, and uses the information of the above blood donation location determined at the time of the blood donation location determination as the objective variable to perform machine learning to generate a blood donation location determination model for determining a blood donation location (step S11A in FIG. 11). Then, as shown in FIGS. 10(b) and 11, the blood donation location determination unit 22 inputs into the generated blood donation location determination model 22A the location information of blood donation locations (blood donation vehicle parking positions and fixed blood donation sites) within each mesh on the target day, blood inventory information for each sex / age group and each blood type, blood order information for each sex / age group and each blood type, and the predicted blood donation volume per hour per mesh for each sex / age group and each blood type, to determine a blood donation location including the time-series arrangement locations of blood donation vehicles on the target day (step S11B in FIG. 11).

[0040] According to Modification 1 as described above, while basing on the predicted blood donation volume per hour per mesh that can be accurately predicted, a blood donation location determination model for determining a blood donation location is generated, and by using the generated blood donation location determination model, among the two examples shown in FIGS. 7 to 8 described above, a blood donation location including the optimized time-series arrangement locations of blood donation vehicles as shown in FIG. 8 can be determined.

[0041] (Modification 2) The blood donation operation system 1 does not necessarily have to be configured to include a blood donation volume prediction device 10 and a blood donation location determination device 20 as shown in FIGS. 1 and 9. The blood donation operation system 1 may be configured as a single device as shown in FIG. 12. In that case, the information transmission unit 13 and the information acquisition unit 21 for transmitting and receiving information between devices may be omitted, and the blood donation location determination unit 22 may be configured to directly acquire information from each of the blood donation volume prediction unit 12, the blood inventory management server 50, and the blood order management server 60. Even in the blood donation operation system 1 configured in this way, it is possible to execute the same processing as the blood donation operation system 1 shown in FIGS. 1 and 9, and the same effects can be obtained.

[0042] (Explanation of terms, explanation of hardware configuration (FIG. 13), etc.) Note that both the aforementioned blood donation volume prediction model 12A (FIGS. 1, 9, 12) and the blood donation location determination model 22A (FIG. 9) are so-called learned models and are assumed to be used as program modules that are part of artificial intelligence software. That is, these learned models are "instructions for a computer" used in a computer equipped with a processor (CPU) and a memory as shown in FIG. 13 described later, and are combined so that one result can be obtained (a predetermined process can be executed), that is, a computer program that causes the computer to function. In other words, the above-mentioned learned models are a combination of the structure of a neural network and parameters (weight coefficients) that are the strengths of the connections between the neurons of the neural network. Specifically, the processor (CPU) of the computer operates to output a predicted blood donation volume value for each mesh of the target blood type with the population transition data for each mesh in the prediction target period as an input value according to the instructions from the blood donation volume prediction model 12A stored in the memory. Also, the processor (CPU) of the computer operates to output information on the blood donation location including the arranged locations of the blood donation vehicles in time series with the predicted blood donation volume value for each mesh of a certain blood type, location information on the parking positions of the blood donation vehicles and the locations of the blood donation centers for each mesh, blood inventory information, and blood order information as input values according to the instructions from the blood donation location determination model 22A stored in the memory.

[0043] In addition, the block diagrams used in the description of the above embodiments and modified examples show blocks of functional units. These functional blocks (components) are realized by any combination of at least one of hardware and software. Also, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using one physically or logically combined device, or two or more physically or logically separated devices may be directly or indirectly (for example, using wired, wireless, etc.) connected and realized using these multiple devices. The functional block may be realized by combining software with the above one device or the above multiple devices.

[0044] Functions include, but are not limited to, judgment, decision, determination, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, solution, selection, selection, establishment, comparison, assumption, expectation, regarded as, notification (broadcasting), notification (notifying), communication (communicating), forwarding, configuration (configuring), reconfiguration (reconfiguring), allocation (allocating, mapping), assignment (assigning), etc. For example, a functional block (component) that functions as transmission is called a transmitting unit, a transmitter. In any case, as described above, the realization method is not particularly limited.

[0045] For example, the blood donation amount prediction device in the blood donation operation system of the present disclosure may function as a computer that performs the processing in the present embodiment. FIG. 13 is a diagram showing an example of the hardware configuration of the blood donation amount prediction device 10. Physically, the above-described blood donation amount prediction device 10 may be configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like. Hereinafter, the blood donation amount prediction device 10 will be described as an example, but the same applies to other devices (blood donation location determination device 20) constituting the blood donation operation system.

