Generative model-based pharmaceutical sales support platform program, operating system therefor, operating server and method therefor, program for executing method, and recording medium on which program is recorded

Abstract

The present invention provides a system for operating a platform program for supporting pharmaceutical sales. The system comprises: one or more servers having a program control unit for controlling the platform program, a pharmaceutical sales support database for operating the platform program, and a generative model for generating a response on the basis of the pharmaceutical sales support database; and one or more terminals communicably connected to the server through the network, wherein the platform program can be provided in the terminal and / or the server.

Description

A generative model-based pharmaceutical sales support platform program, the operating system, the operating server, and the method, and a program for executing the method and a recording medium on which the program is recorded.

[0001] The present invention relates to a platform program for supporting pharmaceutical sales activities, and more specifically, to a platform program that performs question answering, record management, training, analysis, and data provision using a generative model, a system, an operating server, and a method for operating the same, a program for executing the method, and a recording medium on which the program is recorded.

[0002] Sales activities of pharmaceutical companies center around visits (calls) targeting doctors and medical institutions. These activities follow a cycle of repeatedly performing visit planning, execution, and post-visit record keeping and data analysis. However, these activities incur complex costs per call, including labor, transportation, promotional expenses, and system operating costs. This results in a high-cost structure requiring annual expenditures exceeding hundreds of millions of won.

[0003] Furthermore, Medical Representatives (MRs) repeat a series of processes involving preparing materials tailored to the purpose of the visit, managing records afterward, and establishing sales strategies through the analysis of customer, market, and sales data, followed by planning future visits. To enhance the capabilities of MRs, pharmaceutical companies continuously provide education and training to improve their visit planning, execution, and customer development strategy skills.

[0004] However, existing methods suffered from low efficiency and insufficient objective and systematic analysis and training due to manual-based data management, fragmented information resources, and feedback methods relying on subjective evaluations.

[0005] The present invention has one objective of providing a generative model-based pharmaceutical sales support platform program, an operating system, an operating server, and a method thereof, which can improve inefficient data management and analysis by automating the question-and-answer, record management, training, analysis, and provision of data of sales representatives based on a generative model, and a program for executing said method and a recording medium on which said program is recorded.

[0006] In addition, the present invention has one objective of providing a generative model-based pharmaceutical sales support platform program, an operating system, an operating server, and a method thereof, which can reduce sales costs and enable efficient resource allocation by integrally supporting a series of processes including customer visits, visit records, data analysis, and strategy formulation in pharmaceutical sales activities, and a program for executing said method and a recording medium on which said program is recorded.

[0007] In addition, the present invention aims to provide a generative model-based pharmaceutical sales support platform program, an operating system and method thereof, and a recording medium on which a program for executing said method is recorded, which can achieve the improvement of sales representatives' capabilities and the systematization of performance management by simulating conversations with customers during the education and training process of sales representatives and providing objective evaluations and feedback based on data.

[0008] The objectives of the present invention are not limited thereto, and other unmentioned objectives will be clearly understood by a person skilled in the art from the description below.

[0009] The present invention provides a system for operating a platform program that supports pharmaceutical sales. The system comprises: at least one server having a program control unit that controls the platform program, a pharmaceutical sales support database for operating the platform program, and a generation model that generates a response based on the pharmaceutical sales support database; and at least one terminal connected to communicate with the server through the network, wherein the platform program may be provided in at least one of the terminal and the server.

[0010] According to one embodiment, the pharmaceutical sales support database may include at least one of drug information, disease and treatment information, customer information, company schedule, and regulatory compliance information.

[0011] According to one embodiment, the at least one server comprises an internal server and a model providing server, wherein the internal server comprises a storage unit having a program control unit that controls the platform program and the database; a processor unit that executes the program control unit provided in the storage unit; and a network interface unit that communicates the internal server with the terminal and the model providing server through the network, and the model providing server may comprise a model storage unit having the generated model; a model execution unit that executes the generated model provided in the model storage unit; and an external communication unit that communicates the model providing server with the terminal and the internal server through the network.

[0012] According to one embodiment, when the terminal requests a task from the internal server, the internal server measures data required for the task from the pharmaceutical sales support database, and the model providing server generates a response for the requested task based on the specified data, and the response may be displayed on the terminal via the platform program.

[0013] According to one embodiment, the platform program includes a query processing unit that receives a query input through the terminal, and the query processing unit can transmit the query to the generation model and control the generation model to display a response generated based on the pharmaceutical sales support database on the terminal.

[0014] According to one embodiment, the platform program includes a record management unit that manages sales record information input through the terminal, and the record management unit controls the generation model to display a query regarding the missing item on the terminal when there is a missing item in the sales record information from a predefined set of items, and controls the display of the query to end when all the missing items are input.

[0015] According to one embodiment, the predefined set of items may include at least one of a customer name, purpose of visit, result of visit, next visit schedule and goal, information confirmed today, and follow-up activities.

[0016] According to one embodiment, the platform program includes a training unit that simulates a conversation with a customer, and the training unit can specify data regarding a customer selected through the terminal from the pharmaceutical sales support database and control the generation model to generate a response simulating the customer based on the specified data in response to an interactive input entered through the terminal.

[0017] According to one embodiment, when the simulation is terminated, the training unit calculates an evaluation score for at least one interactive input entered through the terminal, and the evaluation score may be calculated based on the degree of consistency with customer information recorded in the pharmaceutical sales support database.

[0018] According to one embodiment, the degree of match may be calculated based on at least one of: whether there is a match between the keyword included in the interactive input and the attribute value of the customer information recorded in the pharmaceutical sales support database; the cosine similarity between the embedding vectors after converting the interactive input and the customer information into embedding vectors; or the probability value that the interactive input belongs to a specific customer group by a classification model pre-trained on the customer information.

[0019] According to one embodiment, the platform program may include an analysis unit that processes data recorded in the pharmaceutical sales support database and produces an analysis result.

[0020] According to one embodiment, the analysis unit can control the generation model to generate the market share and growth rate of the target product on a national basis by period based on sales data recorded in the pharmaceutical sales support database.

[0021] According to one embodiment, the analysis unit can control the generation model to generate at least one of total sales by region, average hospital sales, number of clients, an HHI index representing sales concentration, monthly sales standard deviation, and recent sales growth rate based on data recorded in the pharmaceutical sales support database.

[0022] According to one embodiment, the analysis unit can control the generation model to generate multiple region groups based on the K-Means clustering technique, based on region unit data recorded in the pharmaceutical sales support database.

[0023] According to one embodiment, the analysis unit can control the generation model to classify the plurality of region groups generated by the generation model into core regions, growth-promising regions, crisis regions, and potential regions based on at least one indicator among total regional sales, average hospital sales per region, monthly sales standard deviation, and recent sales growth rate.

[0024] According to one embodiment, the analysis unit can calculate the sales proportion, sales growth rate, or market share relative to competing products of a specific product for the plurality of region groups calculated by the generation model, and control the identification of regions requiring sales reinforcement of the specific product based on the calculated results.

[0025] According to one embodiment, the platform program may include a data providing unit that displays data recorded in the pharmaceutical sales support database on a terminal.

[0026] In addition, the present invention provides a method for operating a platform program to support pharmaceutical sales. The method comprises the steps of: receiving a query or request from at least one terminal that is communicated via a network; executing a program control unit that controls the platform program using at least one server that is communicated with the terminal; specifying data from a pharmaceutical sales support database provided in the server; and executing a generation model based on the specified data to generate a response and providing the response to the terminal, wherein the platform program may be provided in at least one of the terminal and the server.

[0027] According to one embodiment, the pharmaceutical sales support database may include data including at least one of drug information, disease and treatment information, customer information, company schedule, and regulatory compliance information.

[0028] According to one embodiment, the server includes an internal server and a model providing server, and may include the steps of: managing the pharmaceutical sales support database by executing the program control unit on the internal server; communicating between the internal server, the terminal, and the model providing server through a network interface; executing a generated model on the model providing server; and communicating between the model providing server, the terminal, and the internal server through an external communication unit.

[0029] According to one embodiment, the method may include the steps of: the terminal requesting a task from the internal server; the internal server specifying data required for the task from the pharmaceutical sales support database; the model providing server executing the generated model based on the specified data to generate a response to the requested task; and displaying the response on the terminal through the platform program.

[0030] According to one embodiment, the method may include the steps of: receiving a query input through the terminal; transmitting the query to the generation model; generating a response based on the pharmaceutical sales support database of the generation model; and controlling the response to be displayed on the terminal.

