Online and offline fusion system based on medical instrument sales data analysis

Abstract

The invention discloses an online and offline fusion system based on medical instrument sales data analysis, relates to the technical field of online Internet, and is used for solving the problems that inventory information cannot be synchronized in real time, inventory overstock or stockout phenomena are easily caused, and the service life of the medical instrument is affected due to the fact that existing medical instrument sales data, online sales data and offline sales data are always independent from one another. And the sales efficiency is influenced. The system integrates online and offline sales, inventory and other data and performs systematic analysis, optimizes the supply chain efficiency, ensures that the product inventory can better match the market demand, and meets the purchase demand of consumers.

Description

Technical Field

[0001] The present invention relates to the field of online Internet technology, and more specifically, to an online and offline integration system based on medical device sales data analysis. Background Art

[0002] The primary goal of medical device sales data analysis is to understand market demand, optimize inventory management, and enhance sales strategies through data collection, processing, and analysis, thereby improving a company's operational efficiency and profitability. Online-to-offline (O2O) integration is a business model that integrates online and offline resources to achieve optimal resource allocation, thereby improving a company's operational efficiency and user experience.

[0003] In existing medical device sales, online and offline sales data are often independent of each other, resulting in the inability to synchronize inventory information in real time, which can easily lead to inventory backlogs or stockouts, affecting sales efficiency. Therefore, this application proposes an online and offline integration system based on medical device sales data analysis, aiming to dynamically adjust the inventory level of medical device products and allocate medical device inventory online and offline, thereby reducing inventory costs, improving operational efficiency, optimizing supply chain efficiency, and meeting consumer purchasing needs.

[0004] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an online and offline fusion system based on medical device sales data analysis to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An online and offline integrated system based on medical device sales data analysis, including:

[0008] A data acquisition module is used to collect product information, sales data and inventory data of medical devices, and to pre-process and store the data;

[0009] a data classification module, signal-connected to the data acquisition module, for classifying medical device products using a clustering algorithm based on the sales data and calculating inventory turnover;

[0010] a data matching module, signal-connected to the data classification module, configured to predict future online and offline sales data using a smoothing method based on historical sales data, and to allocate online and offline inventory based on the predicted sales data;

[0011] The feedback adjustment module is respectively connected to the data classification module and the data matching module for dynamically adjusting the inventory level of medical devices through fuzzy rules in combination with the product classification results and the inventory turnover rate.

[0012] In a preferred embodiment, the product information collected by the data acquisition module includes product number, name and type; sales data includes online sales data and offline sales data, the online sales data comes from e-commerce platforms and corporate websites, and the offline sales data comes from physical stores, distributors and agents, and both include sales quantity, frequency, cost and date; inventory data includes beginning inventory, ending inventory and inventory change records.

[0013] In a preferred embodiment, the clustering algorithm used by the data classification module includes the following steps:

[0014] Standardize the monthly sales volume, quarterly sales volume, annual sales volume, sales growth rate and seasonal index in the sales data to form a feature matrix;

[0015] Perform cluster analysis based on the preset number of clusters to obtain at least three product categories;

[0016] The characteristic average of each cluster is calculated, and the cluster with the largest average seasonal index is defined as seasonal products. The remaining clusters are divided into best-selling products and stagnant products according to the sum of the average values ​​of non-seasonal characteristics.

[0017] In a preferred embodiment, the data classification module calculates the inventory turnover rate by: based on the sales cost and average inventory in a preset time period, using the formula "inventory turnover rate = sales cost / average inventory", where the average inventory is the average of the inventory at the beginning and the inventory at the end of the time period.

[0018] In a preferred embodiment, the smoothing method adopted by the data matching module includes:

[0019] Calculate the average sales data of the past preset time period by moving average method as the preliminary forecast value;

[0020] Performing exponential smoothing on the preliminary forecast value based on a smoothing coefficient to obtain final online and offline forecast sales data;

[0021] Inventory is allocated based on the online sales ratio and the offline sales ratio, where the sales ratio is the ratio of the corresponding channel's forecast sales data to the total forecast sales data.

