Information processing system and program

Abstract

To take into account the attributes of doctors and hypothetical influences from a total sales number of products per medical facility and estimate a plausible individual sales number per doctor of these products.SOLUTION: An adaptive information acquisition unit 111 acquires adaptive information on products. An attribute information acquisition unit 112 acquires the attribute information of each of doctors belonging to a medical facility. A total sales number acquisition unit 113 acquires information on the total sales number of products sold in the medical facility. A hypothesis acquisition unit 114 acquires a hypothesis content. A derivation unit 115 refers to a derivation program database 125 and derives a program conforming to each hypothesis. A generation unit 116 incorporates a total sales number of a product into the program conforming to the hypothesis and generates an objective function for determining a constraint condition to be imposed on an individual sales number per doctor of the product and a most likelihood individual sales number. An estimation unit 117 incorporates the constraint condition and the adaptive information and attribute information into the objective function and estimates the most likelihood individual sales number.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to an information processing system and a program for use in a sales survey of products used in the medical field. [Background technology]

[0002] Methods of clarifying targets in sales and marketing using statistical information on pharmaceuticals are being investigated.

[0003] For example, Patent Document 1 describes a farming support system that calculates the cost price of agricultural products to be sold based on preparation costs, farm work costs, shipping costs, cultivated area, and the like. Patent document 2 also describes a computer system that determines the number of patients to whom the target drug is prescribed in the entire area during a unit period by dividing the prescription quantity of the target drug in the entire area during a unit period by the average number of days the target drug is administered in the unit period and the average daily dosage of the target drug per person, and then determines the number of patients to whom the target drug is prescribed in the target area during a unit period by multiplying the determined number of patients to whom the drug is prescribed by the ratio of the number of prescriptions in the target area to the number of prescriptions in the entire area, and transmits this to a terminal device. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-197940 A [Patent Document 2] JP 2022-142063 A Summary of the Invention [Problem to be solved by the invention]

[0005] The technology described in Patent Document 1 can determine which plots of agricultural crops have a high profit margin by calculating the expected sales amount for each plot, and can optimally select the number of plots into which the field is divided, the area of ​​each plot, and the agricultural crops for each plot, thereby enabling optimal farming of profitable fields. However, this technology is for agricultural crops, and cannot be applied to verify various hypotheses regarding product sales, taking into account products such as pharmaceuticals, their applications, and the attributes of the doctors who sell them to patients, who are the end users. In addition, the technology described in Patent Document 2 can correct the estimation results by considering the ratio of medical departments in the target area among the entire area and the ratio of the number of doctors belonging to each medical department. However, this technology cannot reflect the influence of other doctor attributes such as job title, or various hypotheses regarding product sales, on the estimation results.

[0006] One of the objectives of the present invention is to estimate the plausible individual sales volume for each doctor for a product based on the total sales volume for each medical facility, taking into account the influence of doctor attributes and hypotheses. [Means for solving the problem]

[0007] In one aspect, the present invention provides an information processing system having an indication information acquisition unit that acquires information on the indications of a product used in medical care, an attribute information acquisition unit that acquires attribute information about doctors belonging to a medical facility, a total sales acquisition unit that acquires information on the total sales volume of the product sold at the medical facility, a hypothesis acquisition unit that acquires a plurality of hypotheses regarding the indications of the product and the attributes of the doctors, a generation unit that generates, from a formula derived from each of the plurality of hypotheses and the total sales volume, constraints to be imposed on an individual sales volume of the product for each doctor or for each indication for which the product is used, and an objective function to obtain the maximum likelihood individual sales volume, an estimation unit that estimates the maximum likelihood individual sales volume based on the generated constraints and objective function, and an output unit that outputs information on the estimated maximum likelihood individual sales volume. Effect of the Invention

[0008] The present invention can estimate the likely individual sales volume for each doctor for a product by taking into consideration the influence of doctor attributes and hypotheses from the total sales volume of the product for each medical facility. [Brief description of the drawings]

