Hospital pharmacy inventory intelligent management system and method

Abstract

The invention discloses a hospital pharmacy inventory intelligent management system and method, and belongs to the technical field of data management. Medicine basic information and multiple batches of inventory records are acquired; an application risk grade R is extracted based on the medicine application label; combining the remaining validity period, the stability parameter and the risk factor to construct a batch ex-warehouse priority scoring model P; generating a sorting ex-warehouse list and executing allocation and ex-warehouse operation; when it is monitored that the medicine is in time or the inventory is abnormal, dynamically adjusting the score and updating the sorting result; according to the invention, fine control of the drug delivery sequence can be realized, clinical availability of high-risk-purpose drugs is guaranteed preferentially, and inventory use efficiency and medication safety are improved at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to an intelligent inventory management system and method for a hospital pharmacy. Background Art

[0002] Distributed fault detection for distribution control equipment involves deploying intelligent modules with detection capabilities across multiple device nodes within the distribution system to achieve collaborative sensing, real-time analysis, and localized fault detection. Compared to centralized detection, distributed detection improves the system's fault response speed and accuracy, reduces information transmission latency, and enhances the system's self-healing capabilities in the face of localized faults or communication anomalies. This approach is particularly well-suited for the complex, multi-source, multi-node distribution network structures found in modern smart grids.

[0003] The existing technology has the following shortcomings:

[0004] In distribution network fault detection, high-risk faults like high-resistance ground faults occur very infrequently, leading to a severe shortage of training data samples. As a result, existing fault detection models, such as decision tree models, tend to favor common fault types when constructing classification paths, neglecting to model rare but potentially more damaging events, creating a fault blind spot. This data imbalance prevents the model from identifying critical anomalies, creating the risk of underreporting and directly threatening the safety of grid operations and system reliability. Summary of the Invention

[0005] The purpose of the present invention is to provide a hospital pharmacy inventory intelligent management system and method to solve the shortcomings of the background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solutions: a hospital pharmacy inventory intelligent management method, comprising:

[0007] Obtain basic information of the target drug, including drug name, dosage form, strength, batch number, expiration date, and corresponding use label information;

[0008] Obtain multiple batch inventory records of drugs in the hospital pharmacy inventory and construct a batch inventory information set I = {I1, I2, ..., I i ,...,I n}, where I i Including batch number, inventory quantity, remaining validity period and storage time; n represents the number of batches of the target drug in inventory;

[0009] Extract the corresponding use risk level R based on the drug's corresponding use label information. The risk level is used to indicate the urgency and safety sensitivity of the drug in actual clinical use.

[0010] Based on the remaining validity period and the use risk level R, a drug batch release priority scoring model P is constructed. The model comprehensively considers the batch validity period, stability weight and use risk factor, and outputs the release priority score P for each batch. i ;

[0011] The priority score P i Applied to the inventory outbound sorting rules, generate the sorted outbound list of target drugs L = {L1, L2, ..., L i ,...,L n}, where L i It is the priority batch for shipment;

[0012] According to the sorted outbound list L, complete the inventory transfer and outbound operations of the corresponding drugs, and update the inventory information set I in real time;

[0013] If a batch of drugs is detected to be nearing expiration or with abnormal inventory, its outbound priority score P will be dynamically adjusted based on the use risk level R. i , update the sorted outbound list L.

[0014] Preferably, the step of extracting the use risk level R includes: establishing a corresponding mapping relationship table between the drug use label and the risk level, wherein the mapping relationship table sets the level classification according to the drug's use time limit requirements, safety tolerance and alternative availability factors in typical clinical scenarios of emergency, surgery and chronic disease treatment; and searching the corresponding risk level R from the mapping relationship table according to the drug's use label.

[0015] Preferably, the construction of the drug batch release priority scoring model P includes: obtaining the remaining expiration date of each batch of drugs, calculating the expiration date score S1 of the batch, and the expiration date score is inversely proportional to the remaining expiration date; obtaining the drug use risk level R, and setting a use risk factor F for different risk levels, and the risk factor F is used to adjust the weight of the release priority score to the expiration date score.

[0016] Preferably, the entry time and drug type of each batch of drugs are obtained, and the stability coefficient W is extracted in combination with the drug stability reference table. The stability coefficient reflects the performance fluctuation risk of the drug during different storage time periods in the inventory, and is used to correct the expiration date score S1 to generate a stability-corrected score S2.

[0017] Preferably, the stability correction score S2 and the usage risk factor F are weighted and synthesized according to the set weights to calculate the final delivery priority score P. The score calculation method is: P = A×S2+B×F, where A and B are preset weighting coefficients, satisfying A+B=1.

[0018] Preferably, completing the inventory transfer and delivery operations according to the sorted delivery list L includes:

[0019] According to the sorted outbound list L, the drug batch information in the current inventory is preferentially selected according to the batch order in the list, the inventory quantity of the batch is obtained, and compared with the quantity to be shipped; if the current batch inventory meets the requirements, the outbound operation is initiated; if it is insufficient, the subsequent batches are selected in sequence according to the list order to supplement until the demand is met.

[0020] Preferably, the inventory information update includes:

[0021] According to the outbound results, the remaining quantity field of the corresponding batch in the inventory information set I is updated in real time; if the inventory of a batch is cleared, the status of the batch is set to shipped out and it is removed from the sorted outbound list L; if there is still a remainder, the position of the batch in L is retained and the latest inventory value is recorded.