[0031] According to one embodiment, the method may include: a step of managing sales record information input through the terminal; a step of determining whether there are any missing items in the sales record information from a predefined set of items; a step of controlling the generation model to display a query regarding the missing items on the terminal if there are any missing items; and a step of controlling the display of the query to end when all the missing items are input.

[0032] According to one embodiment, the method may include the steps of: identifying data regarding a customer selected through the terminal from the pharmaceutical sales support database; receiving an interactive input input through the terminal; and performing an interactive simulation for the interactive input, wherein the generating model generates a response simulating the selected customer based on the identified data.

[0033] According to one embodiment, when the interactive simulation is terminated, the method may include the step of calculating an evaluation score for at least one interactive input entered through the terminal; and the step of calculating the evaluation score based on the degree of alignment with customer information recorded in the pharmaceutical sales support database.

[0034] According to one embodiment, the method may include a step of processing data recorded in the pharmaceutical sales support database to produce an analysis result.

[0035] According to one embodiment, the step of producing the analysis results may include a step of controlling the generation model to generate the market share and growth rate of the target product on a national basis by period, based on sales data recorded in the pharmaceutical sales support database.

[0036] According to one embodiment, the step of producing the analysis result may include controlling the generation model to generate at least one of total sales by region, average hospital sales, number of clients, an HHI index representing sales concentration, monthly sales standard deviation, and recent sales growth rate based on data recorded in the pharmaceutical sales support database.

[0037] According to one embodiment, the step of producing the analysis result may include the step of controlling the generation model to generate a plurality of region groups according to the K-Means clustering technique based on region unit data recorded in the pharmaceutical sales support database.

[0038] According to one embodiment, the step of calculating the analysis results may include the step of calculating the sales proportion, sales growth rate, or market share relative to competing products of a specific product for the plurality of region groups calculated by the generation model, and controlling to identify regions where sales of the specific product need to be strengthened based on the calculated results.

[0039] According to one embodiment, the method may include the step of displaying data recorded in the pharmaceutical sales support database on the terminal.

[0040] In addition, the present invention may provide a recording medium storing a program for executing the above methods.

[0041] In addition, the present invention may provide a program stored in a recording medium that executes the above methods.

[0042] In addition, the present invention provides a server that operates a platform program to support pharmaceutical sales. The server includes a program control unit that controls the platform program; and a pharmaceutical sales support database for operating the platform program, wherein the program control unit can control a response generated by a generation model based on the pharmaceutical sales support database to be output through the platform program.

[0043] According to one embodiment, the pharmaceutical sales support database may include at least one of drug information, disease and treatment information, customer information, company schedule, and regulatory compliance information.

[0044] According to one embodiment, the platform program may include at least one of a query processing unit, a record management unit, a training unit, an analysis unit, and a data provision unit.

[0045] According to one embodiment, the query processing unit receives a query input by a user and can control the output of a response to the query, calculated by the generation model based on the pharmaceutical sales support database, through the platform program.

[0046] According to one embodiment, the record management unit manages sales record information entered by a user, and if there is an item missing from a predefined set of items in the sales record information entered by the user, the generation model can control the generation model to generate a query regarding the missing item and output it through the platform program.

[0047] According to one embodiment, the predefined set of items may include at least one of a customer name, purpose of visit, result of visit, next visit schedule and goal, information confirmed today, and follow-up activities.

[0048] According to one embodiment, the training unit can train interactive input entered by a user in the form of a simulated conversation by specifying customer data recorded in the pharmaceutical sales support database and controlling the generating model to generate a response that simulates an actual customer based on the specified customer data.

[0049] According to one embodiment, when the simulation is terminated, the training unit calculates an evaluation score for at least one interactive input entered by the user, and the evaluation score may be calculated based on the degree of consistency with customer information recorded in the pharmaceutical sales support database.

[0050] According to one embodiment, the analysis unit may control the generation model to calculate at least one of market share, growth rate, number of customers, average hospital sales, HHI index representing sales concentration, monthly sales standard deviation, or recent sales growth rate based on sales data recorded in the pharmaceutical sales support database.

[0051] According to one embodiment, the analysis unit can control the generation model to calculate a plurality of region groups according to the K-Means clustering technique based on region unit data recorded in the pharmaceutical sales support database.

[0052] According to one embodiment, the analysis unit can control the generation model to classify the plurality of region groups into core regions, growth-promising regions, crisis regions, and potential regions based on at least one indicator among total regional sales, average hospital sales, monthly sales standard deviation, and recent sales growth rate.

[0053] According to one embodiment, the data providing unit displays data recorded in the pharmaceutical sales support database to a user, and when a user's query is entered in relation to the displayed data, the generating model can be controlled to generate and output a response to the query.

[0054] According to one embodiment of the present invention, the Q&A, record management, training, analysis, and provision of data for sales representatives are automated based on a generative model, thereby improving inefficient manual-centered data management and analysis.

[0055] In addition, according to one embodiment of the present invention, by integrally supporting a series of processes including customer visits, visit records, data analysis, and strategy formulation in pharmaceutical sales activities, it is possible to reduce sales costs and enable efficient resource allocation.

[0056] In addition, according to one embodiment of the present invention, by simulating conversations with customers during the education and training process of sales representatives and providing objective evaluations and feedback based on data, it is possible to achieve the improvement of sales representatives' capabilities and the systematization of performance management.

[0057] The effects of the present invention are not limited to the effects described above, and unmentioned effects will be clearly understood by those skilled in the art from this specification and the attached drawings.

[0058] FIG. 1 is an example of a system configuration diagram for providing a support platform service according to an embodiment of the present disclosure.

[0059] FIG. 2 is a block diagram for explaining the configuration of a platform program that implements a support platform service according to an embodiment of the present disclosure.

[0060] FIG. 3 is a flowchart illustrating an example of a method for providing a support platform service according to an embodiment of the present disclosure.

[0061] FIG. 4 is a flowchart illustrating another example of a method for providing a support platform service according to an embodiment of the present disclosure.

[0062] FIG. 5 is a drawing showing the main screen of a support platform service according to an embodiment of the present disclosure.

[0063] FIG. 6 is a drawing showing a query processing unit of a support platform service according to an embodiment of the present disclosure.

[0064] FIGS. 7 to 10 are drawings showing a record management unit of a support platform service according to an embodiment of the present disclosure.

[0065] FIGS. 11 to 28 are drawings showing a training section of a support platform service according to an embodiment of the present disclosure.

[0066] FIG. 29 is a drawing showing an analysis unit of a support platform service according to an embodiment of the present disclosure.

[0067] FIGS. 30 to 35 are drawings for explaining the analysis data that can be provided by the analysis unit of FIG. 29.

[0068] FIG. 36 is a drawing showing an example of the analysis results of the analysis unit of a support platform service according to an embodiment of the present disclosure.

[0069] FIGS. 37 and 38 are drawings for explaining a data providing unit of a support platform service according to an embodiment of the present disclosure.

[0070] The various features and benefits of the non-limiting embodiments of this specification may become more apparent from a review of the detailed description in conjunction with the accompanying drawings. The accompanying drawings are provided for illustrative purposes only and should not be construed as limiting the claims. Unless expressly stated otherwise, the accompanying drawings are not to be drawn to scale. For clarity, various dimensions in the drawings may be exaggerated.

[0071] Various embodiments are now described with reference to the drawings. In this specification, various descriptions are provided to provide an understanding of the present disclosure. However, it is evident that these embodiments can be practiced without such specific descriptions.

[0072] As used herein, terms such as “component,” “module,” “system,” etc. refer to computer-related entities, hardware, firmware, software, combinations of software and hardware, or executions of software. For example, a component may be, but is not limited to, a procedure executed on a processor, a processor, an object, an execution thread, a program, and / or a computer. For example, both an application executed on a computing device and the computing device itself may be a component. One or more components may reside within a processor and / or an execution thread. A component may be localized within a single computer. A component may be distributed among two or more computers. Additionally, these components may be executed from various computer-readable media having various data structures stored therein. Components may communicate through local and / or remote processes, for example, according to signals having one or more data packets (e.g., data from a component interacting with another component in a local system or distributed system, and / or data transmitted through signals to other systems and networks such as the Internet).

[0073] Furthermore, the term "or" is intended to mean an implicit "or" rather than an exclusive "or." That is, unless otherwise specified or evident from the context, "X uses A or B" is intended to mean one of the natural implicit substitutions. In other words, if X uses A; if X uses B; or if X uses both A and B, "X uses A or B" may apply to any of these cases. Additionally, the term "and / or" as used herein should be understood to refer to and include all possible combinations of one or more of the enumerated related items.

