System and method for managing target hospital drug inventory through drug demand prediction

Abstract

The invention provides a system and a method for managing target hospital drug inventory by predicting required drugs. The system is composed of the following modules. The inventory management module automatically generates current medicine inventory information of a hospital, and the patient information calculation module generates expected patient information in a specific time period according to the current patient information and the reserved patient information. The medicine calculation module predicts the required medicine according to the medicine inventory information and the expected patient information, and the supplier account module allows the registered medicine supplier to automatically receive the required medicine information. The contract management module manages ordering contracts and updates medicine inventory information through the contract management module. A patient information calculation module predicts expected patient information using a first machine learning model, and a medication calculation module provides a system for predicting a necessary medication using a second machine learning model.

Description

Technical Field

[0001] This invention relates to a system for managing the drug inventory of a target hospital based on the prediction of required drugs, and a method for using it. Background Technology

[0002] The conventional drug ordering method monitors the inventory of drugs prescribed (sold) by the hospital. When the inventory falls below a certain level, the hospital terminal is notified and a certain quantity of drugs is ordered to the pharmaceutical company or drug distributor.

[0003] In other words, in the past, simply relying on drug inventory and considering current prescription trends for that drug, such as whether more or less drugs were prescribed than before (e.g., during seasonal changes), would not allow for an increase or decrease in cold medicine orders.

[0004] In this regard, Korean Patent Publication No. 10-2016-0085513 is cited as prior art. Based on this, a hospital terminal, pharmaceutical factory server, and pharmaceutical distributor server disclose an invention that allows for the implementation of drug inventory and order management by providing drug inventory and order management, based on the prediction of drug prescription trends and the adjustment of order volumes for alternative drugs according to the predicted prescription trends.

[0005] In recent years, artificial intelligence has played a vital role in various fields of science and technology. Specifically, in the field of pharmaceutical inventory management systems, research is underway to use medical data to perform machine learning on artificial intelligence models, and to use the learned artificial intelligence models to obtain various predictive results from the input medical data.

[0006] However, research on technologies that use medical data to accurately predict necessary drugs remains scarce. Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] The purpose of this invention is to address the aforementioned problems by enabling the analysis of required medications based on target hospitals, usage patterns, and trends, and to monitor each key indicator in real time through a user-friendly dashboard. We aim to propose a technology that can customize and adjust automated ordering and inventory management strategies according to the target hospital's operational methods, specific departmental requirements, and seasonal changes.

[0009] The technical challenges to be addressed in the various embodiments are not limited to those described above, and those skilled in the art may consider other technical challenges not mentioned in the various embodiments described below.

[0010] means for solving problems

[0011] To achieve the above objectives, a target hospital drug inventory management system based on required drug prediction, according to an embodiment of the present invention, includes: an inventory management module that automatically generates current drug inventory information for the target hospital; a patient information calculation module that generates expected patient information for a predetermined time period based on current patient information and scheduled patient information of the target hospital; a drug calculation module that calculates necessary drug information within the target time period, taking into account information including drug inventory information and expected patient information; a supplier account module provided for pre-registered drug suppliers, who can log in via a preset method and automatically receive the required drug information from the drug calculation module; and a contract management module that manages drug ordering contracts between drug suppliers and the target hospital and updates drug inventory information according to the content of the ordering contracts. The patient information calculation module includes a first machine learning model for predicting expected patient information, and the first machine learning model uses information including patient information from the past preset time period as learning information to perform machine learning using current patients. A system is also provided, including a second machine, wherein the drug calculation module is used to generate the expected patient information output by the first machine learning model. A learning model is trained to use this information to predict the required drug information.

[0012] In addition, the patient information calculation module generates expected patient information for the target time period based on a preset cycle, the inventory confirmation date for generating expected patient information, current outpatient and inpatient patients, and arranged outpatient and inpatient appointments; by taking into account the generated expected patient information, and by taking into account the characteristics of the target hospital departments, the correction coefficients for each outpatient and inpatient patient can be adjusted.

[0013] In addition, the patient information calculation module pre-stores the types and quantities of drugs required for each disease. The second machine learning model of the drug calculation module takes into account the diseases of outpatients and inpatients and can generate required drug information output.

[0014] In addition, the contract management module allows at least one pharmaceutical supplier to apply for a supply contract based on the required pharmaceutical information provided in a preset manner. When there are multiple pharmaceutical suppliers, one supplier is selected according to the preset method. The supplier screening department also includes a supplier selection unit that independently inputs the supply application information of each pharmaceutical supplier for the required pharmaceutical information through the supplier account module. The supply application information may include the expected supply price and the expected quantity.

[0015] In addition, within the limits of the quantity or capacity included in the necessary drug information It, the supplier selection unit converts multiple drug suppliers from a contract waiting state to a contract signed state based on the order of the lowest expected supply price. The required quantity included in the supply application information of each of the multiple drug suppliers can be set to be accumulated.

