A Method and System for Generating Procurement Plans for Chain Pharmacies Based on Big Data

Abstract

A method and system for generating procurement plans for chain pharmacies based on big data, relating to the field of procurement management, is disclosed. The method includes: first, collecting historical sales data, real-time inventory data, and external environmental data to construct a multidimensional database; then, standardizing the data and constructing a drug classification matrix using a data mining model; next, using causal analysis to identify key factors influencing drug demand and establishing a weight matrix of demand influencing factors; calculating procurement parameters such as safety stock levels, economic order quantities, procurement trigger points, and procurement cycles based on this weight matrix; and finally, combining a supplier evaluation system to generate a procurement plan including a list of drugs to be procured, procurement quantities, procurement time, and supplier selection. This method, through multidimensional data analysis and causal relationship mining, achieves scientific and precise procurement decisions, improves procurement efficiency, and reduces inventory costs.

Description

Technical Field

[0001] This application relates to the field of procurement management, specifically to a method and system for generating procurement plans for chain pharmacies based on big data. Background Technology

[0002] Procurement management is a crucial aspect of the pharmaceutical retail industry, with its core task being to ensure timely drug supply and reasonable inventory levels. Currently, chain pharmacies commonly employ methods based on historical sales data analysis to formulate procurement plans. Specifically, procurement personnel analyze historical sales records, combining this with experience to predict future demand and determine procurement quantities and timing accordingly. However, this method suffers from a significant drawback in practical application: it fails to adequately consider the causal relationships between multiple data dimensions, leading to insufficient accuracy in procurement decisions. For example, the demand for certain drugs is influenced by various factors such as seasonal variations, regional demographics, and disease prevalence, often involving complex causal relationships. Traditional methods, relying solely on simple data statistics and experience, cannot accurately identify and quantify the impact of these factors on drug demand, easily resulting in inventory buildup or stockouts, impacting pharmacy operational efficiency and service quality. Therefore, a technological approach capable of generating precise procurement plans is urgently needed. Summary of the Invention

[0003] This application provides a method and system for generating procurement plans for chain pharmacies based on big data, which can improve the accuracy of procurement plans.

[0004] Firstly, this application provides a method for generating procurement plans for chain pharmacies based on big data, the method comprising:

[0005] Historical sales data, real-time inventory data, and external environment data of chain pharmacies are collected, and a multidimensional database is constructed based on the historical sales data, real-time inventory data, and external environment data; the external environment data includes seasonal information, regional population characteristics, epidemic data, and medical insurance policy information.

[0006] The data in the multidimensional database is standardized to obtain standardized multidimensional data. Based on the standardized multidimensional data, a drug classification matrix is ​​constructed using a preset data mining model.

[0007] Based on the drug classification matrix, the key factors affecting the demand for various drugs are identified using causal analysis, and a weight matrix of demand influencing factors is established based on the key factors.

[0008] Based on the weight matrix of demand influencing factors, parameter calculations are performed to obtain procurement parameters including safety stock level, economic order quantity, procurement trigger point and procurement cycle.

[0009] Based on the procurement parameters and a preset supplier evaluation system, a procurement plan is generated that includes a list of drugs to be procured, the quantity to be procured, the procurement time, and the selection of suppliers.

[0010] By adopting the above technical solution, this application constructs a multidimensional database by collecting historical sales data, real-time inventory data, and external environmental data from chain pharmacies, thereby achieving comprehensive data collection and systematic management. The collected external environmental data includes seasonal information, regional population characteristics, epidemic data, and medical insurance policy information, ensuring the comprehensiveness and relevance of the data. By standardizing the data in the multidimensional database and constructing a drug classification matrix using a pre-set data mining model, unified analysis of data from different sources and of different types is achieved. Based on the drug classification matrix, causal analysis is used to identify key factors influencing the demand for various drugs, and a weight matrix of demand influencing factors is established, thus scientifically quantifying the degree of influence of each factor on drug demand. Furthermore, based on the weight matrix of demand influencing factors, parameter calculations are performed to obtain procurement parameters including safety stock levels, economic purchase batches, procurement trigger points, and procurement cycles, achieving precise and scientific procurement decisions. Finally, by combining the procurement parameters with a pre-set supplier evaluation system, a procurement plan is generated, including a list of drugs to be procured, procurement quantities, procurement time, and supplier selection, thereby achieving intelligent management of the entire procurement process, improving procurement efficiency, reducing inventory costs, and ensuring the timeliness and accuracy of drug supply. The technical solution of this invention, by constructing a complete technical chain from data collection to solution generation, realizes data-driven and intelligent optimization of procurement decisions, significantly improving the operational efficiency and service quality of chain pharmacies.

[0011] A second aspect of this application provides a big data-based chain pharmacy procurement scheme generation system, the system comprising: a data acquisition module, a first data processing module, a second data processing module, a third data processing module, and a procurement scheme determination module;

[0012] The data acquisition module is used to collect historical sales data, real-time inventory data, and external environment data of chain pharmacies, and to construct a multidimensional database based on the historical sales data, real-time inventory data, and external environment data; the external environment data includes seasonal information, regional population characteristics, epidemic data, and medical insurance policy information.

[0013] The first data processing module is used to standardize the data in the multidimensional database to obtain standardized multidimensional data and to construct a drug classification matrix based on the standardized multidimensional data through a preset data mining model.

[0014] The second data processing module is used to identify key factors affecting the demand for various types of drugs based on the drug classification matrix using causal analysis methods, and to establish a weight matrix of demand influencing factors based on the key factors.

[0015] The third data processing module is used to perform parameter calculations based on the weight matrix of demand influencing factors to obtain procurement parameters including safety stock level, economic purchase batch, procurement trigger point and procurement cycle.

