A regional market-oriented medicine automatic screening and replenishment recommendation method

Abstract

The application discloses a medicine automatic selection and replenishment recommendation method for regional markets, and relates to the technical field of computer processing, and comprises the following steps: S01, collecting multi-source characteristic data of a target region, extracting regional characteristic data reflecting regional market demand and medicine operation data reflecting medicine circulation state, wherein the multi-source characteristic data comprises regional population characteristic data, regional disease prevalence data, medicine historical sales data, medicine inventory data and medicine attribute data; S02, based on matching analysis of the regional characteristic data and the medicine attribute data, combining a preset medicine access principle and a synergistic mechanism of a medicine selection model fusing regional characteristics, and dynamically adjusting a medicine selection strategy according to the regional characteristic data. Through multi-dimensional adaptive calculation and comprehensive weighted scoring of regional characteristics and medicine attributes, the demand-oriented selection and priority classification output of candidate medicines in the target region are realized.

Description

Technical Field

[0001] This invention relates to the field of computer processing technology, and in particular to a method for automatic drug selection and replenishment recommendation for regional markets. Background Technology

[0002] Pharmaceutical distribution is a crucial link in the pharmaceutical and healthcare industry, and its supply chain management efficiency directly affects the accessibility and safety of medications for the public. With the rapid development of pharmaceutical e-commerce and the continuous expansion of regional pharmaceutical distribution networks, how to accurately match the differentiated medication needs of regional markets and achieve intelligent management of drug inventory has become a core issue that urgently needs to be addressed in the pharmaceutical distribution sector.

[0003] Currently, traditional drug selection and replenishment decisions mainly rely on manual experience or simple rule-based processing. Pharmacy managers manually formulate drug selection catalogs and replenishment plans based on rough statistics of historical sales data, personal experience, and seasonal factors. This model has significant limitations: on the one hand, different regional markets have significant differences in population structure, disease spectrum, medication preferences, and purchasing power, making it difficult for a uniform selection strategy to meet the personalized needs of each region; on the other hand, the large variety of drugs, their sensitive expiration dates, and fluctuating demand make it difficult for manual decision-making to achieve accurate screening and dynamic replenishment of a massive number of drugs. Summary of the Invention

[0004] To address the aforementioned technical problems, the present invention provides a method for automatic drug selection and replenishment recommendation for regional markets, comprising the following steps: S01. Collect multi-source feature data of the target area, and extract regional feature data reflecting regional market demand and drug operation data reflecting drug circulation status. The multi-source feature data includes regional population feature data, regional disease prevalence data, historical drug sales data, drug inventory data and drug attribute data. S02. Based on the matching analysis of regional feature data and drug attribute data, combined with the synergistic mechanism of the preset drug access principles and the drug selection model that integrates regional features, the drug selection strategy is dynamically adjusted according to the regional feature data. S03. Based on the adaptation evaluation results of the regional feature data and the candidate drugs, output the target selection drug set and the corresponding drug priority, wherein the candidate drugs are configured as structured drug units, and each drug corresponds to at least one quantifiable regional adaptation index set. S04. For each drug in the target drug selection set, collect real-time inventory status data and demand forecast data, and generate a replenishment recommendation strategy for each drug based on a preset replenishment strategy and a replenishment quantity optimization model that integrates dynamic inventory status. The replenishment recommendation scheme includes at least the replenishment quantity and replenishment timing. S05. Collect replenishment execution feedback data and real-time sales data, update regional feature data and drug operation data, and evaluate changes in market and inventory status through a state transition judgment mechanism. When a state change is determined, trigger a structural adjustment to the target selected drug set or replenishment recommendation scheme. The structural adjustment includes at least regenerating at least one of the following: drug selection results, drug priority, replenishment quantity, or replenishment timing. If the process termination condition is not met, return to S02 for iterative execution; if it is met, archive the execution results.

[0005] Preferably, the regional feature data in step S01 includes defining regional feature dimensions, which include, but are not limited to, regional demographic features, regional disease spectrum features, regional medication preference features, regional consumption capacity features, and regional seasonal fluctuation features. Standardize and perform feature engineering on multi-source feature data to eliminate format and dimensional differences between data from different sources; Based on the weights of each feature dimension's impact on regional drug demand, unified regional feature data is generated through weighted integration.

[0006] Preferably, the drug operation data in step S01 includes: Collect historical sales time-series data, current inventory level data, in-transit replenishment data, and drug expiration date data for drugs in the target region; Demand trend analysis is conducted based on historical sales time series data to extract the demand cycle characteristics, fluctuation amplitude characteristics, and trend direction characteristics of pharmaceuticals. By integrating the aforementioned demand cycle characteristics, fluctuation range characteristics, trend direction characteristics, current inventory level data, in-transit replenishment data, and drug expiration date data, drug operation data is generated.