[0022] Preferably, the number is based on the use risk level R, and the outbound priority score P is dynamically adjusted. i , updating the sorted outbound list L, including: scanning the inventory information set I, detecting whether the remaining validity period of each drug batch is less than the preset expiration threshold, or whether the batch has abnormal inventory quantity, frozen status and quality complaint mark. If so, mark the batch as a risk batch;

[0023] After identifying the risky batch, the use risk factor F is found according to the drug use risk level R corresponding to the batch, and the stability correction score S2 in the original scoring model is re-weighted according to the weight ratio, so that drugs with high use risk levels are given priority to retain a higher outbound weight when an abnormality occurs, and drugs with low use risk levels have their outbound priority increased to speed up inventory clearance.

[0024] Preferably, an abnormal batch adjustment factor E is introduced, and E is assigned different values ​​according to the abnormality type, including a positive value for expiring E and a negative value for frozen E. It is adjusted with the original score P to obtain a dynamic score P'. The adjustment logic is: P' = original score P plus adjustment factor E multiplied by use risk factor F, and then multiplied by the correction coefficient α set by the system, and finally limited to the maximum score range; the adjusted score P' is re-substituted into the outbound sorting list update algorithm to generate a new sorted outbound list L', and the difference between the lists before and after the update is compared; if the batch sorting order is changed, the update prompt will be automatically pushed to the pharmacist end or the outbound task scheduling system to ensure that subsequent drug delivery tasks are executed according to the latest sorting.

[0025] The present invention also provides a hospital pharmacy inventory intelligent management system, comprising:

[0026] The information acquisition module is used to obtain basic information about the target drug, including the drug name, dosage form, strength, batch number, expiration date, and corresponding use label information; and obtain multiple batch inventory records of the drug in the hospital pharmacy inventory to construct a batch inventory information set, which includes the batch number, inventory quantity, remaining expiration date, and storage time of each batch;

[0027] A risk level identification module is used to extract the corresponding use risk level R based on the use label information of the drug, and the use risk level R is used to indicate the urgency and safety sensitivity of the drug in actual clinical use;

[0028] A priority scoring module is used to construct a delivery priority scoring model P for each batch of the drug based on the remaining shelf life, stability parameters and the use risk level R of the batch, and output the priority score value corresponding to each batch;

[0029] A sorting list generation module is used to apply the priority score P to the outbound sorting rules to generate a sorted outbound list L of the target drugs to determine the outbound priority order, wherein the first batch in the list is the current priority outbound batch;

[0030] The outbound transfer execution module is used to complete the inventory transfer and outbound operations of drugs according to the sorted outbound list L, and update the batch inventory data in real time, including updating the inventory quantity, batch status and inventory information set I;

[0031] The dynamic monitoring and adjustment module is used to periodically monitor the inventory status of drugs. When a batch is identified as nearing expiration, inventory abnormality, frozen, or with quality issues, a dynamic score adjustment is performed based on the use risk level R, and an updated priority score P' is calculated. The adjustment result is used to regenerate the sorted outbound list L';

[0032] The task synchronization and feedback module is used to push the changes to the pharmacist side or the outbound task scheduling system after the sorting list is updated.

[0033] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0034] 1. This invention incorporates the use risk level R and stability weight into the release priority scoring, breaking through the traditional logic of drug release based solely on expiration date. It implements a multi-dimensional release sorting mechanism based on "clinical risk guidance + quality stability assessment." This solution prioritizes drug availability in high-risk scenarios such as emergency and surgery, while accelerating the removal of near-expiry drugs for low-risk uses, reducing drug waste and improving the accuracy and safety of release strategies.

[0035] 2. This invention also uses a dynamic scoring adjustment mechanism to achieve real-time identification and score updates for near-expiry and abnormal inventory status, building a self-aware and self-regulating inventory allocation model. Combined with real-time updates to outbound delivery lists and task synchronization, this effectively avoids medication risks such as misdelivery and missed shipments, improving the automation level of the pharmacy management system and clinical response efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0037] Figure 1 Flow chart of the method of the present invention.

[0038] Figure 2 It is a flow chart of the system modules of the present invention. DETAILED DESCRIPTION

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0040] Example 1, please refer to Figure 1 As shown, the hospital pharmacy inventory intelligent management method described in this embodiment includes:

[0041] Obtain basic information of the target drug, including drug name, dosage form, strength, batch number, expiration date, and corresponding use label information;

[0042] Obtain multiple batch inventory records of drugs in the hospital pharmacy inventory and construct a batch inventory information set I = {I1, I2, ..., I i ,...,I n}, where I i Including batch number, inventory quantity, remaining validity period and storage time; n represents the number of batches of the target drug in inventory;

[0043] Extract the corresponding use risk level R based on the drug's corresponding use label information. The risk level is used to indicate the urgency and safety sensitivity of the drug in actual clinical use.

[0044] Based on the remaining validity period and the use risk level R, a drug batch release priority scoring model P is constructed. The model comprehensively considers the batch validity period, stability weight and use risk factor, and outputs the release priority score P for each batch. i ;

[0045] The priority score P i Applied to the inventory outbound sorting rules, generate the sorted outbound list of target drugs L = {L1, L2, ..., L i ,...,L n}, where L i It is the priority batch for shipment;

[0046] According to the sorted outbound list L, complete the inventory transfer and outbound operations of the corresponding drugs, and update the inventory information set I in real time;

[0047] If a batch of drugs is detected to be nearing expiration or with abnormal inventory, its outbound priority score P will be dynamically adjusted based on the use risk level R. i , update the sorted outbound list L.