[0046] In the following description, the term "device" can be read as a circuit, a device, a unit, etc. The hardware configuration of the blood donation volume prediction device 10 may be configured to include one or more of each device shown in the figure, or may be configured without including some devices.

[0047] Each function in the blood donation volume prediction device 10 is realized by causing a processor 1001 to perform calculations and control communication by a communication device 1004, or by controlling at least one of reading and writing data in a memory 1002 and a storage 1003, by loading a predetermined software (program) onto hardware such as the processor 1001 and the memory 1002.

[0048] The processor 1001 controls the entire computer by operating an operating system, for example. The processor 1001 may be constituted by a central processing unit (CPU: Central Processing Unit) including an interface with peripheral devices, a control device, an arithmetic device, a register, and the like.

[0049] Further, the processor 1001 reads a program (program code), a software module, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002, and executes various processes according to these. As the program, a program for causing a computer to execute at least a part of the operations described in the above-described embodiments is used. Although it has been described that the above-described various processes are executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. Note that the program may be transmitted from a network via a telecommunication line.

[0050] The memory 1002 is a computer-readable recording medium and may be composed of at least one of, for example, ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), etc. The memory 1002 may also be referred to as a register, a cache, a main memory (main storage device), etc. The memory 1002 can store a program (program code), a software module, etc. that can execute the wireless communication method according to an embodiment of the present disclosure.

[0051] The storage 1003 is a computer-readable recording medium and may be composed of at least one of, for example, an optical disc such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disc, a digital versatile disc, a Blu-ray (registered trademark) disc), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. The storage 1003 may also be referred to as an auxiliary storage device. The above-described recording medium may be, for example, a database including at least one of the memory 1002 and the storage 1003, or other appropriate media.

[0052] The communication device 1004 is hardware (a transmission / reception device) for performing communication between computers via at least one of a wired network and a wireless network, and is also referred to as, for example, a network device, a network controller, a network card, a communication module, etc.

[0053] The input device 1005 is an input device that receives external input (for example, a keyboard, mouse, microphone, switch, button, sensor, etc.). The output device 1006 is an output device that performs external output (for example, a display, speaker, LED lamp, etc.). Note that the input device 1005 and the output device 1006 may have an integrated configuration (for example, a touch panel). Also, each device such as the processor 1001 and the memory 1002 is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus or may be configured using different buses for each device.

[0054] Each aspect / embodiment described in the present disclosure may be used alone, in combination, or switched and used during execution. Also, the notification of predetermined information (for example, the notification of "being X") is not limited to being explicitly performed, and may be performed implicitly (for example, by not performing the notification of the predetermined information).

[0055] As described above in detail about the present disclosure, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described in the present disclosure. The present disclosure can be implemented as a modified and changed aspect without departing from the spirit and scope of the present disclosure determined by the description of the claims. Therefore, the description of the present disclosure is for the purpose of illustrative explanation and has no restrictive meaning for the present disclosure.

[0056] The processing procedures, sequences, flowcharts, etc. of each aspect / embodiment described in the present disclosure may be rearranged as long as there is no contradiction. For example, regarding the methods described in the present disclosure, the elements of various steps are presented using an exemplary order and are not limited to the specific order presented.

[0057] The input / output information etc. may be stored in a specific location (e.g., memory), or may be managed using a management table. The information etc. to be input / output may be overwritten, updated, or appended. The output information etc. may be deleted. The input information etc. may be transmitted to other devices.

[0058] In the present disclosure, the description "based on" used herein does not mean "only based on" unless otherwise specified. In other words, the description "based on" means both "only based on" and "at least based on".

[0059] In the present disclosure, when terms such as "include", "including" and their variants are used, these terms are intended to be inclusive, similar to the term "comprising". Further, the term "or" used in the present disclosure is not intended to be an exclusive disjunction.

[0060] In the present disclosure, for example, when articles are added by translation like a, an and the in English, the present disclosure may include that the nouns following these articles are in the plural form.

[0061] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other". Note that the term may also mean "A and B are each different from C". Terms such as "separate", "coupled" etc. may also be interpreted in the same way as "different".

Description of Reference Numerals

[0062] 1…Blood donation operation system, 10…Blood donation volume prediction device, 11…Data acquisition unit, 12…Blood donation volume prediction unit, 12A…Blood donation volume prediction model, 13…Information transmission unit, 20…Blood donation location determination device, 21…Information acquisition unit, 22…Blood donation location determination unit, 22A…Blood donation location determination model, 23…Information output unit, 30…Demographic server, 40…Blood donation history management server, 50…Blood inventory management server, 60…Blood order management server, 1001…Processor, 1002…Memory, 1003…Storage, 1004…Communication device, 1005…Input device, 1006…Output device, 1007…Bus.