[0007] Preferably, the matching analysis based on regional feature data and drug attribute data in step S02, combined with the synergistic mechanism of preset drug access principles and a drug selection model integrating regional features, dynamically adjusts the drug selection strategy according to the regional feature data, including: S21. Obtain regional characteristic data of the target area and attribute data of candidate drugs. The regional characteristic data includes regional population characteristics, regional disease spectrum characteristics, regional drug preference characteristics, and regional consumption capacity characteristics. The attribute data of candidate drugs includes drug indications, drug dosage forms and specifications, drug price ranges, and drug supply chain parameters. S22. Input the regional feature data and the attribute data of the candidate drugs into the preset drug judgment model, perform deterministic access verification, and generate rule evaluation results. The deterministic access verification includes at least drug compliance verification, drug safety verification and drug effectiveness verification. If a candidate drug fails any of the verifications, it is marked as an unacceptable drug and removed from the selection candidate set. S23. Input the attribute data of the candidate drugs that have passed the deterministic access verification and the regional feature data into the drug selection model that integrates regional features, perform differential matching evaluation, and generate model evaluation results. The differential matching evaluation includes: S231. Based on the attention mechanism, perform feature cross-referencing on the regional feature data, and calculate the influence weight of each feature dimension in the regional feature data on the current drug matching decision; S232. Reconstruct the regional feature data by weighting according to the influence weights to generate a regional feature weighted vector; S233. Calculate the multi-dimensional similarity between the weighted vector of regional features and the attribute feature vector of the candidate drug, wherein the multi-dimensional similarity includes disease treatment matching degree, medication habit matching degree, price range matching degree and supply chain adaptability. S234. Based on the weighted summation of the multi-dimensional similarity, generate the region-drug fit score for each candidate drug, which is used as the model evaluation result; S24. Configure the evaluation switching logic and dynamically adjust the weight ratio of the rule evaluation result and the model evaluation result in the final adaptation evaluation according to the preset trigger conditions. The preset trigger conditions include at least one of the regional market stability index, candidate drug quantity index and historical decision confidence index. S25. Based on the adjusted weight ratio, the rule evaluation result and the model evaluation result are weighted and fused to generate a comprehensive adaptation evaluation result. The candidate drugs are then sorted according to the comprehensive adaptation evaluation result, and the target candidate drug set and its corresponding regional adaptation level are output.

[0008] Preferably, step S03, which involves outputting the target drug selection set and corresponding drug priorities based on the adaptation evaluation results of the regional feature data and candidate drugs, includes: S31. Obtain regional feature data and a set of candidate drugs for the target area. The regional feature data includes regional population characteristics, regional disease spectrum characteristics, regional drug preference characteristics, and regional consumption capacity characteristics. Each candidate drug in the set of candidate drugs is configured as a structured drug unit. Each structured drug unit contains drug attribute data and a corresponding set of quantifiable regional adaptation indicators. S32. Based on the regional feature data and the drug attribute data of each candidate drug, calculate the multidimensional fit between each candidate drug and the target region. The multidimensional fit includes at least the disease treatment matching degree, medication habit matching degree, and price range matching degree. S33. Based on the multidimensional fit, combined with the preset drug priority evaluation indicators, a comprehensive weighted score is calculated for each candidate drug to generate a comprehensive quantitative score for each candidate drug. The drug priority evaluation indicators include the historical sales contribution of the drug in the target area, the drug profit contribution, and the drug supply stability indicator. S34. Sort the candidate drugs in the candidate drug set in descending order according to the comprehensive quantitative score, and select the candidate drugs whose ranking is within the preset range or whose comprehensive quantitative score is not lower than the preset selection threshold to form the target selection drug set. S35. Based on the ranking of the comprehensive quantitative scores of each drug in the target selection drug set, assign a corresponding drug priority level to each drug, and generate and output a drug selection recommendation list according to the drug priority level.

[0009] Preferably, the replenishment recommendation strategy for each drug in step S04 includes: Based on historical sales data, demand forecast data, and current inventory data of pharmaceuticals, the estimated out-of-stock time and safety stock threshold of pharmaceuticals are calculated. Based on the estimated out-of-stock time and safety stock threshold, combined with the procurement lead time and minimum order quantity constraints of the drugs, the replenishment quantity and timing of each drug are determined. When multiple drugs are detected to be replenishing at similar times, replenishment order aggregation optimization is performed to generate a joint replenishment recommendation plan to reduce the total replenishment cost.

[0010] Preferably, step S05, which involves assessing changes in market and inventory status through a state transition determination mechanism, includes: Preset market status change thresholds and inventory status change thresholds, and determine market status stability and inventory status stability by quantifying the differences between updated regional characteristic data and current regional characteristic data, and the differences between updated drug operation data and current drug operation data, respectively. in: If the degree of difference in regional characteristic data is lower than the market status change threshold and the degree of difference in drug operation data is lower than the inventory status change threshold, the current target selection drug set and replenishment recommendation scheme shall be maintained. If the degree of difference in regional feature data is not lower than the market state change threshold, it is determined to be a sudden change in market state, triggering a structural-level regeneration of the target selected drug set and drug priority. If the degree of difference in drug operation data is not lower than the aforementioned inventory status change threshold, it is determined to be a sudden change in inventory status, triggering a structural-level regeneration of the replenishment recommendation scheme.

[0011] Preferably, the process termination condition includes at least one of the following: The coverage and suitability of drugs in the target drug selection set reach the preset target threshold. The degree of difference between the regional feature data and the drug operation data in multiple consecutive iterations is lower than the corresponding preset stability threshold; If an irreversible drug supply anomaly or regional market anomaly is triggered, and the conditions for drug selection and replenishment cannot be met even after the anomaly is handled; The core objectives of automatic drug selection and replenishment recommendation are achieved, including reaching the preset standard for drug demand coverage in the target area or increasing the inventory turnover rate in the target area to the preset level.