[0048] The system first obtains the following structured basic information of the target drug from the hospital information system (HIS), pharmaceutical management system (PIS) or hospital ERP system:

[0049] Drug name: includes generic name and trade name, used to uniquely identify the drug;

[0050] Dosage form: such as injection, tablet, liquid, lyophilized powder, etc.;

[0051] Specifications: such as "0.5g / vial", "10mg / tablet", etc., to accurately represent the drug dosage;

[0052] Batch number: used to trace drug production information;

[0053] Expiration date (expiration date): used to determine the remaining available time of the drug.

[0054] This information is usually included as standard fields in each inventory unit (Lot) record in the hospital drug database.

[0055] While obtaining the above basic information, the system also needs to bind the corresponding usage label information for each drug. The usage label is used to reflect the usage scenario of the drug in the actual medical process. The label can come from the following sources:

[0056] Physician prescription record statistics: The system analyzes the disease diagnosis (ICD code) corresponding to the drug in the electronic prescription (eRx), the frequency of use by department, etc., and extracts common usage scenarios;

[0057] Predefined categories by the hospital pharmacy management department: The hospital or drug regulatory department can standardize the use labels of specific drugs, such as "emergency drugs", "high-alert drugs", "adjuvant treatment drugs", etc.

[0058] Extracting structured information from drug instructions: Terms such as "suitable for severe infection" and "for intraoperative sedation" can be mapped to labels;

[0059] Manual labeling / knowledge graph support: Pharmacists can manually supplement or revise the usage labels of certain key drugs based on their clinical experience.

[0060] Examples of common usage labels include:

[0061] "Emergency rescue", "surgical anesthesia", "hospital maintenance medication", "long-term medication for chronic diseases", "pediatric / elderly medication", "high-alert medication", "cold chain drugs" and "anti-tumor drugs".

[0062] The system calls the database interface in the hospital pharmacy inventory management system (such as PIS, HIS, ERP system) to query all the batch information of the target drugs currently in stock at various inventory points (such as the main drug warehouse, branch pharmacies, and emergency pharmacies) for the target drugs identified in S100, and filters out invalid data that has been shipped out or is about to be scrapped.

[0063] The system takes "drug batch" as the smallest management unit, extracts the inventory information of each batch in a structured manner, and constructs a batch information set: I={I1,I2,...,I i ,...,I n}, where I i Including batch number, inventory quantity, remaining validity period and storage time; n represents the number of batches of the target drug in inventory;

[0064] Drug batch, which contains the following key fields:

[0065] Batch ID: A unique batch identification code assigned by a pharmaceutical manufacturer, often used to trace production sources and manage quality issues.

[0066] Inventory Quantity: The number of drugs currently available for distribution in the warehouse (units such as bottles, boxes, tubes, etc.);

[0067] Expiration Date: The legal expiration date for the use of this batch of drugs;

[0068] Remaining Shelf Life (RSL): The number of days between the current system time and the expiration date, used for sorting and comparison;

[0069] Stock-in Date: The time when the batch of drugs was officially registered and put into storage. It is used together with the expiration date information to analyze storage stability and inventory turnover cycle.

[0070] The system will perform field integrity checks on the acquired batch data (for example, the validity period cannot be NULL); and perform logical checks on the "warehousing time is earlier than the validity period"; if a data conflict is detected (such as duplicate batch numbers or negative inventory), the system will mark the exception and suspend the batch for model calculation.

[0071] The mapping relationship between use labels and risk levels is established as follows: First, to automatically extract the risk level of drug use, the system needs to build a standardized mapping relationship table between use labels and risk levels. This mapping table is based on the typical use scenarios of drugs in actual clinical practice and comprehensively considers the following dimensions:

[0072] The urgency of the medication scenario (e.g., emergency rescue, surgical anesthesia, etc.);

[0073] Possible adverse consequences of medication errors or delays;

[0074] Availability of alternative medicines and difficulty in switching;

[0075] The sensitivity of the patient population (e.g., children, ICU patients, cancer patients, etc.);

[0076] Whether it is a cold chain drug, a high-alert drug or a controlled drug.

[0077] In this example, the drug use risk level R is divided into four levels: R1 (high risk), R2 (medium-high risk), R3 (medium risk), and R4 (low risk). R1 ​​indicates that the drug could be life-threatening if its dispensing is delayed, such as emergency medications like epinephrine, lidocaine, and norepinephrine; R2 indicates that the drug is used in non-interruptible situations such as preoperative anesthesia or cancer treatment; R3 is for general inpatient medications, such as antibiotics and antihypertensive drugs; and R4 is for auxiliary medications with strong substitutability and a flexible use period, such as vitamins and cough suppressants.

[0078] In this system, when drugs are entered into the inventory system, they are associated with one or more corresponding usage tags, such as "emergency medication," "adjuvant therapy," "cancer chemotherapy drug," "intraoperative analgesia," and "pediatric use only." Usage tags can be derived from the standard drug usage classifications of the hospital's pharmacy management committee, or automatically recommended by data mining the prescription usages in the hospital's prescription history over the past three years.

[0079] The system quickly extracts the initial risk level R of the drug by searching the correspondence table between the usage label and the risk level, and records it in the risk parameter field of the drug for subsequent scoring model calls.

[0080] In order to improve the system's adaptability to changes in actual clinical use, this embodiment further provides a data-driven dynamic risk level correction mechanism for correcting the determination of the risk level R based on the original label mapping and combining historical behavior data.

[0081] Specifically, the system counts the frequency of prescriptions for each drug in high-risk departments (such as the emergency department, ICU, and operating room) over a set time period (e.g., 30 or 90 days) and analyzes the average doctor's order response time (i.e., the average time difference between the doctor's order and the pharmacy's completion of dispensing). Based on these two core indicators, the system introduces the usage urgency index (U) to quantify the urgency of a drug's use.