[0074] Additionally, the terms “comprising” and / or “comprising” should be understood to mean that such features and / or components are present. However, the terms “comprising” and / or “comprising” should be understood not to exclude the presence or addition of one or more other features, components and / or groups thereof. Furthermore, unless otherwise specified or clearly evident from the context to indicate a singular form, the singular in this specification and claims should generally be interpreted to mean “one or more.”

[0075] And, the term "at least one of A or B" should be interpreted to mean "a case including only A," "a case including only B," or "a combination of A and B."

[0076] Those skilled in the art should recognize that the various exemplary logical blocks, configurations, modules, circuits, means, logics, and algorithmic steps described in connection with the embodiments disclosed herein may be implemented in electronic hardware, computer software, or a combination of both. To clearly exemplify the interchangeability of hardware and software, various exemplary components, blocks, configurations, means, logics, modules, circuits, and steps have been generally described above in terms of their functionality. Whether such functionality is implemented in hardware or software depends on the specific application and design constraints imposed on the overall system. Skilled technicians may implement the described functionality in various ways for each specific application. However, such decisions regarding implementation should not be construed as going beyond the scope of this disclosure.

[0077] The description of the presented embodiments is provided to enable those skilled in the art to use or practice the present invention. Various modifications to these embodiments will be apparent to those skilled in the art. The general principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present invention is not limited to the embodiments presented herein. The present invention should be interpreted in the broadest possible scope consistent with the principles and novel features presented herein.

[0078] Hereinafter, with reference to FIGS. 1 to 38, a system for providing a support platform service according to an embodiment of the present disclosure will be described in detail. The support platform service according to an embodiment of the present disclosure may be implemented by a platform program (50) that supports pharmaceutical sales. The system for providing a support platform service according to an embodiment of the present disclosure may be a system that operates the platform program (50).

[0079] FIG. 1 is an example of a system configuration diagram for providing a support platform service according to an embodiment of the present disclosure, and FIG. 2 is a block diagram for explaining the configuration of the platform program of FIG. 1.

[0080] Referring to FIG. 1 and FIG. 2, an operating system (1) according to an embodiment of the present disclosure may operate a platform program (50). The operating system (1) may include a first terminal (10, an example of a user terminal), a second terminal (20, an example of an operator terminal), a network (30), and a server (40, 70). The server (40, 70) may include an internal server (40) and a second server (70).

[0081] The platform program (50) may be provided on an internal server (40). In this case, the first terminal (10) may be connected to the platform program (50) via a network (30). The user may use the platform program (50) on the first terminal (10) without a separate installation process. Alternatively, the platform program (50) may be provided in the form of an application that is directly installed on the first terminal (10). For example, if the first terminal (10) is a device such as a desktop PC, laptop, smartphone, or tablet, the platform program (50) may be provided in the form of an executable file, a mobile app, or a web application so that it can operate in a local environment.

[0082] According to another embodiment, the platform program (50) may be provided on a remote server, such as a cloud server or an external server. In this case, the first terminal (10) can execute the platform program (50) by connecting to the remote server. As such, the platform program (50) is not limited to specific hardware and may be provided in various environments, such as server-based, terminal-based, or cloud-based environments.

[0083] The first terminal (10) may be a user terminal for a user to access and execute a platform program (50), or to perform functions such as inputting a query, checking an answer, inputting sales records, performing training, and viewing materials. The first terminal (10) may include various electronic devices capable of transmitting and receiving data with a server (40, 70) via a network (30), such as a desktop computer, a laptop computer, a smartphone, a tablet PC, a portable terminal, or a wearable device. Additionally, the first terminal (10) may be linked with the platform program (50) through an operating system (OS), a web browser, or an application program. Furthermore, the first terminal (10) may be equipped with an input device (keyboard, touchscreen, microphone, etc.) and an output device (display, speaker, etc.).

[0084] The second terminal (20) may be an operator terminal that an administrator accesses to manage the pharmaceutical sales support database (60), set up the platform program (50), manage user accounts, control education and training courses, check sales performance analysis results, and process data for strategy formulation. The second terminal (20) may include an electronic device connected to the server (40, 70) via a wired or wireless network. The second terminal (20) may also be implemented in various forms such as a PC, laptop, tablet, smartphone, or dedicated console device, and may be equipped with security functions, data management functions, or an application dedicated to the administrator depending on the purpose of operation.

[0085] At least one server (40, 70) may be provided. The server (40, 70) may include a first server (40, an example of an internal server) and a second server (70, an example of a model providing server). The first server (40) and the second server (70) may be different servers.

[0086] The first server (40) may be an internal server managed by an operator. The first server (40) may be a configuration that performs control and operation of the platform program (50). The first server (40) may include a processor unit (41), a storage unit (42), and a network interface unit (43).

[0087] The processor unit (41) may be configured to control various programs executed on the first server (40) and perform calculations. The processor unit (41) may be implemented as a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), or one or more calculation units combined with these. By calling and executing a program provided in the storage unit (42), the processor unit (41) can perform control of the platform program (50), database access, and control of linkage with the creation model.

[0088] The storage unit (42) can store the control program (44) and the pharmaceutical sales support database (60). For example, the storage unit (42) may be implemented as read-only memory (ROM), random access memory (RAM), hard disk drive (HDD), solid-state drive (SSD), flash memory, EEPROM, optical disk, magnetic disk, or a combination thereof. Additionally, the storage unit (42) may be a locally mounted storage medium or an external storage device accessible via the network (30), such as network storage or cloud storage. The control program (44) and the pharmaceutical sales support database (60) are recorded and maintained in this storage unit (42) and can be called and executed by the processor unit (41).

[0089] The control program (44) may be a program for controlling the overall operation of the platform program (50), linking with the first terminal (10) and the second terminal (20), and performing data input, output, and management functions with the pharmaceutical sales support database (60). The processor unit (41) and the control program (44) may be collectively defined as the program control unit as a functional unit that controls and operates the platform program (50).

[0090] The pharmaceutical sales support database (60) may be a database that stores various data necessary to support pharmaceutical sales activities, such as drug information, disease and treatment information, customer information, company schedule, and regulatory compliance information.

[0091] Pharmaceutical information may include key product-related information (e.g., insert papers), information on manufacturers or distributors by ingredient, pricing information, and insurance guidelines (reimbursement / non-reimbursement criteria). Disease and treatment information may include the latest domestic and international treatment guidelines for related diseases, comparative information on major treatments, general physiological information, recent clinical papers, and the latest news and issues related to the disease. Client information may include specific product information, product marketing strategies, sales information by region / customer, customer and client information held by the company / Medical Representatives (MRs), ongoing sales activities, sales analysis data, and strategies for allocating sales resources. Company schedule information may include company event schedules exclusively for MRs, event schedules for customers, and domestic and international academic conference schedules. Regulatory information (SOP, Standard Operating Procedure) refers to standard business procedures or guidelines that sales representatives must adhere to, which can be utilized in sales activities and customer interactions.

[0092] The pharmaceutical sales support database (60) may be a set of data collected in advance. Additionally, the pharmaceutical sales support database (60) may be updated by new data entered by a user, who is a sales representative, through the first terminal (10). Furthermore, the pharmaceutical sales support database (60) may include not only structured data but also unstructured data (e.g., reports, images, voice data, etc.).

[0093] The second server (70) may be a model providing server that supports the operation of the platform program (50) by having a generated model (80). The second server (70) may be an external server managed by a company that provides the generated model (80). The second server (70) may include a model execution unit (71), a model storage unit (72), and an external communication unit (73).

[0094] The model execution unit (71) is a component responsible for the computation of the second server (70) and may include one or more computation units such as a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), or a semiconductor device specialized for artificial intelligence computation (NPU, ASIC, etc.). The model execution unit (71) can control the processing of a query or request transmitted from the first server (40) or terminal (10, 20) and the production of a response by calling and executing a generated model (80) provided in the model storage unit (72).

[0095] The model storage unit (72) is configured to store the generated model (80) and may include both non-volatile memory and volatile memory. For example, the model storage unit (72) may be implemented as ROM, RAM, flash memory, SSD, HDD, or a combination thereof, and may be managed in conjunction with cloud storage. The generated model (80) provided in the model storage unit (72) may be a machine learning / deep learning based model capable of processing various data types such as text, voice, and images, and may be called and executed by the model execution unit (71).

[0096] The external communication unit (73) is configured to enable the second server (70) to communicate with the first server (40), the first terminal (10), and the second terminal (20) through a network (30). The external communication unit (73) may include a wired LAN, a wireless LAN (Wi-Fi), a mobile communication module (3G, 4G, 5G, etc.), an optical communication module, or a Bluetooth module, and can perform functions such as transmitting and receiving data packets, encryption and decryption, and detection of transmission errors and retransmission. Additionally, the external communication unit (73) may apply security protocols such as VPN, TLS, and SSL to enhance security, and can perform data exchange between the first server (40) and the model storage unit (72) and transmit model call requests / responses.

