JP6709483B1 - Systems, methods, and programs for predicting drug dosage - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system and the like for predicting a dispensing amount of a drug. The present invention provides a computer system for predicting a dispensing amount of a drug, the computer system acquiring data on a pharmacy (step S601), and based on the data on the pharmacy, It is configured to predict the dispensing amount of the drug in the pharmacy (step S602) and output the predicted dispensing amount (step S603). [Selection diagram] Fig. 6

Description

The present invention relates to systems, methods, and programs for predicting the dosage of a drug.

Conventionally, there has been a drug ordering system (Patent Document 1, etc.).

The pharmacy orderer using the conventional drug ordering system refers to the system that separately manages the dispensing history and the system that separately manages inventory information, and determines the order amount based on the dispensing history and inventory information on that day. ..

JP, 2001-43279, A

However, if the orderer decides the order quantity of the drug based on the dispensing history and inventory information on that day, an appropriate order for the drug cannot be achieved, resulting in an excessive inventory condition or an insufficient inventory condition. It was

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a system or the like for predicting the dispensing amount of a drug, which can support an appropriate ordering of a drug.

The present invention provides the following items, for example.
(Item 1)
A computer system for predicting the dispensing amount of a medicine,
An acquisition means for acquiring data about the pharmacy,
Prediction means for predicting the future dispensing amount of the drug in the pharmacy, based on the data related to the pharmacy,
An output unit that outputs the predicted dispensing amount.
(Item 2)
The computer system according to item 1, wherein the data related to the pharmacy includes the month of the day when the drug is dispensed.
(Item 3)
The computer system of item 2, wherein the pharmacy data further includes the day of the week.
(Item 4)
The computer system according to Item 2 or Item 3, wherein the data regarding the pharmacy further includes a next week holiday flag indicating that one week after the day is a holiday.
(Item 5)
The data related to the pharmacy includes a previous weekly holiday flag indicating that one week before the day is a holiday, a current day holiday flag indicating that the day is a holiday, and a next day holiday flag indicating that the day after the day is a holiday. The computer system of any of items 2-4, further comprising at least one of:
(Item 6)
The data related to the pharmacy includes a previous week holiday flag indicating that one week before the day is a holiday, a current day holiday flag indicating that the day is a holiday, and a next day indicating that the next day is a holiday. The computer system according to any one of Items 2 to 4, further including a holiday flag.
(Item 7)
Any one of Items 1 to 6, wherein the predicting unit predicts the amount of a drug dispensed in the pharmacy by using a learned model, and the learned model is a model on which learning processing is performed based on data on the pharmacy. The system according to paragraph 1.
(Item 8)
The trained model is a linear regression model,
8. The computer system according to item 7, wherein the predicting unit obtains a future dispensing amount of the drug by inputting data regarding the pharmacy into an explanatory variable of the linear regression model.
(Item 9)
9. The computer system according to any one of Items 1 to 8, wherein the predicting unit predicts a dosage amount of the drug for at least one day to three days after three days.
(Item 10)
The acquisition means further acquires drug inventory data,
The computer system further comprises second predicting means for predicting a recommended value of the quantity to be ordered for the medicine, based on the predicted dispensing amount and the inventory data.
The computer system according to any one of Items 1 to 9, wherein the output unit outputs the predicted recommended value of the quantity to be ordered.
(Item 11)
The second prediction means is
Comparing a value obtained by subtracting the predicted dispensing amount from the current inventory amount of the drug, and an appropriate inventory amount of the drug,
When the value obtained by subtracting the predicted dispensing amount from the current inventory amount of the drug is less than the appropriate inventory amount of the drug, the recommended value of the amount to be ordered of the drug is predicted. The computer system according to item 10, which is:
(Item 12)
A method for predicting the amount dispensed of a drug, comprising:
To get data about the pharmacy,
Predicting a future dispensed amount of a drug at the pharmacy based on data about the pharmacy;
Outputting the predicted dispensed amount.
(Item 13)
A program for predicting a dispensing amount of a drug, wherein the program is executed in a system including a processor unit, and the program is
To get data about the pharmacy,
Predicting a future dispensed amount of a drug at the pharmacy based on data about the pharmacy;
A program that causes the processor unit to perform processing including outputting the predicted dispensing amount.
(Item 14)
A computer system for building a model used to predict a drug dosage, comprising:
An acquisition means for acquiring data about the pharmacy,
A learning system that learns data about the pharmacy as teacher data.
(Item 15)
15. The computer system of item 14, wherein the pharmacy data includes data for a period of 4 days to 90 days.
(Item 16)
A method for building a model used to predict a drug dosage
To get data about the pharmacy,
Learning the data about the pharmacy as teacher data.
(Item 17)
A program for constructing a model used for predicting a dispensing amount of a drug, wherein the program is executed in a system including a processor unit, and the program is
To get data about the pharmacy,
A program that causes the processor unit to perform processing including learning data related to the pharmacy as teacher data.

According to the present invention, it is possible to provide a system or the like for predicting the amount of a drug to be dispensed. By using this system for a drug ordering system, it is possible to recommend an appropriate order amount and support an appropriate drug order.

The figure which shows an example of the screen displayed on the terminal device of the dispensing pharmacy which uses the service for supporting the ordering of pharmaceuticals. The figure which shows an example of a structure of the computer system 100 for estimating the dispensing amount of a pharmaceutical. The figure which shows an example of the specific structure of the computer system 100 for estimating the dispensing amount of a pharmaceutical. The figure which shows an example of the data structure of dispensing data stored in the database part 200. The figure which shows another example of the data structure of dispensing data stored in the database part 200. The figure which shows an example of a structure of the processor part 120. The figure which shows an example of a structure of processor part 120' which is an alternative embodiment of processor part 120. The figure which shows an example of the structure of the neural network 1220 which the prediction means 122 can utilize. The flowchart which shows an example of the process in the computer system 100 for estimating the dispensing amount of a pharmaceutical. The flowchart which shows an example of the process in the computer system 100 for estimating the dispensing amount of a pharmaceutical. The flowchart which shows an example of the process in the computer system 100 for estimating the dispensing amount of a pharmaceutical.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

1. New Service to Support Ordering of Drugs The inventor of the present invention has developed a new service to support ordering of drugs. One of the services is a service that presents a future dispensing amount at a dispensing pharmacy to a client in order to assist in deciding a medicine to be ordered. In this service, for example, the orderer is presented with the dispensing amount of a drug that can be dispensed the day after the ordering date. By using this service, the orderer can determine the order quantity in consideration of the future dispensing quantity. As a result, the accuracy of ordering is improved, and it is possible to avoid an excessive inventory state or an insufficient inventory state.

The service can also present the recommended value of the ordered quantity of the medicine to be ordered to the orderer. By using this service, the orderer can reduce the burden of ordering work.

FIG. 1 shows an example of a screen displayed on a terminal device of a dispensing pharmacy that uses a service for supporting ordering of medicines. For example, the terminal device of the dispensing pharmacy installs an application for supporting the ordering of the drug and activates the application or supports the ordering of the drug in order to use the service for supporting the ordering of the drug. It is possible to access a server that provides a web application for executing the web application and start the web application. That is, the terminal device of the dispensing pharmacy can use the service for supporting the ordering of the drug offline or online.

The ordering business of a dispensing pharmacy is a regular business that is carried out daily to prevent shortages and excess inventory and maintain proper inventory. The orderer will order the drug after the business closes on the day.

The screen 10 shown in FIG. 1 is a screen for ordering a drug. On the screen 10, the medicine dispensed on that day is automatically displayed as the medicine to be ordered. As a result, the orderer does not need to pick up the medicine to be ordered from the medicine list. In addition, leakage of ordered medicines is prevented.

The screen 10 is provided with a drug information display column 11, a daily dose display column and 12, a stock amount display column 13, and a predicted dose display column 14. The screen 10 further includes a check box 15, a quantity input field 16 and an order input field 17 for each drug. The screen 10 also includes an order execution button 18.

Information on each drug is displayed in the drug information display column 11. The drug information is, for example, a drug name, a packaging unit (packaging), a drug price per packaging unit (packaging drug price), a selling company, a drug price, and other information about the drug to be sold. For example, in the example shown in FIG. 1, for the drug name “Zaizar tablet 5 mg”, the packaging unit is 10 tablets×10 PTP, the drug price per packaging unit is 8780 yen, and GSK sells 87 tablets per tablet. It is displayed that it is .8 yen.