[0082] The calculation method of the usage urgency index U is as follows:

[0083] The first step is to calculate the proportion of prescriptions of the drug in high-risk departments among all prescriptions, recorded as F1;

[0084] The second step is to calculate the average response time T1 of all prescriptions of the drug;

[0085] The third step is to normalize F1 and T1 (using the maximum and minimum values ​​of the drugs of this type in the hospital for linear normalization) to obtain two standardized scores U1 and U2;

[0086] The fourth step is to perform a weighted calculation based on the set weights to obtain the comprehensive use urgency index U. This is represented by: U = W1 × U1 + W2 × (1-U2), where W1 and W2 are the corresponding weight coefficients, and W1 + W2 = 1. For example, if a drug is frequently prescribed in the ICU and emergency department (F1 is large) and has a short average response time (T1 is short), then the U value is high, indicating high clinical urgency. Based on the set segmentation rules, the system can dynamically increase its original R3 risk level to R2 or even R1, ensuring that it receives a higher weight in inventory priority sorting.

[0087] Taking into account the specific usage characteristics and policy management requirements of some drugs, the system also provides a manual intervention mechanism. If the risk level automatically determined by the system is inconsistent with the "Critical Drug List" or "High Alert Drug List" issued by the hospital's Pharmacy Management Committee, the system will trigger a manual comparison prompt, allowing authorized pharmacists to confirm or force corrections.

[0088] During the revision process, the pharmacist can choose to:

[0089] Retain the risk level automatically calculated by the system;

[0090] Select mandatory coverage of pharmaceutical standard level;

[0091] After adding remarks, set the "Temporary Level for This Quarter" and submit for approval.

[0092] The intervention information will be recorded in the system log for audit tracing and continuous optimization of risk assessment rules.

[0093] For example, taking a drug called "Dopamine for Injection" as an example, the system identifies its use as "emergency medicine" through the label, and the initial corresponding risk level R is R1. According to the prescription data analysis in the past 30 days, the frequency of prescriptions for this drug in ICU and emergency departments accounts for 82%, with an average response time of 5 minutes. After normalization calculation, the system obtains an emergency index U of 0.92, verifying that its R1 level is valid. Another "Vitamin B6 Injection", although occasionally used for postoperative supplementation, has a low prescription rate in high-risk scenarios (F1 is 10%), a long average response time (T1 is 45 minutes), and a U value of only 0.35. The system finally confirmed its risk level as R4.

[0094] In this method, the system first extracts the remaining expiration date information for each batch of drugs to be processed, that is, the number of days remaining until the expiration date of the drug batch. This value can be calculated by subtracting the drug expiration date from the current system time.

[0095] To properly reflect the principle of "shorter expiration dates prioritize shipments" in the scoring model, the system converts the remaining expiration date into an expiration score, S1, which is inversely proportional to the remaining expiration date. In other words, the shorter the remaining expiration date, the higher the corresponding S1 score. For example, the system can preset a standardized reference period, such as 90 days, normalize the remaining expiration date to this period, and generate S1 in an inverse manner. This way, even if a drug is not yet nearing expiration, as long as its relative expiration date is short, the system can identify it as requiring priority shipment.

[0096] This scoring method effectively solves the drawback of the traditional binary judgment of "only judging whether it is about to expire", giving the system a stronger gradient sorting capability.

[0097] Based on the aforementioned scoring, and to address the issue of prior art failing to fully consider differences in drug safety across clinical uses, the present invention introduces a use risk factor, F. This factor, derived from the drug use label and its corresponding risk level R (e.g., R1 to R4), previously extracted by the system, represents the potential adverse consequences of delayed dispensing during clinical use.

[0098] The risk factor F is set to a specific value according to the level, for example:

[0099] High-risk drugs (such as emergency drugs and intraoperative drugs) correspond to an F value of 1.0;

[0100] The corresponding F value for medium- and high-risk drugs (such as anti-tumor drugs and severe anti-infective drugs) is 0.8;

[0101] The corresponding F value for medium-risk drugs is 0.5;

[0102] Low-risk drugs (such as nutritional supplements) correspond to an F value of 0.3 or lower.

[0103] By introducing this factor, the system will automatically increase the priority of high-risk drugs in the sorting process, reflecting a drug dispensing strategy guided by clinical risks.

[0104] Considering that some drugs may have decreased stability due to changes in chemical properties or fluctuations in storage conditions even though they have not expired after long-term storage, the present invention further introduces a drug stability coefficient W.

[0105] Specifically, the system will search the preset "Drug Stability Reference Table" based on the drug category, dosage form, warehousing time and current time of each batch. This table is maintained by the hospital's pharmacy department or pharmacopoeia guidelines and records the "critical storage period" of different drugs under standard storage conditions, that is, the length of time that vigilance should be increased after exceeding this period.

[0106] The system compares the actual storage time of a drug with its critical storage period and calculates a stability factor, W. For example, if a batch of drugs has been stored for longer than its critical storage period, the W value decreases proportionally. The lower the W value, the higher the stability risk of the drug batch, and the sooner it should be released from storage.

[0107] Based on this, the system multiplies the initially calculated expiration score S1 with the stability coefficient W to generate a stability-corrected score S2. In other words, if a batch of drugs has a reasonable expiration date but a low stability coefficient, its final S2 value will also be high, suggesting that it should be processed in advance to prevent loss of efficacy or medication risks.

[0108] To comprehensively consider the triple factors of shelf life, stability, and usage risk, the present invention has designed a configurable multi-factor weighted scoring model. This model calculates the final release priority score P by weighting the stability-corrected score S2 with the usage risk factor F.