[0097] The generative model (80) can be implemented as a Large Language Model (LLM). The generative model (80) can receive a query or command expressed in natural language as input, understand it, and generate a response based on data stored in the pharmaceutical sales support database (60).

[0098] The generative model (80) does not merely perform general language generation functions, but can also operate by restricting response generation for content that does not exist in the database (60) or by requiring supplementary input when the supporting data is insufficient. Through this, the generative model (80) can reduce uncertain outputs such as hallucinations and output only highly reliable responses based on verified information in the database (60).

[0099] In some cases, the model execution unit (71) and model storage unit (72) provided by the second server (70) may be provided to the first server (40). That is, functions related to the generated model (80) may be integrated into the first server (40) to be managed and controlled.

[0100] The platform program (50) may include a query processing unit (51), a record management unit (52), a training unit (53), an analysis unit (54), and a data provision unit (55).

[0101] The platform program (50) may include a query processing unit (51) that receives a query entered by a user through a terminal. The query processing unit (51) may control the received query to be transmitted to a generation model (80) provided in the second server (70) via a program control unit of the internal server (40). The generation model (80) generates a response to the query based on the pharmaceutical sales support database (60), and the query processing unit (51) may control the generated response to be displayed on the first terminal (10) through the program control unit.

[0102] In one embodiment, the query processing unit (51) can preprocess the input query to control the specification of necessary data items from the database (60), and the program control unit can provide the specified data to the generation model (80) to perform response generation.

[0103] The platform program (50) may include a record management unit (52) for managing record information related to sales activities. The record management unit (52) receives sales record information input through the first terminal (10) and can control the storage or updating of the record information in the pharmaceutical sales support database (60) via the program control unit of the internal server (40). Sales record information may be input into the first terminal (10) in voice form. The record management unit (52) may operate to verify whether the input sales record information satisfies a predefined set of items.

[0104] The above set of items may consist of, for example, customer name, purpose of visit, results of visit, next visit schedule and goals, information confirmed today, follow-up activities, etc. If there are any missing items, the record management unit (52) may control the generation model (80) provided in the second server (70) through the program control unit of the internal server (40) to display a query regarding the missing items on the first terminal (10). Additionally, when all missing items are filled in, the record management unit (52) may control the program control unit to terminate the display of the corresponding query.

[0105] In the past, the method and form of expression for recording sales activities varied among sales representatives, resulting in a lack of uniformity in the format of collected data. Consequently, there was a problem in that it was difficult to consistently integrate and manage data related to sales activities. However, by providing a record management unit (52), data related to sales activities can be collected and stored in the same format, and accordingly, the records reflected in the database (60) can be managed in a standardized form. Therefore, sales activity data can be compared / analyzed within the database (60) or reprocessed into meaningful information and utilized in combination with other functions, such as the analysis unit (54) or training unit (53) of the platform program (50).

[0106] The platform program (50) may include a training unit (53) for simulating a conversation with a customer. When a specific customer is selected through the first terminal (10), the training unit (53) can identify the corresponding customer data recorded in the pharmaceutical sales support database (60) via the program control unit of the internal server (40). Subsequently, the training unit (53) receives an interactive input input through the first terminal (10) and controls a generation model (80) provided in the second server (70) via the program control unit of the internal server (40) to generate a response that simulates the actual customer based on the identified customer data. Here, simulation means that the generation model (80) generates a response by reflecting the characteristics of the corresponding customer recorded in the pharmaceutical sales support database (60). For example, it can be implemented in a way that generates a response that the customer might show in a conversational situation based on the customer's preferred medicine, past consultation history, areas of interest, frequently used expressions, or decision patterns recorded in the database (60).

[0107] When the interactive simulation is terminated, the training unit (53) can control the calculation of an evaluation score for one or more interactive inputs entered through the first terminal (10). The evaluation score may be calculated based on the degree of alignment with customer information recorded in the pharmaceutical sales support database (60). The calculation of the degree of alignment may be performed in various ways.

[0108] For example, the keyword matching rate between the conversation content entered by the user and the customer-related data recorded in the database (60) can be calculated, or text embedding vectors can be calculated for the conversation sentences and customer data, and then the similarity can be evaluated using cosine similarity or Euclidean distance. In addition, the degree of consistency can be evaluated by checking whether major named entities such as customer names, product names, disease names, treatment names, and medical institution names match, or by verifying the consistency of question-and-answer patterns related to customer characteristics in the conversation flow.

[0109] The calculated consistency value can be scored according to predefined criteria. For example, if the consistency exceeds a certain standard, a high score is assigned; if the consistency is low, points may be deducted or a feedback message may be provided. Additionally, results obtained from multiple consistency calculation methods can be converted into a single integrated score through a weighted average or rule-based combination. The evaluation score is not merely provided as a numerical result; supplementary instructions or points for improvement regarding sections or items within the conversation that were rated as having low consistency may also be presented.

[0110] The platform program (50) may include an analysis unit (54) for processing data recorded in the pharmaceutical sales support database (60) to produce various analysis results.

[0111] The analysis unit (54) can identify sales data, customer information, regional sales information, etc. recorded in the database (60) via the program control unit of the internal server (40), and control the generation model (80) provided in the second server (70) to generate analysis results based on the identified data. For example, the analysis unit (54) can control the generation model (80) to calculate the market share and growth rate of the target product on a national basis by period based on the sales data recorded in the database (60).

[0112] Additionally, the analysis unit (54) can control the generation model (80) to calculate various indicators based on regional unit data, such as total sales, average hospital sales, number of customers, HHI index representing sales concentration, monthly sales standard deviation, and recent sales growth rate. In one embodiment, the analysis unit (54) can control the generation model (80) to generate multiple regional groups by applying a clustering algorithm, such as the K-Means clustering technique. The generated regional groups can be further classified into core regions, growth-promising regions, crisis regions, and potential regions based on indicators such as total regional sales, average sales per hospital, monthly sales volatility, and recent sales trends.

[0113] In another embodiment, the analysis unit (54) controls the generation model (80) to calculate the sales proportion, sales growth rate, or market share relative to competing products of a specific product, and can identify regions requiring sales reinforcement of the product based on the results.

[0114] Additionally, the analysis unit (54) can control the generation model (80) to predict future sales performance of a specific product, region, and sales representative based on past sales data, marketing activity history, changes in the competitive environment, and activity data by representative recorded in the database (60).

[0115] The analysis unit (54) can also predict the response rate of target customers (doctors, patients) to specific marketing channels (digital, academic conferences, advertisements, etc.) or messages and the correlation of prescription changes based on past data. These analysis results allow for the efficient allocation of future marketing budgets and can contribute to improving the return on investment (ROI) by focusing on the predicted channels and messages.

[0116] The analysis unit (54) can predict changes in a doctor's future prescription preference for a specific disease by analyzing the doctor's past prescription history, characteristics of the medical department, regional factors, and patterns of acquiring the latest research and academic society information. That is, the generative model (80) can function as a predictive AI.

[0117] The platform program (50) may include a data provision unit (55) for displaying data recorded in the pharmaceutical sales support database (60) on the first terminal (10). The data provision unit (55) can control various data stored in the database (60), such as pharmaceutical information, disease and treatment information, customer information, and company schedule information, through the program control unit of the internal server (40), so that the data is displayed on the first terminal (10) as is without processing. As a result, the user can directly check the original information recorded in the database (60) without the intervention of the generation model (80).

[0118] Additionally, when a user inputs a query through the first terminal (10) in relation to the displayed data, the data providing unit (55) can transmit the query to the second server (70) via the program control unit of the internal server (40) so that the generating model (80) can generate a response to the query based on the data recorded in the database (60). The generated response can then be displayed on the first terminal (10) again through the platform program (50).

[0119] FIG. 3 is a flowchart illustrating an example of a method for providing a support platform service according to an embodiment of the present disclosure. FIG. 3 illustrates a first process (S10) in which a generation model (80) responds to a task requested by a user.

[0120] Referring to FIG. 3, the user can request tasks from the platform program (50) using the first terminal (10), which is a user terminal (S11). The tasks may include requesting a response to a query based on the pharmaceutical sales support database (60) through the query processing unit (51), inputting, modifying, and managing records related to sales activities through the record management unit (52), simulating conversations with customers or performing training through the training unit (53), requesting analysis of various data such as sales, market share, and customer information through the analysis unit (54), and requesting a response to an additional query from the user regarding information stored in the database (60) through the data provision unit (55).

