A medical product sales data intelligent management method and system

Abstract

The present invention relates to the field of data management technology, and specifically to an intelligent management method and system for medical product sales data, including: a product data acquisition module, a product data storage management module, a data processing automatic judgment module and sales data transmission data, wherein the product data acquisition module collects medical product type information and historical sales data of medical products, and the product data storage management module performs data storage planning management on the use information of medical products after sales, and the data processing automatic judgment module establishes a data processing selection model for automatically judging whether to trigger a data storage management operation, and performs an automatic storage management operation when the triggering condition is met, and the sales data transmission module transmits data to a requesting terminal after receiving a product data retrieval request, thereby providing auxiliary functions for relevant personnel to analyze product sales data, and helping relevant personnel to quickly understand product sales data so as to better analyze the sales situation of medical products.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a method and system for intelligent management of medical product sales data. Background Art

[0002] During the sales process of medical products, sales and post-sales data are generated. Sales and post-sales data refers to the data generated when customers need follow-up services after purchasing and starting to use the products. During the follow-up services, customer feedback on product usage is usually generated. When analyzing product sales and usage, quickly obtaining customer feedback on product usage information can, to a certain extent, understand the product sales and usage in order to understand product quality issues, customer satisfaction and other information, thereby helping to improve products and enhance customer experience.

[0003] However, the existing technology generally stores the usage information of the same type of medical products sold to users together, and when relevant personnel need to retrieve the product usage information, all the information is transmitted to the terminal where the relevant personnel is located. The relevant personnel can only obtain the user's approximate feedback information on the corresponding medical product. The existing technology does not perform storage planning and management on sales and after-sales data, and cannot help relevant personnel accurately and quickly obtain and understand the product usage information fed back by users at different times or stages to analyze the usage effects of the product at different stages of use, which is not conducive to helping relevant personnel better analyze the sales of medical products.

[0004] Therefore, people need an intelligent management method and system for medical product sales data to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide a medical product sales data intelligent management method and system to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: an intelligent management system for medical product sales data, the system comprising: a product data acquisition module, a product data storage management module, a data processing automatic judgment module and a sales data transmission module;

[0007] Collecting medical product type information and historical sales data of medical products through the product data collection module;

[0008] Retrieving historical sales data of medical products through the product data storage management module, analyzing the time concentration of user feedback on the use of medical products, selecting preliminary storage management measures for the use information based on the concentration, analyzing the data management results after the preliminary storage management, and performing secondary storage management based on the data management results;

[0009] A data processing selection model for automatically judging whether to trigger a data storage management operation is established through the data processing automatic judgment module, and an automatic storage management operation is performed when the triggering condition is met;

[0010] After receiving the product data retrieval request, the sales data transmission module transmits the data to the requesting terminal.

[0011] Further, the product data collection module includes a product type collection unit, a usage data collection unit and an after-sales information collection unit;

[0012] The product type collection unit is used to collect information on the types of medical products sold in the past;

[0013] The usage data collection unit is used to collect the start time information of the user's previous use of different types of medical products;

[0014] The after-sales information collection unit is used to collect the time when the user feedbacks the product usage information after starting to use the product. The feedback product usage information refers to the use experience or use effect information feedback by the user after using a certain product, for example: the use experience or use effect information of the implant feedback by the user after using the dental implant;

[0015] The data collected by the product data collection module is collected after the user grants permission.

[0016] Further, the product data storage management module includes a feedback concentration analysis unit, a division preprocessing unit, a preprocessing data dispersion judgment unit and a data integration processing unit;

[0017] The feedback concentration analysis unit is used to analyze the feedback time concentration of different types of product usage information;

[0018] The division preprocessing unit is used to set a concentration threshold, compare the feedback time concentration and the concentration threshold: if the feedback time concentration does not exceed the concentration threshold, the product usage information initially stored in the same location by default is divided and preprocessed: the product usage information is divided according to the feedback time, and the product usage information divided into the same group is stored in the same location, and the division preprocessing operation is a preliminary storage management measure; if the feedback time concentration exceeds the concentration threshold, no division preprocessing is performed;