[0009] [Figure 1] 1 is a diagram showing an example of the configuration of an information processing device 1 according to the present invention. [Diagram 2] FIG. 4 is a diagram showing an example of an adaptive DB 121. [Diagram 3] FIG. 4 is a diagram showing an example of an attribute DB 122. [Figure 4] FIG. 13 shows an example of a total sales number DB123. [Diagram 5] FIG. 13 is a diagram showing an example of a hypothesis DB 124. [Figure 6] FIG. 13 is a diagram showing an example of a derivation program DB 125. [Figure 7] FIG. 2 is a diagram showing an example of the functional configuration of an information processing device 1. [Figure 8] FIG. 4 is a flowchart showing an example of the operation flow of the information processing device 1. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] <Embodiment> <Configuration of information processing device> Fig. 1 is a diagram showing an example of the configuration of an information processing device 1 according to the present invention. The information processing device 1 according to the present invention is a device that performs information processing to analyze the actual sales volume of products used in medical care, such as medicines and medical devices, and utilize the analysis results in sales activities and marketing activities for the products. The information processing device 1 is, for example, a computer. The information processing device 1 shown in Fig. 1 has a processor 11, a memory 12, a communication unit 13, an input unit 14, and an output unit 15. These components are connected to each other so that they can communicate with each other, for example, by a bus.

[0011] The processor 11 reads and executes a program stored in the memory 12 to control each part of the information processing device 1. The processor 11 is, for example, a CPU (Central Processing Unit).

[0012] The input unit 14 includes operators such as operation buttons, a keyboard, a touch panel, and a mouse for various inputs, and receives an operation and sends a signal according to the operation content to the processor 11. This operation is, for example, pressing a button, a gesture on a touch panel, etc.

[0013] The output unit 15 is a device for performing various outputs. The output unit 15 has a display screen such as a liquid crystal display, and displays images for conveying information to the user under the control of the processor 11. A transparent touch panel of the input unit 14 may be placed on top of this display screen. The output unit 15 may also have a speaker for conveying information to the user by synthesized voice.

[0014] The communication unit 13 is a communication circuit that communicatively connects the information processing device 1 to an external device or the like by wire or wirelessly. The above-mentioned input unit 14 or output unit 15 may be an external device via the communication unit 13. When the input unit 14 or the output unit 15 is substituted by an external device via the communication unit 13, the information processing device 1 does not need to have these units.

[0015] The memory 12 is a storage means for storing an operating system, various programs, data, etc., which are loaded into the processor 11. The memory 12 includes a RAM (Random Access Memory) and a ROM (Read Only Memory).

[0016] The memory 12 may include a solid state drive, a hard disk drive, etc. The memory 12 also stores an adaptation DB121, an attribute DB122, a total sales number DB123, a hypothesis DB124, and a derivation program DB125.

[0017] <Adaptive DB configuration> Fig. 2 is a diagram showing an example of the adaptation DB 121. The adaptation DB 121 is a database that stores information (called adaptation information) about the adaptation of each product. The adaptation DB 121 shown in Fig. 2 has the items of "product ID", "product name", and "adaptation information". The product ID is identification information for uniquely identifying a product used in the above-mentioned medical treatment. The product name is the name of the product. The adaptation information is information about the adaptation of the product.

[0018] Here, "indication" means that the effect of the above-mentioned product is medically recognized and it is the subject of use. Indication information is information on symptoms for which the product is mainly effective, symptoms for which the product is effective as a side effect, etc., symptoms for which the product is contraindicated, etc. For example, indication information is information such as "Indication diseases for which a doctor can prescribe include HER2-positive inoperable or recurrent breast cancer with a history of chemotherapy, BRCA gene mutation-positive and HER2-negative inoperable or recurrent breast cancer with a history of cancer chemotherapy, and esophageal cancer with recurrence or distant metastasis." In addition, the indication DB 121 may store information on the approval of the product by public institutions such as the Ministry of Health, Labor and Welfare.

[0019] <Attribute DB configuration> Fig. 3 is a diagram showing an example of the attribute DB 122. The attribute DB 122 is a database that stores, for each facility (clinic, hospital, nursing home, etc.) to which the above-mentioned products are sold, doctors who belong to the facility and their attributes in association with each other. The attribute DB 122 shown in Fig. 3 has a facility list 1221 and an attribute table 1222.

[0020] The facility list 1221 is a list of the above-mentioned facilities. The facility list 1221 has the items of "facility ID" and "facility name". The facility ID is identification information for uniquely identifying the above-mentioned facility. The facility name is the name of the facility.