The daily dose display field 12 displays the daily dose of each drug. For example, in the example shown in FIG. 1, the amount of the "Alamyst nasal solution 27.5 μg 56 for spraying" that day is 1 kit, and the amount of "Clarithromycin tablets 200 mg "Towa"" that day is 14 tablets. It is displayed that the daily dosage of "Zaizar tablet 5 mg" is 7 tablets and the daily dosage of "Zingurea tablet 10 mg" is 7 tablets. The orderer can confirm the dispensing record of the day by only looking at the same-day dispensing amount display column 12 on the screen 10 without accessing the system for separately managing the dispensing record.

The inventory amount display column 13 displays the inventory amount of each drug on that day, that is, the inventory amount after the end of business on that day. The stock amount may be displayed as, for example, an absolute amount or a relative amount with respect to the proper stock amount. In the example shown in FIG. 1, it is displayed as a relative amount with respect to the appropriate inventory amount. For example, in the example shown in FIG. 1, the stock amount of “Alamist nasal solution 27.5 μg 56 for spraying” on that day is (appropriate stock amount-1 kit), and that day of “clarithromycin tablet 200 mg “Towa”” Is (appropriate inventory +12 tablets), "Zaizar Tablets 5mg" is the inventory of that day (appropriate inventory -7 tablets), and "Singleair Tablets 10mg" is the inventory of that day (appropriate inventory Amount-7 tablets) is displayed. The orderer can confirm the stock amount of the day by only looking at the stock amount display column 13 of the screen 10 without accessing the system for separately managing the stock information. The appropriate stock amount is a value set for each drug and can be set by the orderer.

In the predicted dispensing amount display field 14, the predicted dispensing amount of the next day of each drug is displayed. For example, in the example shown in FIG. 1, the predicted dose for the next day of “Alamist nasal solution 27.5 μg 56 for spraying” is 3 kits and the next day of “Clarithromycin tablets 200 mg “Towa””. It is displayed that the predicted dosage is 5 tablets, the predicted dosage for the next day for "Zaizar tablet 5 mg" is 18 tablets, and the predicted dosage for the next day for "Singleair tablets 10 mg" is 10 tablets. ing. The orderer can confirm the predicted dispensing amount for the next day only by looking at the predicted dispensing amount display field 14 on the screen 10.

The check box 15 is operated to select a drug to be actually ordered from the drugs to be ordered displayed on the screen 10. For example, by checking the check box 15, the medicine is ordered.

The quantity input field 16 is a field for inputting the quantity of medicines to be ordered. In the quantity input field 16, for example, the quantity may be input by key input, or the quantity may be input by selecting the quantity from a pull-down list.

The order entry field 17 is a field for entering the order recipient of the medicine to be ordered. In the order entry destination input column 17, for example, the order entry party may be entered by key input, or the order entry party may be entered by selecting the order entry party from the pull-down list.

The order execution button 18 is a button operated to execute an order. For example, by using the check box 15, the quantity input field 16 and the order input field 17, the drug to be ordered and the order quantity and the order destination are determined, and then the order execution button 18 is operated, whereby the order of the drug is executed. It

Since the information necessary for ordering is collected on the screen 10, the orderer can complete the drug ordering work only by using the screen 10. The orderer considers, for example, the dispensing amount displayed in the dispensing amount display column 12 on the day, the inventory amount displayed in the inventory amount display column 13, and the predicted dispensing amount displayed in the predicted dispensing amount display column 14. Thus, it is possible to easily determine how much the medicine displayed on the screen 10 should be ordered. This can reduce the time required for ordering work. Further, the accuracy of ordering is improved, and an excessive stock state or an insufficient stock state can be avoided.

In another example, the recommended value of the quantity to be ordered for the medicine may be displayed in the quantity input field 16 by default. This allows the orderer to place an order simply by accepting the recommended value displayed by default, which also reduces the time required for ordering work, and excessive inventory or shortage. It leads to avoiding stock status.

The new service for supporting the ordering of the above-mentioned medicine can be realized by, for example, a computer system for predicting the dispensing amount of the medicine described below.

2. Configuration of Computer System for Predicting Dispensed Amount of Pharmaceutical Product FIG. 2A shows an example of the configuration of a computer system 100 for predicting a dispensed amount of medical product.

The computer system 100 is connected to the database unit 200. Further, the computer system 100 is connected to at least one pharmacy terminal device 300 via a network 400.

The network 400 can be any type of network. The network 400 may be the Internet or a LAN, for example. The network 400 may be a wired network or a wireless network.

An example of the computer system 100 is a computer (for example, a server device) installed in a service provider that provides a new service for supporting ordering of medicines, but is not limited to this. An example of the pharmacy terminal device 300 is a computer (for example, a terminal device) installed in a dispensing pharmacy, but is not limited to this. For example, the pharmacy terminal device 300 may be connected via a network to a receipt computer system (not shown) that manages dispensing data. The receipt computer system may be located, for example, in the local network of the dispensing pharmacy to which the pharmacy terminal device 300 is connected, or may be located outside the local network of the dispensing pharmacy. In the latter case, the receipt computer system can be connected to the pharmacy terminal device 300 via the network 400. Here, the computer (server device or terminal device) may be any type of computer. For example, the terminal device may be any type of terminal device such as a smartphone, a tablet, a personal computer, smart glasses, or the like.

The database unit 200 may store, for example, data about a pharmacy transmitted from at least one pharmacy terminal device 300.

FIG. 2B shows an example of a specific configuration of the computer system 100 for predicting the amount of medicine to be dispensed.

The computer system 100 includes an interface unit 110, a processor unit 120, and a memory unit 130.

The interface unit 110 exchanges information with the outside of the computer system 100. The processor unit 120 of the computer system 100 can receive information from the outside of the computer system 100 via the interface unit 110, and can send the information to the outside of the computer system 100. The interface unit 110 can exchange information in any format.

The interface unit 110 includes, for example, an input unit that enables information to be input to the computer system 100. It does not matter how the input unit enables information to be input to the computer system 100. For example, when the input unit is a touch panel, the user may input the information by touching the touch panel. Alternatively, when the input unit is a mouse, the user may input the information by operating the mouse. Alternatively, when the input unit is a keyboard, the user may input information by pressing a key on the keyboard. Alternatively, when the input unit is a microphone, the user may input information by inputting voice into the microphone. Alternatively, when the input unit is a camera, information captured by the camera may be input. Alternatively, when the input unit is a data reading device, the information may be input by reading the information from a storage medium connected to the computer system 100. Alternatively, when the input unit is a receiver, the receiver may input information by receiving information from outside the computer system 100 via a network.

The interface unit 110 includes, for example, an output unit that enables the computer system 100 to output information. It does not matter how the output unit enables the computer system 100 to output information. For example, when the output unit is a display screen, information may be output on the display screen. Alternatively, when the output unit is a speaker, the information may be output by voice from the speaker. Alternatively, when the output unit is a data writing device, the information may be output by writing the information in a storage medium connected to the computer system 100. Alternatively, when the output unit is a transmitter, the transmitter may output the information by transmitting the information to the outside of the computer system 100 via a network. For example, the output unit of the interface unit 110 can transmit information to the pharmacy terminal device 300 via the network 400 and display the information on the output unit (eg, display screen) of the pharmacy terminal device 300. For example, a screen such as the screen 10 in FIG. 1 can be displayed on the display screen of the pharmacy terminal device 300. The output unit of the interface unit 110 can output the dispensing amount predicted by the processor unit 120, for example. In one embodiment, the output unit of the interface unit 110 may output, for example, a recommended value of the quantity to be ordered of the drug predicted by the processor unit 120.

The processor unit 120 executes the processing of the computer system 100 and controls the operation of the entire computer system 100. The processor unit 120 reads the program stored in the memory unit 130 and executes the program. This allows the computer system 100 to function as a system that executes desired steps. The processor unit 120 may be implemented by a single processor or multiple processors.