[0016] In addition, the inventory management module further considers the individual expiration information of the current medicines and generates the current medicine inventory information. When the target time period changes from the kth target time period to the k+1th target time period, the medicines whose expiration information can be obtained can be classified as emergency medicines.

[0017] Meanwhile, the present invention is a method using the above system, comprising the following steps: (a1) automatically generating current drug inventory information of the target hospital through the inventory management module; (a2) generating expected patient information for a preset time period based on the current patient information and appointment patient information of the target hospital through the patient information calculation module; (a3) ​​calculating the drug information required for the target time period by taking into account information including drug inventory information and expected patient information through the drug calculation module; (a4) automatically providing the necessary drug information from the drug calculation module to the pre-registered drug supplier through the supplier account module; (a5) signing a drug purchase contract between the drug supplier and the target hospital through the contract management module, and updating the drug inventory information according to the content of the purchase contract.

[0018] Additionally, in step (a2), the desired patient information is output through the first machine learning model included in the patient information calculation module (a20). The first machine learning model is patient information from a preset time A in the past. The steps are: performing machine learning on information containing this information as learning information, using current patient information and reserved patient information as input information, and learning to generate desired patient information as output information; step (a3) ​​includes, (a30) outputting necessary drug information through the second machine learning model included in the drug calculation module, wherein the second machine learning model predicts necessary drug information based on learning from the first machine learning model. The expected patient information is then output.

[0019] This invention relates to a drug inventory management system based on necessary drug demand forecasting to effectively manage drug inventory in hospitals. The system according to embodiments of the invention is configured to automatically identify and manage the current drug inventory status of a target hospital, predict future drug demand using patient data, and support smooth contracts and supply procedures with suppliers. To achieve this, the invention includes the following components.

[0020] It provides an inventory management module to track and manage the drug inventory of target hospitals in real time. The module automatically generates the inventory status of drugs within the hospital and updates information such as the quantity, expiration date, and stock shortages of each drug in real time. This is crucial for ensuring timely treatment for patients and an adequate supply of necessary medications.

[0021] Additionally, a patient information calculation module is included to predict the expected number of patients in a specific future period (hereinafter referred to as the "target period") based on the current and reserved patient information of the target hospital. The patient information calculation module generates information about the expected patients within the target period by applying a first machine learning model. This first machine learning model is configured to learn from past patient information and treatment history within a specific time period, receiving current and reserved patient information as input, and deriving expected patient information. This expected patient information includes detailed data on each patient's treatment type, treatment plan, and expected treatment process.

[0022] A drug calculation module is provided to predict the types and quantities of drugs needed by the hospital during a target period. Based on the output of a first machine learning model, the module uses a second machine learning model to derive necessary drug information and leverages the prediction results to manage the hospital's drug inventory and establish ordering plans. The second machine learning model aims to accurately predict drug demand based on the expected number of patients by learning from past patient groups and drug usage history.

[0023] This invention includes a supplier account module for smooth collaboration between hospitals and drug suppliers. The supplier account module allows pre-registered drug suppliers to access the system, and suppliers can log in using preset authentication methods. By configuring the supplier account module, it automatically receives predicted drug demand information from the drug calculation module, enabling suppliers to understand the hospital's needs in real time and supply drugs promptly.

[0024] In addition, this invention includes a contract management module for signing and managing purchase contracts between hospitals and suppliers. The contract management module efficiently manages the drug ordering and supply process and automatically updates the hospital's drug inventory information based on the content of the purchase contracts. In this way, suppliers can supply drugs according to the contracts, and hospitals can ensure an adequate supply of drugs when needed.

[0025] The system of this invention can prevent drug shortages and inventory backlogs by optimizing hospital drug inventory and responding promptly to patient needs. Furthermore, the use of machine learning models for sophisticated predictions can maximize hospital operational efficiency and ensure the stability of the drug supply chain by supporting smooth collaboration with suppliers.

[0026] The patient information calculation module includes a first machine learning model for predicting expected patient information. This model is configured to perform machine learning using patient information from a specific past period as learning data. The learning data includes detailed information such as past outpatient and inpatient medical records, appointment information, patient age, gender, disease type, and treatment duration. This information allows analysis of past patient numbers and treatment patterns to predict future patient numbers.

[0027] The first machine learning model takes current outpatient and inpatient information, as well as patient appointment information based on future schedules, as input and is trained to generate expected patient information for a target time period as output. This model combines past data with current information, deriving patterns based on patients' treatment cycles and disease tendencies, and can predict the expected number of patients and disease distribution within a certain timeframe. For example, it can make predictions reflecting periods of increased seasonal illness or considering increased visitation frequency in specific healthcare sectors.

[0028] In addition, the drug calculation module includes a second machine learning model for predicting necessary drug information based on anticipated patient information derived from the first machine learning model. The second machine learning model is configured to accurately predict the type and dosage of drug required for each patient type. To this end, the second machine learning model learns from past patient information and medication data, and generates necessary drug information by reflecting differences in medication use between outpatients and inpatients, characteristics of each department, and prescription trends for specific diseases.