[0109] The synthesis process adopts the following strategy:

[0110] The system sets two weighting coefficients, A and B, whose sum is 1 (i.e., A + B = 1). System administrators can adjust A and B based on the hospital's specific management requirements. For example, if shelf life and stability are prioritized, A can be set to 0.7 and B to 0.3; if clinical use risk is prioritized, A can be set to 0.5 and B to 0.5. Substituting these two scoring factors and performing a weighted addition yields the final priority score, P. This scoring model not only has a clear structure and adjustable logic, but also supports on-demand adjustment of scoring strategies to accommodate the dispensing priorities of different hospitals and drug categories, making it highly practical.

[0111] The scoring model, P, is written into the database as a key ranking parameter for each batch. When drugs are shipped out, the system sorts them from high to low based on their corresponding P values, prioritizing batches with higher scores. This scoring is used not only for routine shipments but also for scenarios such as automated replenishment and near-expiry drug screening.

[0112] Taking "Dopamine for Injection" as an example, the drug's use risk level is R1, with an F value set at 1.0. A batch currently has a remaining shelf life of 30 days, corresponding to an S1 of 0.9. This batch has been stored for 20% of its critical period, and the stability factor W is 0.85, resulting in an S2 of approximately 0.765. Assuming A is 0.6 and B is 0.4, then P = 0.6 × 0.765 + 0.4 × 1.0 ≈ 0.859. This score is relatively high among all batches, and the system prioritizes shipment of this batch.

[0113] After completing the outbound priority score P for each batch of drugs in the previous step (this score comprehensively considers the remaining validity period, stability coefficient and use risk factor), the score is applied to the inventory outbound logic to automatically generate the sorted outbound list L of the drug under the current inventory environment, that is, L={L1,L2,...,L i ,...,L n}, where L i It is the priority batch for shipment.

[0114] This list not only reflects the order of drug delivery between batches, but is also used to guide the system to perform specific outbound operations, such as: picking list generation, automatic drug delivery equipment task scheduling and replenishment batch selection.

[0115] The system calculates the outbound priority score P for each batch and calculates the set of all available batches {B1, B2, ..., B i ,...,B n Sort by score from high to low to generate the corresponding sorted outbound list L. The sorting algorithm supports the following rules:

[0116] Main sorting field: priority score P, with higher scores being ranked first;

[0117] Secondary sorting field (trade-off mechanism): When the scores P of multiple batches are the same or the difference is very small, other parameters can be further referenced for secondary sorting, for example:

[0118] Priority is given to items with earlier arrivals (to prevent long-term backlogs); priority is given to items with smaller remaining quantities (to facilitate clearing batches); and priority is given to items with closer physical storage locations (to facilitate quick delivery). After sorting is completed, the system forms a structured outbound list L.

[0119] This step usually includes the following sub-processes:

[0120] The system calls the inventory interface to query all valid batches of the drug in stock and filters out batches that are sealed, scrapped, frozen, etc. and cannot be shipped out.

[0121] Call the scoring model output results and extract the current priority score P of each batch as the basic data for sorting.

[0122] The primary sorting is performed based on the rating value. If there is a tie in rating, the secondary sorting rule is called for determination.

[0123] The system generates a list L with sequential numbers based on the sorting results and writes it into a cache or task queue for subsequent allocation, picking, or dispensing operations.

[0124] The sorted outbound list L can be directly called by the following modules in the hospital pharmacy system:

[0125] Pharmaceutical Dispensing System (PIS): When a doctor's prescription triggers a medication dispensing request, the system automatically matches the first batch L1 in the list L for delivery. If L1 is insufficient, it switches to L2 in sequence until the total prescription quantity is met.

[0126] Automatic medicine dispensing machine task scheduling: List L is transmitted to the automatic medicine dispensing equipment interface module to implement accurate medicine picking logic, avoiding repeated scanning and invalid inventory calls.

[0127] Expiration warning and replenishment assistance: When there are multiple batches in list L with high P values ​​but short expiration dates, the system can warn of the potential depletion risk of the current drugs and submit replenishment suggestions to the inventory management module.

[0128] Inventory optimization decision support: The system can compare the historical sorting list with the outbound data to analyze the consistency between the drug batch issuance rate and the scoring model, which is used for fine-tuning the subsequent model weights (such as A and B).

[0129] After the system completes the calculation of the priority score P for each batch of target drugs, it will generate an ordered list L of outbound batches according to the score from high to low.

[0130] When a doctor's prescription request or automatic replenishment instruction is received, the system will search the current inventory quantity of each batch in the order of the list L and compare them one by one with the quantity to be shipped. If the first batch L1 in the list is in sufficient stock, the required quantity will be directly allocated from that batch. If the stock of L1 is insufficient, the system will automatically replenish the remaining quantity from the next batch L2 in the list, and so on, until the shipment demand is met.

[0131] This mechanism solves the problem in traditional systems that relies heavily on a single batch and is prone to "failure to ship out due to insufficient inventory of the preferred batch", and supports "combined shipping" across batches, improving the system's drug delivery efficiency and task completion rate.

[0132] Once the system determines the selected outbound batch and the corresponding outbound quantity, it starts the outbound operation. During the outbound process, the system performs the following registration tasks:

[0133] Record the time of shipment;

[0134] Record drug name, dosage form, strength, and batch number;

[0135] Record the actual quantity shipped out;

[0136] Record the purpose of use (e.g., "clinical dispensing," "internal transfer," "scrap disposal," etc.);

[0137] Record the department information used (e.g., "Emergency Department," "Internal Medicine Department," "Operating Room");

[0138] Record the operator or automatic dispensing equipment ID;

[0139] Generate a unique outbound event number.