[0021] The attribute table 1222 is a table that lists doctors belonging to each facility in association with their attributes. The attribute table 1222 shown in Fig. 3 is associated with each facility listed in the facility list 1221. This attribute table 1222 has items of "Doctor ID" and "Attribute". The "Attribute" item has sub-items such as "Medical department", "Position (variable 1)", "Predicted value of indication 1 (variable 2)", "Predicted value of indication 2 (variable 3)", etc.

[0022] The doctor ID is identification information for identifying each doctor who belongs to the corresponding medical facility. The doctor ID may be associated with the doctor's name.

[0023] The attribute is information indicating the attribute of the doctor identified by the corresponding doctor ID. As described above, the attribute may have multiple sub-items. For example, "department" indicates the department to which the doctor belongs. That is, the department to which the doctor identified by doctor ID "A" in FIG. 3 belongs is "urology."

[0024] "Position" is information indicating the position of the doctor. The position is the position in the medical facility, the responsibilities, the rank, the decision-making authority, the job duties, etc. The position is expressed, for example, on an ordinal scale, an interval scale, a ratio scale, etc. The position may be expressed on a nominal scale.

[0025] The "predicted value for indication 1" and "predicted value for indication 2" are indexes showing predicted values ​​such as the proportion of doctors in charge of the symptoms identified in indication 1 and indication 2, their expertise, and reliability. These predicted values ​​for each indication are also expressed on an ordinal scale, interval scale, ratio scale, etc. For example, the higher the predicted value, the more likely a doctor is to treat patients with the corresponding symptoms. Therefore, the higher the predicted value, the more likely a doctor is to use (sell) products that are suitable for the corresponding symptoms.

[0026] 3 in the memory 12, acquires information on the medical department or position of each doctor as attribute information of the doctor. That is, the information processing device 1 is an example of an information processing system that acquires information on the medical department or position of each doctor as attribute information.

[0027] <Total Sales DB Composition> 4 is a diagram showing an example of the total sales number DB 123. The total sales number DB 123 is a database that stores information on the total number of products (also called the total sales number) actually sold to each facility, together with the classification of the facility, the importance of the hypothesis, and the like. The total sales number DB 123 may store the total number of products sold to each facility over a determined period, such as one month, two months, or quarterly. The total sales number DB 123 shown in FIG. 4 has a product ID list 1231, a facility ID list 1232, and sales information 1233.

[0028] The product ID list 1231 is a list that lists identification information of products, and is common information to the information listed in the adaptive DB 121. The facility ID list 1232 is a list that lists identification information of facilities, and is common information to the information listed in the facility list 1221 of the attribute DB 122. For example, the facility ID list 1232 is associated with each of the product IDs listed in the product ID list 1231.

[0029] The sales information 1233 stores the total number of products actually sold to a certain facility, that is, the total number of sales. In addition to the total number of sales, the sales information 1233 also stores information about the combination of a certain product and a certain facility described above. The information about the combination is, for example, the regional division to which the facility belongs, the main indication indicating the main indication of the product, the importance of each hypothesis, and the like. These pieces of information are described as character strings in a serialized format such as JSON (JavaScript (registered trademark) Object Notation). The sales information 1233 shown in FIG. 4 is associated with each combination of a product ID listed in the product ID list 1231 and a facility ID listed in the facility ID list 1232.

[0030] <Hypothesis DB Configuration> FIG. 5 is a diagram showing an example of hypothesis DB 124. Hypothesis DB 124 is a database that stores trends, rules, etc. discovered by sales representatives, etc. regarding product sales. Hypothesis DB 124 shown in FIG. 5 has the items of "hypothesis ID" and "hypothesis content." The hypothesis ID is identification information for identifying each hypothesis. The hypothesis ID may be associated with identification information of the sales representative who discovered the hypothesis. The hypothesis content is a character string or the like indicating the content of the hypothesis. For example, in the hypothesis DB 124 shown in FIG. 5, the content of the hypothesis identified by the hypothesis ID "R1" is "Sales of the main indication (indication 1) cover at least 80% or more of the total."

[0031] Other examples of hypotheses include the following: "The sales of a physician or indication are equal to or less than the sales of the facility to which the physician or indication belongs." "The apportionment results will be closer to the results of the predictive model depending on the prediction confidence." "The upper and lower sales bounds are based on the prediction interval of the forecast value of the forecast model." "The sum of the physician sales tally and the facility sales tally equals the facility sales." "At the clinic, sales in the oncology field are zero." "This facility has a policy of using other companies' products for certain illnesses." "Facilities located in aging areas have shorter treatment periods per patient and lower revenue per patient."