[0121] The task requested by the user can be transmitted to the internal server (30) via the platform program (50). The program control unit of the internal server (30) can identify the data required for the task requested by the user from the pharmaceutical sales support database (60) (S12). Here, the identification of data can be performed by searching for items stored in the database (60) based on keywords or query conditions entered by the user, or by selecting the corresponding data by matching predefined indexes and identifiers. Additionally, if necessary, it can be performed by deriving highly relevant data through similarity calculation or rule-based filtering. In some cases, the identification of data required for the task may be performed directly by the generation model (80) rather than the program control unit.

[0122] When data required for a task is specified, the specified data can be connected to a generation model (80) (S14). The generation model (80) can perform analysis, calculations, etc. required for a response based on the connected data.

[0123] When the analysis, calculation, etc. required for the response are completed, the generation model (80) can respond to the task requested by the user (S15). The response content can be transmitted from the second server (70) to the first server (30). The program control unit of the internal server (30) can analyze and / or process the response of the generation model (80) into a form required by the user, for example, a form that can be displayed through the screen of the first terminal (10) (S16). Analyzing and processing the response of the generation model (80) into a form required by the user may be omitted depending on the case. Subsequently, the response of the generation model (80) can be displayed to a user, who may be a salesperson, through the first terminal (10).

[0124] As can be seen by referring to the first process (S10), the generation model (80) generates a response based on the pharmaceutical sales business database (60). That is, the present invention may include a Retrieval-Augmented Generation (RAG) structure linked with the pharmaceutical sales support database (60). According to the RAG structure, the generation model (80) first searches for relevant information from the database (60) before generating a response to a user query, and utilizes the searched information as a context for response generation to produce a reliable response. Since the generation model (80) can reflect the latest information recorded in the database (60) or verified internal data even for areas not included in the training data, the problem of inaccurate or false responses being output can be minimized. In addition, when using the RAG structure, if the database (60) is updated, the generation model (80) can reflect the updated information in response generation without a separate retraining process.

[0125] FIG. 4 is a flowchart illustrating another example of a method for providing a support platform service according to an embodiment of the present disclosure. FIG. 4 illustrates a second process (S20) that responds to a task requested by a user without using a generation model (80).

[0126] Referring to FIG. 4, the user can request a task from the platform program (50) using the first terminal (10), which is a user terminal (S21). The task may include requesting information stored in the database (60) to be displayed on the first terminal (10) through the data provision unit (55).

[0127] The task requested by the user can be transmitted to the internal server (30) via the platform program (50). The program control unit of the internal server (30) can identify the data required for the task requested by the user from the pharmaceutical sales support database (60) (S22). Here, the identification of data can be performed by searching for items stored in the database (60) based on keywords or query conditions entered by the user, or by selecting the corresponding data by matching predefined indexes and identifiers. Additionally, if necessary, it can be performed by deriving highly relevant data through similarity calculation or rule-based filtering.

[0128] When the data required for the task is specified, the system can respond to the task requested by the user according to the algorithm stored in the platform program (50) (S24). That is, the internal server (30) can display the data itself from the database (60) as a response. In some cases, the generated response can be analyzed and processed into a form required by the user according to the algorithm stored in the platform program (50) (S25). This step may be omitted as necessary. Subsequently, the response to the user's task request can be displayed through the first terminal (10) (S26).

[0129] Below, examples of how a platform program (50) according to an embodiment of the present invention is implemented are described.

[0130] FIG. 5 is a drawing showing the main screen of a support platform service according to an embodiment of the present disclosure. FIG. 5 may show a screen that can be viewed by a user (e.g., a pharmaceutical salesperson) through a first terminal (10).

[0131] Referring to FIG. 5, the query processing unit (51), record management unit (52), training unit (53), analysis unit (54), and data provision unit (55) of the platform program (50) can be implemented as asking, recording, challenging, analyzing, and viewing data, respectively. Asking, recording, challenging, analyzing, and viewing data can be displayed in the form of icons, and when a user clicks the corresponding icons, the parts of the platform program (50) corresponding to them can be executed.

[0132] FIG. 6 is a drawing showing a query processing unit (51) of a support platform service according to an embodiment of the present disclosure. When a user selects a ask icon on the main screen using an input device of the first terminal (10), the screen of FIG. 6 may be switched.

[0133] Referring to FIG. 6, the user can inquire about basic information regarding pharmaceuticals, disease / treatment information, and customer information through a chat window. In this case, the inquiry processing unit (51) can generate a response to the user's inquiry and display it on the screen of the first terminal (10). The response by the inquiry processing unit (51) can be generated by either the first process (S10) or the second process (S20) described above, and preferably by the first process (S10).

[0134] FIGS. 7 to 10 are drawings showing a record management unit (52) of a support platform service according to an embodiment of the present disclosure. When a user selects a record icon on the main screen using an input device of the first terminal (10), the screen of FIG. 7 may be switched.

[0135] As illustrated in FIG. 7, the user can input sales record information, which is information collected while conducting sales activities, into the platform program (50). The sales record information can be input through the input device of the first terminal (10). The sales record information can be recorded by the user using their voice. At this time, if the input sales record information has missing items in a predefined set of items, the record management unit (52) can display a query regarding the missing items on the first terminal (10). When the missing items are entered, the record management unit (52) can stop displaying the query.

[0136] In addition, as shown in FIGS. 8 and 9, such business record information can be accumulated and recorded by business date.

[0137] In addition, as illustrated in FIG. 10, sales record information entered by the user can be automatically saved in the form of a Daily Report. The Daily Report is recorded in the storage unit (42) of the server (40) described above, and the operator can check the Daily Report recorded in the storage unit (42) through the second terminal (20). In some cases, a conversation between the user and the customer via voice can be recorded as sales record information. When a conversation with a customer is recorded as sales record information, the user can perform input for items missing from a predefined set of items after the conversation ends. The record management unit (52) of the present invention can enhance the work efficiency of sales representatives and provide a foundation for performing high-quality customer relationship management (CRM) activities in the future by managing data related to the user's sales activities in a unified format.

[0138] FIGS. 11 to 28 are drawings showing a training section (53) of a support platform service according to an embodiment of the present disclosure. When a user selects a challenge icon on the main screen using an input device of the first terminal (10), the screen may be switched to the screen of FIG. 11.

[0139] As illustrated in FIG. 11, the user can select either [Level 1: Achieving a goal through Role-Play with one customer] or [Level 2: Developing business partners by taking on Role-Play challenges sequentially according to the customer Adoption Ladder] using the input device of the first terminal (10). Additionally, the user can return to the home screen of the training while performing the training by clicking the return icon (53H) at the top left of the screen using the input device of the first terminal (10).

[0140] When the user selects Level 1, a customer selection screen as illustrated in FIG. 12 may be displayed. The user can select a target customer to perform a conversation simulation using the input device of the first terminal (10). At this time, the character icons configured to allow the selection of a customer may be varied according to the needs of the pharmaceutical company. In addition, for each character icon, personality, speech style, level of familiarity, affiliated hospital, major subject, related background knowledge (medical knowledge), and prescription patterns (current prescription regimens for the company's own medicines and competing products, etc.) may be set differently, and such settings may be made based on the aforementioned pharmaceutical sales support database (60). Each character may reflect customer information stored in the pharmaceutical sales support database (60).

[0141] When a user selects a character (customer) to perform a conversation simulation, a customer information screen as illustrated in FIG. 13 may be displayed. The user can view the customer information as illustrated in FIG. 13. The visit goal may be displayed at the top, and the customer's basic information may be displayed at the bottom. Icons for selecting Customer Information, Start Visit, Stop Visit, End Visit, and View Results may be arranged on the buttons below the character. When the user selects Customer Information, the Customer Information may be displayed. If Start Visit is selected, the conversation simulation may be executed. If Stop Visit is selected, the conversation simulation may be terminated. If End Visit is selected, the conversation simulation may be terminated. The View Results icon may be activated when the conversation simulation is terminated. It may remain inactive until the End Visit icon is selected.

[0142] As illustrated in FIG. 14, when a visit is selected, the user can perform a conversation simulation with a customer through the input device of the first terminal (10). At this time, the user can perform the conversation simulation by inputting text, or can perform the conversation by voice through an input device such as a microphone of the first terminal (10). The training unit (53) can also output a response to the user's conversational input in the form of text or voice. Input content entered into the training unit (53) and output content output from the training unit (53) can be converted into text and displayed on the screen of the first terminal (10).

[0143] As illustrated in FIG. 15, when Stop Visit is selected, the training unit (53) may pause the recording of user input. When Stop Visit is selected, the Stop Visit icon may be switched to Resume Visit. In some cases, when Stop Visit is selected, the Start Visit icon may be switched to Resume Visit. When the activated Resume Visit icon is selected, the interactive simulation may be resumed.