[0012] The present invention has at least the following beneficial effects 1. By collecting multi-dimensional data such as regional population characteristics, disease prevalence trends, and medication preferences, and constructing structured regional characteristic data, combined with the collaborative evaluation of drug access rules and drug selection models that integrate regional characteristics, it is possible to adaptively output differentiated drug selection solutions for different regional markets. 2. Based on the output of the target drug selection set and its priority, the system further generates differentiated replenishment quantity and timing plans for each drug based on its real-time inventory status and demand forecasts, and supports the aggregation and optimization of replenishment orders. This effectively avoids inventory backlog or shortage problems caused by mismatch between drug selection and replenishment, ensuring regional drug accessibility while reducing overall inventory holding costs and stockout losses.

[0013] 3. By continuously collecting replenishment execution feedback and real-time sales data, the system quantitatively assesses changes in market and inventory status. When a sudden change in status is detected, it automatically triggers a structural regeneration of drug selection results or replenishment plans. This ensures that drug selection and replenishment decisions are always synchronized with actual regional market demand and inventory conditions, significantly improving the responsiveness and resilience of the pharmaceutical supply chain. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1This is a flowchart of an automatic drug selection and replenishment recommendation method for regional markets provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of S02 provided in Embodiment 1 of the present invention; Figure 3 A flowchart of S23 provided in Embodiment 1 of the present invention; Figure 4 The flowchart of S03 provided in Embodiment 1 of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0018] Example 1 This embodiment provides a method for automatic drug selection and replenishment recommendation for regional markets. The method includes the following steps: Figure 1 As shown: S01. Collect multi-source feature data of the target area, and extract regional feature data reflecting regional market demand and drug operation data reflecting drug circulation status. Multi-source feature data includes regional population feature data, regional disease prevalence data, historical drug sales data, drug inventory data and drug attribute data.

[0019] Furthermore, the regional feature data in the above embodiments includes defining regional feature dimensions, which include, but are not limited to, regional demographic features, regional disease spectrum features, regional medication preference features, regional consumption capacity features, and regional seasonal fluctuation features. Standardize and perform feature engineering on multi-source feature data to eliminate format and dimensional differences between data from different sources; Based on the weights of each feature dimension's impact on regional drug demand, unified regional feature data is generated through weighted integration.

[0020] Furthermore, the drug operation data in the above embodiments includes: Collect historical sales time-series data, current inventory level data, in-transit replenishment data, and drug expiration date data for drugs in the target region; Demand trend analysis is conducted based on historical sales time series data to extract the demand cycle characteristics, fluctuation amplitude characteristics, and trend direction characteristics of pharmaceuticals. By integrating demand cycle characteristics, fluctuation range characteristics, trend direction characteristics, current inventory level data, in-transit replenishment data, and drug expiration date data, drug operation data is generated.

[0021] Specifically, the system collects five types of raw data from multiple sources: regional population characteristics, regional disease prevalence, historical drug sales, drug inventory, and drug attributes. A unified data access layer is established to standardize and clean the heterogeneous data sources, laying the foundation for subsequent feature construction. Secondly, in constructing the regional feature data, the system defines five core dimensions: regional demographic characteristics, regional disease spectrum characteristics, regional medication preference characteristics, regional consumption capacity characteristics, and regional seasonal fluctuation characteristics. The system standardizes the original distribution of data for each dimension to eliminate dimensional differences and uses feature engineering methods to denoise and extract features from the raw data, transforming the high-dimensional raw data into structured feature vectors. Based on this, the system presets influence weights according to the actual impact of each feature dimension on regional drug demand, and integrates multi-dimensional features into unified regional feature data through weighted aggregation, achieving a quantitative representation of drug demand in the target region. In the process of constructing pharmaceutical operation data, the system synchronously collects historical sales time-series data, current inventory level data, in-transit replenishment data, and drug expiration date data for each drug within the target area, forming the original data foundation for pharmaceutical operations. Based on historical sales time-series data, the system uses trend analysis algorithms to extract the demand cycle characteristics (such as monthly or quarterly periodic fluctuations), fluctuation amplitude characteristics (such as the standard deviation and range of sales volume), and trend direction characteristics (such as rising, falling, or stable demand), thereby depicting the dynamic evolution of drug demand. Finally, the above three types of dynamic characteristics are integrated and correlated with static status data such as current inventory level, in-transit replenishment, and expiration date to generate structured pharmaceutical operation data, providing real-time and accurate data support for subsequent drug selection and replenishment decisions.