[0019] The pre-processing data dispersion judgment unit is used to analyze the time dispersion index of the product usage information stored in different locations after the division pre-processing is performed, and the time dispersion index obtained by the analysis is the data management result after taking the preliminary storage management measures;

[0020] The data integration processing unit is used to set a dispersion index threshold and compare the time dispersion index and the threshold: if the time dispersion index exceeds the threshold, the data located in different storage locations after preprocessing is reintegrated, and the product usage information in the same group after the integration is stored in the same location; if the time dispersion index does not exceed the threshold, data integration is not performed, and the process of selecting whether to reintegrate the data and store the product usage information in the same group after the integration in the same location is a secondary storage management process.

[0021] Further, the data processing automatic judgment module includes a medical product number creation unit, a processing necessity analysis unit, a processing selection model establishment unit and a data automatic processing unit;

[0022] The medical product number creation unit is used to create numbers for different types of medical products;

[0023] The processing necessity analysis unit is used to count the number of times and time that the usage information of medical products with different numbers is divided and preprocessed, and analyze the necessity index of dividing and preprocessing the usage information of different types of medical products and the frequency of processing of the usage information based on the statistical results;

[0024] The processing selection model building unit is used to map the processing frequency and necessary index of the product usage information to form data points in a two-dimensional coordinate system, and to establish a data processing selection model after performing linear fitting on the data points;

[0025] The data automatic processing unit is used to substitute the frequency of processing of the current product usage information into the data processing selection model to predict the necessary index for dividing and preprocessing the usage information of the current product, set the necessary index threshold, and if the predicted necessary index exceeds the necessary index threshold, the product usage information is automatically divided and preprocessed when the user who uses the current product subsequently feedbacks the product usage information. After the product data storage management operation has been performed a certain number of times, the regularity of the feedback of usage information of different types of products can be observed to a certain extent. The usage information of some products is divided and preprocessed many times and frequently, and it is judged that the user's feedback information on the corresponding product is generally too much and the feedback time is relatively scattered, which needs to be processed. The more necessary it is to perform data division and preprocessing, the purpose of establishing the data processing selection model is that there is no need to analyze the feedback time concentration of different types of product usage information to determine whether it is necessary to divide and preprocess the product usage information fed back by the user, but to automatically determine whether to divide and preprocess the product usage information based on the model output results, thereby reducing the workload of the management system data processing.

[0026] Further, the sales data transmission module includes a data retrieval request receiving unit and a product data transmission unit;

[0027] The data retrieval request receiving unit is used to receive a request for retrieving user feedback information of any product sent by a terminal where a relevant person is located, and the terminal where the relevant person is located is the requesting terminal;

[0028] The product data transmission unit is used to retrieve all the product usage information stored in the corresponding locations and transmit them to the terminal where the relevant personnel are located after receiving the request, if the corresponding product usage information fed back by all users is stored in the same location; if the corresponding product usage information fed back by the users is stored in different locations, retrieve and transmit the product usage information stored in different locations to the terminal where the relevant personnel are located in a random order, and at the same time transmit the average interval duration information between the time when the product usage information stored in the corresponding location is fed back and the time when the user starts using the product to the terminal where the relevant personnel are located. The purpose of feeding back the average interval duration while feeding back the product usage information is to provide the relevant personnel with the product usage information fed back at a similar time and the specific feedback time. The feedback time reflects the time when the user uses the product, which is conducive to helping the relevant personnel accurately know and analyze the usage information fed back by users at different times or stages to analyze the usage effects of the product at different stages of use.

[0029] A method for intelligent management of medical product sales data, comprising the following steps:

[0030] S100: Collect medical product type information and historical sales data of medical products;

[0031] S200: Retrieving historical sales data of medical products, analyzing the time concentration of user feedback on the use of medical products, selecting preliminary storage management measures for the use information based on the concentration, analyzing the data management results after the preliminary storage management, and performing secondary storage management based on the data management results;

[0032] S300: Establishing a data processing selection model for automatically determining whether to trigger a data storage management operation, and performing an automatic storage management operation when a trigger condition is met;

[0033] S400: After receiving the product data retrieval request, the data is transmitted to the requesting terminal.