[0029] For example, the second machine learning model can predict the demand for drugs such as antibiotics, painkillers, and anticancer drugs based on the expected number of outpatients and inpatients and their disease types, and use these predictions for hospital inventory management. This is directly used to formulate ordering plans. The drug calculation module assesses the risk of hospital stockouts in advance based on predicted drug demand and can automatically place orders with suppliers when necessary.

[0030] Therefore, the organic link between the first and second machine learning models can help hospitals ensure the appropriate medications are available based on anticipated patient numbers and treatment needs, minimizing unnecessary inventory and maximizing operational efficiency. Furthermore, these machine learning models continuously learn from new patient information and treatment data to progressively improve predictive accuracy and enable rapid response to evolving healthcare environments.

[0031] In addition, the patient information calculation module is configured to periodically generate anticipated patient information for a target time period according to a preset cycle, in order to maximize the hospital's operational efficiency. The anticipated patient information is generated based on the inventory confirmation date, taking into account information on all currently visiting outpatients and inpatients, as well as those with scheduled appointments. This allows the hospital to appropriately allocate necessary resources based on treatment and schedules, and to predict drug demand in advance based on the expected number of patients within a certain timeframe.

[0032] The patient information calculation module applies correction coefficients to the number of outpatients and inpatients, reflecting the characteristics of hospital departments and hospital operation types. These correction coefficients can be adjusted based on patient type, treatment frequency, treatment mode, and patient severity, improving the accuracy of patient number predictions and enabling personalized forecasting for each hospital. These correction coefficients can be variably set based on hospital operator policies or statistical analysis results, allowing for flexible responses to real-time changes in the hospital environment.

[0033] In addition, the patient information calculation module pre-stores the required drug type and dosage for each patient's disease in the database, and allows it to be used in conjunction with the drug calculation module. Specifically, the second machine learning model of the drug calculation module is used to predict the type and dosage of drugs required within a target time period, taking the disease information of outpatients and inpatients provided by the patient information calculation module as input.

[0034] The second machine learning model can predict the type and exact dosage of medication needed to treat a specific disease by learning from past patient treatment history and corresponding drug usage data. For example, it can predict the consumption of antibiotics or anticancer drugs that are repeatedly used in a particular department. In this way, hospitals can ensure adequate drug supplies in advance based on anticipated patient numbers and treatment trends. Furthermore, the second machine learning model can reflect the differences between outpatient and inpatient treatment in hospitals, deriving necessary drug information suitable for each situation, thereby preventing over-ordering of drugs and optimizing inventory.

[0035] By organically linking the patient information calculation module and the drug calculation module, hospitals can accurately predict drug demand based on the expected number of patients and treatment characteristics, and obtain supplies in a timely manner when needed. This allows hospitals to improve the quality of patient treatment, reduce unnecessary inventory costs, and ensure a stable drug supply chain.

[0036] The system provides required drug information in a pre-defined manner to apply for a drug supply contract. This contract application process includes a procedure where suppliers express their intention to supply drugs and specify necessary conditions; this process is conducted through the supplier account module. The contract management module includes a supplier selection unit. When multiple drug suppliers exist, this unit ultimately selects a suitable supplier based on pre-defined evaluation criteria and procedures.

[0037] The supplier selection unit allows each pharmaceutical supplier to independently input and submit supply request information based on the required drug information. This request information includes the supplier's expected supply price and quantity of the drugs. The system manages the request information entered by each supplier through the supplier account module independently, ensuring fair competition among suppliers.

[0038] The supplier selection department screens suppliers based on their supply requests, according to pre-defined evaluation criteria. These criteria may include the supplier's suggested price, suggested quantity, supply history, reputation, ability to fulfill contracts, and past transaction history. In particular, depending on the urgency of the specific drug or the hospital's demand, in addition to price, supply time or the supplier's inventory status may also be considered as key evaluation factors.

[0039] The supplier selection department automatically creates the final contract terms between the selected suppliers and the hospital, and manages them by integrating them into the contract management module. If multiple suppliers are selected and partial supply is required, the supplier selection department clearly specifies the quantity and conditions allocated to each supplier, and this is reflected in the contract. This process ensures transparency in the supply process and helps the hospital receive the required medicines in a timely manner.

[0040] Furthermore, the supplier selection department monitors supplier-entered supply request information in real time and notifies hospital administrators when necessary to expedite the supplier selection process. This efficiently supports contract signing between suppliers and hospitals, contributing to the stability of the pharmaceutical supply chain.