[0022] In a preferred embodiment, the fuzzy rules of the feedback adjustment module include:

[0023] Divide the product classification results into at least three fuzzy sets, and divide the inventory turnover rate into at least three fuzzy sets;

[0024] The inventory level adjustment results are divided into three fuzzy sets: increase, no change and decrease;

[0025] Based on the combined relationship between product classification fuzzy sets and inventory turnover rate fuzzy sets, the adjustment method of inventory level is determined.

[0026] The technical effects and advantages of the online and offline fusion system based on medical device sales data analysis of the present invention are as follows:

[0027] The present invention solves the problems of independent online and offline data and inability to synchronize inventory information in real time in existing medical device sales, avoids inventory backlogs or stockouts, and improves sales efficiency.

[0028] By dynamically adjusting the inventory levels of medical device products and precisely controlling them based on product categories (best-selling, stagnant, and seasonal products) and inventory turnover rates, inventory costs are reduced.

[0029] Based on the smoothing forecast of historical sales data and the analysis of the online and offline sales ratio, we have achieved a reasonable allocation of online and offline inventory, optimized resource allocation, and improved operational efficiency.

[0030] Finally, we integrated online and offline sales, inventory and other data and conducted systematic analysis to optimize supply chain efficiency, ensure that product inventory can better match market demand and meet consumer purchasing needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a schematic diagram of the structure of the online and offline fusion system based on medical device sales data analysis of the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] Example 1, Figure 1 An online and offline integration system based on medical device sales data analysis is proposed, which includes the following steps:

[0034] Step 1: Collect analytical data used to analyze and adjust medical device product inventory levels, including medical device product categories and inventory turnover rates, ensure the accuracy and consistency of the data, pre-process the data, and store it in the database.

[0035] Step 2: Use a clustering algorithm to calculate the medical device product categories and divide them into three clusters: hot-selling products, stagnant products, and seasonal products. Also calculate the inventory turnover rate.

[0036] Step 3: Use fuzzy definition rules to comprehensively analyze the categories and inventory turnover rates of medical device products, thereby achieving dynamic adjustment of medical device product inventory levels.

[0037] Step 4: Based on historical sales data, use the smoothing method to predict future online and offline sales data.

[0038] Step 5: Allocate online and offline inventory based on the predicted online and offline sales data.

[0039] Specifically, in step 1, medical device product categories primarily include best-selling products, stagnant products, and seasonal products. Best-selling products are products with high sales volumes and stable demand. These products typically enjoy high market recognition and market share. Stagnant products are products with low sales volumes, low demand, or unstable demand. These products may be outdated or have low market demand. Seasonal products are products whose demand is significantly affected by the seasons. These products experience high demand during certain time periods and low demand during other time periods. Inventory turnover ratio is an important indicator of a company's inventory management efficiency, indicating the number of times inventory is consumed and replenished within a certain period.

[0040] The data required to analyze medical device product categories and inventory turnover primarily includes product information, sales data, and inventory data. Product information includes product number, name, and type (e.g., consumables, equipment, reagents). Large enterprises typically use ERP systems (software systems that integrate and manage core business processes) to manage product information, including product number, name, and type. This information can be directly exported from the ERP system's product management module or database.

[0041] Sales data includes sales quantity, sales frequency, sales cost, and sales date. Online and offline sales data are collected and integrated to form a unified sales database. Online data collection includes obtaining sales data from online sales channels such as e-commerce platforms and corporate websites. Offline data collection includes obtaining sales data from offline sales channels such as physical stores, distributors, and agents. This data may be obtained through POS systems or manual sales records.

[0042] Inventory data includes opening and closing inventory numbers, inventory changes, and more. Most businesses use an EPR system to manage inventory, which allows for easy export of this data. Small businesses or those without automated systems may manually record inventory data. Inventory management can also be done through Excel spreadsheets, where a simple spreadsheet can be created to record opening and closing inventory numbers, as well as the date, quantity, and reason for each inventory change.

[0043] Specifically, in step 2, based on online and offline sales data, a clustering algorithm is used to calculate the medical device product categories and divide them into three clusters: best-selling products, stagnant products, and seasonal products. The specific process is as follows:

[0044] S1: Collect data. Collect characteristic data of medical device products (data describing a certain characteristic of the station), including but not limited to monthly sales volume, quarterly sales volume, annual sales volume, sales growth rate, seasonal index, etc. The characteristic data is formed into a feature matrix. Assuming there are n medical device products in total and each product has m features, then the feature matrix X is obtained. represents the jth eigenvalue of the i-th product. A sample is a vector containing all eigenvalues. Each sample represents a medical device product and contains a vector of all the features of the product. That is, each sample is an m-dimensional feature vector, where m is the number of features.