[0022] S02. Based on the matching analysis of regional characteristic data and drug attribute data, and combined with the synergistic mechanism of the preset drug access principles and the drug selection model that integrates regional characteristics, the drug selection strategy is dynamically adjusted according to the regional characteristic data. Details include (e.g.) Figure 2(as shown) S21. Obtain regional characteristic data and candidate drug attribute data for the target area. Regional characteristic data includes regional population characteristics, regional disease spectrum characteristics, regional drug preference characteristics, and regional consumption capacity characteristics. Candidate drug attribute data includes drug indications, drug dosage forms and specifications, drug price range, and drug supply chain parameters. S22. Input the regional feature data and the attribute data of the candidate drugs into the preset drug judgment model, perform deterministic access verification, and generate rule evaluation results. The deterministic access verification includes at least drug compliance verification, drug safety verification and drug effectiveness verification. If a candidate drug fails any of the verifications, it is marked as an unacceptable drug and removed from the selection candidate set. S23. Input the attribute data and regional feature data of candidate drugs that have passed the deterministic access verification into the drug selection model that integrates regional features, perform differential matching evaluation, and generate model evaluation results. The differential matching evaluation includes (e.g.) Figure 3 (as shown) S231. Based on the attention mechanism, perform feature cross-interaction on the regional feature data and calculate the influence weight of each feature dimension in the regional feature data on the current drug matching decision. S232. Reconstruct the regional feature data by weighting according to the influence weights, and generate a weighted vector of regional features; S233. Calculate the multi-dimensional similarity between the weighted vector of regional features and the attribute feature vector of candidate drugs. The multi-dimensional similarity includes disease treatment matching degree, medication habit matching degree, price range matching degree and supply chain adaptability. S234. Based on the weighted summation of multi-dimensional similarity, generate the region-drug fit score for each candidate drug, which serves as the model evaluation result; S24. Configure the evaluation switching logic and dynamically adjust the weight ratio of the rule evaluation result and the model evaluation result in the final adaptation evaluation according to the preset trigger conditions. The preset trigger conditions include at least one of the regional market stability indicators, candidate drug quantity and scale indicators and historical decision confidence indicators. S25. Based on the adjusted weight ratio, the rule evaluation results and model evaluation results are weighted and fused to generate a comprehensive adaptation evaluation result. The candidate drugs are then ranked according to the comprehensive adaptation evaluation result, and the target candidate drug set and its corresponding regional adaptation level are output.

[0023] Specifically, the first step is to prepare data by acquiring regional characteristic data of the target area (including regional population characteristics, regional disease spectrum characteristics, regional drug preference characteristics, and regional consumption capacity characteristics) and attribute data of each candidate drug (including drug indications, dosage forms and specifications, price ranges, and supply chain parameters), providing a structured input basis for subsequent evaluation.

[0024] The evaluation system employs a two-stage architecture that combines rules and models. The first stage is deterministic access verification, where regional characteristic data and candidate drug attribute data are input into a pre-set drug judgment model to perform drug compliance verification (e.g., whether it is a product included in the medical insurance catalog, whether it has legal marketing qualifications), drug safety verification (e.g., whether there are known serious adverse reactions, whether it belongs to specially controlled drugs), and drug efficacy verification (e.g., whether the indication is accurate, whether there is sufficient evidence to support clinical efficacy). The output of this stage is a binary judgment result; drugs that fail any verification are marked as ineligible for access and directly removed from the selection candidate set to ensure that drugs entering subsequent stages all meet the basic compliance, safety, and efficacy requirements.

[0025] The second stage is the differentiated matching assessment, which involves using a drug selection model that integrates regional characteristics to conduct a refined quantitative evaluation of candidate drugs that have passed the deterministic admission verification. The core processing flow of this model includes four sub-steps: (1) Based on the attention mechanism, feature cross-operation is performed on regional feature data to adaptively identify and calculate the differentiated influence weights of various dimensions such as regional population characteristics, disease spectrum characteristics, drug preference characteristics and consumption capacity characteristics on the current drug matching decision; (2) The original regional feature data are reconstructed by weighting according to the calculated influence weights to generate regional feature weighted vectors, so that the feature dimensions that contribute highly to the current drug matching decision can obtain higher representation intensity. (3) Calculate the multi-dimensional similarity between the weighted vector of regional features and the attribute feature vector of candidate drugs, specifically covering the matching degree of disease treatment (the degree of correspondence between drug indications and diseases prevalent in the region), the matching degree of medication habits (the degree of fit between drug dosage form and specifications and common medication habits in the region), the matching degree of price range (the degree of fit between drug pricing and regional consumption capacity) and the matching degree of supply chain (the degree of fit between drug supply parameters and regional logistics conditions). (4) The above multi-dimensional similarity is weighted and summed to generate the region-drug suitability score of each candidate drug, which is then used as the output of the model evaluation result.

[0026] S03. Based on the regional feature data and the adaptation evaluation results of candidate drugs, output the target selection drug set and the corresponding drug priority. The candidate drugs are configured as structured drug units, and each drug corresponds to at least one quantifiable set of regional adaptation indicators.

[0027] Furthermore, the steps in the above embodiments include (such as) Figure 4 (as shown) S31. Obtain regional characteristic data and candidate drug set for the target area. The regional characteristic data includes regional population characteristics, regional disease spectrum characteristics, regional drug preference characteristics, and regional consumption capacity characteristics. Each candidate drug in the candidate drug set is configured as a structured drug unit. Each structured drug unit contains drug attribute data and a corresponding set of quantifiable regional adaptation indicators. S32. Based on regional feature data and drug attribute data of each candidate drug, calculate the multidimensional fit between each candidate drug and the target region. The multidimensional fit includes at least the disease treatment matching degree, medication habit matching degree, and price range matching degree. S33. Based on the multidimensional fit, combined with the preset drug priority evaluation indicators, a comprehensive weighted score is calculated for each candidate drug to generate a comprehensive quantitative score for each candidate drug. The drug priority evaluation indicators include the historical sales contribution of the drug in the target region, the drug profit contribution, and the drug supply stability indicators. S34. Sort the candidate drugs in the candidate drug set in descending order according to the comprehensive quantitative score, and select the candidate drugs whose ranking is within the preset range or whose comprehensive quantitative score is not lower than the preset selection threshold to form the target selection drug set. S35. Based on the ranking of the comprehensive quantitative scores of each drug in the target drug selection set, assign a corresponding drug priority level to each drug, and generate and output a drug selection recommendation list according to the drug priority level.