[0041] Furthermore, the supplier selection unit is configured to proceed with the contract signing process in order of the lowest expected supply price among multiple pharmaceutical suppliers. Specifically, the supplier selection department starts with the supplier with the lowest expected supply price, sequentially switching from a contract waiting state to a contract signing state, and selecting the required pharmaceuticals based on the expected quantities included in the supply requests submitted by each supplier. The hospital is established to accumulate relevant supplier request information until the required quantity or capacity is met.

[0042] The supplier selection department adjusts the supplier's supply request information, using the total quantity or capacity included in the required drug information as the upper limit. If the quantity requested by any supplier exceeds the hospital's needs, the excess will be excluded. The system is configured to reflect only the required quantity. For example, if the total demand for a certain drug is 1000 units, and the first supplier requests 600 units and the second supplier requests 500 units, then the contract with the first supplier will be for 600 units, the contract with the second supplier will be for 600 units, and the remaining 400 units will be contracted.

[0043] In this way, the supplier selection unit can secure the required quantity under optimal conditions, prevent hospital inventory shortages, and maximize cost savings by merging multiple supplier proposals in order of desired supply price. If multiple suppliers apply for the same price, the supplier selection department can determine priority by applying additional evaluation criteria (such as supply history, credit rating, delivery time, etc.).

[0044] Furthermore, the contract signing process is automated, allowing for quick and accurate contract signing without intervention from hospital administrators. This automated signing process improves the efficiency of supply chain operations and enables hospitals to respond rapidly to unexpected changes in demand.

[0045] In addition, the inventory management module is configured to take into account the expiration dates of each drug when generating inventory information, enabling more accurate management of current drug inventory. The module records the arrival date, expiration date, and inventory quantity of each drug held by the hospital in the database and updates the current inventory status in real time. Specifically, it supports optimized inventory depletion planning by predicting the expiration dates of specific drugs in advance and prioritizing their use.

[0046] Additionally, the inventory management module is configured to automatically identify expired medications and classify them as emergency medications during the transition from the kth target period to the (k+1)th target period. Hospital administrators will automatically receive notifications for medications falling into the emergency category and will take immediate action to ensure the medications are used correctly or reassigned.

[0047] Specifically, the inventory management module can be configured to create a list of medicines nearing their expiration date and automatically deliver them to necessary departments within the hospital, or prioritize their delivery to medical departments where drug use can be optimized. For example, if a specific medicine's expiration date is within the k+1 target period, the inventory management module classifies the medicine as an emergency medicine and provides plans to supply it to other hospitals or external organizations if necessary.

[0048] In this way, the inventory management module can systematically manage medications nearing their expiration date, thereby minimizing medication waste and avoiding financial losses for the hospital. Furthermore, through the automatic classification and notification functions for emergency medications, hospitals can reduce unnecessary inventory burdens, achieve efficient inventory operations, and ensure that necessary medications are available in a timely manner.

[0049] This invention provides a system and method for maximizing hospital operational efficiency and improving the accuracy of inventory and order management based on the analysis of usage patterns and trends in each hospital. To this end, the invention includes functions for customizing inventory management and automated ordering strategies by reflecting each hospital's operational methods, specific departmental requirements, seasonal variations, etc., and provides key metrics to support real-time monitoring through a user-friendly dashboard.

[0050] The system is equipped with modules for analyzing medication patterns and trends in various hospitals. It analyzes the medication history and trends of each department within each hospital. The analysis results reflect the specific needs of outpatient and inpatient departments, as well as the hospital as a whole, and are used to develop strategies that take into account seasonal changes or peak periods for specific diseases. For example, during flu season, the demand for antiviral drugs or antipyretics is expected to increase, so appropriate ordering plans should be developed based on this period.

[0051] The ability to monitor key metrics in real time through a user-friendly dashboard. The dashboard intuitively visualizes various data points, including hospital drug inventory, drugs nearing expiration, departmental drug usage, estimated patient numbers, and current order status, allowing users to quickly identify the information they need and make informed decisions. This dashboard can be customized based on administrator privileges, providing a tailored interface to meet the specific needs of each hospital.

[0052] This invention allows for the dynamic adjustment of automated ordering and inventory management strategies based on usage data and patterns collected for each hospital. This approach proactively prevents shortages or overstocking of specific medications and optimizes inventory management according to the characteristics of each hospital. For example, if the inventory of commonly used medications in a particular department falls below a set threshold, the automated ordering function is activated, and an order request is automatically sent to the supplier. At this point, the ordering strategy is optimized by comprehensively considering each hospital's budget, past ordering history, and the supplier's delivery schedule.

[0053] Furthermore, this invention also has the function of predicting seasonal changes and specific needs of various departments, and flexibly adjusting the ordering cycle and ordering quantity accordingly. For example, by predicting the specific demand growth that occurs at a certain time each year, strategies can be adjusted to obtain the medicines needed at that time in advance. In this way, hospitals can respond quickly to unexpected changes in demand, while achieving efficient use of medicines and cost savings.