[0016] The procurement scheme determination module is used to generate a procurement scheme that includes a list of drugs to be procured, the quantity to be procured, the procurement time, and the selection of suppliers, based on the procurement parameters and a preset supplier evaluation system.

[0017] A third aspect of this application provides a computer storage medium storing a plurality of instructions adapted for loading by a processor and executing the method steps described above.

[0018] A fourth aspect of this application provides an electronic device comprising: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to execute the method steps described above.

[0019] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0020] 1. This application constructs a multidimensional database by collecting historical sales data, real-time inventory data, and external environmental data from chain pharmacies, thereby achieving comprehensive data collection and systematic management. The collected external environmental data includes seasonal information, regional population characteristics, epidemic data, and medical insurance policy information, ensuring the comprehensiveness and relevance of the data. By standardizing the data in the multidimensional database and constructing a drug classification matrix using a pre-defined data mining model, unified analysis of data from different sources and of different types is possible. Based on the drug classification matrix, causal analysis is used to identify key factors influencing the demand for various drugs, and a weight matrix of demand influencing factors is established, thereby scientifically quantifying the degree of influence of each factor on drug demand.

[0021] 2. This application calculates parameters based on a weight matrix of demand influencing factors, obtaining procurement parameters including safety stock level, economic order quantity, procurement trigger point, and procurement cycle, thus achieving precise and scientific procurement decisions. Finally, by combining the procurement parameters with a pre-set supplier evaluation system, a procurement plan is generated that includes a list of drugs to be procured, procurement quantity, procurement time, and supplier selection. This enables intelligent management of the entire procurement process, improves procurement efficiency, reduces inventory costs, and ensures the timeliness and accuracy of drug supply. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating a method for generating procurement plans for chain pharmacies based on big data, as provided in an embodiment of this application.

[0023] Figure 2 This is a schematic diagram of a big data-based chain pharmacy procurement scheme generation system provided in an embodiment of this application;

[0024] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0025] Explanation of reference numerals in the attached figures: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0027] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0028] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0029] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0030] Please refer to Figure 1This paper presents a flowchart illustrating a method for generating procurement plans for chain pharmacies based on big data. This method can be implemented using a computer program, a microcontroller, or run on a big data-based chain pharmacy procurement plan generation system. The computer program can be integrated into a computer device or run as a standalone application. Specifically, this method is applied to non-fixed fan dehumidifiers. The method includes steps 10 to 60, as follows:

[0031] Step 10: Collect historical sales data, real-time inventory data, and external environment data from chain pharmacies, and construct a multidimensional database based on the historical sales data, real-time inventory data, and external environment data; the external environment data includes seasonal information, regional population characteristics, epidemic data, and medical insurance policy information;

[0032] Specifically, the data acquisition system simultaneously acquires historical sales data, real-time inventory data, and external environment data from chain pharmacies. Historical sales data includes daily sales records for each store over the past three years, recording information such as drug number, sales quantity, and sales amount. Real-time inventory data, recorded in real-time by the pharmacy management system, tracks inventory changes at each store, including inventory quantity, inventory value, and expiration date. External environment data includes key external factors affecting drug sales, including seasonal information (temperature, humidity, etc.), regional demographic information (population density, age structure, etc.), epidemiological data (incidence and spread trends of various diseases), and medical insurance policy information (updates to the medical insurance catalog and changes in reimbursement policies). After collecting this data, a multidimensional database is constructed using data warehouse technology, organizing and storing data from different sources according to multiple dimensions such as time, space, and product. This multidimensional database allows the system to analyze and mine data from multiple perspectives, providing comprehensive data support for subsequent data analysis and decision-making. For example, when analyzing the sales patterns of a certain drug in a specific season or region, only the appropriate dimension needs to be selected in the multidimensional database for data extraction and analysis to obtain accurate results. In addition, the establishment of multidimensional databases has enabled efficient data management and rapid retrieval, improving data processing efficiency.

[0033] Step 20: Standardize the data in the multidimensional database to obtain standardized multidimensional data, and construct a drug classification matrix based on the standardized multidimensional data using a preset data mining model;

[0034] Specifically, the data in the multidimensional database is first standardized. Since the multidimensional database contains data from different sources and with different units of measurement, standardization is necessary to make the data comparable. Standardization is achieved by converting the data into a unified numerical range, specifically using the maximum-minimum standardization method to map various data types to the [0,1] interval. For sales data, standardization is performed by calculating the average daily sales volume; for inventory data, standardization is performed by calculating the inventory turnover rate; for external environment data, standardization is performed according to the data type, such as converting temperature data into standard scores and population density data into relative density values. After standardization, a drug classification matrix is ​​constructed using a pre-set data mining model. This model includes a hierarchical clustering algorithm to cluster drugs with similar characteristics into the same category. The rows of the drug classification matrix represent different drugs, the columns represent various characteristics of the drugs, and each element of the matrix represents the standardized value of a drug in a certain feature dimension. In this way, the system establishes a multidimensional classification system for drugs, providing a foundation for subsequent causal analysis. This classification method based on standardized data eliminates the differences in data dimensions and improves the accuracy of classification. At the same time, the construction of the drug classification matrix enables the system to understand and analyze the characteristics of drugs from multiple dimensions, providing a scientific basis for subsequent procurement decisions.

[0035] Based on the above embodiments, as an optional embodiment, the step of constructing a drug classification matrix based on the standardized multidimensional data using a preset data mining model includes:

[0036] S110, extract features from the sales data records in the standardized multidimensional data to obtain sales frequency features, sales fluctuation features and sales seasonality features;

[0037] Specifically, the system extracts features from sales data records in the standardized multidimensional data. First, it extracts the sales frequency feature by dividing the number of sales of each drug within a certain time window by the length of the time window; this feature reflects the sales activity of the drug. Then, it extracts the sales volatility feature by calculating the ratio of the standard deviation to the mean of sales volume; this feature reflects the stability of drug sales volume. Finally, it extracts the sales seasonality feature by performing time series decomposition on the sales data to separate the seasonal component; this feature reflects the cyclical changes in drug sales. Through the extraction of these three features, the system comprehensively captures the key characteristics of drugs at the sales level, providing accurate feature indicators for subsequent drug classification. For example, cold and flu drugs typically exhibit high sales frequency, large sales volatility, and obvious sales seasonality, while chronic disease medications exhibit lower sales volatility and weaker sales seasonality. This feature extraction method enables the system to accurately identify the sales patterns of different drugs, laying the foundation for constructing a scientific drug classification matrix.