The memory unit 130 stores a program required to execute the processing of the computer system 100, data required to execute the program, and the like. The memory unit 130 stores a program for causing the processor unit 120 to perform a process for predicting a dispensing amount of a drug (for example, a program that implements a process shown in FIGS. 6, 7, and 8 described later). May be. Here, it does not matter how the program is stored in the memory unit 130. For example, the program may be preinstalled in the memory unit 130. Alternatively, the program may be installed in the memory unit 130 by being downloaded via the network. In this case, the type of network does not matter. The memory unit 130 can be implemented by any storage means.

The database unit 200 may store, for example, data regarding a pharmacy. The database unit 200 may store data about the pharmacy transmitted via the network 400 by the pharmacy terminal device 300 of each pharmacy or the accompanying receipt computer system.

The data on the pharmacy includes, for example, dispensing data on dispensing at the pharmacy. The dispensation data includes, for example, data of which pharmacy dispensed which medicine to which patient at what time and how much, and is generated for each dispensing by the receipt computer system of each pharmacy. The dispensing data may include, for example, a pharmacy ID, a patient ID, a dispensing date and time, a drug ID, and a dispensing amount. For example, the dispensing date and time may be in arbitrary units, may be in seconds, may be in minutes, may be in hours, may be in days, week Although it may be in units, it is preferably in units of days in order to predict the dosage amount on a daily basis. For example, the drug ID may be a dispensing unit code that is associated with the drug one-to-one, and may be, for example, a YJ code.

The dispensing data may further include the month of the dispensed day (January, February, March,... Etc.) and the day of the week (Sunday, Mon, Tue, Wed..., Etc.).

The dispensing data may further include holiday flags. The holiday flag is a flag indicating that a specific day is a holiday. The holiday flag is, for example, a current day holiday flag indicating that the dispensing day (dispensing day) is a holiday, a next week holiday flag indicating that one week after the dispensing day is a holiday, and one week before the dispensing day is a holiday. It includes a previous-week holiday flag indicating that there is a day, a next-day holiday flag indicating that the next day of the dispensing day is a holiday, and a previous-day holiday flag indicating that the day before the dispensing day is a holiday. Here, the holiday may include a day when the pharmacy is closed (store holiday) in addition to a holiday in a general sense defined by law.

Data relating to pharmacies include, for example, weather information on the day of dispensing (eg, temperature, humidity, etc.), the amount of pollen scattered on the day of dispensing, the number of patients with a particular disease on the day of dispensing (eg, the number of influenza patients), If the particular disease is viral, it may include a viral type (eg, influenza virus type). For example, the number of influenza patients can use information published by the National Institute of Infectious Diseases. Data related to pharmacies include, for example, location information/location conditions of pharmacies (eg, distance from station, distance from hospital, etc.), location information/location conditions of hospitals around pharmacies (eg, distance from station, population of the surrounding area). Density, etc.).

The database unit 200 may further store order data, for example. The order data is data generated for each order. The order data includes, for example, an order unit code, order date and time, order quantity, order destination, and the like. The ordering unit code is a code that is associated with the packaging unit (ordering unit) on a one-to-one basis, and may be, for example, a GTIN (Global Trade Item Number) code or a JAN code.

The database unit 200 may further store inventory data, for example. The inventory data includes, for example, information indicating the current inventory amount and the proper inventory amount of the pharmacy. The current stock amount can be calculated, for example, based on the initial stock amount, the dispensing data, and the order data. For example, the dispensing amount is known from the dispensing data, and the receipt amount is known from the inventory data. Therefore, the current inventory amount is calculated by cumulatively calculating the relationship between the initial inventory amount and the receipt amount and the delivery amount. .. This eliminates the need for frequent inventory checks and reduces the labor of inventory management. Of course, the current inventory quantity may be the result of the actual inventory check. The appropriate stock amount is a value set for each drug and can be set by each pharmacy.

FIG. 3A shows an example of the data structure of the dispensing data stored in the database unit 200. In FIG. 3A, an example of the dispensing data 310 generated for each dispensing is shown.

For example, the receipt computer system assigns an ID (8ze2g4, ixd6f3, v89smz, etc. in the example shown in FIG. 3A) to each dispensing to generate the data 310. For example, the receipt computer system can generate the dispensing data 310 by parsing the input receipt data. The generated dispensing data 310 is sent to the database unit 200 and stored for each dispensing. Each dispensing data 310 includes items of dispensing date, pharmacy ID, drug ID, patient ID, and dispensing amount, for example.

FIG. 3B shows another example of the data structure of the dispensing data stored in the database unit 200. FIG. 3B shows data 320 in which the dispensing data 310 shown in FIG. 3A is aggregated for each pharmacy and for each drug.

In the example shown in FIG. 3B, the pharmacy dispensing data 310 of the pharmacy ID: 5DJ1, the pharmacy dispensing data 310 of the pharmacy ID: J7F5, and the pharmacy dispensing data 310 of the pharmacy ID: E1XH are aggregated. The dispensed subtotal represents the total dispensed amount of a particular drug dispensed at that pharmacy that day.

The example of the data described above is an example, and the database unit 200 may store other arbitrary data.

In the example shown in FIGS. 2A and 2B, the database unit 200 is provided outside the computer system 100, but the present invention is not limited to this. It is also possible to provide at least a part of the database unit 200 inside the computer system 100. At this time, at least a part of the database unit 200 may be implemented by the same storage unit that implements the memory unit 130, or may be implemented by a storage unit different from the storage unit that implements the memory unit 130. May be. In any case, at least a part of the database unit 200 is configured as a storage unit for the computer system 100. The configuration of the database unit 200 is not limited to a particular hardware configuration. For example, the database unit 200 may be composed of a single hardware component or may be composed of a plurality of hardware components. For example, the database unit 200 may be configured as an external hard disk device of the computer system 100, or may be configured as a cloud storage connected via a network.

For example, instead of the computer system 100 receiving the dispensing data from the database unit 200, the computer system 100 is connected to the receiving computer system (not shown) via the interface unit 110 and receives the dispensing data directly from the receiving computer system. It is also within the scope of the present invention.

FIG. 4A shows an example of the configuration of the processor unit 120.

The processor unit 120 includes an acquisition unit 121 and a prediction unit 122.

The acquisition means 121 is configured to acquire data regarding a pharmacy. The data regarding the pharmacy includes, for example, dispensing data. The dispensing data includes, for example, data of which pharmacy dispensed what medicine to which patient, when, and how much. The dispensing data may have, for example, a data structure as described above with reference to FIG. 3A or a data structure as described above with reference to FIG. 3B.

The acquisition unit 121 may perform filtering by using, for example, a time stamp added to the data on the pharmacy or the dispensing date and time in the dispensing data to obtain only the data on the pharmacy within a predetermined date and time in the past or a predetermined period. .. For example, the acquisition unit 121 includes data on a pharmacy for 1 week to 1 year before the dispensing day, for example, data on a pharmacy for 14 days before the dispensing day, data on a pharmacy for 30 days before the dispensing day, data on a pharmacy for 60 days before the dispensing day, and a dispensing day before Only data for 90 days of pharmacies or 180 days before the dispensing day can be obtained. For example, the acquisition unit 121 includes data on a pharmacy from 4 days before to 1 week before the dispensing day, for example, data on a pharmacy from 4 days to 14 days before the dispensing day, data on a pharmacy from 4 days before to 30 days before the dispensing day, and dispensing. Only data on pharmacies between 4 and 60 days before the day, data on pharmacies between 4 and 90 days before the dispensing day, or data on pharmacies between 4 and 180 days before the dispensing date can be obtained. For example, the acquisition unit 121 preferably acquires data on a pharmacy for 90 days before the dispensing day, or data on a pharmacy for 4 days to 90 days before the dispensing day. 90 days is the maximum number of days that a drug can be prescribed, and the tendency of dispensing is best reflected up to 90 days before. By using the data on the pharmacy up to 90 days before for learning, This is because the accuracy is improved.

The acquisition unit 121 may acquire, for example, the data about the pharmacy stored in the database unit 200 via the communication interface unit 110. Alternatively, the acquisition unit 121 may acquire the data regarding the pharmacy received from the pharmacy terminal device 300 or the attendant receipt computer system via the communication interface unit 110. The acquired data about the pharmacy is passed to the determining unit 122.