[0054] In summary, this invention ensures the stability of the pharmaceutical supply chain and maximizes hospital operational efficiency by customizing inventory management and ordering strategies to reflect the characteristics of each hospital. Furthermore, it provides real-time monitoring capabilities through a user-friendly dashboard, helping hospital administrators immediately identify critical information and make rapid decisions, thereby improving the overall quality of hospital inventory and order management.

[0055] Invention Effects

[0056] It can analyze the required drugs by target hospital, usage pattern and trend, and can be designed to monitor key indicators in real time through a user-friendly dashboard.

[0057] Each system can be customized with automatic sorting and inventory management strategies based on the target hospital's operating methods, specific departmental requirements, and seasonal changes.

[0058] The effects that can be obtained in the various embodiments are not limited to those described above, and based on the following detailed description, those skilled in the art will clearly understand other effects not mentioned. Attached Figure Description

[0059] As with the other aspects described above in certain ideal embodiments of the invention, the features and benefits will become more apparent from the following description, which is processed together with the accompanying drawings.

[0060] Figure 1 This is a block diagram of the system according to an example of the present invention.

[0061] Figure 2 This is an example system of the present invention, which schematically represents a pattern of the process of predicting patient information and required drug information.

[0062] Figure 3 This is a schematic representation of the processing pattern in the contract management module of the system according to an embodiment of the present invention.

[0063] Figure 4 This is a sequence diagram of a method using the system, according to an example of the present invention.

[0064] Figures 5 to 9 An example of the UI of a system actually implemented according to an example of the present invention is shown.

[0065] Figure 10 This is a drawing showing the composition of an electronic device according to an embodiment of the present invention.

[0066] Figure 11 It is a drawing illustrating a procedure according to an example of the present invention.

[0067] It should be noted that, through the above graphics, similar reference numbers are used to display the same or similar elements, features, and structures. Detailed Implementation

[0068] Because the present invention can be modified and has various embodiments, specific embodiments will be shown in the accompanying drawings and described in detail in the detailed description. However, this is not intended to limit the invention to the specific embodiments, and it should be understood to include all modifications, equivalents, and substitutions contained within the spirit and scope of the invention. In describing each drawing, similar reference numerals are used for similar parts.

[0069] Terms such as first, second, A, and B may be used to describe various components, but components should not be limited by these terms. Terms are used only to distinguish one component from another. For example, a first component may be named a second component, and similarly, a second component may be named a first component, without departing from the scope of the invention. The term "and / or" includes any combination of one or more related statements.

[0070] When a component is referred to as "connected" or "connected" to another component, it can be understood as being able to connect directly to or be linked to another component, but other components should also exist in between. On the other hand, when it is said that a component is "directly connected" or "directly linked" to another component, it should be understood that no other components exist in between.

[0071] The terminology used in this application is for describing specific embodiments only and is not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as “comprising” or “having” are intended to indicate the presence of features, numbers, steps, operations, components, portions, or combinations thereof described in the specification, but are not intended to indicate the presence of: it should be understood that this does not preclude the possibility of the presence or addition of elements, numbers, steps, operations, components, portions, or combinations thereof.

[0072] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and should not be interpreted in an ideal or overly formal sense, unless expressly defined in this application No. 2007 / 2007.

[0073] The system according to an embodiment of the present invention will now be described with reference to the accompanying drawings. The present invention is a system for managing the drug inventory of a target hospital based on the prediction of required drugs, and provides a solution capable of analyzing each target hospital and providing analysis of usage patterns and trends.

[0074] Modern healthcare systems face the challenge of improving the efficiency and accuracy of drug management and supply. In particular, hospitals must accurately predict patient demand and stock the right medications at the right time. However, existing methods are limited in areas such as patient forecasting, drug inventory management, and effective communication with suppliers.

[0075] This invention proposes a system and method for managing drug inventory in a target hospital based on forecasts of required drugs.

[0076] One embodiment of the present invention develops a dynamic prediction model based on patient information and appointment information, instead of existing static methods, and can accurately predict the future drug demand of each hospital.

[0077] One embodiment of the present invention may provide a dashboard that can appropriately manage drug inventory based on predicted demand, monitor key indicators in real time, and take action when necessary.

[0078] One embodiment of the present invention can customize and adjust automatic ordering and inventory management strategies by taking into account each hospital's operating methods, patient usage patterns, seasonal changes, etc.

[0079] One embodiment of the present invention can provide functionality for managing contracts with suppliers to facilitate smooth communication with pharmaceutical suppliers and update pharmaceutical inventory based on order details.

[0080] This approach allows hospitals to improve drug management and supply efficiency, provide patients with the necessary medications, and enhance the quality of medical services.

[0081] The present invention includes a target hospital account 110, an inventory management module 120, a patient information calculation module 130, a drug calculation module 140, a contract management module 150, and a supplier account module 160.

[0082] Target hospital account 110 refers to the account assigned to each target hospital. Each target hospital (including its administrator) is set to log in using a default method.