[0144] As illustrated in FIG. 16, when the end of the visit is selected, the training unit (53) can transmit the interactive input collected through the conversation simulation and the response of the training unit (53) to the server (40). The server (40) can transmit the interactive input and response to the generation model (80). The generation model (80) can perform analysis on the received interactive input and response.

[0145] As illustrated in FIG. 17, when the analysis is complete, a “Yes” icon and a “First” icon may be displayed. If the user selects “Yes,” the analysis result of the generation model (80) is output. If the user selects “First,” the user may return to the Level 1 character selection screen.

[0146] As illustrated in FIG. 18, if there are multiple analysis results related to a selected customer (character), the analysis results may be displayed as a list. The user can select an analysis result displayed in the list to view the analysis result.

[0147] As illustrated in FIG. 19, the training unit (53) may display an evaluation score of the interactive simulation performed by the user based on interactive input and response. The evaluation score may be calculated based on whether the goal has been achieved. As described above, the evaluation score may be calculated based on the degree of alignment with customer information recorded in the pharmaceutical sales support database (60). If necessary, whether the user's selling skills were appropriately demonstrated in the interactive input may be reflected in the evaluation score. For example, by verifying whether expressions that form a relationship through icebreaking at the beginning of a conversation with a customer, questions that identify the customer's needs or interests, explanations that clearly convey the product's features and benefits, or closing remarks that induce follow-up actions at the end of the conversation are included in the interactive input,

[0148] The results can be added to or deducted from the evaluation score. Additionally, internal company rankings can be displayed in the form of (Self / All Applicants).

[0149] If a specific customer (character) obtains an evaluation score of 90 points or higher, the customer (character) may be dimmed as shown in FIG. 20. However, even in the case of a dimmed customer (character), a conversation simulation can be performed again, and the previous results can be checked again.

[0150] When the user selects Level 2, an initial Level 2 setup screen as shown in FIGS. 21 and FIGS. 22 may be displayed. Level 2 is determined to have achieved the final goal only when N visits are successfully completed in sequence. N can be set to approximately 3 to 10, and a goal can be set for each visit. Additionally, the current number of visits may be displayed on the screen of the first terminal (10).

[0151] The function of the training unit (53) implemented in FIGS. 23 to 27 may be the same or similar to the function of the training unit (53) implemented in FIGS. 13 to 19 described above, so a repeated description is omitted.

[0152] If an interactive simulation is performed at least once at Level 2, the user's own achieved position and the average position of other salespeople can be displayed together, as shown in FIG. 28.

[0153] FIG. 29 is a diagram showing an analysis unit of a support platform service according to an embodiment of the present disclosure. When a user selects an analysis icon on the main screen using an input device of the first terminal (10), the screen of FIG. 29 may be switched. The user may request an analysis of a specific topic through the input device of the first terminal (10), and a generation model (80) may generate a response to the user's request.

[0154] Below, examples of analysis content that the platform program (50) can present through analysis (53) are described.

[0155] According to the first embodiment of the analysis, the analysis unit (53) can generate the market share and growth rate of the target product on a national basis by period based on sales data recorded in the pharmaceutical sales support database (80) by the generation model (80).

[0156] According to the second embodiment of the analysis, the analysis unit (53) can generate at least one of the total sales by region, average sales by hospital, number of customers, HHI index, standard deviation of monthly sales, and recent growth rate based on data recorded in the pharmaceutical sales support database (80) by the generation model (80).

[0157] Total sales at the regional level can be calculated by summing the total monthly sales of a specific region, including the company's own products, competitor products, and products with the same ingredients, and this can be used as a standard for evaluating the market size of that region.

[0158] The average hospital revenue at the regional level can be calculated by averaging the total sales of the company's own products, competitor products, and products with the same ingredients based on the number of clients within a specific region; this allows for the identification of revenue levels at the hospital level and the performance of regional comparisons.

[0159] The number of regional business partners can be determined by calculating the number of partners in a specific region that handle the company's products, competitor products, and products with the same ingredients, respectively, and this can be used as an indicator of the extent of product distribution and the status of securing business partners.

[0160] The HHI index can be calculated to indicate the degree to which sales in a specific region are concentrated in certain hospitals; a higher value signifies a greater concentration of sales to specific clients. Therefore, the HHI index can be utilized as an indicator to simultaneously evaluate the stability and risk of the sales structure within a region.

[0161] The monthly sales standard deviations by region, company, competitor, and products with the same ingredients can be calculated to quantitatively represent the variability of monthly sales data, thereby allowing for the determination of sales stability.

[0162] In addition, the recent sales growth rates of regional, company, competitor, and products with the same ingredients can be calculated by comparing the average sales of the last three months with the average sales of the previous 12 months, and can be used as a standard to evaluate the short-term growth trend and market changes of a specific product or ingredient.

[0163] According to the third embodiment of the analysis, the analysis unit (54) can calculate sales results by customer grade, the correlation between customer grade and regional characteristics, the company's market share and growth rate within the region, and the distribution and growth trends of sales-based accounts based on data recorded in the pharmaceutical sales support database (60). These analysis results can be provided to terminals (10, 20) in the form of visualizations such as bar graphs, distribution charts, and tables, through which sales representatives can verify the sales contribution and growth potential of each customer.

[0164] According to the fourth embodiment of the analysis, the analysis unit (54) can control the generation model (80) to generate multiple regional groups based on regional unit data recorded in the pharmaceutical sales support database (60). For example, the analysis unit (54) can apply the K-Means clustering technique to classify regions with similar sales volume and sales growth / decrease trends into the same group. FIG. 30 and [Table 1] illustrate the results of such K-Means clustering as an example. In this embodiment, regions are classified into four types: stable growth regions, underperforming / stagnant regions, core regions, and high-risk concentrated regions.

[0165] Cluster Area Average Revenue Average Revenue per Hospital Monthly Revenue Standard Deviation Recent Revenue Growth Rate Meaning of Key Indicator Interpretation 0- Stable Growth Regions (53 locations) 0.8 billion 0.12 billion 830k -0.45 Revenue scale is not large, but efficiency per hospital is good, and volatility is low. Market is small, but management is efficient, and stability can be maintained. 1- Poor Performance - Stagnant Regions (227 locations) 0.15 billion 0.09 billion 180k -0.86 Both revenue scale and efficiency are low, and the downward trend is clear. Low efficiency and low growth expectations. 2- Core Regions (8 locations) 4.47 billion 0.75 billion 337k -0.47 Both scale and efficiency are excellent, however, recent growth has stagnated. Current performance is excellent, but intensive management is required. 3- High-Risk Concentration Regions (2 locations) 3.97 billion 3.97 billion 333k -0.94 Entirely dependent on specific hospitals, volatility is very high, high risk.

[0166] As shown in [Table 1], each regional group can be classified based on user-specified indicators such as the average regional sales, average hospital sales per region, standard deviation of monthly sales, and recent sales growth rate, and the generating model (80) can define the nature of the regions belonging to each group by reflecting the key characteristics of each group. For example, a stable growth region has a small sales volume but low volatility and high stability, while a core region has both high sales volume and efficiency but may show a recent stagnation in growth. On the other hand, a high-risk concentration region can be classified as having very high volatility due to excessive concentration of sales in specific hospitals. Additionally, the analysis unit (54) can derive additional factors affecting each cluster as shown in FIG. 31. Specifically, the generating model (80) can calculate the importance of variables that contributed to the classification of each regional group by utilizing the SHAP (Shapley Additive exPlanations) technique. For example, the number of clients, the sales growth rate of the company's own products and other companies' products, the share of sales in the top 10%, and the HHI index can be identified as factors that significantly influence the clustering results. FIG. 31 is an example illustrating the results of such SHAP-based factor analysis, showing that each regional group is determined by variables such as the sales proportion of a specific product or the number of customers. Furthermore, the analysis unit (54) can identify regions where sales of a specific product need to be strengthened based on the calculated regional groups and the results of the influencing factor analysis.

[0167] For example, regions with a low sales proportion of a specific product but high growth potential can be identified as regions for sales reinforcement, while conversely, regions with an excessively high sales proportion of a specific product and high risk can be identified as regions requiring management and diversification strategies.

[0168] According to the fifth embodiment of the analysis, the analysis unit (54) can analyze sales and growth-related indicators at the regional level based on data recorded in the pharmaceutical sales support database (60) by the generation model (80), and classify multiple regional groups based on this. For example, as shown in FIG. 32 and [Table 2], the analysis unit (54) can classify each region into Hot regions, promising growth regions, crisis regions, and potential regions by utilizing indicators such as total regional sales, average hospital sales per region, monthly sales standard deviation, and recent sales growth rate.