The acquisition means 121 may be further configured to acquire order data. The order data is data generated for each order. The acquisition means 121 may be further configured to acquire inventory data including information indicating inventory of a dispensing pharmacy. The inventory data may be, for example, data that can be generated each time the inventory is checked, or may be data that is calculated based on the initial inventory amount, the dispensing data, and the order data, for example. For example, the acquisition unit 121 may perform filtering by using the date and time information in the order data or the inventory data to acquire only the order data or the inventory data of a specific date and time or period. As a result, it becomes possible to handle only the inventory data as of the day when the medicine is ordered, and the subsequent processing load is reduced.

The acquisition unit 121 may acquire the order data and/or inventory data stored in the database unit 200 via the communication interface unit 110, for example. The acquired order data and inventory data can be passed to the conversion means 123 for later processing.

The prediction unit 122 is configured to predict the dispensing amount of the drug at the pharmacy, for example, based on the data about the pharmacy acquired by the acquisition unit 121.

In one embodiment, the prediction means 122 can predict the dispensed amount of the drug at the pharmacy, for example, using machine learning. The prediction unit 122 may predict the dispensing amount of the drug in the pharmacy by machine learning using a linear regression model, for example. The linear regression model is a learned model that has been subjected to a learning process in advance using the data acquired by the acquisition means 121. A linear regression model can be defined for each drug.

For example, the linear regression model y _i of drug i is
y _i =w ₁ x ₁ +w ₂ x ₂ +...+w _k x _k +α
It is represented by. In the learning process based on the data acquired by the acquisition means 121, for example, the month of the dispensing day, the day of the week of the dispensing day, and the holiday flag related to the dispensing day in the data related to the pharmacy are set as the explanatory variables x _k, and the data related to the pharmacy is selected. The learned model is constructed by determining the coefficient w _k with the subtotal of the dispensing amount as the target variable y _i . For example, the month of some dispensing date, day of the week, the holiday flag as explanatory variable x _k, the next day of the dispensing amount subtotal of the preparation date for the purpose of variable y _i, learned model that has been built, predict the next day of the dispensing amount It becomes a learned model that can be done. For example, the month of some dispensing date, day of the week, the holiday flag as explanatory variable x _k, the dispensing amount subtotal of two days later of its preparation date as objective variable y _i, trained model that has been built, predict the dispensing amount of two days later It becomes a learned model that can be done. In this way, it is possible to construct a learned model that can predict the dispensing amount of a desired future day by changing the dispensing amount subtotal used for learning according to the future day for which the dispensing amount is desired to be predicted. it can. At this time, the learned model (dedicated learned model) peculiar to the pharmacy may be constructed using the data on the pharmacy acquired from the specific pharmacy, or the pharmacy acquired from the unspecified pharmacy A general trained model (general-purpose trained model) may be constructed using the data. For example, in the learning process, pre-processing may be performed on the data on the pharmacy, and then the learned model may be constructed using the pre-processed data on the pharmacy. Arbitrary processing can be performed as preprocessing. For example, the dispensing subtotal used as the objective variable may be increased or decreased based on the presence or absence of a holiday flag. As a result, for example, it is possible to make a more precise forecast corresponding to a specific change in demand and required stock amount of a drug due to a time factor such as the end of the month or the end of the year.

For example, if the data about the pharmacy of the day obtained at the operation stage is input to the explanatory variable of the learned linear regression model capable of predicting the dispensing amount of the next day, the predicted dispensing amount of the next day can be output. For example, when the data about the pharmacy of the day obtained at the operation stage is input to the explanatory variable of the learned linear regression model capable of predicting the dose of the day after the next day, the dose of the day after the forecast can be output.

The predicting means 122 may predict the amount of a drug dispensed in a pharmacy by machine learning using a neural network, for example. A neural network can be defined for each drug.

FIG. 5 shows an example of the structure of the neural network 1220 that can be used by the prediction means 122.

The neural network 1220 has an input layer, at least one hidden layer, and an output layer. The number of nodes in the input layer of the neural network 1220 corresponds to the number of dimensions of input data. The hidden layer of neural network 300 may include any number of nodes. The number of nodes in the output layer of the neural network 1220 corresponds to the number of dimensions of output data. For example, when predicting the dispensing amount of a drug on the next day, the number of nodes in the output layer is 1. For example, when predicting the amount of medicine dispensed the next day, the amount of drug dispensed two days later, and the amount of drug dispensed three days later, the number of nodes in the output layer is three.

The neural network 1220 can be subjected to a learning process in advance using the data acquired by the acquisition means 121. The learning process is a process of calculating the weight coefficient of each node in the hidden layer of the neural network 1220 using the data acquired by the acquisition unit 121. In the learning process, for example, data about the pharmacy (for example, the month when the drug is dispensed, the day when the drug is dispensed, the holiday flag next week when the drug is dispensed, the previous week holiday flag when the drug is dispensed, the drug The same day holiday flag on the day the drug was dispensed, the holiday flag on the day after the drug was dispensed, and the dispensing amount subtotal on the day the drug was dispensed) were used as input teacher data, and the dispensing amount subtotal for the next day was used as output teacher data. By calculating the weighting coefficient of each node of the hidden layer of the neural network 1220 using a plurality of data within a period (for example, 90 days, or 87 days from 4 days before to 90 days before), the amount of dispensing on the next day A trained model capable of predicting can be constructed. In the learning process, for example, the data about the pharmacy on the day when the drug is dispensed is used as the input teacher data, and the dispensing amount subtotal on the next day, the dispensing amount subtotal after 2 days, and the dispensing amount subtotal after 3 days are output as teacher data for output. By calculating the weighting coefficient of each node in the hidden layer of the neural network 1220 using a plurality of data within a predetermined period, it is possible to predict the dispensing amount on the next day, 2 days, and 3 days later, respectively. A trained model can be built. At this time, the learned model (dedicated learned model) peculiar to the pharmacy may be constructed using the data on the pharmacy acquired from the specific pharmacy, or the pharmacy acquired from the unspecified pharmacy A general trained model (general-purpose trained model) may be constructed using the data.

For example, for a losartan potassium tablet 25 mg “Amel” represented by drug ID: 622144801, when using the data shown in FIG. 3B to construct a learned model capable of predicting the amount to be dispensed the next day, the pharmacy ID : (DJ teacher data for output, teacher data for output) for constructing a trained model peculiar to pharmacy A represented by 5DJ1 (data about pharmacy A on September 10, 2019, September 9, 2019 (Pharmacy A subtotal on March 11th), (Data regarding pharmacy A on September 9, 2019, Dispensing subtotal on pharmacy A on September 10, 2019), ... (June 2019 The data may be the data on the pharmacy A on the 12th, the dispensing amount subtotal on the pharmacy A on June 13, 2019). Further, in this case, a set of (input teacher data, output teacher data) for constructing a general learned model is (data relating to pharmacy A on September 10, 2019, September 11, 2019). (Pharmacy A subtotal of Pharmacy A), (Data on pharmacy B on September 10, 2019, Pharmacy B subtotal on September 11, 2019), ... (September 9, 2019 (Data on pharmacy A, subtotal amount of dispensing on September 10, 2019), (Data on pharmacy B on September 9, 2019, subtotal amount of dispensing on pharmacy B on September 10, 2019),... ( Data for pharmacy A on June 12, 2019, sub-dispensed amount for pharmacy A on June 13, 2019), (Data for pharmacy B on June 12, 2019, Pharmacy B on June 13, 2019) The sub-dispensing amount in the above),...