[0028] Specifically, after acquiring the regional characteristic data and candidate drug set for the target area, each candidate drug is first configured as a structured drug unit containing drug attribute data and a set of quantifiable regional suitability indicators. Based on this, multi-dimensional suitability is calculated using the regional characteristic data and the drug attribute data of each candidate drug, including disease treatment matching degree, medication habit matching degree, and price range matching degree, to quantify the degree of matching between each drug and the needs of the target area. Subsequently, based on the multi-dimensional suitability, the historical sales contribution of the drug in the target area, drug profit contribution, and drug supply stability indicators are further introduced as priority evaluation weights to comprehensively weight and score each candidate drug, generating a unified comprehensive quantitative score. Finally, the candidate drug set is sorted in descending order according to the comprehensive quantitative score, and the target selection drug set is selected according to a preset ranking range or preset selection threshold. A corresponding drug priority level is assigned according to the ranking of each drug, and the final result is output in the form of a drug selection recommendation list. The above process realizes a complete link from the input of regional feature data to structured drug units, to multi-dimensional matching and quantification, comprehensive weighted scoring, and then to sorting, screening and priority classification output, ensuring that the drug selection results are accurately matched with regional market demand and have a quantifiable priority layer.

[0029] S04. For each drug in the target drug selection set, collect real-time inventory status data and demand forecast data. Based on the preset replenishment strategy and the replenishment quantity optimization model that integrates dynamic inventory status, generate a replenishment recommendation strategy for each drug. The replenishment recommendation plan should at least include the replenishment quantity and replenishment timing.

[0030] Furthermore, the replenishment recommendation strategy for each drug generated in the above embodiments includes: Based on historical sales data, demand forecast data, and current inventory data of pharmaceuticals, the estimated out-of-stock time and safety stock threshold of pharmaceuticals are calculated. Based on the expected out-of-stock time and safety stock threshold, combined with the procurement lead time and minimum order quantity constraints of the drugs, determine the replenishment quantity and timing of each drug. When multiple drugs are detected to be replenishing at similar times, replenishment order aggregation optimization is performed to generate a joint replenishment recommendation plan to reduce the total replenishment cost.

[0031] Specifically, after obtaining the target selection set of drugs, real-time inventory status data and demand forecast data are collected for each drug. Based on the historical sales data, demand forecast data, and current inventory data of the drugs, the expected stockout time and safety stock threshold for each drug are first calculated to quantitatively assess the inventory risk level and replenishment urgency of each drug. On this basis, according to the expected stockout time and safety stock threshold, and in conjunction with the procurement lead time and minimum order quantity constraints of each drug, the optimal replenishment quantity and best replenishment time for each drug are determined to ensure that the replenishment plan meets both inventory safety requirements and procurement cost constraints. At the same time, when the replenishment time of multiple drugs is detected to be close, replenishment order aggregation optimization is performed to merge the replenishment needs of multiple drugs into a joint replenishment recommendation scheme to reduce the overall total replenishment cost. The above process realizes a complete replenishment recommendation chain, from perceiving the inventory status and forecasting demand of the target drug selection set, to quantifying the risk of expected stockout time and safety stock threshold, and then to differentiating the replenishment quantity and timing and optimizing order aggregation. This ensures that the replenishment plan not only accurately matches the differentiated inventory needs of each drug, but also minimizes the overall supply chain cost through collaborative optimization.

[0032] S05: Collect replenishment execution feedback data and real-time sales data, update regional characteristic data and drug operation data, and evaluate changes in market and inventory status through a state transition judgment mechanism. When a state change is determined, trigger a structural adjustment to the target drug selection set or replenishment recommendation plan. The structural adjustment includes regenerating at least one of the following: drug selection results, drug priority, replenishment quantity, or replenishment timing. If the process termination condition is not met, return to S02 for iterative execution; if it is met, archive the execution results.

[0033] Furthermore, in the above embodiments, the changes in market state and inventory state are assessed through a state transition determination mechanism, including: Preset market status change thresholds and inventory status change thresholds, and determine market status stability and inventory status stability by quantifying the differences between updated regional characteristic data and current regional characteristic data, and the differences between updated drug operation data and current drug operation data, respectively. in: If the degree of difference in regional characteristic data is lower than the market status change threshold and the degree of difference in drug operation data is lower than the inventory status change threshold, the current target drug selection set and replenishment recommendation scheme will be maintained. If the degree of difference in regional characteristic data is not lower than the market state change threshold, it is determined to be a sudden change in market state, triggering a structural-level regeneration of the target selection drug set and drug priority. If the degree of difference in drug operation data is not lower than the inventory status change threshold, it is judged as a sudden change in inventory status, triggering a structural regeneration of the replenishment recommendation plan.