[0034] Further, in S100: information on the types of medical products sold in the past is collected, information on the start time of users using different types of medical products in the past is collected, and the start time of a random medical product being used at random is obtained as T. A total of n users start using the corresponding medical product at time T. After collecting n users starting to use the corresponding medical product at time T, the time set at which the use information of the corresponding medical product is fed back is {t 1 , t 2 ,…,t i ,…tm}, t i It represents the time when the i-th item of usage information is fed back, m represents the number of usage information items fed back after the corresponding medical product is used at time T, and each piece of usage information fed back by the user is recorded as one item of usage information. Since a user may feed back usage information more than once after starting to use the product, there are m items instead of n items in the above set. The feedback data of usage information of different types of medical products after they are started to be used at different times are collected, and the usage information is fed back by users who use the corresponding products.

[0035] Further, in S200: according to the formula E=(1 / m)+[1 / [(∑ m i=1 ((t i -T)-(∑ m i=1 (t i -T)) / m) 2 ) / m] 1 / 2 ] Calculate the time concentration E of the feedback of the use information of a random medical product after it is used randomly once, and set the concentration threshold to E ’ , compare E and E ’ :If E≤E ’ , the product usage information initially stored in the same location by default is divided and preprocessed: the m items of usage information are arranged in the order of the interval time between the feedback time and T from small to large, and the arranged m items of usage information are randomly divided into c groups, and a random grouping result is obtained. The average value set of the interval time between the feedback time and T of each group of usage information in group c is V={V 1 , V 2 , …, V j ,…V c},V j It represents the average time interval between the feedback time of the j-th group usage information and T in a random grouping result. According to the formula Q=(∑ c j=1 (V j -(∑ c j=1 (V j )) / c) 2 ) / cCalculate the goodness metric value Q of a random grouping result, calculate the goodness metric values ​​of different grouping results, select the grouping result with the largest goodness metric value as the optimal grouping result, and store the product usage information of the same group in the optimal grouping result in the same location; if E>E ’ , without partitioning preprocessing; after partitioning preprocessing, the average interval time set between the time when the product usage information stored in different locations is fed back and T is K={K 1, K 2 , …, K j , …K c} , K j represents the average interval time between the time when the product usage information stored in the j-th position is fed back and T. According to W = [(∑ c j=1 (K j -(∑ c j=1 (K j )) / c) 2 ) / c] 1 / 2 calculate the time dispersion index W of the product usage information stored at different positions after the partitioning preprocessing. Set the dispersion index threshold w, and compare W and w: If W > w, re-integrate the data located at different storage positions after the partitioning preprocessing: Arrange the c groups of usage information in ascending order of the average interval time in the set K, and perform secondary grouping on the arranged c groups of usage information: Divide them into f groups in total, where f < c. Among them, the usage information within each of the c groups is not split, that is, the usage information originally in the same group remains in the same group after secondary grouping. Obtain a random secondary grouping result. The set of average values of the interval times between the feedback times of the usage information in each of the f groups and T is v = {v 1 , v 2 , …, v e , …v f}, where v e represents the average value of the interval time between the feedback time of the usage information in the e-th group and T in a random secondary grouping result. According to the formula Q ’ =(∑ f e=1 (v e -(∑ f e=1 (v e )) / f) 2 ) / f calculate the goodness-of-fit measure value Q ’ of a random secondary grouping result. Calculate the goodness-of-fit measure values of different secondary grouping results, and select the secondary grouping result with the largest goodness-of-fit measure value as the optimal secondary grouping result. Store the usage information in the same group in the optimal secondary grouping result at the same position, that is, integrate and store the usage information of some groups in the original c groups of usage information together at the same position; If W ≤ w, choose not to perform data integration processing;