For example, with respect to losartan potassium tablet 25 mg “Amel” represented by drug ID: 622144801, a learned model capable of predicting the amount to be dispensed the next day, 2 days, and 3 days after using the data shown in FIG. 3B. , The set of [input teacher data, output teacher data] for building the learned model peculiar to pharmacy A represented by pharmacy ID: 5DJ1 is [pharmacy on September 8, 2019] Data on A, (Subtotal amount of dispensing at Pharmacy A on September 9, 2019, Subtotal amount of dispensing at Pharmacy A on September 10, 2019, Subtotal amount of dispensing at Pharmacy A on September 11, 2019) ], [Data on Pharmacy A on September 7, 2019, (Subtotal amount of dispensing at Pharmacy A on September 8, 2019, Subtotal amount of dispensing at Pharmacy A on September 9, 2019, September 2019 10: Pharmacy A subtotal on day 10]],... [Data on pharmacy A on June 10, 2019, (Subtotal dosage on pharmacy A on June 11, 2019, June 12, 2019] Sub-total amount dispensed at Pharmacy A for the day, Sub-total amount dispensed at Pharmacy A on June 13, 2019)] and the like. In this case, the set of [input teacher data, output teacher data] for constructing a general learned model is [data relating to pharmacy A on September 8, 2019, (September 9, 2019 Sub-total of the amount dispensed at the pharmacy A on Sunday, Sub-total of the amount dispensed at the pharmacy A on September 10, 2019, Sub-total amount dispensed at the pharmacy A on September 11, 2019)], [September 8, 2019 Data on pharmacy B, (Subtotal amount of pharmacy A on September 9, 2019, Subtotal amount of pharmacy B on September 10, 2019, Subtotal amount of pharmacy B on September 11, 2019 )],... [Data on Pharmacy A on September 7, 2019, (Subtotal amount of pharmacy A at September 8, 2019, Subtotal amount of pharmacy A at September 9, 2019, Subtotal of the amount dispensed at Pharmacy A on September 10, 2019]], [Data for Pharmacy B on September 7, 2019, (Subtotal amount dispensed at Pharmacy B on September 8, 2019, September 2019] 9: Pharmacy B subtotal on September 9, 2019 Pharmacy B pharmacy B on September 10, 2019]],... [Data on pharmacy A on June 10, 2019, (June 11, 2019 Sub-total amount dispensed at pharmacy A for the day, Sub-total total dose for pharmacy A on June 12, 2019, Sub-total dose for pharmacy A on June 13, 2019)], [June 10, 2019 Data on pharmacy B, (Subtotal amount of pharmacy B on June 11, 2019, Subtotal amount of pharmacy B on June 12, 2019, Subtotal amount of pharmacy B on June 13, 2019 )],..., etc.

For example, if the data of the pharmacy of the day obtained in the operation stage is input to the input layer of the learned neural network 1220 capable of predicting the dose of the next day, the dose of the next day can be output to the output layer. .. For example, if the input layer of the learned neural network 1220 capable of predicting the dispensing amount on the following day, 2 days, and 3 days is input with the data on the pharmacy of the day obtained at the operation stage, the next day is input to the output layer. The amount dispensed after 2 days, the amount dispensed after 2 days, and the amount dispensed after 3 days can be output.

For example, by using "month of the day on which a drug is dispensed" as input data, it becomes possible to predict the amount of drug dispensed in consideration of the season. Considering the season is particularly useful, for example, in predicting the dosage of a drug that is effective against seasonal diseases. This is because the number of patients required for a drug that is effective against seasonal diseases varies depending on the season. For example, by using “the day of the week when a drug is dispensed” as input data, the dose of the drug can be predicted in consideration of the behavior of the patient. In general, the number of patients in the pharmacy tends to increase at the end of holidays such as Monday after new week. The accuracy of the prediction can be improved by taking into consideration such patient behavior related to the day of the week. For example, by using the “holiday flag” as input data, the dispensing amount of the drug can be predicted in consideration of the holidays of the pharmacy. Since the dispensed amount becomes zero on a pharmacy holiday, the accuracy of the prediction can be improved by taking into account such a specific day when the dispensed amount becomes zero. For example, by using the "next week's holiday flag" out of the "holiday flags" as input data, in addition to the pharmacy holidays described above, it is possible to predict the dispensing amount of a drug in consideration of the patient's anomalous behavior. Become. For example, a patient who receives regular dispensing may receive the dispensing when one week later becomes a holiday due to a holiday or the like, even though the medicine remains. Since even such irregular behavior of the patient can be taken into consideration, the accuracy of prediction can be improved. For example, by using the "previous day holiday flag" of the "holiday flag" as input data, the dispensing amount of the drug can be predicted in consideration of the behavior of the patient in addition to the above-mentioned closed days of the pharmacy. As described above, since the number of patients in the pharmacy tends to increase at the end of the holiday, the accuracy of the prediction can be improved by taking into consideration the behavior of the patients related to the holiday.

The timing at which the computer system 100 acquires the data about the pharmacy used for learning does not matter. For example, the computer system 100 may periodically (for example, every 1 day, every 7 days, every 14 days, every 30 days, every 90 days, every 180 days, etc.) acquire data regarding a pharmacy from the pharmacy terminal device 300, Data relating to the pharmacy may be acquired from the pharmacy terminal device 300 irregularly (for example, when a drug is dispensed in the pharmacy and data of the receipt is generated by the pharmacy terminal device 300). For example, the computer system 100 may not acquire the data on the pharmacies used for the learning after the learning process is completed once, or may update the data on the pharmacies used for the learning to update the learned model. You may make it continue to acquire.

The dispensing amount predicted by the prediction unit 122 is output to the outside of the computer system 100 via the interface unit 110. In one example, the dispensing amount predicted by the prediction unit 122 can be transmitted to the pharmacy terminal device 300 via the interface unit 110 and displayed on the output unit (eg, display screen) of the pharmacy terminal device 300. For example, the predicted dispensing amount can be displayed in the predicted dispensing amount display field 14 of the screen 10 of FIG. As a result, the orderer can consider the future predicted dispensing amount when ordering the drug, and the ordering accuracy is improved.

FIG. 4B shows an example of the configuration of a processor unit 120 ′ that is an alternative embodiment of the processor unit 120. In FIG. 4B, the same elements as those shown in FIG. 4A are designated by the same reference numerals, and the description thereof will be omitted here. The processor unit 120' shown in FIG. 4B has a configuration including a second prediction unit 123 in addition to the configuration shown in FIG. 4A.

The second predicting means 123 predicts a recommended value of the quantity to be ordered for the drug, based on the inventory data acquired by the acquiring means 121 and the dispensing amount predicted by the predicting means 122. The second predicting unit 123 can predict the recommended value of the quantity to order the drug based on, for example, the current stock quantity indicated by the stock data, the appropriate stock quantity, and the predicted dispensing quantity. .. The appropriate stock amount is a value set for each drug and can be set by each pharmacy.

The second predicting means 123 compares, for example, a value obtained by subtracting the predicted dispensing amount from the current inventory amount with an appropriate inventory amount, and a value obtained by subtracting the predicted dispensing amount from the current inventory amount is It may be configured to predict a recommended value for the quantity of medicinal product to be ordered when the quantity is below the proper stock quantity. This is because if the value obtained by subtracting the predicted dispensing amount from the current inventory amount is less than the appropriate inventory amount, it may be expected to fall below the appropriate inventory amount due to future pharmaceutical dispensing.

The second predicting unit 123 can predict the recommended value of the quantity to be ordered by using, for example, the current stock quantity, the proper stock quantity _0, and the predicted dispensing quantity. For example, the recommended value E of the quantity to be ordered is calculated using the current stock quantity S ₁ , the proper stock quantity S _0, and the predicted dispensing quantity S _E ,
E=S _E +(S ₀ −S ₁ )+α
Can be expressed as Here, α is an index of how much margin is available for the appropriate inventory amount, and is a value that can be set by each pharmacy for each drug.

The recommended value predicted by the second predicting unit 123 is output to the outside of the computer system 100 via the interface unit 110. In one example, the recommended value predicted by the second prediction unit 123 is transmitted to the pharmacy terminal device 300 via the interface unit 110 and displayed on the output unit (for example, display screen) of the pharmacy terminal device 300. You can For example, the predicted recommended value may be displayed by default in the quantity input field 16 of the screen 10 of FIG. At this time, for example, the recommended value may be converted into an order unit and displayed. The orderer can simply place an order by accepting the recommended value displayed by default, and the time required for ordering work can be reduced.

Each component of the computer system 100 described above may be configured by a single hardware component or may be configured by a plurality of hardware components. When it is configured with a plurality of hardware components, it does not matter how each hardware component is connected. Each hardware component may be connected wirelessly or by wire. The computer system 100 of the present invention is not limited to a particular hardware configuration. It is within the scope of the invention to configure the processor sections 120, 120' with analog circuits rather than digital circuits. The configuration of the computer system 100 of the present invention is not limited to that described above as long as its function can be realized.