[0038] S120, extract features from the demand data records in the standardized multidimensional data to obtain demand stability features, demand forecasting difficulty features, and demand suddenness features;

[0039] Specifically, the system extracts features from demand data records in the standardized multidimensional data. First, it extracts the demand stability feature by calculating the coefficient of variation (COP) of demand over a continuous time period. The COP is the ratio of the standard deviation to the mean of demand, reflecting the degree of fluctuation in drug demand. Second, it extracts the demand forecasting difficulty feature by calculating the average absolute percentage error between historical forecasts and actual demand, reflecting the complexity of drug demand forecasting. Finally, it extracts the demand surge characteristic by calculating the ratio of the peak demand to the average demand, reflecting the degree of suddenness in drug demand. Through the extraction of these three features, the system accurately characterizes the key attributes of drugs at the demand level. For example, commonly used drugs exhibit high demand stability and low demand forecasting difficulty, while emergency drugs exhibit high demand surge characteristics. This multidimensional demand feature extraction provides a precise description of demand characteristics for subsequent drug classification and procurement decisions, improving the accuracy of procurement decisions.

[0040] S130, extract features from the inventory data records in the standardized multidimensional data to obtain storage condition features, shelf life features and inventory cost features;

[0041] Specifically, the system extracts features from inventory data records in the standardized multidimensional data. First, it extracts storage condition features by analyzing environmental parameters such as temperature, humidity, and light required for the drug, quantifying the stringency of storage conditions into numerical values. This feature reflects the complexity of drug storage. Second, it extracts shelf-life features by calculating the time from the drug's production date to its expiration date, reflecting the drug's storage time limitations. Finally, it extracts inventory cost features by calculating the unit storage costs, insurance costs, and capital occupation costs of the drug, reflecting the economic cost of drug storage. Through the extraction of these three features, the system comprehensively characterizes the key attributes of drugs at the inventory management level. For example, ordinary drugs stored at room temperature have lower storage condition feature values, while drugs requiring cold chain storage have higher storage condition feature values ​​and inventory cost feature values. This feature extraction method enables the system to accurately assess the difficulty and cost of inventory management for different drugs, providing an important basis for formulating reasonable procurement strategies.

[0042] S140, the sales frequency characteristics, sales fluctuation characteristics, sales seasonality characteristics, demand stability characteristics, demand forecasting difficulty characteristics, demand suddenness characteristics, storage condition characteristics, shelf life characteristics, and inventory cost characteristics are combined to form a drug feature vector;

[0043] Specifically, the system combines the extracted features to construct a drug feature vector. First, sales frequency, sales fluctuation, and sales seasonality features from the sales perspective are arranged in order, representing the drug's sales activity, sales stability, and cyclical variation patterns, respectively. Next, demand stability, demand forecasting difficulty, and demand suddenness features from the demand perspective are added to the feature sequence, reflecting the degree of demand fluctuation, forecasting complexity, and suddenness, respectively. Finally, storage conditions, shelf life, and inventory costs from the inventory perspective are added to the feature sequence, representing storage requirements, time constraints, and economic costs, respectively. By combining these nine features in a fixed order, a complete drug feature vector is formed. Each drug corresponds to a feature vector, and each dimension of the vector represents a quantified value of the drug on a specific feature. Through this feature vector construction method, the system integrates the multi-dimensional attributes of drugs into a unified mathematical expression, providing a standardized input data format for subsequent drug classification using data mining models.

[0044] S150, input the drug feature vector into the preset data mining model to obtain multi-dimensional drug data, and construct the drug classification matrix based on the multi-dimensional drug data.

[0045] Specifically, the pre-defined data mining models include clustering analysis, association rule mining, and pattern recognition. First, the drug feature vectors are input into the clustering analysis model. By calculating the distance between sample points in the feature space, drugs with similar features are grouped into classes, resulting in initial drug categories. Next, the drug feature vectors are input into the association rule mining model to mine the relationships between different features, obtaining drug associations. Then, the drug feature vectors are input into the pattern recognition model to identify patterns in drug demand, obtaining drug demand patterns. The system combines the initial drug categories, drug associations, and drug demand patterns to form multi-dimensional drug data. Based on this multi-dimensional drug data, the system constructs a drug classification matrix. Rows in this matrix represent different drugs, columns represent various features and category attributes of the drugs, and matrix elements represent specific feature values ​​or category identifiers. In this way, the system achieves multi-dimensional classification of drugs, giving each category a clear feature description and management strategy, providing a scientific classification basis for subsequent procurement decisions.

[0046] Based on the above embodiments, as an optional embodiment, the preset data mining model includes a preset clustering analysis model, a preset association rule mining model, and a preset pattern recognition model; the drug feature vector is input into the preset data mining model to obtain multi-dimensional drug data, including:

[0047] S151, Input the drug feature vector into the preset clustering analysis model to obtain the initial drug category;

[0048] Specifically, the system inputs drug feature vectors into a pre-defined clustering analysis model for classification. The pre-defined clustering analysis model uses the K-means clustering algorithm, which measures the similarity between drugs by calculating the Euclidean distance between drug feature vectors in the feature space. First, the system sets the number of cluster centers, K, to 5, representing 5 initial categories for the drugs. Then, K sample points are randomly selected as initial cluster centers, and the distance from each drug feature vector to these K cluster centers is calculated, assigning each drug feature vector to the category of the nearest cluster center. Next, the center point of each category is recalculated, i.e., the mean of all drug feature vectors in that category, as the new cluster center. The system repeats the process of dividing sample points and updating cluster centers until the cluster centers no longer change or the preset number of iterations is reached, ultimately obtaining the initial drug categories. This clustering-based classification method automatically groups drugs with similar characteristics into the same category, forming a scientifically sound initial classification system, providing a classification basis for subsequent procurement strategy formulation.