[0083] In order to use the solution applied in this invention, a fee can be charged to the target hospital, or it can be designed as a subscription payment, in which a certain fee is paid periodically.

[0084] The inventory management module 120 automatically generates the target hospital's current drug inventory information. To this end, all drugs delivered to the target hospital are assigned a unique identification code and configured to be identified and managed according to a preset method. Drugs can be divided into primary drugs and secondary drugs. Primary drugs are consumed through direct administration to patients, while secondary drugs are considered consumed when their preset expiration date arrives.

[0085] As an example, it can be configured to clearly distinguish between first-line and second-line medications for users. Here, first-line medications primarily refer to essential drugs continuously used in routine hospital care and treatment, such as antipyretics, analgesics, antibiotics, anti-inflammatory drugs, intravenous infusions, and basic saline solutions. Second-line medications refer to adjunctive drugs used for irregular needs or under specific conditions. For example, antiviral drugs (such as oseltamivir) are mainly used during influenza epidemics and rarely used at other times. Furthermore, anticancer drugs and immunosuppressants are used only to treat specific diseases, with limited demand in general practice. Additionally, special anesthetics required for rare disease treatment or surgery may be examples of adjunctive medications. This classification aims to improve the efficiency of drug management, optimizing ordering and inventory strategies for each drug group based on hospital operations and patient needs.

[0086] The system according to embodiments of the present invention can classify and provide first and second drugs to users. If the proportion of second drugs is high, the amount of discarded drugs can be reduced by decreasing the storage cycle and the amount of storage loss.

[0087] For example, the system can automatically detect when the proportion of secondary drugs is high and allow appropriate action to be taken. For instance, if the proportion of secondary drugs in the drugs received by the target hospital exceeds a first preset threshold, the system i) stops the automatic ordering of secondary drugs and / or ii) approves the addition of secondary drug purchase process control lines.

[0088] The first threshold can be preset to 30%, 50%, etc.

[0089] If the proportion of a second-tier drug exceeds the first threshold, the inventory management module can temporarily suspend the automatic ordering process for that drug. This suspension is to maintain a balanced inventory of primary drugs and prevent unnecessary inventory accumulation of less frequently used secondary drugs. For secondary drugs whose automatic ordering is suspended, the ordering process will be resumed if necessary after a manual check of the inventory status, and the hospital administrator will be notified through the system.

[0090] If the proportion of a second drug exceeds the first threshold, the system can set up an additional payment procedure during the purchase of that drug. At this point, an additional approval line, comprised of hospital department heads, financial managers, or hospital operations managers, re-examines the necessity and budgetary appropriateness of the drug purchase. This minimizes potential inefficiencies in drug ordering and ensures transparency in budget management. Furthermore, the final approval process for the second drug purchase considers current inventory levels, expiration dates, substitution possibilities, and budget balance. Adding this payment line allows for more careful handling of purchases of drugs with low usage frequency or limited seasonal demand, thereby maximizing hospital operational efficiency. If the proportion of a second drug exceeding the first threshold persists for an extended period (e.g., a pre-set second threshold), the system issues a warning to administrators and, if necessary, manages the hospital's overall inventory, providing procurement strategies and recommendations for re-review. This enables hospitals to achieve cost-effective inventory management while maintaining a stable drug supply.

[0091] If the proportion of secondary drugs exceeds the first threshold and continues to increase, the system can adjust its inventory management strategy based on usage patterns and consumption analysis to reduce drug lead times and inventory levels. This prevents over-ordering of secondary drugs with unpredictable demand and minimizes disposal losses due to expiration dates.

[0092] For example, if the system identifies a trend of low usage frequency (below a preset third threshold) or use only during specific time periods for a minor medication, it can extend the ordering period for that medication or maintain only that minor medication's ordering period. Notifications can be controlled to maintain a minimum safety stock. Furthermore, if a medication is about to expire, the system can provide an urgent notification to hospital administrators, recommending rapid use or redeployment to other departments.

[0093] Meanwhile, to improve the efficiency of drug management, the system of this invention determines whether the unit price of a drug exceeds a preset standard (e.g., a preset fourth threshold) and / or is within its expiration date. Predefined critical periods (e.g., various operations can be automatically determined by considering whether the level remains below a fifth threshold) are also defined.

[0094] Specifically, the system can determine the appropriate operation based on the following conditions:

[0095] i) Whether to omit emergency notification: If the unit price of the drug involved is low, the impact of the loss is minor, or the remaining shelf life exceeds the critical period (such as the 5th threshold) and there is sufficient time, emergency notification can be omitted. This can reduce the confusion caused by unnecessary notifications and help managers focus on more important matters.

[0096] ii) Whether to recommend (and / or suggest) reassignment to other departments: If a drug that is unused or in low demand in a particular department is identified as nearing its expiration date, it can be recommended to reassign it to another department where the drug can be used. During this process, the medication history and demand of each department are analyzed, and a suitable department is selected. When a relocation occurs, the system automatically updates the inventory information of that department.