3. Processing of Computer System for Predicting Dispensed Amount of Pharmaceuticals FIG. 6 shows an example of processing in the computer system 100 for predicting a dispensed amount of pharmaceuticals. The process 600 is executed in the processor unit 120 of the computer system 100. In the following, a case will be described as an example in which the dispensing amount of the drug a that can be dispensed at the pharmacy A is predicted.

In step S601, the acquisition unit 121 of the processor unit 120 acquires data regarding the pharmacy A. The data on the pharmacy A includes, for example, dispensing data on the dispensing of the drug a at the pharmacy A, for example, data on how much the drug a was dispensed on the day of the prediction process at the pharmacy A. The data relating to the pharmacy A is, for example, the month of the day on which the prediction process is performed (the day when the drug a is dispensed), the day of the week when the drug a is dispensed, the holiday flag next week on the day when the drug a is dispensed, the date when the drug a is dispensed. It may include, but is not limited to, the previous week's holiday flag, the current day's holiday flag on the day when the drug a is dispensed, the next day holiday flag on the day when the drug a is dispensed, and the dispensing amount subtotal on the day when the drug a is dispensed.

The acquisition unit 121 can acquire the data on the pharmacy A on the day of performing the prediction process by filtering using the time stamp added to the data on the pharmacy or the dispensing date and time in the dispensing data, for example. The acquisition unit 121 may acquire the data on the pharmacy A stored in the database unit 200 via the communication interface unit 110, for example. Alternatively, the acquisition unit 121 may acquire the data about the pharmacy A received from the pharmacy terminal device 300 of the pharmacy A or the accompanying receipt computer system via the communication interface unit 110. The acquired data on the pharmacy A is passed to the prediction unit 122.

In step S602, the prediction unit 122 of the processor unit 120 predicts the dispensing amount of the drug a based on the data on the pharmacy A acquired in step S601.

The prediction means 122 predicts the dispensing amount of the drug a at the pharmacy, for example, by machine learning. The learned model used may be a model that has been subjected to a learning process based on data related to pharmacies. The trained model may be a dedicated trained model individually built for the pharmacy A, or may be a general-purpose trained model constructed for all of the pharmacies. Since the dedicated learned model is subjected to the learning process based on the data related to the pharmacy, it is possible to make a prediction specific to the pharmacy, and generally, the accuracy is higher than that of the generalized learned model.

The machine learning may be machine learning using a linear regression model or machine learning using a neural network, for example. For example, when the data on the pharmacy A acquired in step S601 is input to the learned model constructed by machine learning, the dispensing amount of the drug a can be output. For example, the output of the trained model constructed to predict the dispensing amount of the next day is the dispensing amount of the next day of the medicine a, and the learned model constructed to predict the dispensing amount of the next day, 2 days, 3 days later. The output of the model is the dose of the drug a on the next day, the dose of the drug a after 2 days, and the dose of the drug a after 3 days.

The prediction means 121 can predict the dispensing amount of the drug in the pharmacy A by, for example, an arbitrary statistical analysis, for example, regression analysis, instead of machine learning.

When the dispensing amount of the drug a is predicted, in step S603, the processor unit 120 passes the dispensing amount of the drug a to the interface unit 110, and outputs the dispensing amount of the drug a via the output unit of the interface unit 110. ..

In one example, the dispensing amount predicted by the prediction unit 122 can be transmitted to the pharmacy terminal device 300 via the interface unit 110 and displayed on the output unit (eg, display screen) of the pharmacy terminal device 300.

By repeating Step S601 to Step S603 for each of the plurality of medicines, the predicted dispensing amount for each of the plurality of medicines can be output. Thereby, for example, it becomes possible to determine the medicine to be ordered in consideration of the predicted dispensing amounts of the plurality of medicines.

After performing steps S601 to S603, the learning model may be updated by performing the same processing as the learning processing described later. As a result, the learning model is adjusted to correspond to the change in the state of the pharmacy over time, and the accuracy of prediction can be improved.

The timing at which each of step S601 to step S603 of the process 600 described above is performed does not matter. For example, step S601 to step S603 can be performed in response to the computer system 100 receiving the request for the predicted dispensing amount from the pharmacy terminal device 300. For example, regularly (for example, 1 hour, 12 hours, 1 day, 1 week, 1 month, 3 months, 6 months, 1 year, etc.) or irregularly (for example, a drug is dispensed at a pharmacy, and data on the receipt is stored at the pharmacy. When input to the terminal device 300), step S601 is performed so that the acquisition unit 121 of the computer system 100 acquires data regarding a pharmacy, and the computer system 100 requests a predicted dispensing amount from the pharmacy terminal device 300. In response to the reception, the step S602 and the step S603 can be performed using the data about the pharmacy acquired at that time. For example, regularly (for example, 1 hour, 12 hours, 1 day, 1 week, 1 month, 3 months, 6 months, 1 year, etc.) or irregularly (for example, a drug is dispensed at a pharmacy, and data on the receipt is stored at the pharmacy. (When input to the terminal device 300), the prediction means 122 of the computer system 100 predicts the dispensing amount of the drug by performing steps S601 to S602, and requests the predicted dispensing amount from the pharmacy terminal device 300. In response to the reception, step S603 may be performed so as to output the predicted dispensing amount of the medicine at that time.

FIG. 7 shows an example of processing in the computer system 100 for predicting the amount of medicine to be dispensed. The process 700 is executed in the processor unit 120 of the computer system 100. A process 700 is a process for predicting and outputting a recommended value of the quantity of medicines to be ordered. Hereinafter, a case will be described as an example in which the recommended value of the quantity to be ordered of the medicine a that can be dispensed at the pharmacy A is predicted and output.

In step S701, the acquisition unit 121 of the processor unit 120 acquires data regarding the pharmacy A. Since step S701 is the same process as step S601, description thereof will be omitted here.

In step S702, the acquisition unit 121 of the processor unit 120 acquires the inventory data of the drug a in the pharmacy A. The inventory data may be, for example, data that can be generated each time the inventory is checked, or may be data that is calculated based on the initial inventory amount, the dispensing data, and the order data, for example.

For example, the acquisition unit 121 may perform filtering by using the date and time information in the inventory data and acquire only the inventory data on the day when the prediction process is performed. The acquisition unit 121 may acquire the inventory data stored in the database unit 200 via the communication interface unit 110, for example. Alternatively, the acquisition unit 121 may acquire the inventory data received from the pharmacy terminal device 300 of the pharmacy A via the communication interface unit 110. The acquired inventory data is passed to the second prediction means 123.

In step S703, the prediction unit 122 of the processor unit 120 predicts the dispensing amount of the drug a based on the data on the pharmacy A acquired in step S701. Since step S703 is the same process as step S602, description thereof is omitted here.

In step S704, the second predicting means 123 of the processor unit 120 makes a recommended value of the quantity to be ordered for the medicine a based on the dispensing amount predicted in step S703 and the inventory data acquired in step S702. Predict. The second prediction unit 123 should order the drug a based on the current inventory amount of the drug a indicated by the inventory data, the appropriate inventory amount of the drug a, and the dispensing amount predicted in step S703, for example. Recommended values for quantity can be predicted. The appropriate stock amount is a value set for each drug and can be set by each pharmacy.

In step S704, for example, the second predicting means 123 compares the value obtained by subtracting the predicted dispensing amount from the current inventory amount of the drug a with the appropriate inventory amount of the drug a to determine the predicted dispensing amount. If the value subtracted from the current inventory amount of the drug a is less than the appropriate inventory amount of the drug a, the recommended value of the amount to be ordered for the drug a may be predicted. If the value obtained by subtracting the predicted dispensing amount from the current inventory amount of drug a is equal to or greater than the appropriate inventory amount of drug a, it is considered that it is not necessary to order drug a in the first place, so the second predicting means Although 123 does not have to predict the recommended value of the quantity for which drug a should be ordered, it may nevertheless predict the recommended value of the quantity for which drug a should be ordered.