[0049] S152, Input the drug feature vector into the preset association rule mining model to obtain drug association relationships;

[0050] Specifically, the system inputs drug feature vectors into a pre-defined association rule mining model for association analysis. This model employs the Apriori algorithm, analyzing the relationships between features in the drug feature vectors to discover association patterns between drugs. First, the system sets a minimum support threshold of 0.1 and a minimum confidence threshold of 0.6 to filter strong association rules. Then, each feature in the drug feature vector is discretized, dividing continuous feature values ​​into discrete intervals. Next, the system scans all drug feature vectors, calculating the frequency of each feature combination to generate frequent itemsets. Based on these frequent itemsets, the system calculates the conditional probability between feature combinations; when both support and confidence simultaneously meet the threshold requirements, a drug association relationship is established. The resulting drug association relationships reflect the strength of association between different features, such as the association rule between sales frequency and demand stability features. This association rule-based analysis method reveals the inherent connections between drug features, providing a basis for developing joint procurement strategies.

[0051] S153, Input the drug feature vector into the preset pattern recognition model to obtain the drug demand pattern;

[0052] Specifically, the system inputs drug feature vectors into a pre-defined pattern recognition model for demand pattern analysis. The pre-defined pattern recognition model employs a decision tree algorithm, identifying patterns of demand change by analyzing feature combinations within the drug feature vectors. First, the system focuses on demand-related features from the drug feature vectors, including demand stability, demand forecasting difficulty, and demand suddenness. Then, the system calculates the information gain of each feature and selects the feature with the highest information gain as the splitting node of the decision tree. Next, based on the distribution of feature values, the system assigns samples to different child nodes, repeating the feature selection and splitting process at each child node until a termination condition is met. Through each path in the decision tree, the system identifies demand change patterns corresponding to different feature combinations, forming drug demand patterns. This pattern recognition-based analysis method transforms drug demand characteristics into explicit demand patterns, providing a basis for demand forecasting in procurement decisions.

[0053] S154, the initial drug category, the drug association relationship, and the drug demand pattern are used as multi-dimensional data of the drug.

[0054] Specifically, the system integrates the analysis results obtained from different models to form multi-dimensional drug data. First, the system uses the initial drug categories obtained from the clustering analysis model as the first dimension, which includes basic classification information and category characteristics. Then, the system uses the drug association relationships obtained from the association rule mining model as the second dimension, reflecting the strength and rules of association between drug features. Next, the system uses the drug demand patterns obtained from the pattern recognition model as the third dimension, describing the patterns of demand changes and predictive characteristics of drugs. The system combines these three dimensions to form complete multi-dimensional drug data, where each drug includes a category identifier, association rules, and demand patterns. This multi-dimensional data integration method allows the system to comprehensively describe and understand drug characteristics from multiple perspectives, providing a complete data foundation for subsequently constructing a drug classification matrix.

[0055] Based on the above embodiments, as an optional embodiment, the step of constructing the drug classification matrix based on the multi-dimensional drug data includes:

[0056] S155, Establish a drug category dimension based on the initial drug category;

[0057] Specifically, the system establishes drug category dimensions based on initial drug categories. First, the system groups the initial drug categories obtained through cluster analysis according to category characteristics, forming a hierarchical structure. Then, the system analyzes the feature centers of each category to determine its core attributes, including typical values ​​of sales characteristics, demand characteristics, and inventory characteristics. Next, the system assigns a unique identifier to each category and establishes hierarchical relationships between categories, forming a complete drug category dimension. This dimension integrates the drug category attributes, feature centers, and hierarchical relationships, constituting the first dimension of the drug classification matrix. This method of establishing dimensions based on initial drug categories enables the system to clearly describe the main characteristics and category attributes of each drug category, providing a categorized management basis for subsequent procurement decisions.

[0058] S156, Establish demand feature dimensions based on the aforementioned drug association relationships;

[0059] Specifically, a demand feature dimension is established based on drug association relationships. First, the system performs strength analysis on drug association relationships obtained through association rule mining, calculating the association strength value between each pair of associated features. Then, the system filters the association strength values ​​according to a preset threshold, retaining feature pairs with significant associations. Next, based on these significant associations, the system constructs a demand feature network, where nodes represent drug features and connections represent the associations between features. The system transforms this feature network into a demand feature dimension, which describes the mutual influence relationships between drug demand features. This demand feature dimension, established in this way, enables the system to accurately grasp the associations between drug demand features, providing a basis for demand correlation in procurement decisions.

[0060] S157, Establish management strategy dimensions based on the aforementioned drug demand pattern;

[0061] Specifically, the system establishes management strategy dimensions based on drug demand patterns. First, the system categorizes drug demand patterns obtained through a pattern recognition model, grouping patterns with similar demand change patterns together. Then, the system sets corresponding management strategy parameters for each demand pattern, including safety stock coefficients, procurement cycles, and order quantities. Next, the system organizes these management strategy parameters into management strategy dimensions, which describe the specific management strategies under different demand patterns. Through this method of establishing management strategy dimensions based on demand patterns, the system achieves a precise correspondence between demand patterns and management strategies, providing specific strategic guidance for the procurement management of different types of drugs.