[0097] iii) Whether to propose (and / or suggest) relocation to an external hospital (another hospital): If it is determined that the relevant medications in all departments of the target hospital are difficult to use up, and considering the number of doctors and averages in certain situations, the system may suggest relocation to an external hospital or a nearby hospital. In this case, the system will automatically select the relocation hospital (hereinafter referred to as "relocation hospital") based on the medication demand and its cooperative relationship with nearby hospitals. If the relocation plan is approved, the system will update the relevant necessary information, understand the order status, and expedite the relocation process.

[0098] However, the system of the present invention takes into account that if the price of the medicine is too low, relocation to an external hospital (another hospital) may actually incur additional costs, thus relocation recommendations with prices below a predetermined standard (e.g., Section 6) are configured to be omitted. For example, if the transportation cost of the medicine or the administrative cost of relocation exceeds or approaches the unit price of the medicine, the relocation recommendation will be automatically ignored to prevent cost inefficiency due to relocation.

[0099] However, if the drug is classified as having high medical importance, the system may recommend relocation to an external hospital by prioritizing its importance to medical practice over cost. For example, if the drug is life-sustaining or used for emergency treatment, rapid redeployment to a nearby hospital may be recommended, even if it is less expensive. In this case, the system will comprehensively consider the drug's medical use, urgency, and the likelihood of substitutability as criteria for evaluating its importance in deciding whether to relocate.

[0100] Furthermore, if a relocation to an external hospital is planned, the system will analyze the inventory needs and usage patterns of the relevant hospital (i.e., the relocation hospital) in real time, recommend suitable relocation hospitals, and if the relocation is approved, display inventory data and automatically provide order status. This process maximizes the efficiency of drug transfer between hospitals, avoids unnecessary disposal, and improves the utilization rate of medical resources.

[0101] The system is designed to optimize drug inventory management by maintaining a balance between cost-effectiveness and medical importance, and to support minimizing cost waste while maintaining the quality of healthcare services, even when drug reallocation is required.

[0102] These measures, as preventative steps, aim to maximize the effective utilization of medicines and minimize losses from expired disposal. Furthermore, the system monitors the relocation process and the execution of emergency notifications in real time, allowing hospital administrators to clearly understand all actions related to inventory status and improving the efficiency of internal and external medicine management and supply chain operations.

[0103] Thus, the inventory adjustment function based on the classification and proportion of primary and secondary drugs plays a crucial role in optimizing hospital inventory operations, preventing unnecessary inventory accumulation, and helping to reduce drug disposal costs. Furthermore, the system provides users with real-time inventory status and usage rates for each drug through an intuitive dashboard, enabling managers to make effective decisions for each drug group. In this way, hospitals can minimize financial losses while maintaining patient continuity of care and ensuring a stable drug supply chain.

[0104] The patient information calculation module 130 generates expected patient information for a target time period (preset time) based on the current and reserved patient information of the target hospital. In other words, it can generate expected patient information for a target time period. As an example, if the target time period is one week (7 days), expected patient information can be generated weekly using the current and reserved patient information. Expected patient information for week n+1 can be generated using the patient information from week n and the reserved patient information from week n. The expected patient information is generated through a first machine learning model 131, which will be described later.

[0105] Simultaneously, the patient information calculation module 130 can generate expected patient information based on the inventory confirmation date used to generate expected patient information, by taking into account current outpatients and inpatients, as well as scheduled outpatients and inpatients. Outpatients and inpatients may be taking different medications, so it is best to treat them separately. In addition, depending on the severity of the patient's symptoms, the necessary medications may be used differently, and the severity of the patient's symptoms as assessed by the doctor may also be subdivided and applied individually.

[0106] At this point, the patient information calculation module 130 needs to pre-store the types and dosages of necessary drugs corresponding to each disease. The types and quantities of required drugs are pre-stored in a database and can be designed to automatically calculate based on various parameter values.

[0107] Furthermore, the correction coefficients for each outpatient and inpatient patient can be set and adjusted according to the characteristics of the target hospital's departments. This is because the ratio of outpatients to inpatients may vary depending on the characteristics of the treatment departments in the target hospital.

[0108] The drug calculation module 140 calculates the drug information required for a target time period by considering information including drug inventory information and expected patient information. The drug calculation module 140 includes a second machine learning model 141. The second machine learning model 141 uses the expected patient information to output and provide the necessary drug information.

[0109] Thus, the present invention includes a first machine learning model 131 for calculating expected patient information and a separate second machine learning model 141 for calculating necessary drug information. This allows for more accurate predictions by using separate machine learning models instead of a single model. Since the detailed items constituting the current and retained patient information are not directly used as parameters for calculating necessary drug information, more accurate machine learning can be performed. Furthermore, the accuracy of demand value prediction can be further improved by utilizing an ARIMA (Autoregressive Integrated Moving Average) model.