For example, the recommended value E _{a of the} quantity to be ordered for the drug a is calculated by using the current inventory amount S _1a of the drug a, the appropriate inventory amount S _0a of the drug a, and the predicted dispensing amount S _Ea of the drug a. ,
E _a =S _Ea +(S _0a −S _1a )+α _a
Can be expressed as Here, α _a is an index of how much margin is available with respect to the appropriate stock amount of the drug a, and is a value that can be set by each pharmacy.

When the recommended value of the quantity to be ordered for the medicine a is predicted, the processor unit 120 passes the recommended value of the quantity to be ordered for the medicine a to the interface unit 110 through the output unit of the interface unit 110 in step S705. Then, the recommended value of the quantity to be ordered for the medicine a is output.

In one example, the recommended value predicted by the second prediction unit 123 is transmitted to the pharmacy terminal device 300 via the interface unit 110 and displayed on the output unit (for example, display screen) of the pharmacy terminal device 300. You can

By repeating steps S701 to S705 for each of the plurality of medicines, it is possible to output the recommended value of the quantity to be ordered for each of the plurality of medicines. Thereby, for example, it becomes possible to determine the order quantity of the medicine to be ordered in consideration of the recommended value of the quantity to be ordered of the plurality of medicines.

The timing at which each of step S701 to step S705 of the process 700 described above is performed does not matter. Each step S701 to step S605 can be performed at an arbitrary timing.

FIG. 8 shows an example of processing in the computer system 100 for predicting the amount of medicine to be dispensed. The process 800 is a process for constructing a model used for predicting the dispensing amount of a drug, and is executed by the processor unit 120 of the computer system 100.

In step S801, the acquisition unit 121 of the processor unit 120 acquires data regarding a pharmacy. The data regarding the pharmacy includes, for example, dispensing data. The dispensing data includes, for example, data of which pharmacy dispensed what medicine to which patient, when, and how much. The dispensing data may be, for example, data having the data structure described above with reference to FIG. 3A, or may be data having the data structure described above with reference to FIG. 3B. In the case of the data as described above with reference to FIG. 3A, the processor unit 120 is acquired so as to be the data aggregated for each pharmacy and for each drug as described above with reference to FIG. 3B. Data about the pharmacy can be preprocessed.

Data on pharmacies include, for example, the month on which the drug was dispensed, the day of the day on which the drug was dispensed, the next week's holiday flag on the day on which the drug was dispensed, the previous week's holiday flag on the day on which the drug was dispensed, the day on which the drug was dispensed It may include, but is not limited to, a holiday flag, a next-day holiday flag on the day when a medicine is dispensed, a subtotal of the amount of medicine dispensed on the day when a medicine is dispensed, and the like.

The acquisition unit 121 may perform filtering by using, for example, a time stamp added to the data on the pharmacy or the dispensing date and time in the dispensing data to obtain only the data on the pharmacy within a predetermined date and time in the past or a predetermined period. .. For example, the acquisition unit 121 includes data on a pharmacy for 1 week to 1 year before the dispensing day, for example, data on a pharmacy for 14 days before the dispensing day, data on a pharmacy for 30 days before the dispensing day, data on a pharmacy for 60 days before the dispensing day, and a dispensing day before Only data for 90 days of pharmacies or 180 days before the dispensing day can be obtained. For example, the acquisition unit 121 includes data on a pharmacy from 4 days before to 1 week before the dispensing day, for example, data on a pharmacy from 4 days to 14 days before the dispensing day, data on a pharmacy from 4 days before to 30 days before the dispensing day, and dispensing. Only data on pharmacies between 4 and 60 days before the day, data on pharmacies between 4 and 90 days before the dispensing day, or data on pharmacies between 4 and 180 days before the dispensing date can be obtained. For example, the acquisition unit 121 preferably acquires data on a pharmacy for 90 days before the dispensing day, or data on a pharmacy for 4 days to 90 days before the dispensing day. 90 days is the maximum number of days that a drug can be prescribed, and 90 days ago best reflects the tendency of dispensation. By using the data on the pharmacy up to 90 days ago in step S802, the prediction by the model constructed can be performed. This is because the accuracy is improved.

The acquisition unit 121 may acquire, for example, the data about the pharmacy stored in the database unit 200 via the communication interface unit 110. Alternatively, the acquisition unit 121 may acquire the data regarding the pharmacy received from the pharmacy terminal device 300 or the attendant receipt computer system via the communication interface unit 110.

For example, the acquisition unit 121 may acquire only the data regarding a specific pharmacy when constructing a dedicated learned model, or the data regarding a plurality of pharmacies when constructing a general-purpose learned model. May be acquired. The acquired data about the pharmacy is passed to the determining unit 122.

In step S802, the processor unit 120 learns the data about the pharmacy acquired in step S801 as teacher data. As a result, a model (learned model) used for predicting the dispensing amount of the drug is constructed.

For example, when a trained model is constructed using a linear regression model, for example, the linear regression model y _i of the drug _i is
y _i =w ₁ x ₁ +w ₂ x ₂ +...+w _k x _k +α
It is represented by. In step S802, the processor unit 120, for example, the month of the dispensing day, the day of the week of the dispensing day, the next weekday holiday flag of the dispensing day, the previous weekday holiday flag of the dispensing day, and the dispensing day of the data related to the pharmacy acquired in step S801. day holiday flag, and an explanatory variable x _k at least one of the following day holiday flag dispensing date, the dispensing amount subtotal of the data relating to pharmacies as a dependent variable y _i, by determining the coefficients w _k, already learned Models can be built. For example, if a subtotal of the amount of the drug for the day following the drug dispensing day is used as the objective variable y _i , the constructed learned model becomes a model for which the amount of the drug for the next day can be predicted. For example, when the dispensing amount subtotal two days after the dispensing day is used as the objective variable y _i , the constructed learned model becomes a learned model capable of predicting the dispensing amount on the next day. In this way, it is possible to construct a learned model that can predict the dispensing amount of a desired future day by changing the dispensing amount subtotal used for learning according to the future day for which the dispensing amount is desired to be predicted. it can.

For example, in the case of building a learned model using a neural network, in step S802, the processor unit 120, in the data on the pharmacy acquired in step S801, the month of the dispensing day, the day of the dispensing day, and the dispensing day. At least one of the following week's holiday flag, the previous week's holiday flag on the dispensing day, the same day's holiday flag on the dispensing day, the next day's holiday flag on the dispensing day, and the dispensing amount subtotal of the dispensing day of the dispensing day is used as input teacher data, and the pharmacy data The learned model can be constructed by calculating the weighting coefficient of each node in the hidden layer of the neural network using the subtotal of the amount of dispensing on the specific day as the output teacher data. For example, when the dispensing amount subtotal on the day after the dispensing day is used as the output teacher data, the constructed learned model becomes a learned model capable of predicting the dispensing amount on the next day. For example, when the dispensing amount subtotal on the next day of the dispensing day, the dispensing amount subtotal on the 2nd day, and the dispensing amount subtotal on the 3rd day are used as the teacher data for output, the learned model constructed is the next day, 2 days, and 3 days later. It is a learned model that can predict the amount of drug dispensed after day. In this way, it is possible to construct a learned model that can predict the dispensing amount of a desired future day by changing the dispensing amount subtotal used for learning according to the future day for which the dispensing amount is desired to be predicted. it can.

For example, by using "month of the day on which a drug is dispensed" as input data, it becomes possible to predict the amount of drug dispensed in consideration of the season. Considering the season is particularly useful, for example, in predicting the dosage of a drug that is effective against seasonal diseases. This is because the number of patients required for a drug that is effective against seasonal diseases varies depending on the season. For example, by using “the day of the week when a drug is dispensed” as input data, the dose of the drug can be predicted in consideration of the behavior of the patient. In general, the number of patients in the pharmacy tends to increase at the end of holidays such as Monday after new week. The accuracy of the prediction can be improved by taking into consideration such patient behavior related to the day of the week. For example, by using the “holiday flag” as input data, the dispensing amount of the drug can be predicted in consideration of the holidays of the pharmacy. Since the dispensed amount becomes zero on a pharmacy holiday, the accuracy of the prediction can be improved by taking into account such a specific day when the dispensed amount becomes zero. For example, by using the "next week's holiday flag" out of the "holiday flags" as input data, in addition to the pharmacy holidays described above, it is possible to predict the dispensing amount of a drug in consideration of the patient's anomalous behavior. Become. For example, a patient who receives regular dispensing may receive the dispensing when one week later becomes a holiday due to a holiday or the like, even though the medicine remains. Since even such irregular behavior of the patient can be taken into consideration, the accuracy of prediction can be improved. For example, by using the "previous day holiday flag" of the "holiday flag" as input data, the dispensing amount of the drug can be predicted in consideration of the behavior of the patient in addition to the above-mentioned closed days of the pharmacy. As described above, since the number of patients in the pharmacy tends to increase at the end of the holiday, the accuracy of the prediction can be improved by taking into consideration the behavior of the patients related to the holiday.