[0062] S158, combine the drug category dimension, demand characteristic dimension and management strategy dimension to form a drug classification matrix.

[0063] Specifically, the system establishes management strategy dimensions based on drug demand patterns. First, the system categorizes drug demand patterns obtained through a pattern recognition model, grouping patterns with similar demand change patterns together. Then, the system sets corresponding management strategy parameters for each demand pattern, including safety stock coefficients, procurement cycles, and order quantities. Next, the system organizes these management strategy parameters into management strategy dimensions, which describe the specific management strategies under different demand patterns. Through this method of establishing management strategy dimensions based on demand patterns, the system achieves a precise correspondence between demand patterns and management strategies, providing specific strategic guidance for the procurement management of different types of drugs.

[0064] Step 30: Based on the drug classification matrix, use causal analysis to identify key factors affecting the demand for various drugs, and establish a weight matrix of demand influencing factors based on the key factors;

[0065] Specifically, causal analysis is performed based on a drug classification matrix to identify key factors influencing drug demand. First, historical sales data for each drug category in the matrix is ​​acquired. This historical sales data is then time-series matched with external environmental data to establish an initial causal relationship dataset. Based on this dataset, conditional independence tests are used to verify the causal relationships between variables. Conditional independence tests are performed by calculating the conditional mutual information values ​​between variables. When the mutual information value exceeds a preset significance threshold, a causal relationship is confirmed between the two variables. In this way, the system identifies the key factors that truly influence drug demand. After obtaining the key factors, the degree of influence of each key factor on drug demand is calculated, establishing a weight matrix of demand influencing factors. Both rows and columns of this weight matrix represent key factors, and the element values ​​indicate the strength of the influence of the row factor on the column factor. For example, for cold and flu medications, causal analysis reveals that temperature changes and influenza incidence rates are the two most significant key factors, with weights of 0.6 and 0.4 respectively in the weight matrix. By establishing this weight matrix, the system clearly quantifies the influence relationships between various factors, providing a scientific basis for subsequent parameter calculations and improving the accuracy of procurement decisions.

[0066] Based on the above embodiments, as an optional embodiment, the step of identifying key factors affecting the demand for various types of drugs using causal analysis methods based on the drug classification matrix includes:

[0067] S310, Obtain historical sales data corresponding to each type of drug in the drug classification matrix, and perform time-series matching between the historical sales data and external environment data to establish an initial causal relationship dataset;

[0068] Specifically, the system first extracts the identification information of various drugs from the drug classification matrix, and then obtains the corresponding historical sales data based on this identification information. For each drug category, the system extracts daily sales records for the past three years, including sales quantity and sales amount. Simultaneously, the system acquires external environmental data, including seasonal information, regional population characteristics, epidemic data, and medical insurance policy information. Then, the system matches the historical sales data with the external environmental data along the time dimension, combining sales data and environmental data at the same point in time. The system uses a sliding time window approach to pair data at each point in time, ensuring a temporal correspondence between sales data and environmental data. Through this time-series matching, the system establishes an initial causal relationship dataset, in which each record contains sales data and corresponding environmental data at a specific point in time. This data matching method provides a complete data foundation for subsequent causal relationship analysis, enabling the system to accurately analyze the impact of external environmental factors on drug sales.

[0069] S320, Perform a conditional independence test on the initial causal relationship dataset to obtain the target variable pairs that pass the conditional independence test, and calculate the causal strength coefficient of the target variable pairs;

[0070] Specifically, the system performs conditional independence tests on variable pairs in the initial causal relationship dataset. First, the system calculates the conditional mutual information value for each variable pair, measuring their independence by calculating the degree of interdependence between the two variables given other variables. Next, the system sets a significance level of 0.05. When the p-value corresponding to the mutual information value is less than the significance level, the system determines that the variable pair has a significant causal relationship and marks it as the target variable pair that passes the conditional independence test. Then, the system calculates the causal strength coefficient for the target variable pairs that pass the test. This coefficient is calculated using a combination of conditional probability and time series correlation. For example, for a certain type of cold medicine, the system finds that the causal strength coefficient between its sales volume and temperature changes is 0.72, and the causal strength coefficient between its sales volume and the incidence rate of influenza is 0.85. Through this conditional independence test and causal strength calculation, the system screens out variable pairs that truly have a causal relationship and quantifies the degree of their influence, providing a scientific basis for subsequently identifying key factors.

[0071] S330, the causal strength is screened according to a preset coefficient threshold to obtain key factors.

[0072] Specifically, the system filters the calculated causal strength coefficients to identify the true key factors. First, the system sets a threshold of 0.6 for the causal strength coefficient, determined based on historical data analysis and expert experience. Then, the system compares the causal strength coefficient of each target variable pair with the preset threshold. When the causal strength coefficient is greater than the preset threshold, the influencing factor in that variable pair is identified as a key factor. Through this screening method, the system ultimately obtains the key factors affecting drug demand. For example, for cold medicines, the system, through causal strength coefficient screening, identified temperature change (causal strength coefficient 0.72) and influenza incidence rate (causal strength coefficient 0.85) as key factors, while relative humidity (causal strength coefficient 0.45) was excluded because its coefficient was below the threshold. This threshold-based coefficient screening method ensures that the system retains only factors with significant influence, providing a reliable factor basis for subsequently establishing a weight matrix of demand influencing factors.

[0073] Based on the above embodiments, as an optional embodiment, the step of establishing a weight matrix of demand influencing factors according to the key factors includes:

[0074] S340, Calculate the direct impact weight and interactive impact weight of each of the key factors;

[0075] Specifically, the system calculates two types of influence weights for the selected key factors. First, the system calculates the direct influence weight, obtained by analyzing the impact of each key factor acting alone on drug demand. This weight is calculated using the partial correlation coefficient between the key factor and the demand. Then, the system calculates the interaction influence weight, obtained by analyzing the impact of two key factors acting simultaneously on drug demand. This weight is calculated using the interaction coefficient between the factors. For example, for cold and flu medications, the direct influence weight of temperature change is 0.6, the direct influence weight of influenza incidence rate is 0.7, while their interaction influence weight is 0.8, indicating that the combined effect of these two factors has a greater impact on demand. By calculating these two weights, the system not only quantifies the influence of individual factors but also captures the synergistic effect between factors, providing accurate weight data for constructing a complete weight matrix.