[0032] <Configuration of Derivation Program DB> Fig. 6 is a diagram showing an example of the derivation program DB 125. The derivation program DB 125 is a database that stores a program describing a process for executing information processing according to the hypothesis content. The derivation program DB 125 shown in Fig. 6 stores a group of programs derived from the hypothesis content. The number of programs constituting this group of programs is at least the same as the number of hypotheses identified by the hypothesis IDs described in the hypothesis DB 124.

[0033] Each program stored in the derivation program DB125 receives various conditions such as specific total sales volume, attribute information, and adaptation information as arguments, and returns a formula when these conditions are applied in accordance with the associated hypothesis, and a distinction as to whether the formula is an objective function or a constraint condition.

[0034] The association between the hypothesis and the program may be performed by a programmer who has read and understood the content of the hypothesis in advance. The association may also be performed according to a predetermined generation rule. Furthermore, the association may be performed by machine learning from statistical data of a plurality of associations performed by a programmer in the past. In addition, when the hypothesis described in the hypothesis DB124 and the program described in the derivation program DB125 are associated one-to-one, the hypothesis DB124 and the derivation program DB125 do not need to be separated and may be a single database.

[0035] <Functional configuration of information processing device> Fig. 7 is a diagram showing an example of a functional configuration of the information processing device 1. In Fig. 7, the communication unit 13 of the information processing device 1 is omitted.

[0036] The processor 11 of the information processing device 1 executes the programs stored in the memory 12 to function as an adaptation information acquisition unit 111, an attribute information acquisition unit 112, a total sales acquisition unit 113, a hypothesis acquisition unit 114, a derivation unit 115, a generation unit 116, an estimation unit 117, and an output control unit 118.

[0037] The adaptation information acquisition unit 111 acquires adaptation information for products used in medical care from the input unit 14. In this case, the adaptation information is input to the information processing device 1 by a user operating the input unit 14. The adaptation information acquisition unit 111 stores the adaptation information input from the input unit 14 in the adaptation DB 121 of the memory 12. In addition, the adaptation information acquisition unit 111 reads out and acquires already stored adaptation information from the adaptation DB 121.

[0038] In other words, the adaptation information acquisition unit 111 is an example of an adaptation information acquisition unit that acquires information on the adaptation of a product used in medical care.

[0039] The attribute information acquisition unit 112 acquires attribute information about each doctor belonging to the medical facility from the input unit 14. In this case, this attribute information is input to the information processing device 1 by a user operating the input unit 14. The attribute information acquisition unit 112 stores this attribute information input from the input unit 14 in the attribute DB 122 of the memory 12. In addition, the attribute information acquisition unit 112 reads out and acquires attribute information that has already been stored from the attribute DB 122.

[0040] That is, the attribute information acquiring unit 112 is an example of an attribute information acquiring unit that acquires attribute information about doctors who belong to a medical facility.

[0041] The sales total acquisition unit 113 acquires information on the sales total, which is, for example, the total number of products sold in a medical facility during a predetermined period, from the input unit 14, and stores it in the sales total DB 123 of the memory 12. In addition, the sales total acquisition unit 113 reads out and acquires information on the sales total that has already been stored from the sales total DB 123.

[0042] In other words, the sales total acquisition unit 113 is an example of a sales total acquisition unit that acquires information on the total number of products sold in a medical facility.

[0043] The hypothesis acquisition unit 114 acquires the above-mentioned hypothesis content from the input unit 14 through the operation of a user who is a sales representative or the like. The hypothesis acquisition unit 114 adds, for example, identification information to the hypothesis content acquired from the input unit 14 and stores it in the hypothesis DB 124 of the memory 12. In addition, the hypothesis acquisition unit 114 reads out and acquires information on hypotheses that have already been stored from the hypothesis DB 124.

[0044] In other words, the hypothesis acquiring unit 114 is an example of a hypothesis acquiring unit that acquires a plurality of hypotheses regarding the indications of products and the attributes of doctors.

[0045] The derivation unit 115 refers to the derivation program DB 125 and derives, for each hypothesis acquired by the hypothesis acquisition unit 114, a program in accordance with that hypothesis.