In the above-described example, it has been described that the processing is performed in a specific order, but the order of each processing is not limited to that described, and the processing may be performed in any logically possible order.

In the example described above with reference to FIGS. 6 and 7, it is explained that the processing of each step shown in FIGS. 6, 7, and 8 is realized by the processor unit 120 and the program stored in the memory unit 130. However, the present invention is not limited to this. At least one of the processes of the steps shown in FIGS. 6, 7, and 8 may be realized by a hardware configuration such as a control circuit.

In the above-described example, the case where the computer system 100 is a computer (for example, a server device) connected to the pharmacy terminal device 300 via the network 400 has been described as an example, but the present invention is not limited to this. The computer system 100 may be any information processing device including a processor unit. For example, the computer system 100 may be the pharmacy terminal device 300. In this case, since the data that can be used by the computer system 100 is limited to the data related to the pharmacy, the learned model constructed by the computer system 100 can be a dedicated learned model specific to the pharmacy.

The present invention is not limited to the above embodiments. It is understood that the scope of the present invention should be construed only by the scope of the claims. It is understood that those skilled in the art can implement an equivalent range based on the description of the present invention and common general knowledge from the description of the specific preferred embodiments of the present invention.

INDUSTRIAL APPLICABILITY The present invention is useful as a system and the like for predicting the amount of a drug to be dispensed, which can support the ordering of an appropriate drug.

10 screens 100 computer system 110 interface unit 120 processor unit 130 memory unit

Claims (15)

A computer system for predicting the dispensing amount of a medicine,
An acquisition means for acquiring data regarding a pharmacy, wherein the data regarding the pharmacy includes a next week holiday flag indicating that one week after the day when the drug is dispensed is a holiday ;
A predicting means for predicting a future dispensing amount of a drug in the pharmacy by using a learned model based on data on the pharmacy , the learned model being at least data on the pharmacy and the date of dispensing the drug. It is constructed by a learning process with the dispensing amount subtotal of and the teacher data, the learning process includes a pre-process for the teacher data, the pre-process is based on the presence or absence of a holiday flag next week on the day the drug is dispensed. A prediction means including increasing or decreasing a dispensing subtotal on the day when the drug used in the learning process is dispensed ,
An output unit that outputs the predicted dispensing amount. The computer system according to claim 1, wherein the data related to the pharmacy includes the month of the day when the drug is dispensed. The computer system of claim 2, wherein the pharmacy data further includes the day of the week. The data related to the pharmacy includes a previous weekly holiday flag indicating that one week before the day is a holiday, a current day holiday flag indicating that the day is a holiday, and a next day holiday flag indicating that the day after the day is a holiday. The computer system of claim 2 or claim 3 , further comprising at least one of: The data related to the pharmacy includes a previous week holiday flag indicating that one week before the day is a holiday, a current day holiday flag indicating that the day is a holiday, and a next day indicating that the next day is a holiday. The computer system according to claim 2 or 3 , further comprising a holiday flag. The trained model is a linear regression model,
The computer system according to any one of claims 1 to 5, wherein the prediction unit obtains a future dispensing amount of the drug by inputting data related to the pharmacy into an explanatory variable of the linear regression model. The prediction means predicts the dispensing amount of the medicament for at least one day after 1-3 days after three days, according to any one of claims 1 to 6 computer system. The acquisition means further acquires drug inventory data,
The computer system further comprises second predicting means for predicting a recommended value of the quantity to be ordered for the medicine, based on the predicted dispensing amount and the inventory data.
And the output means, the outputs of the recommended value of the predicted amount to be ordered has, according to any one of claims 1-7 computer system. The second prediction means is
Comparing a value obtained by subtracting the predicted dispensing amount from the current inventory amount of the drug, and an appropriate inventory amount of the drug,
When the value obtained by subtracting the predicted dispensing amount from the current inventory amount of the drug is less than the appropriate inventory amount of the drug, the recommended value of the amount to be ordered of the drug is predicted. The computer system according to claim 8 , which is: A method for predicting a dose of a pharmaceutical product, the method being executed in a computer system including a processor unit, the method comprising:
The processor unit is to obtain data regarding a pharmacy, and the data regarding the pharmacy includes a next-week holiday flag indicating that one week after the day when the drug is dispensed is a holiday, and
The processor unit predicts a future dispensing amount of a drug in the pharmacy using a learned model based on the data related to the pharmacy, and the learned model includes at least data related to the pharmacy and the drug. Is constructed by a learning process with the dispensing amount subtotal of the day when the drug is dispensed as teacher data, the learning process includes a pre-process for the teacher data, and the pre-process includes a holiday flag of the week following the day when the drug is dispensed. Based on the presence or absence, including increasing or decreasing the dispensing amount subtotal of the day of dispensing the drug used in the learning process ,
The processor unit outputting the predicted dispensed amount. A program for predicting a dispensing amount of a drug, wherein the program is executed in a system including a processor unit, and the program is
Obtaining data regarding a pharmacy, wherein the data regarding the pharmacy includes a next-week holiday flag indicating that one week after the day when the drug is dispensed is a holiday, and
Predicting the future dose of a drug in the pharmacy using a learned model based on data about the pharmacy , wherein the learned model is at least data about the pharmacy and the date on which the drug was dispensed. Constructed by a learning process with the dispensing amount subtotal and teacher data, the learning process includes a pre-process for the teacher data, the pre-process, based on the presence or absence of the next week holiday flag of the day the drug was dispensed, Including increasing or decreasing a dispensing subtotal on the day when the drug used in the learning process is dispensed ,
A program that causes the processor unit to perform processing including outputting the predicted dispensing amount. A computer system for building a model used to predict a drug dosage, comprising:
An acquisition means for acquiring data regarding a pharmacy, wherein the data regarding the pharmacy includes a next week holiday flag indicating that one week after the day when the drug is dispensed is a holiday, and a dispensing amount subtotal on the day when the drug is dispensed. Acquisition means, including
A learning means for performing a learning process using the data relating to the pharmacy as teacher data , wherein the learning process includes a pre-process for the teacher data, wherein the pre-process determines whether or not there is a holiday flag next week on the day the drug is dispensed. And a learning means for increasing or decreasing a dispensing subtotal on the day when the drug used in the learning process is dispensed . The computer system according to claim 12 , wherein the data regarding the pharmacy includes data within a period of 4 days to 90 days ago. A method for constructing a model used for predicting a drug dosage, the method being performed in a computer system comprising a processor unit, the method comprising:
The processor unit is to obtain data on a pharmacy, and the data on the pharmacy is a next-week holiday flag indicating that one week after the day on which the drug is dispensed is a holiday and the date on which the drug is dispensed. Including a subtotal of the dispensing amount of
The processor unit is to perform a learning process using the data related to the pharmacy as teacher data, the learning process includes a pre-process for the teacher data, and the pre-process is a holiday next week on the day when the drug is dispensed. Increasing or decreasing a dispensing subtotal on the day when the drug used in the learning process is dispensed based on the presence or absence of a flag . A program for constructing a model used for predicting a dispensing amount of a drug, wherein the program is executed in a system including a processor unit, and the program is
Acquiring data on a pharmacy, the data on the pharmacy is a next week holiday flag indicating that one week after the day when the drug is dispensed is a holiday, and a dispensing subtotal on the day when the drug is dispensed. Including, and
A learning process is performed by using the data related to the pharmacy as teacher data, and the learning process includes pre-processing for the teacher data, and the pre-processing is based on the presence or absence of a holiday flag next week on the day when the drug is dispensed. And increasing or decreasing a dispensing amount subtotal on the day when the drug is dispensed used in the learning process .