[0076] S350, construct a weight matrix of demand influencing factors based on the direct influence weight and the interactive influence weight.

[0077] Specifically, the system constructs a weight matrix of demand influencing factors based on the calculated direct and interactive impact weights. First, the system creates an n×n matrix framework, where n represents the number of key factors, and both rows and columns are labeled with key factors. Then, the system fills the diagonal positions of the matrix with the direct impact weights of each key factor; these weights reflect the degree of independent influence of each factor on demand. Next, the system fills the off-diagonal positions with the interactive impact weights; these weights represent the strength of the interaction between two factors. For example, in the demand influencing factor weight matrix, the direct impact weight of 0.6 for temperature change is filled in the diagonal position, while the interactive impact weight of 0.8 between temperature change and influenza incidence rate is filled in the corresponding off-diagonal position. This weight matrix, constructed in this way, fully records the influence relationships of each key factor, providing a weight basis for subsequent parameter calculations.

[0078] Step 40: Calculate parameters based on the weight matrix of demand influencing factors to obtain procurement parameters including safety stock level, economic order quantity, procurement trigger point and procurement cycle;

[0079] Specifically, in this embodiment, the system calculates procurement parameters based on a weighted matrix of demand influencing factors. First, the weighted average daily demand μ is calculated using the weighted matrix of demand influencing factors. This value is obtained by weighting the impact weights of each key factor on demand with historical demand data. The demand standard deviation σ is calculated based on the weighted average daily demand μ to measure the degree of demand fluctuation. Then, the system obtains the preset service level target value P, replenishment lead time L, unit procurement cost A, and unit inventory holding cost h, which are set by the pharmacy according to its actual operating conditions. Next, the safety factor k is calculated using the formula k=√(-2ln(1-P)), and the safety factor k, demand standard deviation σ, and replenishment lead time L are substituted into the formula SS=k×σ×√L to obtain the safety stock level SS. Based on the weighted average daily demand μ, the system calculates the annual demand D=365×μ, and the annual demand D, unit procurement cost A, and unit inventory holding cost h are substituted into the formula EOQ=√(2AD / h) to obtain the economic order quantity EOQ. Subsequently, the weighted average daily demand μ, replenishment lead time L, and safety stock level SS are substituted into the formula ROP = μL + SS to calculate the procurement trigger point ROP. Finally, based on the economic order quantity EOQ and the weighted average daily demand μ, the procurement cycle T is calculated using the formula T = EOQ / μ. Through this series of parameter calculations, the system transforms the qualitative causal analysis results into quantitative procurement parameters, providing specific numerical basis for procurement decisions and improving the scientific nature and accuracy of procurement decisions.

[0080] Based on the above embodiments, as an optional embodiment, the parameter calculation based on the weight matrix of demand influencing factors to obtain procurement parameters including safety stock level, economic order quantity, procurement trigger point, and procurement cycle includes:

[0081] The weighted average daily demand μ is calculated based on the weight matrix of the demand influencing factors, and the demand standard deviation σ is calculated based on the weighted average daily demand μ.

[0082] The safety factor k is calculated based on the preset service level target value P and the preset safety factor calculation formula, and the safety stock level SS is determined based on the safety factor k, the demand standard deviation σ and the preset replenishment lead time L.

[0083] Based on the weighted average daily demand μ, the annual demand D is determined, and based on the annual demand D, the preset unit purchase cost A, and the preset unit inventory holding cost h, the economic order quantity EOQ is determined.

[0084] Based on the weighted average daily demand μ, the preset replenishment lead time L, and the safety stock level SS, the procurement trigger point ROP is determined;

[0085] The procurement cycle T is determined based on the Economic Order Quantity (EOQ) and the weighted average daily demand (μ).

[0086] Specifically, the system calculates procurement parameters based on a weighted matrix of demand-influencing factors. First, the system calculates the weighted average daily demand μ using this matrix. This involves multiplying the historical daily demand corresponding to each key factor by its weight and then summing the results. For example, the weighted average daily demand μ for a certain drug is 100 units. Based on the weighted average daily demand μ, the system calculates the square root of the sum of squares of the deviations of the demand from the mean, obtaining the standard deviation σ, which measures the degree of demand fluctuation.

[0087] Next, based on the preset service level target value P (e.g., 0.95), the system calculates the safety factor k using the formula k=√(-2ln(1-P)). Substituting the safety factor k, the demand standard deviation σ, and the preset replenishment lead time L (e.g., 7 days) into the formula SS=k×σ×√L, the safety stock level SS is calculated.

[0088] Then, the system calculates the annual demand D based on the weighted average daily demand μ, i.e., D = 365 × μ. Substituting the annual demand D, the preset unit purchase cost A (e.g., 100 yuan), and the preset unit inventory holding cost h (e.g., 10 yuan) into the formula EOQ = √(2AD / h), the economic order quantity EOQ is calculated.

[0089] Subsequently, the system substitutes the weighted average daily demand μ, the preset replenishment lead time L, and the safety stock level SS into the formula ROP = μL + SS to calculate the procurement trigger point ROP. Finally, based on the economic order quantity EOQ and the weighted average daily demand μ, the system calculates the procurement cycle T using the formula T = EOQ / μ.

[0090] Step 50: Based on the procurement parameters and the preset supplier evaluation system, generate a procurement plan that includes a list of drugs to be procured, the quantity to be procured, the procurement time, and the selection of suppliers.