[0046] The generation unit 116 acquires information on the total sales number of the product from the total sales number acquisition unit 113, and acquires a program in line with the hypothesis from the derivation unit 115. The generation unit 116 also acquires the above-mentioned adaptation information from the adaptation information acquisition unit 111, and acquires the above-mentioned attribute information from the attribute information acquisition unit 112. The generation unit 116 then incorporates the total sales number of the product, the adaptation information, and the attribute information into a program in line with the hypothesis, and generates a constraint to be imposed on the sales number of the product per doctor (also referred to as individual sales number), and an objective function for finding the most likely individual sales number.

[0047] In other words, this generation unit 116 is an example of a generation unit that generates, from a formula derived from each of a plurality of hypotheses and the total sales volume, constraints to be imposed on the individual sales volume of the product for each doctor or each indication for which the product is used, and an objective function to find the most likely individual sales volume.

[0048] The estimation unit 117 estimates the above-mentioned maximum likelihood individual sales volume using the constraint conditions and the objective function generated by the generation unit 116. The estimation unit 117 estimates the individual sales volume by a mathematical optimization algorithm so that the above-mentioned constraint conditions are satisfied and the above-mentioned objective function becomes an optimal value. Note that the term "optimal" refers to a directionality previously determined for the objective function, and is either maximum or minimum.

[0049] In other words, the estimation unit 117 is an example of an estimation unit that estimates the most likely individual sales volume based on the generated constraint conditions and objective function.

[0050] The output control unit 118 causes the output unit 15 to output the most likely individual sales volume estimated by the estimation unit 117. In other words, the output control unit 118 is an example of an output control unit that causes the output unit to output information on the estimated most likely individual sales volume.

[0051] <Example of generation> Below, a case A in which the derivation unit 115 and the generation unit 116 derive a constraint condition from a hypothesis and a case B in which the objective function is derived will be described.

[0052] [Case A] When constraints are derived from hypotheses The derivation unit 115 acquires a hypothesis, "Sales of the main indication (indication 1) cover at least 80% or more of the total" from the hypothesis acquisition unit 114. Then, the derivation unit 115 extracts a program that conforms to the acquired hypothesis from the derivation program DB 125 and supplies it to the generation unit 116.

[0053] The generating unit 116 acquires the character string "[{"Sales total number":"100"}, {"Adaptation 1 ratio":"0.8"}]..." as information on the total number of sales. The generating unit 116 then incorporates the information on the total number of sales into the program supplied from the derivation unit 115. At this time, the generating unit 116 determines that the result generated from the content of the hypothesis is a constraint condition, and supplies this constraint condition to the estimation unit 117 together with a flag indicating that fact.

[0054] For example, if the value of “Total Sales Volume” listed in the total sales volume is S, the value of “Adaptation 1 Ratio” is λ, and the allocation result is xi, and the set of indexes corresponding to Adaptation 1 is G, the following formula (1) holds true in accordance with the above-mentioned hypothesis.

[0055]

number

[0056] The generating unit 116 generates a process and calculation procedure according to the above-mentioned formula (1) by incorporating the information on the total number of sales into the formula indicated by the program supplied from the derivation unit 115. Then, since the content of the hypothesis includes a phrase such as "80% or more...", it is determined that the generated process indicates a restrictive condition, and a flag to that effect is generated.

[0057] For example, when a programmer creates one program for one hypothesis and stores it in a database, a hypothesis and a program corresponding to the hypothesis are in one-to-one correspondence, so that the hypothesis DB 124 and the derivation program DB 125 may be represented in one database, as described above. In this case, the derivation unit 115 simply needs to acquire a program corresponding to the hypothesis, and the generation unit 116 needs to acquire a flag associated with the program supplied from the derivation unit 115. In this case, the generation unit 116 does not need to determine the wording of the hypothesis.

[0058] The estimation unit 117 acquires the above-mentioned adaptation information from the adaptation information acquisition unit 111 and the above-mentioned attribute information from the attribute information acquisition unit 112. Then, when the estimation unit 117 further acquires the above-mentioned constraint condition and a flag to that effect from the generation unit 116, the estimation unit 117 reflects the adaptation information and attribute information in this constraint condition. Then, the estimation unit 117 performs a calculation based on all other constraint conditions generated from other hypotheses and the objective function (the sum of the objective functions if there are multiple constraint conditions), and estimates the individual sales volume of each doctor. Note that the estimation unit 117 may use an existing algorithm such as a branch and bound method for this calculation.