Claims

1. A data acquisition unit that acquires population transition data for each pre-divided area and blood donation volume achievement data for each area and each blood type, and A blood donation volume prediction unit that predicts the blood donation volume for each area for each blood type based on the population transition data for each area and the blood donation volume achievement data for each area and each blood type acquired by the data acquisition unit, Comprising, The blood donation volume prediction unit Using the population transition data for each area in a past predetermined period as an explanatory variable and the blood donation volume achievement data for each area and each blood type in the predetermined period as an objective variable, machine learning is performed for each blood type to generate a blood donation volume prediction model for predicting the blood donation volume for each area for each blood type, By inputting the population transition data for each area in the prediction target period into the generated blood donation volume prediction model for each blood type, the blood donation volume for each area in the prediction target period is predicted for each blood type, Blood donation operation system.

2. At least, blood inventory information for each blood type regarding the available blood inventory, blood order information for each blood type regarding the blood order volume, location information regarding the pre-determined blood donation vehicle parking positions and the positions of fixed blood donation sites for each area, and based on the blood donation volume prediction values for each area and each blood type obtained by the prediction of the blood donation volume prediction unit, a blood donation site determination unit that determines a blood donation site including the placement location of the blood donation vehicle in a time series, The blood donation operation system according to claim 1, further comprising.

3. The blood donation site determination unit In an evaluation function that uses the blood donation site including the placement location of the blood donation vehicle as a variable and the shortage blood donation volume obtained based on the blood inventory information for each blood type and the blood order information for each blood type, and the total blood donation volume for each blood type obtained based on the blood donation volume prediction values for each area and each blood type as an output value, an optimization problem of avoiding the total blood donation volume for each blood type from becoming a negative value or minimizing the absolute value of the negative value, By solving the position information regarding the pre-determined blood donation vehicle parking positions and the positions of fixed blood donation sites for each area and the blood donation volume prediction values for each area and each blood type as constraint conditions, the blood donation site including the placement location of the blood donation vehicle in a time series is determined, The blood donation operation system according to claim 2.

4. A conversion coefficient for converting the shortage blood volume, which is the difference value obtained by subtracting the blood order information from the blood inventory information, into the shortage blood donation volume has been obtained in advance. The blood donation location determination unit obtains the shortage blood donation volume by multiplying the shortage blood volume obtained by subtracting the blood order information from the blood inventory information by the conversion coefficient. The blood donation operation system according to claim 3.

5. The blood donation location determination unit Using, as explanatory variables, position information regarding the parking positions of blood donation vehicles and the positions of fixed blood donation sites predetermined for each region at the time of past blood donation location determination, the blood inventory information for each blood type, the blood order information for each blood type, and the blood donation volume prediction values for each region and each blood type, and using the information on the blood donation location determined at the time of the blood donation location determination as the objective variable, machine learning is performed to generate a blood donation location determination model for determining the blood donation location. By inputting the position information regarding the parking positions of blood donation vehicles and the positions of fixed blood donation sites predetermined for each region on the target day, the blood inventory information for each blood type, the blood order information for each blood type, and the blood donation volume prediction values for each region and each blood type into the generated blood donation location determination model, the blood donation location including the placement locations of the time-series blood donation vehicles on the target day is determined. The blood donation operation system according to claim 2.

6. The data acquisition unit acquires population transition data for each sex-age group and each region, and blood donation volume achievement data for each sex-age group, each region, and each blood type. The blood donation volume prediction unit predicts the blood donation volume for each region for each sex-age group and each blood type. The blood donation location determination unit determines the blood donation location including the placement locations of the time-series blood donation vehicles based on information including the blood donation volume prediction values for each region, each sex-age group, and each blood type. The blood donation operation system according to claim 2.

7. A learned model for causing a computer to function so as to output a blood donation volume prediction value for each region of a certain blood type, Generated by machine learning using, as explanatory variables, the actual values of past population transition data for each region, and using, as the objective variable, the actual values of the blood donation volume for each region of the blood type, A learned model for causing a computer to function so as to output a blood donation volume prediction value for each region of the blood type, using the population transition data for each region during the prediction target period as the input value.

Images