[0034] Specifically, driven by continuous replenishment execution and real-time sales data, the system first collects replenishment execution feedback data and real-time sales data to dynamically update regional characteristic data and drug operation data. Based on this, using preset market state change thresholds and inventory state change thresholds, the system quantifies the differences between the updated regional characteristic data and the current regional characteristic data, and the differences between the updated drug operation data and the current drug operation data, respectively, to determine market state stability and inventory state stability. The specific judgment logic is as follows: if the difference in regional characteristic data is lower than the market state change threshold and the difference in drug operation data is lower than the inventory state change threshold, it indicates that the current market environment and inventory state have not changed significantly, and the current target drug selection set and replenishment recommendation scheme remain unchanged. If the difference in regional characteristic data is not lower than the market state change threshold, it indicates that the regional market has fluctuated significantly, which is judged as a market state mutation, triggering a structural regeneration of the target drug selection set and drug priorities to re-adapt to the changed regional demand. If the difference in drug operation data is not lower than the inventory state change threshold, it indicates that the inventory state has changed significantly, which is judged as an inventory state mutation, triggering a structural regeneration of the replenishment recommendation scheme to adapt to the new inventory situation. The above mechanism enables real-time monitoring and quantitative assessment of market and inventory status, and triggers targeted structural adjustments based on changes in different dimensions, ensuring that drug selection and replenishment recommendations are always synchronized with the actual needs of the regional market and the actual inventory status. If the process termination conditions are not met, it returns to the previous iteration; if they are met, the current execution results are archived, forming a closed-loop adaptive optimization process.

[0035] Example 2 Based on Embodiment 1, the process termination conditions in this embodiment include at least one of the following: The coverage and suitability of drugs in the target drug selection set reach the preset target threshold. The degree of difference between the regional feature data and the drug operation data in multiple consecutive iterations is lower than the corresponding preset stability threshold; If an irreversible drug supply anomaly or regional market anomaly is triggered, and the conditions for drug selection and replenishment cannot be met even after the anomaly is handled; The core objectives of automatic drug selection and replenishment recommendation are achieved, including reaching the preset standard for drug demand coverage in the target area or increasing the inventory turnover rate in the target area to the preset level.

[0036] Specifically, the process termination conditions are achieved through a pre-set multi-dimensional termination judgment mechanism, including the following scenarios: First, real-time monitoring of the drug coverage and regional suitability of the target drug selection set. When both reach the pre-set target threshold, it indicates that the current selection results have fully met the regional drug demand, and termination is automatically triggered. Second, continuous tracking of the degree of difference between regional characteristic data and drug operation data during multiple iterations. If the degree of difference between each iteration is lower than the pre-set stability threshold, it indicates that the market and inventory status are stabilizing and without significant fluctuations, and no further iteration optimization is needed, so the process is automatically terminated. Third, real-time detection of whether there are any unrecoverable abnormal events on the drug supply side and the regional market side. Once such an anomaly is triggered and the normal execution conditions for drug selection and replenishment cannot be met after the pre-set anomaly handling process, the process is automatically terminated to avoid ineffective iteration. Fourth, real-time evaluation of the achievement of the core objectives of automatic drug selection and replenishment recommendation. When the drug demand coverage rate in the target region increases to the pre-set standard or the inventory turnover rate in the target region increases to the pre-set level, it is determined that the core objective has been achieved, the process is automatically terminated, and the execution results are archived. The above-mentioned multiple termination conditions are configured in parallel. When any condition is met, the termination operation is triggered to ensure that the iteration process converges within a reasonable boundary, avoids infinite loops, and takes into account both execution efficiency and reliability of results.

[0037] Example 3 This invention provides a non-transitory computer-readable storage medium storing at least one instruction or at least one program segment, which is loaded and executed by a processor to implement the following steps: Collect multi-source feature data of the target area, and extract regional feature data reflecting regional market demand and drug operation data reflecting drug circulation status. The multi-source feature data includes regional population feature data, regional disease prevalence data, historical drug sales data, drug inventory data and drug attribute data. Based on the matching analysis of regional feature data and drug attribute data, combined with the synergistic mechanism of the preset drug access principles and the drug selection model that integrates regional features, the drug selection strategy is dynamically adjusted according to the regional feature data. Based on the regional feature data and the adaptation evaluation results of candidate drugs, the target selection drug set and the corresponding drug priority are output. The candidate drugs are configured as structured drug units, and each drug corresponds to at least one set of quantifiable regional adaptation indicators. For each drug in the target drug selection set, real-time inventory status data and demand forecast data are collected. Based on the preset replenishment strategy and the replenishment quantity optimization model that integrates dynamic inventory status, a replenishment recommendation strategy for each drug is generated. The replenishment recommendation plan includes at least the replenishment quantity and replenishment timing. Collect replenishment execution feedback data and real-time sales data, update regional characteristic data and drug operation data, and evaluate changes in market and inventory status through a state transition judgment mechanism. When a state change is determined, a structural adjustment to the target drug selection set or replenishment recommendation plan is triggered. The structural adjustment includes regenerating at least one of the following: drug selection results, drug priority, replenishment quantity, or replenishment timing. If the process termination conditions are not met, return to the "matching analysis" iteration; if they are met, archive the execution results.