[0059] [Case B] When the objective function is derived from the hypothesis The derivation unit 115 acquires a hypothesis from the hypothesis acquisition unit 114, "the apportionment result approaches the result of the prediction model according to the prediction reliability." A method for measuring the prediction reliability is, for example, a Gaussian process. Then, the derivation unit 115 extracts a program that conforms to the acquired hypothesis from the derivation program DB 125 and supplies it to the generation unit 116.

[0060] The generating unit 116 acquires the character string "[{"total sales number":"100"}, ... {"importance of hypothesis R1":"80"}] ..." as information on the total sales number. The generating unit 116 then incorporates the information on the total sales number into the program supplied from the derivation unit 115. At this time, the generating unit 116 determines that the result generated from the contents of the hypothesis is an objective function, and supplies this objective function to the estimation unit 117 together with a flag indicating this fact.

[0061] The estimation unit 117 acquires the adaptation information and attribute information as described above. Then, when the estimation unit 117 determines that the supplied objective function requires a prediction model, it reads and acquires a prediction model composed of a machine-learned parameter group (not shown) from the memory 12. Furthermore, the estimation unit 117 calculates a predicted value for the individual sales volume of each doctor by reflecting the adaptation information and attribute information described above in the acquired prediction model.

[0062] For example, the estimation unit 117 reads out data of doctors belonging to a certain medical facility identified by facility ID "H1" as attribute information, and obtains the predicted value of individual sales volume for the doctors listed n to n+mth among them, and obtains the prediction reliability for the doctors listed n+m+1 to n+2m+1. In this case, if the true allocation result is xi, the allocation result (predicted value) by the prediction model is yi, and the reciprocal of the width of the confidence interval is wi, the following formula (2) holds true as a formula in accordance with the above-mentioned hypothesis.

[0063]

number

[0064] The generating unit 116 generates a process and calculation procedure according to the above-mentioned formula (2) by incorporating the information on the total number of sales into the formula indicated by the program supplied from the derivation unit 115. Then, since the content of the hypothesis includes a phrase such as "approaches according to...", it is determined that the generated process indicates an objective function, and a flag to that effect is generated.

[0065] The estimation unit 117 acquires the above-mentioned adaptation information from the adaptation information acquisition unit 111 and the above-mentioned attribute information from the attribute information acquisition unit 112. Then, when the estimation unit 117 further acquires the above-mentioned objective function and a flag therefor from the generation unit 116, the estimation unit 117 reflects the adaptation information and attribute information in this objective function. Then, the estimation unit 117 performs calculations based on all the constraint conditions and the objective function to estimate the individual sales volume of each doctor.

[0066] <Operation of the information processing device> 8 is a flow diagram showing an example of the flow of operations of the information processing device 1. The processor 11 of the information processing device 1 acquires information on the total number of products used in medical care sold in a medical facility from the input unit 14 operated by a user (step S101).

[0067] In parallel with step S101, the processor 11 acquires hypotheses, which are knowledge related to the suitability of the product and the attributes of the doctor, from the input unit 14 operated by a user such as a salesperson (step S102). Then, in accordance with the acquired hypotheses, the processor 11 derives a formula indicating the hypotheses based on, for example, predetermined rules (step S103).

[0068] When the information on the total number of sales is acquired in step S101 and the formula showing the hypothesis is derived in steps S102 and S103, the processor 11 acquires the above-mentioned adaptation information of the product from the input unit 14 operated by the user (step S104). In parallel with step S105, the processor 11 acquires attribute information for each doctor belonging to the above-mentioned medical facility from the input unit 14 operated by the user (step S105).

[0069] When the fitness information is acquired in step S104 and the attribute information is acquired in step S105, the processor 11 uses these to generate constraint conditions imposed on the individual sales volume of the product and an objective function for finding the most likely individual sales volume (step S106). Then, the processor 11 estimates the individual sales volume of the above-mentioned product for each doctor using these objective functions and constraint conditions (step S107). Then, the processor 11 controls the output unit 15 to output the estimated result (step S108).

[0070] Through the operations described above, the information processing device 1 can estimate the most likely individual sales numbers of the product for each doctor for a product or medical facility being evaluated, based on the attributes of the doctors affiliated with the medical facility and the accumulated hypotheses regarding the medical facility, even if only the total number of products sold at the medical facility over a certain period of time is known.