[0140] This data is structured and stored by the system as outbound records, written into the database, and used in subsequent processes such as hospital pharmacy management, audit tracing, and medical insurance verification. This recording mechanism not only ensures the traceability of drug flows.

[0141] After the outbound operation is completed, the system needs to immediately synchronize and update the inventory information set I. This information set records the current status of all drug batches in the hospital pharmacy and is the core data object of the entire intelligent inventory system.

[0142] Specifically, the system updates the following fields for each batch involved in outbound delivery:

[0143] Inventory quantity field: deduct the current shipment quantity from the original quantity;

[0144] Batch status field: If the quantity after deduction is zero, the batch will be marked as "shipped" and will be set to no longer be selectable;

[0145] Inventory update timestamp field: updated to the current system time;

[0146] Available identification fields: If a drug still has a remaining quantity but is approaching its expiration date or critical inventory line, it can be marked as "near expiration" or "warning" status for subsequent replenishment strategies.

[0147] At the same time, the system also dynamically maintains the sorted outbound list L. Batches that have been shipped out are removed from the list; batches that have been partially shipped out have their remaining quantity information updated and retained in the list so that they can still be selected in the next round of outbound shipments.

[0148] Through this mechanism, the system ensures that the inventory status is highly consistent with the actual logistics, avoiding problems such as "virtual inventory" or "wrong transfer" caused by data delays or inconsistencies.

[0149] After inventory information is updated, the system can also automatically trigger a variety of intelligent feedback operations to improve the initiative and foresight of inventory management. Including but not limited to:

[0150] Inventory warning mechanism: When the total remaining inventory of a certain drug falls below the safety threshold, or the number of available batches falls below the set lower limit, the system will automatically issue an inventory warning notification, prompting the pharmacist to conduct a replenishment assessment or temporary allocation;

[0151] Expiration risk warning mechanism: If a batch has a large remaining quantity after shipment, but the remaining validity period is less than the set expiration threshold (e.g., 30 days), the system will mark the batch as "high priority use" and automatically increase its ranking score P value in subsequent scoring;

[0152] Automatic recalculation mechanism for model scoring: When a batch of inventory is significantly adjusted (e.g., a one-time delivery volume is too large) or key parameters change (e.g., the expiration date is approaching), the system can trigger the re-calling of the scoring model and refresh the ranking result L;

[0153] Report statistics and tracking mechanism: The system records the degree of match between the outbound results and the original score ranking, analyzes the accuracy of the model and the rationality of the drug allocation path, and assists in optimizing the score weight or algorithm strategy.

[0154] For example, a doctor writes a prescription for 20 bottles of antibiotic injection. The system determines from the sorted list L that the highest-rated batch, L1, currently has only 12 bottles remaining. The system automatically ships the 12 bottles from L1 and transfers the remaining 8 bottles from L2. Upon completion, the system updates L1's inventory to 0 and its status to "shipped." It also updates L2's remaining inventory to the original value minus 8.

[0155] After that, the system monitored that the total inventory of the drug had fallen below the preset safety line, and at the same time found that the expiration date of L2 was only 28 days away. L2 was immediately marked as an "expiring batch" and pushed to the pharmacist's workstation, recommending priority use or emergency replenishment.

[0156] The system first sets up a regularly running inventory status monitoring task. This task is responsible for scanning all the drug batches in the current inventory information set I and performing the following conditional judgment for each batch:

[0157] Remaining validity period judgment: If the remaining validity period of a batch is less than the "near-expiry threshold" set by the system (for example, 30 days), the batch will be marked as "near-expiry batch";

[0158] Inventory quantity judgment: If the inventory quantity of a batch suddenly decreases abnormally (such as a decrease of more than 50% in a short period of time), the system will mark it as "inventory abnormality";

[0159] Frozen status identification: If a batch is marked as "frozen" by the system or manually, that is, prohibited from leaving the warehouse, but the system sort list has not been updated, an exception mark will be triggered;

[0160] Identification of quality problem records: If a batch is the subject of multiple complaints or adverse reaction events after clinical use, the system will mark it as a "quality risk batch."

[0161] Batches that meet any of the above conditions will be identified as "risk batches" by the system, and a risk identification field will be added to the inventory information set for subsequent processing flow calls.

[0162] After risk batches are identified, the system calls the drug use risk level R corresponding to each batch, and searches the risk factor F value corresponding to the level from the mapping table of use risk level and risk factor F.

[0163] Examples of system-preset risk factors are as follows:

[0164] High-risk use (e.g., emergency drugs, surgical drugs): F = 1.0;

[0165] Medium- and high-risk uses (e.g., tumor treatment, antibiotics for severe illness): F = 0.8;

[0166] Medium-risk use (such as conventional anti-inflammatory and antihypertensive drugs): F = 0.5;

[0167] Low-risk uses (such as nutritional supplements and vitamins): F = 0.3.

[0168] Based on this risk factor, the system redistributes the weights of the stability correction score S2 calculated in the original scoring model according to the risk sensitivity, and implements the following delivery strategy:

[0169] For drugs with a high use risk level (F high), if they are nearing their expiration date, avoid excessively lowering their scores to prevent delays in drug dispensing due to a drop in ranking;

[0170] For drugs with low use risk level (F low), if they are nearing their expiration date, their delivery priority will be proactively increased and they will be cleared first to avoid waste.

[0171] This strategy reflects the "risk-oriented + scenario adaptation" concept emphasized by this invention, that is, the priority of shipping out not only considers the expiration date and inventory status, but also dynamically adjusts it based on the actual use value of the medicine.