[0038] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0039] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0040] Example 4 This invention provides an electronic device, including a processor and a memory, wherein the memory stores at least one instruction or at least one program segment, and the at least one instruction or the at least one program segment is loaded and executed by the processor to implement the following steps: Collect multi-source feature data of the target area, and extract regional feature data reflecting regional market demand and drug operation data reflecting drug circulation status. The multi-source feature data includes regional population feature data, regional disease prevalence data, historical drug sales data, drug inventory data and drug attribute data. Based on the matching analysis of regional feature data and drug attribute data, combined with the synergistic mechanism of the preset drug access principles and the drug selection model that integrates regional features, the drug selection strategy is dynamically adjusted according to the regional feature data. Based on the regional feature data and the adaptation evaluation results of candidate drugs, the target selection drug set and the corresponding drug priority are output. The candidate drugs are configured as structured drug units, and each drug corresponds to at least one set of quantifiable regional adaptation indicators. For each drug in the target drug selection set, real-time inventory status data and demand forecast data are collected. Based on the preset replenishment strategy and the replenishment quantity optimization model that integrates dynamic inventory status, a replenishment recommendation strategy for each drug is generated. The replenishment recommendation plan includes at least the replenishment quantity and replenishment timing. Collect replenishment execution feedback data and real-time sales data, update regional characteristic data and drug operation data, and evaluate changes in market and inventory status through a state transition judgment mechanism. When a state change is determined, a structural adjustment to the target drug selection set or replenishment recommendation plan is triggered. The structural adjustment includes regenerating at least one of the following: drug selection results, drug priority, replenishment quantity, or replenishment timing. If the process termination conditions are not met, return to the "matching analysis" iteration; if they are met, archive the execution results.

[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for automatic drug selection and replenishment recommendation for regional markets, characterized in that, Includes the following steps: S01. Collect multi-source feature data of the target area, and extract regional feature data reflecting regional market demand and drug operation data reflecting drug circulation status. The multi-source feature data includes regional population feature data, regional disease prevalence data, historical drug sales data, drug inventory data and drug attribute data. S02. Based on the matching analysis of regional feature data and drug attribute data, combined with the synergistic mechanism of the preset drug access principles and the drug selection model that integrates regional features, the drug selection strategy is dynamically adjusted according to the regional feature data. S03. Based on the adaptation evaluation results of the regional feature data and the candidate drugs, output the target selection drug set and the corresponding drug priority, wherein the candidate drugs are configured as structured drug units, and each drug corresponds to at least one quantifiable regional adaptation index set. S04. For each drug in the target drug selection set, collect real-time inventory status data and demand forecast data, and generate a replenishment recommendation strategy for each drug based on a preset replenishment strategy and a replenishment quantity optimization model that integrates dynamic inventory status. The replenishment recommendation scheme includes at least the replenishment quantity and replenishment timing. S05. Collect replenishment execution feedback data and real-time sales data, update regional feature data and drug operation data, and evaluate changes in market and inventory status through a state transition judgment mechanism. When a state change is determined, trigger a structural adjustment to the target selected drug set or replenishment recommendation scheme. The structural adjustment includes at least regenerating at least one of the following: drug selection results, drug priority, replenishment quantity, or replenishment timing. If the process termination condition is not met, return to S02 for iterative execution; if it is met, archive the execution results.

2. The method for automatic drug selection and replenishment recommendation for regional markets according to claim 1, characterized in that, The regional feature data in step S01 includes defining regional feature dimensions, which include, but are not limited to, regional demographic features, regional disease spectrum features, regional medication preference features, regional consumption capacity features, and regional seasonal fluctuation features. Standardize and perform feature engineering on multi-source feature data to eliminate format and dimensional differences between data from different sources; Based on the weights of each feature dimension's impact on regional drug demand, unified regional feature data is generated through weighted integration.

3. The method for automatic drug selection and replenishment recommendation for regional markets according to claim 1, characterized in that, The drug operation data mentioned in step S01 includes: Collect historical sales time-series data, current inventory level data, in-transit replenishment data, and drug expiration date data for drugs in the target region; Demand trend analysis is conducted based on historical sales time series data to extract the demand cycle characteristics, fluctuation amplitude characteristics, and trend direction characteristics of pharmaceuticals. By integrating the aforementioned demand cycle characteristics, fluctuation range characteristics, trend direction characteristics, current inventory level data, in-transit replenishment data, and drug expiration date data, drug operation data is generated.