[0071] In particular, this information processing device 1 determines the type of hypothesis by applying predetermined rules to the content of the hypothesis and performs calculation processing according to the type, so that it does not have to distinguish, for example, whether a hypothesis regarding sales is qualitative or quantitative.

[0072] In addition, the information processing device 1 of the present invention is capable of expressing the relationship between total sales numbers and individual sales numbers through machine learning, but in addition to machine learning, it is also possible to incorporate partial knowledge acquired by sales representatives, etc. about the internal affairs of medical facilities, into the rules and handle them in a unified manner, thereby improving the accuracy of the estimation results compared to when information is handled solely through machine learning.

[0073] <Modification> The above is a description of the embodiment, but the contents of this embodiment may be modified as follows. In addition, the following modifications may be combined with each other.

[0074] <1> In the above-described embodiment, the information processing device 1 has the processor 11 configured as a CPU, but the control means for controlling the information processing device 1 may have other configurations.

[0075] That is, the information processing device 1 may have various processors as the processor 11, such as a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), a programmable logic device, etc., in addition to a CPU.

[0076] <2> The operations of the processor in the above-described embodiments may not only be performed by one processor, but may also be performed by a plurality of processors located at physically separate locations working together.

[0077] When the above-mentioned operations are performed by the respective processors of one or more information processing devices working together, the information processing system constituted by this one or more information processing devices is an example of an information processing system having an adaptation information acquisition unit that acquires information on the adaptation of a product used in medical care, an attribute information acquisition unit that acquires attribute information about doctors belonging to the medical facility, a total sales acquisition unit that acquires information on the total sales volume of the product sold at the medical facility, a hypothesis acquisition unit that acquires multiple hypotheses regarding the adaptation of the product and the attributes of the doctors, a generation unit that generates, from a formula derived from each of the multiple hypotheses and the total sales volume, constraints to be imposed on the individual sales volume of the product for each doctor or for each indication for which the product is used, and an objective function to obtain the most likely individual sales volume, an estimation unit that estimates the most likely individual sales volume based on the generated constraints and objective function, and an output unit that outputs information on the estimated most likely individual sales volume.

[0078] Furthermore, the order of each operation of the processor is not limited to the order described in the above embodiment, and may be changed as appropriate.

[0079] <3> In the above-described embodiment, the processor 11 functioning as the estimation unit 117 performs calculations by reflecting the adaptive information and attribute information in the constraint conditions and objective function generated by the generation unit 116, and estimates the individual sales volume of the target product for each doctor in the target medical facility. However, the processor 11 may determine the type of objective function and perform different calculations depending on the determination result. For example, the processor 11 may change the calculation process based on whether the generated objective function is constructed as a quadratic expression. In this case, if any of the objective functions includes a quadratic expression, the processor 11 executes calculation processing prepared for a mixed integer quadratic programming problem, and if no such function exists, the processor 11 executes calculation processing prepared for a quadratic programming problem.

[0080] According to this modification, the information processing device 1 can select an algorithm suitable for the problem of estimating individual sales figures from the content of the hypothesis.

[0081] <4> In the above-described embodiment, the program executed by the processor 11 of the information processing device 1 is an example of a program for causing a computer having an output unit to function as an indication information acquisition unit that acquires information on the indication of a product used in medical care, an attribute information acquisition unit that acquires attribute information on each of the doctors belonging to a medical facility, a sales total acquisition unit that acquires information on the total sales volume of the product sold at the medical facility, a hypothesis acquisition unit that acquires multiple hypotheses regarding the indication of the product and the attributes of the doctors, a generation unit that generates, from a formula derived from each of the multiple hypotheses and the total sales volume, constraints to be imposed on the individual sales volume of the product for each doctor or for each indication for which the product is used, and an objective function to determine the maximum likelihood individual sales volume, an estimation unit that estimates the maximum likelihood individual sales volume based on the generated constraints and objective function, and an output control unit that outputs information on the estimated maximum likelihood individual sales volume to the output unit.

[0082] This program may be provided in a state where it is stored in a computer-readable recording medium, such as a magnetic recording medium such as a magnetic tape or a magnetic disk, an optical recording medium such as an optical disk, a magneto-optical recording medium, a semiconductor memory, etc. This program may also be downloaded via a communication line such as the Internet.