[0169] Category Name | Criteria | Characteristics HOT Regions (53) | Total regional revenue ≥ Top 40% AND Average hospital revenue per region ≥ Top 40% AND Recent revenue growth rate ≥ Top 40% | Core regions with excellent scale and growth potential Growth-Promising Regions (47) | Total regional revenue < Top 40% AND Recent revenue growth rate ≥ Top 40% | Small-scale regions with high growth potential Risk Regions (49) | Total regional revenue ≥ Top 40% AND (Average hospital revenue per region < Top 40% OR Monthly revenue standard deviation ≥ Top 40%) | Large-scale regions with efficiency or stability issues Potential Regions (141) | Does not meet any of the above conditions | No performance or growth

[0170] Additionally, for each classified regional group, the analysis unit (54) can additionally calculate factors such as the sales proportion of a specific product, the sales growth rate, and the market share relative to competing products, as illustrated in FIG. 33, and can identify regions where sales of a specific product need to be strengthened by synthesizing the calculated results. For example, if the sales proportion of a specific product (e.g., Davi Duo) is low despite being a growth region, it can derive the need to strategically re-select target customers for that product. On the other hand, if the sales proportion of a specific product is excessively high in a crisis region, it can derive the result that a sales strategy to diversify risk needs to be established. Additionally, the analysis unit (54) can automatically generate a performance analysis report for individual sales representatives based on activity data, sales performance, customer response history, and other relevant information of sales representatives recorded in the pharmaceutical sales support database (60). The performance analysis report may include results of score and grade calculation, quantitative and qualitative analysis results regarding strengths and improvement points, and comprehensive analysis results, as illustrated in FIG. 33 and FIG. 34, for example. In particular, the analysis unit (54) can visualize key performance indicators, such as the number of customer visits by sales representatives, the contribution to sales by clients, growth rates, and stability indicators (HHI index, etc.), and provide them in a graphic form such as a radar chart. According to one embodiment, as illustrated in FIG. 36, the analysis unit (54) can predict the short-term and medium-term performance of major clients based on sales data and growth trends by client recorded in the pharmaceutical sales support database (60). For example, the analysis unit (54) can calculate the current sales performance by client and, based on this, predict the growth rates for 3 months and 6 months later to provide graded forecast results. The forecast results can distinguish and display clients with high growth potential and clients expected to decline.In addition, the analysis unit (54) can generate customized coaching points based on the prediction results for each customer.

[0171] FIGS. 37 and 38 are drawings for explaining the data provision unit (55) of a support platform service according to an embodiment of the present disclosure. When a user selects the data viewing icon on the main screen using the input device of the first terminal (10), the screen may be switched to the screen of FIG. 36.

[0172] The user can check the necessary data by selecting an icon related to a data item as shown in FIG. 37. When a data item icon is selected, the screen switches to FIG. 38. The user can input a query regarding the data through the input device of the first terminal (10). In this case, the generation model (80) can display an answer to the user's query on the first terminal device (10).

[0173] The generative model-based pharmaceutical sales support platform program, its operating system, and method according to the present invention can integrate and manage data generated from pharmaceutical sales activities and provide responses and analysis results based on this data. Accordingly, sales representatives can perform customer response, sales record management, conversation simulation training, data analysis, and prediction, thereby increasing sales efficiency and supporting organizational decision-making.

[0174] Furthermore, the present invention links a pharmaceutical sales support database with a generative model to ensure the accuracy and reliability of responses and to derive predictions and response strategies regarding changes in the sales environment. This can reduce the workload of sales representatives, improve the level of customer management, and contribute to increasing sales and strengthening the competitiveness of pharmaceutical companies.

[0175] Those skilled in the art of the present disclosure will understand that information and signals may be represented using any various different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced in the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

[0176] Those skilled in the art will understand that the various exemplary logic blocks, modules, processors, means, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented by electronic hardware, various forms of programs or design code (referred to herein as software for convenience), or a combination of all such. To clearly illustrate this interoperability between hardware and software, various exemplary components, blocks, modules, circuits, and steps have been generally described above in relation to their functions. Whether such functions are implemented as hardware or software depends on the design constraints imposed on the specific application and the overall system. Those skilled in the art may implement the functions described in various ways for each specific application, but such implementation decisions should not be interpreted as being outside the scope of this disclosure.

[0177] The various embodiments presented herein may be implemented as methods, devices, or articles manufactured using standard programming and / or engineering techniques. The term "article manufactured" includes a computer program, a carrier, or a medium accessible from any computer-readable storage device. For example, computer-readable storage media include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical discs (e.g., CDs, DVDs, etc.), smart cards, and flash memory devices (e.g., EEPROMs, cards, sticks, key drives, etc.). Additionally, the various storage media presented herein include one or more devices and / or other machine-readable media for storing information.

[0178] It should be understood that the specific order or hierarchy of steps in the presented processes is an example of exemplary approaches. It should be understood that the specific order or hierarchy of steps in the processes may be rearranged within the scope of this disclosure based on design priorities. The appended method claims provide elements of various steps in a sample order, but do not imply being limited to the specific order or hierarchy presented.

[0179] Description of the presented embodiments is provided so that a person skilled in the art may use or practice the present disclosure. Various modifications to these embodiments will be apparent to a person skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present disclosure is not limited to the embodiments presented herein, but should be interpreted in the broadest possible scope consistent with the principles and novel features presented herein.

[0180] [Explanation of the symbol]

[0181] User terminal: 10

[0182] Operator Terminal: 20

[0183] Network: 30

[0184] Internal Server: 40

[0185] Processor Section: 41

[0186] Storage: 42

[0187] Network Interface Section: 43

[0188] Control Program: 44

[0189] Database: 60

[0190] Model providing server: 70

[0191] Model Execution Unit: 71

[0192] Model Storage: 72

[0193] External Communications Unit: 73

[0194] Generative Model: 80

Claims

1. In a system operating a platform program that supports pharmaceutical sales, At least one server having a program control unit for controlling the platform program, a pharmaceutical sales support database for operating the platform program, and a generation model for generating a response based on the pharmaceutical sales support database; and It includes at least one terminal connected to communicate with the server through the above network, and The above platform program is a system equipped in at least one of the terminal and the server.

2. In Paragraph 1, The above pharmaceutical sales support database is, A system containing at least one of drug information, disease and treatment information, customer information, company schedule, and regulatory compliance information.

3. In Paragraph 1, The above at least one server is, Internal server; and Includes a model provision server, The above internal server is, A program control unit for controlling the above platform program and a storage unit having the above database; A processor unit for executing the program control unit provided in the storage unit; and It includes a network interface unit that communicates the internal server with the terminal and the model providing server through the network, and The above model providing server is, A model storage unit having the above-mentioned generated model; A model execution unit that executes the generated model provided in the above model storage unit; and A system comprising an external communication unit that communicates the above-mentioned model providing server with the above-mentioned terminal and the above-mentioned internal server through the above-mentioned network.

4. In Paragraph 3, A system in which, when the terminal requests a task from the internal server, the internal server measures data required for the task from the pharmaceutical sales support database, the model providing server generates a response to the requested task based on the specified data, and the response is displayed on the terminal via the platform program.

5. In any one of paragraphs 1 through 4, The above platform program is, It includes a query processing unit that receives a query input through the above terminal, and A system in which the above query processing unit transmits the above query to the above generation model and controls the generation model to display the response generated based on the above pharmaceutical sales support database on the terminal.

6. In any one of paragraphs 1 through 4, The above platform program is, It includes a record management unit that manages sales record information input through the above terminal, and The above record management unit controls the generation model to display a query regarding the missing item on the terminal when there is a missing item in the sales record information in a predefined set of items, and controls the display of the query to end when all the missing items are entered.

7. In Paragraph 6, The above-mentioned predefined set of items is, A system including at least one of the following: customer name, purpose of visit, results of visit, next visit schedule and goals, information confirmed today, and follow-up activities.

8. In any one of paragraphs 1 through 4, The above platform program is, It includes a training unit that simulates conversations with customers, and The above training department, Data regarding a customer selected through the above terminal is identified from the above pharmaceutical sales support database, and A system that controls the generation model to generate a response simulating the customer based on specified data for an interactive input entered through the terminal.

9. In Paragraph 8, The above training department, When the above simulation is terminated, an evaluation score for at least one of the above interactive inputs entered through the terminal is calculated, and The above evaluation score is a system calculated based on the degree of consistency with customer information recorded in the above pharmaceutical sales support database.