[0110] The supplier account module 160 is granted to pre-registered drug suppliers, who can log in according to a preset method and automatically receive the required drug information from the drug calculation module 140. The supplier account module 160 can also be granted to drug suppliers who have previously subscribed to the solution applying this invention. In other words, a qualified drug supplier can obtain the drug information required by the target hospital. The system of this embodiment can provide a platform service as a link between multiple target hospitals and multiple drug suppliers.

[0111] The contract management module 150 manages drug purchase contracts between drug suppliers and target hospitals and updates drug inventory information based on the content of the purchase contracts. Drug inventory information is not updated automatically; instead, a double check is performed using a matching operation with the drug inventory information calculated in the inventory management module 120.

[0112] The contract management module 150 includes a supplier selection unit 151, the handling of which will be described later when multiple suppliers are present.

[0113] This invention can combine advanced technologies and system methodologies to propose the following methods to solve the above-mentioned complex problems.

[0114] One embodiment of the present invention proposes introducing a machine learning model for accurate drug demand prediction. Additionally, an embodiment of the present invention proposes a machine learning model that generates expected patient information using patient information and appointment information. This model learns from past data and receives current patient information and appointment information as input for predicting future drug demand. This allows for accurate predictions based on the characteristics of each hospital.

[0115] One embodiment of the present invention provides a user-friendly dashboard. Additionally, embodiments of the present invention provide a user-friendly dashboard that can monitor key metrics in real time. In this way, hospital officials can quickly check drug inventory levels and take necessary actions.

[0116] One embodiment of the present invention proposes an automated ordering and inventory management algorithm. Furthermore, embodiments of the present invention propose an algorithm that customizes automated ordering and inventory management strategies by taking into account each hospital's operational methods, specific departmental requirements, seasonal variations, etc. This enables hospitals to reduce costs while maintaining optimal inventory levels.

[0117] This invention facilitates the establishment of a supplier management platform. Furthermore, this invention also establishes a supplier management platform for smooth communication with pharmaceutical suppliers. This platform manages contracts with pharmaceutical suppliers and shares order details to ensure rapid drug supply.

[0118] These measures will improve the efficiency of our system in managing and supplying medicines, enabling hospitals to provide better care for patients.

[0119] The invention has been described above with reference to preferred embodiments; however, those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention as set forth in the appended claims. I understand you can do it.

Claims

1. A system for managing drug inventory in a target hospital based on forecasts of drug demand, wherein, The system includes: The inventory management module automatically generates current drug inventory information for the target hospital; The patient information calculation module generates expected patient information for the target time period (preset time) based on the current patient information and appointment patient information of the target hospital. The drug calculation module takes into account information including drug inventory information and expected patient information, and calculates the necessary drug information within the target time period. The supplier account module is provided for pre-registered drug suppliers. Suppliers can log in using preset methods and use the account to automatically receive necessary drug information from the drug calculation module. The contract management module manages drug purchase contracts between drug suppliers and target hospitals, and updates drug inventory information based on the content of the purchase contracts.

2. The system according to claim 1, wherein, The patient information calculation module includes the first machine learning model for predicting expected patient information. In the first machine learning model, Machine learning is performed using information including patient data from past preset time periods as learning information, and current and retained patient data as input information, and the desired patient data is learned and generated as output information. The drug calculation module includes a second machine learning model that is learned to predict necessary drug information using the expected patient information output by the first machine learning model.

3. The system according to claim 2, wherein, The patient information calculation module generates expected patient information for the target time period based on a preset cycle. Based on the inventory confirmation date for generating the aforementioned expected patient information. The system generates expected patient information by considering current outpatient and inpatient numbers and scheduled outpatient and inpatient appointments, and adjusts the correction coefficients for each outpatient and inpatient patient based on the characteristics of the target hospital departments. The patient information calculation module pre-stores the types and dosages of necessary medications for each disease. The second machine learning model in the drug calculation module considers the diseases of outpatients and inpatients, and generates the necessary drug information as output information.

4. The system according to claim 3, wherein, The contract management module also includes applying for a supply contract based on the required drug information provided in a preset manner. When there are multiple drug suppliers, the supplier selection unit selects one drug supplier according to a preset method. The supplier screening department independently enters the supply application information for each pharmaceutical supplier through the supplier account module, including the required supply price and quantity through the system.

5. The system according to claim 4, wherein, The supplier screening department, based on the lowest expected supply price, changed multiple pharmaceutical suppliers from a contract waiting status to a contract signing status. The system is configured to accumulate the required quantity included in the supply application information of each of multiple drug suppliers, within the limits of the quantity or capacity included in the required drug information.

6. The system according to claim 5, wherein, The inventory management module further considers the individual expiration date information of each drug to generate current drug inventory information. When the target time period changes from k-th to k+1-th, drugs whose expiration dates have expired are classified as emergency treatment drugs.

Images