[0172] To achieve more detailed and quantitative priority control of risky batches, the system further introduces an abnormal batch adjustment factor E to reflect the specific impact of different abnormal types on the scoring model. This factor can be set as follows:

[0173] Abnormal near-term: E is a positive value (such as +0.2) to improve the score;

[0174] Inventory freeze: E is a negative value (such as -0.5), which is used to significantly reduce the score;

[0175] Quality issues: E is a negative value (such as -0.3), indicating that there is a potential drug safety problem;

[0176] Other exceptions (such as inventory errors) can be set after manual review.

[0177] The score adjustment logic is as follows:

[0178] The system first obtains the batch original score P;

[0179] Then calculate the new score P′, and its calculation rule is:

[0180] The new score P′ is equal to the original score P plus (adjustment factor E multiplied by risk factor F), multiplied by a system-set correction coefficient α; among them, α is used to control the overall adjustment range (for example, adjustment within the range of 0.8 to 1.2) to avoid excessive changes in the scoring factor that affect the stability of the ranking.

[0181] For example, for a batch, P = 0.7, E = +0.2, F = 0.5, α = 1.0, then:

[0182] E×F=0.2×0.5=0.1; P′=(0.7+0.1)×1.0=0.8;

[0183] That is, the score of this batch increased from 0.7 to 0.8.

[0184] In order to control the score within the range set by the system (such as 0 to 1), the system limits the P′ value to the maximum and minimum boundary values ​​after calculation.

[0185] After the score adjustment is completed, the system re-substitutes the new scores P′ of all affected batches into the outbound sorting model to generate a new sorted outbound list L′. This list re-sorts the batches in the original sorted list L according to the new scores and compares and analyzes them with the original list.

[0186] If the ranking change affects the current or next round of drug dispatch (for example, a batch is promoted from 3rd to 1st), the system will automatically generate a ranking update notification and push it to:

[0187] Pharmacist workbench (for manual review);

[0188] Automatic medicine dispensing system task queue (used for medicine dispensing scheduling and reordering);

[0189] Medical order execution queue (used to rematch the correspondence between drug batches and medical orders).

[0190] This update mechanism ensures that changes in inventory status are consistent with the task execution logic, avoiding problems such as wrong shipments or missed shipments caused by the old sorting list being called.

[0191] In addition, the system retains the change records of the old and new sorting lists, and records the reasons for the adjustments, score changes, trigger time and other information to form a complete "dynamic score traceability chain" for hospital audits, management optimization and model performance analysis.

[0192] Example 2, please refer to Figure 2 As shown, the hospital pharmacy inventory intelligent management system described in this embodiment includes:

[0193] The information acquisition module is used to obtain basic information about the target drug, including the drug name, dosage form, strength, batch number, expiration date, and corresponding use label information; and obtain multiple batch inventory records of the drug in the hospital pharmacy inventory to construct a batch inventory information set, which includes the batch number, inventory quantity, remaining expiration date, and storage time of each batch;

[0194] A risk level identification module is used to extract the corresponding use risk level R based on the use label information of the drug, and the use risk level R is used to indicate the urgency and safety sensitivity of the drug in actual clinical use;

[0195] A priority scoring module is used to construct a delivery priority scoring model P for each batch of the drug based on the remaining shelf life, stability parameters and the use risk level R of the batch, and output the priority score value corresponding to each batch;

[0196] A sorting list generation module is used to apply the priority score P to the outbound sorting rules to generate a sorted outbound list L of the target drugs to determine the outbound priority order, wherein the first batch in the list is the current priority outbound batch;

[0197] The outbound transfer execution module is used to complete the inventory transfer and outbound operations of drugs according to the sorted outbound list L, and update the batch inventory data in real time, including updating the inventory quantity, batch status and inventory information set I;

[0198] The dynamic monitoring and adjustment module is used to periodically monitor the inventory status of drugs. When a batch is identified as nearing expiration, inventory abnormality, frozen, or with quality issues, a dynamic score adjustment is performed based on the use risk level R, and an updated priority score P' is calculated. The adjustment result is used to regenerate the sorted outbound list L';

[0199] The task synchronization and feedback module is used to push the changes to the pharmacist side or the outbound task scheduling system after the sorting list is updated.

[0200] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0201] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0202] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0203] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A hospital pharmacy inventory intelligent management method, characterized by: include: Obtain basic information of the target drug, including drug name, dosage form, strength, batch number, expiration date, and corresponding use label information; Obtain multiple batch inventory records of drugs in the hospital pharmacy inventory and construct a batch inventory information set I = {I1, I2, ..., I i ,...,I n }, where I i Including batch number, inventory quantity, remaining validity period and storage time; n represents the number of batches of the target drug in inventory; Extract the corresponding use risk level R based on the drug's corresponding use label information. The risk level is used to indicate the urgency and safety sensitivity of the drug in actual clinical use. Based on the remaining validity period and the use risk level R, a drug batch release priority scoring model P is constructed. The model comprehensively considers the batch validity period, stability weight and use risk factor, and outputs the release priority score P for each batch. i ; The priority score P i Applied to the inventory outbound sorting rules, generate the sorted outbound list of target drugs L = {L1, L2, ..., L i ,...,L n }, where L i It is the priority batch for shipment; According to the sorted outbound list L, complete the inventory transfer and outbound operations of the corresponding drugs, and update the inventory information set I in real time; If a batch of drugs is detected to be nearing expiration or with abnormal inventory, its outbound priority score P will be dynamically adjusted based on the use risk level R. i , update the sorted outbound list L.