[0045] S2: Data preprocessing. Standardize the eigenvalues ​​to ensure they have the same scale and dimension. A common method is to subtract the mean from each eigenvalue and then divide it by its standard deviation to make its mean equal to 0 and standard deviation equal to 1. The specific formula is Where is the standardized eigenvalue, x is the original eigenvalue, α is the mean of the feature, is the standard deviation of the feature.

[0046] S3: Select an appropriate number of clusters. The number of clusters indicates how many clusters or groups the data will be divided into when using a clustering algorithm. Selecting an appropriate number of clusters is a very important step in cluster analysis, which directly affects the quality and interpretability of the final clustering results. According to this embodiment, medical device products are divided into three categories, namely, best-selling products, stagnant products, and seasonal products. Therefore, the number of clusters can be determined to be 3. By selecting the number of clusters as 3, the desired three clusters can be obtained, each cluster representing a product category. At the same time, further adjustments can be made through the elbow rule or silhouette coefficient, which are not described in detail here.

[0047] S4: Select and apply a clustering algorithm. Common clustering algorithms include K-means clustering, hierarchical clustering, and density clustering. Based on the station's characteristic data, this application uses K-means clustering as an example. The advantages of this algorithm are its simplicity, ease of understanding and implementation, and its applicability to large-scale datasets.

[0048] Randomly select k initial cluster centers and assign each sample to the cluster with the closest cluster center. Then, calculate the new center of each cluster (i.e., the mean of the samples within the cluster) and use it as the new cluster center. Repeat the two steps of assigning samples and updating centers until the cluster centers no longer change or the maximum number of iterations is reached. Assigning samples means assigning each sample to the cluster with the closest cluster center, and updating centers means calculating the new center of each cluster (i.e., the mean of the samples within the cluster) and using it as the new cluster center.

[0049] Mark the randomly selected k initial cluster centers as μ1, μ2..., μk. For each sample, calculate its distance to each cluster center. The specific formula is ;In the formula Represents the distance between sample xi and cluster center μj. Assign it to the nearest cluster Cu (u=1, 2, 3), specifically For each cluster Cu, calculate its new cluster center μj, that is, the sample mean within the cluster, , where xi represents the i-th product and contains the feature vector of all the features of the product.

[0050] S5: Explain the clustering results. For each cluster, calculate the average value of each feature expressed as ; where u represents the index of the cluster and j represents the index of the feature. The specific formula is Where is the number of products in cluster Cu, is the jth eigenvalue of the qth product in cluster Cu. For the seasonal index feature, the average seasonal index in each cluster is calculated, and the cluster corresponding to the maximum average seasonal index is defined as the seasonal product. For other features, the average value of the feature in a cluster is added to get the sum . Sum the remaining features excluding the seasonal index feature, that is ; where E represents the sum of the average values ​​of other features (excluding the seasonal index) in the cluster, and m-1 is the number of other features excluding the seasonal index.

[0051] By comparing the size of E, we define two clusters: best-selling products and stagnant products. The best-selling product cluster is the one with the larger sum of the average values ​​of other features (excluding the seasonality index). For example, if E1 is greater than E2, the cluster corresponding to E1 is a best-selling product cluster.

[0052] It should be noted that the seasonal product cluster is defined only based on the seasonal index. Therefore, this application provides a seasonal index calculation method as follows:

[0053] First, collect historical sales data through methods such as the EPR system and organize them by month. For each month, calculate the average sales data for each year. For example, calculate the average sales data for all January, the average sales data for all February, and so on. The specific formula is ;in is the number of years, i.e. the number of years collected in the past Historical sales data.

[0054] Next, calculate the overall average sales volume, specifically Finally, the seasonal index is calculated. The seasonal index of each month is equal to the average sales volume of that month divided by the overall average sales volume, which is ; where j represents the month, ranging from 1 to 12.