[0083] <5> In the above-described embodiment, the processor 11 functioning as the attribute information acquisition unit 112 acquires information such as "medical department," "job title," and "predicted value of indication 1" as attribute information of doctors belonging to a medical facility, but may also acquire information on the number of patients actually examined by the doctor. Also, the attribute information acquisition unit 112 may acquire, as attribute information, information on the estimated number of patients that the doctor is estimated to have examined, for each doctor, according to a predetermined ratio calculated from experience by a sales representative, for example.

[0084] In this case, the information processing device 1 is an example of an information processing system that acquires, as attribute information, information on the number of patients actually examined by each of a plurality of doctors, or the estimated number of patients that are estimated to have been examined by each of the doctors.

[0085] <6> In the above-described embodiment, there is no limitation on the source of the attribute information input by the user, but this attribute information may be information that the user, such as a sales representative, directly obtains from a doctor. In this case, the information processing device 1 is an example of an information processing system that obtains, from the user, attribute information that the user obtained from the doctor.

[0086] <7> In the above-described embodiment, there is no limitation on the source of the hypothesis input by the user, but this hypothesis may be information obtained directly from a doctor by the user who is a sales representative, etc. In this case, the information processing device 1 is an example of an information processing system that obtains a hypothesis based on information obtained by the user from a doctor. [Explanation of symbols]

[0087] 1...information processing device, 11...processor, 111...adaptation information acquisition unit, 112...attribute information acquisition unit, 113...total sales acquisition unit, 114...hypothesis acquisition unit, 115...derivation unit, 116...generation unit, 117...estimation unit, 118...output control unit, 12...memory, 121...adaptation DB, 122...attribute DB, 1221...facility list, 1222...attribute table, 123...total sales DB, 1231...product ID list, 1232...facility ID list, 1233...sales information, 124...hypothesis DB, 125...derivation program DB, 13...communication unit, 14...input unit, 15...output unit.

Claims

1. an adaptation information acquisition unit that acquires information on adaptation of a product used in medical treatment; An attribute information acquisition unit that acquires attribute information about doctors belonging to a medical facility; a sales total acquisition unit that acquires information on the total number of the products sold at the medical facility; a hypothesis acquisition unit that acquires a plurality of hypotheses regarding the indication of the product and the attributes of the doctor; A generating unit that generates, from a formula derived from each of the multiple hypotheses and the total sales volume, a constraint condition imposed on an individual sales volume of the product for each of the doctors or each of the indications for which the product is used, and an objective function for determining the maximum likelihood of the individual sales volume; an estimation unit that estimates the maximum likelihood of the individual sales volume based on the generated constraint condition and the objective function; an output unit that outputs information on the estimated maximum likelihood individual sales volume; An information processing system having the above configuration.

2. The attribute information acquisition unit acquires, as the attribute information, information on the number of patients actually examined by each of the plurality of doctors or the estimated number of patients that are estimated to have been examined by each of the plurality of doctors. The information processing system according to claim 1 .

3. The attribute information acquisition unit acquires information on the medical department or position of each of the plurality of doctors as the attribute information. The information processing system according to claim 1 .

4. The attribute information acquisition unit acquires, from the user, information on the attributes that the user has obtained from the doctor. The information processing system according to claim 1 .

5. The plurality of hypotheses includes a hypothesis based on information obtained by the user from the physician. Characterized by The information processing system according to claim 1 .

6. A computer having an output unit, an adaptation information acquisition unit that acquires information on adaptation of a product used in medical treatment; an attribute information acquisition unit that acquires attribute information for each doctor belonging to a medical facility; a sales total acquisition unit that acquires information on the total number of the products sold at the medical facility; a hypothesis acquisition unit that acquires a plurality of hypotheses regarding the indication of the product and the attributes of the doctor; A generating unit that generates, from a formula derived from each of the multiple hypotheses and the total sales volume, a constraint condition imposed on an individual sales volume of the product for each of the doctors or each of the indications for which the product is used, and an objective function for determining the maximum likelihood of the individual sales volume; an estimation unit that estimates the maximum likelihood of the individual sales volume based on the generated constraint condition and the objective function; an output control unit that outputs information on the estimated maximum likelihood individual sales volume to the output unit; A program to function as a

Images