[0091] Specifically, the final procurement plan is generated based on procurement parameters. First, the system extracts drugs whose current inventory is below the procurement trigger point ROP, forming a list of drugs to be procured. For each drug in the list, the system determines the procurement quantity based on the Economic Order Quantity (EOQ), while adjusting for the safety stock level (SS) to ensure that the post-procurement inventory is not lower than the safety stock level SS. When determining the procurement time, the system sets a specific procurement execution time point based on the procurement cycle T. Next, the system selects suitable suppliers based on a pre-set supplier evaluation system, which includes four dimensions: on-time delivery rate, price competitiveness, quality assurance capability, and emergency response capability. On-time delivery rate is calculated by dividing the number of timely deliveries by the total number of deliveries; price competitiveness is calculated by subtracting the supplier's price from the highest bid and then dividing by the difference between the highest and lowest bids; quality assurance capability is calculated by dividing the number of qualified deliveries by the total number of deliveries; and emergency response capability is calculated by dividing the number of successful emergency deliveries by the number of emergency requests. Based on the scores of these four dimensions, the system calculates the supplier's comprehensive score and selects the supplier with the highest score as the procurement target. Ultimately, the system generates a complete procurement plan, including a detailed list of medicines to be procured, specific quantities to be purchased, a clear procurement schedule, and the optimal supplier selection. In this way, the system achieves scientific and automated procurement decision-making, improving procurement efficiency and reducing procurement costs.

[0092] Please see Figure 2 The above is a schematic diagram of a big data-based chain pharmacy procurement scheme generation system provided in an embodiment of this application. The big data-based chain pharmacy procurement scheme generation system may include: a data acquisition module 1, a first data processing module 2, a second data processing module 3, a third data processing module 4, and a procurement scheme determination module 5.

[0093] The data acquisition module 1 is used to collect historical sales data, real-time inventory data, and external environment data of chain pharmacies, and to construct a multidimensional database based on the historical sales data, real-time inventory data, and external environment data; the external environment data includes seasonal information, regional population characteristics, epidemic data, and medical insurance policy information.

[0094] The first data processing module 2 is used to standardize the data in the multidimensional database to obtain standardized multidimensional data and to construct a drug classification matrix based on the standardized multidimensional data through a preset data mining model.

[0095] The second data processing module 3 is used to identify key factors affecting the demand for various types of drugs based on the drug classification matrix using causal analysis methods, and to establish a weight matrix of demand influencing factors based on the key factors.

[0096] The third data processing module 4 is used to perform parameter calculations based on the weight matrix of demand influencing factors to obtain procurement parameters including safety stock level, economic purchase batch, procurement trigger point and procurement cycle.

[0097] The procurement scheme determination module 5 is used to generate a procurement scheme that includes a list of drugs to be procured, the quantity to be procured, the procurement time, and the selection of suppliers, based on the procurement parameters and a preset supplier evaluation system.

[0098] It should be noted that the system provided in the above embodiments is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0099] This application also provides a computer storage medium that can store multiple instructions. The instructions are adapted to be loaded and executed by a processor. The above-described method for generating a chain pharmacy procurement plan based on big data is described in detail below. The specific execution process can be found in the detailed description of the above-described embodiments, which will not be repeated here.

[0100] Please refer to Figure 3 This application also discloses an electronic device. Figure 3 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. The electronic device 300 may include: at least one processor 301, at least one network interface 304, a user interface 303, a memory 305, and at least one communication bus 302.

[0101] The communication bus 302 is used to enable communication between these components.

[0102] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0103] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0104] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 305, and by calling data stored in memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0105] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory 305 may include non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. (Refer to...) Figure 3 The memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for generating procurement plans for chain pharmacies based on big data.

[0106] exist Figure 3In the illustrated electronic device 300, the user interface 303 is mainly used to provide an input interface for the user and acquire user input data; while the processor 301 can be used to call an application stored in the memory 305 that generates a chain pharmacy procurement plan based on big data. When executed by one or more processors 301, the electronic device 300 performs one or more of the methods described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0107] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0108] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0109] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0110] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0111] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0112] The above description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the specification and the disclosure of practical truths.

[0113] This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A big data based chain pharmacy procurement plan generation method, characterized in that, The method comprises the following steps: Collecting historical sales data, real-time inventory data and external environment data of a chain drugstore, and constructing a multidimensional database based on the historical sales data, the real-time inventory data and the external environment data; the external environment data comprises seasonal information, regional population characteristics, epidemic data and medical insurance policy information; Standardizing the data in the multidimensional database to obtain standardized multidimensional data, and constructing a drug classification matrix based on the standardized multidimensional data by using a preset data mining model; the data mining model comprises a preset clustering analysis model, a preset association rule mining model and a preset pattern recognition model; Based on the drug classification matrix, using a causal analysis method to identify key factors affecting the demand of various drugs, and establishing a demand influencing factor weight matrix according to the key factors; Based on the demand influencing factor weight matrix, performing parameter calculation to obtain procurement parameters including safety stock level, economic procurement batch, procurement trigger point and procurement cycle; According to the procurement parameters, combining a preset supplier evaluation system to generate a procurement scheme including a procurement drug list, a procurement quantity, a procurement time and a supplier selection; Wherein, based on the drug classification matrix, using a causal analysis method to identify key factors affecting the demand of various drugs, comprises: Obtaining historical sales data corresponding to various drugs in the drug classification matrix, and matching the historical sales data with the external environment data according to the time dimension, combining the historical sales data and the external environment data at the same time point, pairing the data at each time point by using a time window sliding, ensuring the corresponding relationship between the historical sales data and the external environment data in time, to realize time sequence matching and establish an initial causal relationship data set; Performing conditional independence test on the initial causal relationship data set to obtain a target variable pair passing the conditional independence test, and calculating the causal strength coefficient of the target variable pair; According to a preset coefficient threshold, the causal strength is screened to obtain key factors; According to the key factors, the demand influencing factor weight matrix is established, which comprises: Calculating the direct influence weight and the interaction influence weight of each key factor; Based on the direct influence weight and the interaction influence weight, a demand influencing factor weight matrix is constructed.