2. The intelligent inventory management method for a hospital pharmacy according to claim 1, characterized in that: The step of extracting the usage risk level R includes: establishing a corresponding mapping relationship table between the drug usage label and the risk level, wherein the mapping relationship table sets the level classification according to the usage time requirements, safety tolerance and alternative availability factors of the drug in typical clinical scenarios of emergency, surgery and chronic disease treatment; and searching the corresponding risk level R from the mapping relationship table according to the usage label of the drug.

3. The intelligent inventory management method for a hospital pharmacy according to claim 1, characterized in that: The method for constructing a drug batch release priority scoring model P includes: obtaining the remaining expiration date of each batch of drugs, calculating an expiration date score S1 for the batch, wherein the expiration date score is inversely proportional to the remaining expiration date; obtaining the drug use risk level R, and setting a use risk factor F for different risk levels, wherein the risk factor F is used to adjust the weight of the release priority score to the expiration date score.

4. The intelligent inventory management method for a hospital pharmacy according to claim 3, characterized in that: The entry date and drug type of each batch of drugs are obtained, and the stability coefficient W is extracted in combination with the drug stability reference table. The stability coefficient reflects the performance fluctuation risk of the drug during different storage time periods in inventory and is used to correct the expiration date score S1 to generate the stability-corrected score S2.

5. The intelligent inventory management method for a hospital pharmacy according to claim 4, characterized in that: The stability correction score S2 and the usage risk factor F are weighted and synthesized according to the set weights to calculate the final delivery priority score P. The score calculation method is: P = A×S2+B×F, where A and B are preset weighting coefficients, satisfying A+B=1.

6. The intelligent inventory management method for a hospital pharmacy according to claim 1, characterized in that: The inventory transfer and delivery operations according to the sorted delivery list L include: According to the sorted outbound list L, the drug batch information in the current inventory is preferentially selected according to the batch order in the list, the inventory quantity of the batch is obtained, and compared with the quantity to be shipped; if the current batch inventory meets the requirements, the outbound operation is initiated; if it is insufficient, the subsequent batches are selected in sequence according to the list order to supplement until the demand is met.

7. The intelligent inventory management method for a hospital pharmacy according to claim 6, characterized in that: The inventory information update includes: According to the outbound results, the remaining quantity field of the corresponding batch in the inventory information set I is updated in real time; if the inventory of a batch is cleared, the status of the batch is set to shipped out and it is removed from the sorted outbound list L; if there is still a remainder, the position of the batch in L is retained and the latest inventory value is recorded.

8. The intelligent inventory management method for a hospital pharmacy according to claim 1, characterized in that: Based on the usage risk level R, the outbound priority score P is dynamically adjusted. i , updating the sorted outbound list L, including: scanning the inventory information set I, detecting whether the remaining validity period of each drug batch is less than the preset expiration threshold, or whether the batch has abnormal inventory quantity, frozen status and quality complaint mark. If so, mark the batch as a risk batch; After identifying the risky batch, the use risk factor F is found according to the drug use risk level R corresponding to the batch, and the stability correction score S2 in the original scoring model is re-weighted according to the weight ratio, so that drugs with high use risk levels are given priority to retain a higher outbound weight when an abnormality occurs, and drugs with low use risk levels have their outbound priority increased to speed up inventory clearance.

9. The intelligent inventory management method for a hospital pharmacy according to claim 8, characterized in that: An abnormal batch adjustment factor E is introduced. E is assigned different values ​​according to the abnormality type, including a positive value for expiring E and a negative value for frozen E. It is adjusted with the original score P to obtain a dynamic score P′. The adjustment logic is: P′ = original score P plus adjustment factor E multiplied by the use risk factor F, and then multiplied by the correction coefficient α set by the system, and finally limited to the maximum score range; the adjusted score P′ is re-substituted into the outbound sorting list update algorithm to generate a new sorted outbound list L′, and the differences between the lists before and after the update are compared; if the batch sorting order is changed, the update prompt will be automatically pushed to the pharmacist end or the outbound task scheduling system to ensure that subsequent drug delivery tasks are executed according to the latest sorting.

10. A hospital pharmacy inventory intelligent management system, used to implement the hospital pharmacy inventory intelligent management method according to any one of claims 1 to 9, characterized in that: include: The information acquisition module is used to obtain the basic information of the target drug, including the drug name, dosage form, strength, batch number, expiration date and its corresponding use label information; and obtaining multiple batch inventory records of the drug in the hospital pharmacy inventory, and constructing a batch inventory information set, wherein the information set includes the batch number, inventory quantity, remaining validity period, and storage time of each batch; A risk level identification module is used to extract the corresponding use risk level R based on the use label information of the drug, and the use risk level R is used to indicate the urgency and safety sensitivity of the drug in actual clinical use; A priority scoring module is used to construct a delivery priority scoring model P for each batch of the drug based on the remaining shelf life, stability parameters and the use risk level R of the batch, and output the priority score value corresponding to each batch; A sorting list generation module is used to apply the priority score P to the outbound sorting rules to generate a sorted outbound list L of the target drugs to determine the outbound priority order, wherein the first batch in the list is the current priority outbound batch; The outbound transfer execution module is used to complete the inventory transfer and outbound operations of drugs according to the sorted outbound list L, and update the batch inventory data in real time, including updating the inventory quantity, batch status and inventory information set I; The dynamic monitoring and adjustment module is used to periodically monitor the inventory status of drugs. When a batch is identified as nearing expiration, inventory abnormality, frozen, or with quality issues, a dynamic score adjustment is performed based on the use risk level R, and an updated priority score P' is calculated. The adjustment result is used to regenerate the sorted outbound list L'; The task synchronization and feedback module is used to push the changes to the pharmacist side or the outbound task scheduling system after the sorting list is updated.

Images