[0055] The specific process of calculating inventory turnover ratio is as follows:

[0056] Collect data, which includes sales data and inventory data, including sales cost (the total cost of goods sold in a certain period), sales quantity, beginning inventory, and ending inventory. The beginning inventory is the inventory quantity in a specific period, and the ending inventory quantity, and sales cost is the cost of goods sold in a specific period.

[0057] The basic calculation formula using inventory turnover ratio is ; The calculation formula for average inventory is Specifically, you can calculate monthly inventory first, then, depending on your needs, calculate quarterly or annual inventory turnover rates. Calculating inventory turnover rates can be used to analyze a company's inventory management efficiency. A higher inventory turnover rate indicates more frequent inventory updates and better inventory management efficiency; a lower inventory turnover rate indicates an inventory backlog and lower management efficiency.

[0058] It should be noted that the calculation based on medical device product categories and inventory turnover rates requires data collected within the same time frame, such as product information, sales data, inventory data, etc. in the past year, to facilitate subsequent comprehensive analysis of the two.

[0059] Specifically, in step 3, a comprehensive analysis of medical device product categories and inventory turnover rates is conducted to dynamically adjust the inventory levels of medical device products. The specific process is as follows:

[0060] Define medical device product category and inventory turnover rate as input variables and divide them into different fuzzy sets. For example, "Bad", "Medium", and "Good" are for medical device product category, and "Low", "Medium", and "High" are for inventory turnover rate.

[0061] The adjustment of the inventory level of medical device products is defined as the output variable and divided into different fuzzy sets. For example, "Decrease", "Unchanged", and "Increase" are used for the inventory level of medical device products. "Decrease" means reducing the inventory level of medical device products, "Unchanged" means that the inventory level of medical device products remains unchanged, and "Increase" means increasing the inventory level of medical device products.

[0062] Develop fuzzy rules to describe the impact of medical device product category and inventory turnover on inventory levels. Rules can be defined based on industry expertise or obtained through data analysis and experimentation. For example, label the medical device product category η, the inventory turnover rate A, and the adjusted medical device product inventory level R. Then, we can define:

[0063] Rule1: IF (η is Good) AND (A is High) THEN(R is Increase)

[0064] Rule2: IF (η is Bad) AND (A is Low) THEN(R is Decrease)

[0065] Rule3: IF (η is Good) AND (A is Medium) THEN(R is Unchanged) ......

[0067] Fuzzy reasoning is performed based on fuzzy rules to dynamically adjust the operating time of ventilation equipment.

[0068] It should be noted that the division of fuzzy sets can be adjusted according to actual conditions. For example, this embodiment takes three fuzzy sets as an example. In fact, the medical device product categories, inventory turnover rates, and adjusted inventory levels can be divided into more than three sets, so that the inventory levels of medical device products can be adjusted dynamically more conveniently.

[0069] Furthermore, for the judgment of whether the medical device product categories are good, medium, or bad, all medical device products in the best-selling product cluster can be marked as "Good", etc., as shown in step S5, which will not be repeated here; for the judgment of whether the inventory levels are low, medium, or high, the inventory turnover rate threshold can be set to make a judgment. The inventory turnover rates of each month or quarter can be sorted from large to small, and the 66th percentile (the value at the 66th percentile position) is selected as the first inventory turnover rate threshold, and the 33% percentile is selected as the second inventory turnover rate threshold. When the inventory turnover rate is greater than or equal to the first inventory turnover rate threshold, it is marked as "High", etc., which will not be repeated here.

[0070] Specifically, in step 4, smoothing is a statistical method used to process time series data. The two most common methods are simple moving average and simple exponential smoothing. In this embodiment, simple moving smoothing can be used to smooth the historical sales data first, and then simple exponential smoothing can be used to further smooth and predict the moving average data. The details are as follows:

[0071] B1: Data Collection. Using methods such as an EPR system, collect historical online and offline sales data for all medical device products over the past month (a month) and organize it by month. For each medical device product, use its historical sales data from the past year to predict sales data for a future period (one month in this example).

[0072] B2: Calculate the initial forecast value. Select a time window (the past a month) and calculate the average monthly sales data for the past a month for a given time t (the next month). The specific calculation formula is: Where is the moving average sales data at time t, is the actual sales data at time ti, and a is the length of the time window. The calculated moving average is used as the preliminary sales data forecast.