10. In Paragraph 9, The above alignment diagram is, Whether there is a match between the keywords included in the above interactive input and the attribute values ​​of the customer information recorded in the above pharmaceutical sales support database; After converting the above interactive input and the above customer information into embedding vectors, the cosine similarity between the embedding vectors; or A system in which the above interactive input is calculated based on at least one probability value of belonging to a specific customer group by a classification model pre-trained on the above customer information.

11. In any one of paragraphs 1 through 4, The above platform program is, A system comprising an analysis unit that processes data recorded in the above pharmaceutical sales support database to produce an analysis result.

12. In Paragraph 11, The above analysis unit is a system that controls the above generation model to generate market share and growth rate of target products on a national basis by period, based on sales data recorded in the above pharmaceutical sales support database.

13. In Paragraph 11, The analysis unit is a system that controls the generation model to generate at least one of total sales by region, average hospital sales, number of clients, an HHI index representing sales concentration, monthly sales standard deviation, and recent sales growth rate based on data recorded in the pharmaceutical sales support database.

14. In Paragraph 13, The analysis unit above is a system that controls the generation model to generate multiple regional groups according to the K-Means clustering technique based on regional unit data recorded in the pharmaceutical sales support database.

15. In Paragraph 14, The analysis unit is a system that controls the generation model to classify the plurality of region groups generated by the generation model into core regions, growth-promising regions, crisis regions, and potential regions based on at least one indicator among total regional sales, average hospital sales per region, monthly sales standard deviation, and recent sales growth rate.

16. In Paragraph 14, The analysis unit is a system that calculates the sales proportion, sales growth rate, or market share relative to competing products of a specific product for the plurality of region groups generated by the generation model, and controls the identification of regions requiring sales reinforcement of the specific product based on the calculated results.

17. In any one of paragraphs 1 through 4, The above platform program is, A system including a data providing unit that displays data recorded in the above pharmaceutical sales support database on a terminal.

18. Regarding the method of operating a platform program to support pharmaceutical sales, A step of receiving a query or request from at least one terminal connected to communicate through a network; A step of executing a program control unit that controls the platform program using at least one server connected to the terminal; A step of identifying data from a pharmaceutical sales support database provided in the above server; and The method includes the step of generating a response by executing a generation model based on specific data and providing the response to the terminal. The above platform program is a method provided in at least one of the terminal and the server.

19. In Paragraph 18, A method in which the above pharmaceutical sales support database includes data comprising at least one of drug information, disease and treatment information, customer information, company schedule and regulatory compliance information.

20. In Paragraph 18, The above server includes an internal server and a model providing server, and A step of managing the pharmaceutical sales support database by executing the program control unit on the internal server; A step of performing communication between the internal server, the terminal, and the model providing server through a network interface; Step of executing the generated model on the above-mentioned model providing server; and A method comprising the step of performing communication between the above-mentioned model providing server and the above-mentioned terminal and internal server through an external communication unit.

21. In Paragraph 20, A step in which the above terminal requests a task from the above internal server; A step in which the internal server identifies data required for the business from the pharmaceutical sales support database; A step in which the above model providing server executes the above generation model based on specified data to generate a response to the above requested task; and A method comprising the step of displaying the above response on the terminal through the platform program.

22. In Paragraph 18, A step of receiving a query input through the above terminal; A step of transmitting the above query to the above generation model; The step of the above-mentioned generation model generating a response based on the above-mentioned pharmaceutical sales support database; and A method comprising the step of controlling the above response to be displayed on the terminal.

23. In Paragraph 18, A step of managing sales record information input through the above terminal; A step of determining whether the above sales record information has any missing items in a predefined set of items; If there is a missing item, a step of controlling the generation model to display a query regarding the missing item on the terminal; and A method comprising a step of controlling the display of the above query to end when all the above missing items are entered.

24. In Paragraph 18, A step of identifying data regarding a customer selected through the above terminal from the above pharmaceutical sales support database; A step of receiving interactive input input through the above terminal; and A method comprising the step of performing an interactive simulation for the above interactive input, wherein the generating model generates a response simulating the selected customer based on the above specific data.

25. In Paragraph 24, When the above interactive simulation is terminated, a step of calculating an evaluation score for at least one of the above interactive inputs entered through the terminal; and A method comprising the step of calculating the above evaluation score based on the degree of consistency with customer information recorded in the above pharmaceutical sales support database.

26. In Paragraph 18, A method comprising the step of processing data recorded in the above pharmaceutical sales support database to produce an analysis result.

27. In Paragraph 26, The step of producing the above analysis results is, A method comprising the step of controlling the above-mentioned generation model to generate market share and growth rate of target products on a national basis by period, based on sales data recorded in the above-mentioned pharmaceutical sales support database.

28. In Paragraph 26, The step of producing the above analysis results is, A method comprising the step of controlling the above-mentioned generation model to generate at least one of total sales by region, average sales by hospital, number of clients, HHI index representing sales concentration, monthly sales standard deviation, and recent sales growth rate based on data recorded in the above-mentioned pharmaceutical sales support database.

29. In Paragraph 26, The step of producing the above analysis results is, A method comprising the step of controlling the above-mentioned generation model to generate multiple region groups according to the K-Means clustering technique based on region unit data recorded in the above-mentioned pharmaceutical sales support database.

30. In Paragraph 29, The step of producing the above analysis results is, A method comprising the step of calculating the sales proportion, sales growth rate, or market share relative to competing products of a specific product for the plurality of region groups calculated by the generation model, and controlling to identify regions where sales of the specific product need to be strengthened based on the calculated results.

31. In Paragraph 18, A method comprising the step of displaying data recorded in the above pharmaceutical sales support database on the terminal.

32. A recording medium storing a program that executes the method of any one of paragraphs 18 through 31.

33. A program stored on a recording medium that executes the method of any one of paragraphs 18 through 31.

34. In a server operating a platform program to support pharmaceutical sales, A program control unit that controls the above platform program; and It includes a pharmaceutical sales support database for operating the above platform program, and A server characterized in that the above program control unit controls the response generated by the generation model based on the pharmaceutical sales support database to be output through the platform program.

35. In Paragraph 34, The above pharmaceutical sales support database is, A server characterized by including at least one of drug information, disease and treatment information, customer information, company schedule, and regulatory compliance information.

36. In Paragraph 34, The above platform program is, A server characterized by including at least one of a query processing unit, a record management unit, a training unit, an analysis unit, and a data provision unit.

37. In Paragraph 36, The above query processing unit is, Receive a query entered by the user, and A server characterized by controlling the output of a response to the query, calculated by the above-mentioned generation model based on the above-mentioned pharmaceutical sales support database, through the above-mentioned platform program.

38. In Paragraph 36, The aforementioned record management department, It manages sales record information entered by users, and If the sales record information entered by the above user has missing items in the predefined set of items, A server characterized by the above-mentioned generation model generating a query regarding the above-mentioned missing item and controlling it to be output through the above-mentioned platform program.

39. In Paragraph 38, The above-mentioned predefined set of items is, A server characterized by including at least one of the following: customer name, purpose of visit, results of visit, schedule and goals for the next visit, information confirmed today, and follow-up activities.

40. In Paragraph 36, The above training department, Identifying customer data recorded in the above pharmaceutical sales support database, and By controlling the generation model to generate a response that simulates an actual customer based on the specified customer data, A server characterized by enabling interactive input entered by a user to be trained in a simulated conversational format.

41. In Paragraph 40, The above training department, When the above simulation ends, Calculate an evaluation score for at least one of the interactive inputs entered by the user, and A server characterized in that the above evaluation score is calculated based on the degree of consistency with customer information recorded in the above pharmaceutical sales support database.

42. In Paragraph 36, The above analysis unit is, Based on the sales data recorded in the above pharmaceutical sales support database, A server characterized by controlling the generation model to calculate at least one of market share, growth rate, number of customers, average hospital revenue, HHI index representing revenue concentration, monthly revenue standard deviation, or recent revenue growth rate.

43. In Paragraph 36, The above analysis unit is, A server characterized by controlling the generation model to calculate multiple regional groups according to the K-Means clustering technique based on regional unit data recorded in the above pharmaceutical sales support database.

44. In Paragraph 43, The above analysis unit is, A server characterized by controlling the generation model to classify the above plurality of region groups into core regions, growth-promising regions, crisis regions, and potential regions based on at least one indicator among total regional sales, average hospital sales, monthly sales standard deviation, and recent sales growth rate.

45. In Paragraph 36, The above data provider is, Displays data recorded in the above pharmaceutical sales support database to the user, and If a user query is entered regarding the displayed data, A server characterized by controlling the above-mentioned generation model to generate and output a response to the above-mentioned query.

Images