2. The method of claim 1, wherein, Based on the standardized multidimensional data, a drug classification matrix is constructed by using a preset data mining model, which comprises: Extracting features from the sales data records in the standardized multidimensional data to obtain sales frequency features, sales fluctuation features and sales seasonality features; Extracting features from the demand data records in the standardized multidimensional data to obtain demand stability features, demand prediction difficulty features and demand burstiness features; Extracting features from the inventory data records in the standardized multidimensional data to obtain storage condition features, shelf life features and inventory cost features; The sales frequency feature, the sales fluctuation feature, the sales seasonality feature, the demand stability feature, the demand prediction difficulty feature, the demand suddenness feature, the storage condition feature, the shelf life feature, and the inventory cost feature are combined to form a drug feature vector; The drug feature vector is input into the preset data mining model to obtain drug multi-dimensional data, and the drug classification matrix is constructed based on the drug multi-dimensional data.

3. The method of claim 2, wherein, The drug feature vector is input into the preset data mining model to obtain drug multi-dimensional data, and the drug classification matrix is constructed based on the drug multi-dimensional data. The drug feature vector is input into the preset clustering analysis model to obtain an initial drug category; The drug feature vector is input into the preset association rule mining model to obtain a drug association relationship; The drug feature vector is input into the preset pattern recognition model to obtain a drug demand pattern; The initial drug category, the drug association relationship, and the drug demand pattern are used as the drug multi-dimensional data.

4. The method of claim 3, wherein, The drug feature vector is input into the preset data mining model to obtain drug multi-dimensional data, and the drug classification matrix is constructed based on the drug multi-dimensional data. The drug feature vector is input into the preset data mining model to obtain drug multi-dimensional data, and the drug classification matrix is constructed based on the drug multi-dimensional data. The drug feature vector is input into the preset data mining model to obtain drug multi-dimensional data, and the drug classification matrix is constructed based on the drug multi-dimensional data. The demand influence factor weight matrix is used for parameter calculation to obtain procurement parameters including a safety inventory level, an economic procurement lot size, a procurement trigger point, and a procurement cycle, which include: A weighted average daily demand μ calculated based on the demand influence factor weight matrix, and a demand standard deviation σ calculated based on the weighted average daily demand μ; 5. The method of claim 1, wherein, A safety factor k calculated based on a preset service level target value P and a preset safety factor calculation formula, and a safety inventory level SS determined based on the safety factor k, the demand standard deviation σ, and a preset replenishment lead time L; An annual demand D determined based on the weighted average daily demand μ, and an economic procurement lot size EOQ determined based on the annual demand D, a preset unit procurement cost A, and a preset unit inventory holding cost h; A procurement trigger point ROP determined based on the weighted average daily demand μ, a preset replenishment lead time L, and the safety inventory level SS; A procurement cycle T determined based on the economic procurement lot size EOQ and the weighted average daily demand μ. The system includes a data acquisition module, a first data processing module, a second data processing module, a third data processing module, and a procurement scheme determination module; The data acquisition module is configured to acquire historical sales data, real-time inventory data, and external environment data of a chain pharmacy, and construct a multi-dimensional database based on the historical sales data, the real-time inventory data, and the external environment data; the external environment data includes seasonal information, regional population characteristics, epidemic data, and medical insurance policy information; 6. A big data based chain pharmacy procurement plan generation system, characterized by, ​ ​ The first data processing module is configured to perform standardization processing on data in the multidimensional database to obtain standardized multidimensional data, and construct a drug classification matrix based on the standardized multidimensional data by using a preset data mining model, wherein the data mining model comprises a preset clustering analysis model, a preset association rule mining model, and a preset pattern recognition model. The second data processing module is configured to identify key factors affecting the demand of various drugs by using a causal analysis method based on the drug classification matrix, and establish a demand influencing factor weight matrix according to the key factors. The method comprises the following steps: obtaining historical sales data corresponding to various drugs in the drug classification matrix, and matching the historical sales data with the external environment data according to a time dimension; combining the historical sales data and the external environment data at the same time point together, pairing the data at each time point by using a time window sliding, and ensuring the corresponding relationship between the historical sales data and the external environment data in time to realize time sequence matching and establish an initial causal relationship data set; performing conditional independence test on the initial causal relationship data set to obtain a target variable pair passing the conditional independence test, and calculating a causal strength coefficient of the target variable pair; performing coefficient screening on the causal strength according to a preset coefficient threshold to obtain key factors; calculating direct influence weights and interaction influence weights of the key factors; and constructing a demand influencing factor weight matrix based on the direct influence weights and the interaction influence weights. The third data processing module is configured to perform parameter calculation based on the demand influencing factor weight matrix to obtain procurement parameters including a safety stock level, an economic procurement batch, a procurement trigger point, and a procurement cycle. The procurement scheme determination module is configured to generate a procurement scheme including a procurement drug list, a procurement quantity, a procurement time, and a supplier selection according to the procurement parameters and in combination with a preset supplier evaluation system.

7. A computer readable storage medium characterized in that, The computer readable storage medium stores a plurality of instructions, which are suitable for being loaded and executed by the processor to perform the method of any one of claims 1-5.

8. An electronic device, comprising: The electronic device comprises a processor, a memory, a user interface, and a network interface. The memory is configured to store instructions. The user interface and the network interface are configured to communicate with other devices. The processor is configured to execute the instructions stored in the memory to enable the electronic device to perform the method of any one of claims 1-5.

Images