[0073] B3: Simple exponential smoothing. Select a smoothing coefficient (usually between 0 and 1), and use simple exponential smoothing to further smooth the moving average data. The calculation formula is: Where is the exponentially smoothed value at time t, is the moving average sales data value at time t, is the exponential smoothing value at time t-1, is the smoothing coefficient. The calculated exponential smoothing value is used as the final sales data forecast value, that is, .

[0074] Calculate and calculate the predicted sales data for both online and offline sales for further analysis. It's important to note that the predicted sales data is used for online and offline inventory allocation. The sales data here should be quantitative, meaning the number or volume of medical device sales. This is because online and offline inventory allocation requires ensuring that the actual product quantity meets market demand, not just the sales amount or other metrics.

[0075] Specifically, in step 5, the online sales ratio and offline sales ratio are calculated respectively. The calculation formula is: ; ;in and Forecast sales data respectively.

[0076] The online and offline inventory of medical device products can be calculated based on the sales ratio and total inventory. For example If total inventory levels are unknown (e.g., medical devices not yet manufactured), online and offline replenishment can be conducted based on forecasted sales data to ensure inventory reaches at least the forecasted sales value. Safety stock levels can also be set based on specific circumstances and market demand to mitigate impacts of seasonal demand, weather, and other factors. This is not detailed here.

[0077] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0078] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0079] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0080] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0081] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0082] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An online and offline fusion system based on medical device sales data analysis, characterized by: include: A data acquisition module is used to collect product information, sales data and inventory data of medical devices, and to pre-process and store the data; a data classification module, signal-connected to the data acquisition module, for classifying medical device products using a clustering algorithm based on the sales data and calculating inventory turnover; a data matching module, signal-connected to the data classification module, configured to predict future online and offline sales data using a smoothing method based on historical sales data, and to allocate online and offline inventory based on the predicted sales data; The feedback adjustment module is respectively connected to the data classification module and the data matching module for dynamically adjusting the inventory level of medical devices through fuzzy rules in combination with the product classification results and the inventory turnover rate.

2. The online and offline integration system based on medical device sales data analysis according to claim 1 is characterized by: The product information collected by the data collection module includes product number, name and type; sales data includes online sales data and offline sales data. The online sales data comes from e-commerce platforms and corporate websites, and the offline sales data comes from physical stores, distributors and agents, and both include sales quantity, frequency, cost and date; inventory data includes beginning inventory, ending inventory and inventory change records.

3. The online and offline integration system based on medical device sales data analysis according to claim 1 is characterized by: The clustering algorithm used by the data classification module includes the following steps: Standardize the monthly sales volume, quarterly sales volume, annual sales volume, sales growth rate and seasonal index in the sales data to form a feature matrix; Perform cluster analysis based on the preset number of clusters to obtain at least three product categories; The characteristic average of each cluster is calculated, and the cluster with the largest average seasonal index is defined as seasonal products. The remaining clusters are divided into best-selling products and stagnant products according to the sum of the average values ​​of non-seasonal characteristics.

4. The online and offline integration system based on medical device sales data analysis according to claim 1 is characterized in that ； The data classification module calculates inventory turnover rate by using the formula "inventory turnover rate = sales cost / average inventory" based on sales cost and average inventory within a preset time period, where average inventory is the average of the inventory at the beginning and end of the time period.

5. The online and offline integration system based on medical device sales data analysis according to claim 1 is characterized in that: The smoothing method adopted by the data matching module includes: Calculate the average sales data of the past preset time period by moving average method as the preliminary forecast value; Performing exponential smoothing on the preliminary forecast value based on a smoothing coefficient to obtain final online and offline forecast sales data; Inventory is allocated based on the online sales ratio and the offline sales ratio, where the sales ratio is the ratio of the corresponding channel's forecast sales data to the total forecast sales data.

6. The online and offline integration system based on medical device sales data analysis according to claim 1 is characterized in that: The fuzzy rules of the feedback regulation module include: Divide the product classification results into at least three fuzzy sets, and divide the inventory turnover rate into at least three fuzzy sets; The inventory level adjustment results are divided into three fuzzy sets: increase, no change and decrease; Based on the combined relationship between product classification fuzzy sets and inventory turnover rate fuzzy sets, the adjustment method of inventory level is determined.

Images