4. The method for automatic drug selection and replenishment recommendation for regional markets according to claim 1, characterized in that, The matching analysis based on regional feature data and drug attribute data in step S02, combined with the synergistic mechanism of preset drug access principles and a drug selection model that integrates regional features, dynamically adjusts the drug selection strategy according to the regional feature data, including: S21. Obtain regional characteristic data of the target area and attribute data of candidate drugs. The regional characteristic data includes regional population characteristics, regional disease spectrum characteristics, regional drug preference characteristics, and regional consumption capacity characteristics. The attribute data of candidate drugs includes drug indications, drug dosage forms and specifications, drug price ranges, and drug supply chain parameters. S22. Input the regional feature data and the attribute data of the candidate drugs into the preset drug judgment model, perform deterministic access verification, and generate rule evaluation results. The deterministic access verification includes at least drug compliance verification, drug safety verification and drug effectiveness verification. If a candidate drug fails any of the verifications, it is marked as an unacceptable drug and removed from the selection candidate set. S23. Input the attribute data of the candidate drugs that have passed the deterministic access verification and the regional feature data into the drug selection model that integrates regional features, perform differential matching evaluation, and generate model evaluation results. The differential matching evaluation includes: S231. Based on the attention mechanism, perform feature cross-referencing on the regional feature data, and calculate the influence weight of each feature dimension in the regional feature data on the current drug matching decision; S232. Reconstruct the regional feature data by weighting according to the influence weights to generate a regional feature weighted vector; S233. Calculate the multi-dimensional similarity between the weighted vector of regional features and the attribute feature vector of the candidate drug, wherein the multi-dimensional similarity includes disease treatment matching degree, medication habit matching degree, price range matching degree and supply chain adaptability. S234. Based on the weighted summation of the multi-dimensional similarity, generate the region-drug fit score for each candidate drug, which is used as the model evaluation result; S24. Configure the evaluation switching logic and dynamically adjust the weight ratio of the rule evaluation result and the model evaluation result in the final adaptation evaluation according to the preset trigger conditions. The preset trigger conditions include at least one of the regional market stability index, candidate drug quantity index and historical decision confidence index. S25. Based on the adjusted weight ratio, the rule evaluation result and the model evaluation result are weighted and fused to generate a comprehensive adaptation evaluation result. The candidate drugs are then sorted according to the comprehensive adaptation evaluation result, and the target candidate drug set and its corresponding regional adaptation level are output.

5. The method for automatic drug selection and replenishment recommendation for regional markets according to claim 1, characterized in that, Step S03, which involves outputting the target drug selection set and corresponding drug priorities based on the adaptation evaluation results of the regional feature data and candidate drugs, includes: S31. Obtain regional feature data and a set of candidate drugs for the target area. The regional feature data includes regional population characteristics, regional disease spectrum characteristics, regional drug preference characteristics, and regional consumption capacity characteristics. Each candidate drug in the set of candidate drugs is configured as a structured drug unit. Each structured drug unit contains drug attribute data and a corresponding set of quantifiable regional adaptation indicators. S32. Based on the regional feature data and the drug attribute data of each candidate drug, calculate the multidimensional fit between each candidate drug and the target region. The multidimensional fit includes at least the disease treatment matching degree, medication habit matching degree, and price range matching degree. S33. Based on the multidimensional fit, combined with the preset drug priority evaluation indicators, a comprehensive weighted score is calculated for each candidate drug to generate a comprehensive quantitative score for each candidate drug. The drug priority evaluation indicators include the historical sales contribution of the drug in the target area, the drug profit contribution, and the drug supply stability indicator. S34. Sort the candidate drugs in the candidate drug set in descending order according to the comprehensive quantitative score, and select the candidate drugs whose ranking is within the preset range or whose comprehensive quantitative score is not lower than the preset selection threshold to form the target selection drug set. S35. Based on the ranking of the comprehensive quantitative scores of each drug in the target selection drug set, assign a corresponding drug priority level to each drug, and generate and output a drug selection recommendation list according to the drug priority level.

6. The method for automatic drug selection and replenishment recommendation for regional markets according to claim 1, characterized in that, The step S04 of generating replenishment recommendation strategies for each drug includes: Based on historical sales data, demand forecast data, and current inventory data of pharmaceuticals, the estimated out-of-stock time and safety stock threshold of pharmaceuticals are calculated. Based on the estimated out-of-stock time and safety stock threshold, combined with the procurement lead time and minimum order quantity constraints of the drugs, the replenishment quantity and timing of each drug are determined. When multiple drugs are detected to be replenishing at similar times, replenishment order aggregation optimization is performed to generate a joint replenishment recommendation plan to reduce the total replenishment cost.

7. The method for automatic drug selection and replenishment recommendation for regional markets according to claim 1, characterized in that, The step S05, which involves assessing changes in market and inventory status through a state transition determination mechanism, includes: Preset market status change thresholds and inventory status change thresholds, and determine market status stability and inventory status stability by quantifying the differences between updated regional characteristic data and current regional characteristic data, and the differences between updated drug operation data and current drug operation data, respectively. in: If the degree of difference in regional characteristic data is lower than the market status change threshold and the degree of difference in drug operation data is lower than the inventory status change threshold, the current target selection drug set and replenishment recommendation scheme shall be maintained. If the degree of difference in regional feature data is not lower than the market state change threshold, it is determined to be a sudden change in market state, triggering a structural-level regeneration of the target selected drug set and drug priority. If the degree of difference in drug operation data is not lower than the aforementioned inventory status change threshold, it is determined to be a sudden change in inventory status, triggering a structural-level regeneration of the replenishment recommendation scheme.

8. The method for automatic drug selection and replenishment recommendation for regional markets according to claim 1, characterized in that, The process termination condition includes at least one of the following: The coverage and suitability of drugs in the target drug selection set reach the preset target threshold. The degree of difference between the regional feature data and the drug operation data in multiple consecutive iterations is lower than the corresponding preset stability threshold; If an irreversible drug supply anomaly or regional market anomaly is triggered, and the conditions for drug selection and replenishment cannot be met even after the anomaly is handled; The core objectives of automatic drug selection and replenishment recommendation are achieved, including reaching the preset standard for drug demand coverage in the target area or increasing the inventory turnover rate in the target area to the preset level.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the automatic drug selection and replenishment recommendation method for the regional market targeted by any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the automatic drug selection and replenishment recommendation method for regional markets as described in any one of claims 1 to 8.

Images