[0036] The purpose of secondary storage management of data is to take into account the problem that the storage location of product usage information may be too dispersed after division and preprocessing. Excessive dispersion of storage locations may slow down the data retrieval speed when product data needs to be retrieved. Therefore, after the data is divided and preprocessed, the time dispersion index is analyzed to determine whether the preprocessed stored data is too dispersed. When the dispersion index exceeds the threshold, secondary storage management is performed to integrate and process some group usage information of the usage information originally divided into group c, and store them in the same location after integration. This is conducive to helping relevant personnel accurately obtain and understand the usage information fed back by users at different times or stages to analyze the usage effects of the product at different usage stages, and at the same time avoid the situation where the data retrieval speed is slowed down or even the retrieval data is incomplete due to the excessive dispersion of the stored medical product usage information.

[0037] Further, in S300: create numbers for different types of medical products, each number corresponds to a type of medical product, and the number of times the usage information of medical products with different numbers is divided and pre-processed is A={A 1 , A 2 , …A p , …A g}, the number of times in set A is greater than 2, g represents the total number, according to H p =A p / (∑ g p=1 (A p )) Obtain the necessary index H for the pre-processing of the usage information of a randomly numbered medical product p , the necessary index set is {H 1 , H 2 , …H p , …H g}, the time interval set for the collection of the usage information of a randomly numbered medical product to be divided and preprocessed is U={U 1 , U 2 , …U d , …U r}, r+1 represents the number of times the usage information of a randomly numbered medical product is divided and preprocessed, U d The frequency of processing of the usage information of a random medical product with a number L is obtained by representing the time interval between the time when the usage information of a random medical product with a number is divided and preprocessed for the d+1th time and the time when the usage information of the random medical product with a number is divided and preprocessed for the dth time. p , L p =(∑ r d=1 U d) / r, the time intervals for collecting the usage information of medical products with different numbers and dividing them for preprocessing are obtained, and the frequency set of the usage information of medical products with different numbers being processed is obtained as {L 1 , L 2 , …L p , …L g}, map the frequency and necessary index to a two-dimensional coordinate system to form data points {(L 1 , H 1 ), L 2 , H 2 ),…(L p , H p ),…(L g , H g )}, after performing straight line fitting on the data points, a data processing selection model is established: y=β1*x+β2, where β1 and β2 are fitting coefficients. The frequency of processing of the usage information fed back by users of the corresponding type of product currently sold in the past is obtained as J, and J is substituted into the model: Let x=J, and the necessary index for pre-processing the usage information of the current product is predicted to be: β1*J+β2. The threshold of the necessary index is set to G. If β1*J+β2>G, β1*J+β2>G means that the trigger condition is met. After the user who subsequently uses the current product feeds back the product usage information, the product usage information is automatically divided and pre-processed. The automatic division and pre-processing operation method is the same as the operation method in step S200.

[0038] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0039] The present invention uses big data to analyze the feedback time of the use information of sold medical products stored in the same location by default, and determines whether it is necessary to divide and pre-process the product use information. When the feedback time difference is large, that is, the time is too dispersed, the information is divided and pre-processed according to the feedback time of the product use information, which is conducive to helping relevant personnel to retrieve product use information with similar user feedback time when they need to understand the product sales and after-sales situation; considering that the storage location of the product use information after the division and pre-processing may be too dispersed, and the too dispersed storage location may slow down the data retrieval speed when the product data needs to be retrieved, after the data is divided and pre-processed, the time dispersion index is analyzed to determine whether the pre-processed stored data is too dispersed. When the dispersion index exceeds the threshold, secondary storage management is performed, and the use information of some groups in the use information originally divided into c groups is integrated and processed, and stored in the same location after the integration and processing, which is conducive to helping relevant personnel accurately know and understand the use information fed back by users at different times or stages to analyze the use effect of the product at different use stages, and at the same time avoids the situation that the data retrieval speed is slowed down or even the retrieval data is incomplete due to the too dispersed distribution of the stored medical product use information;

[0040] When the product data storage management operation has been performed a certain number of times, the feedback patterns of usage information of different types of products can be observed to a certain extent. The usage information of some products is divided and pre-processed many times and frequently. It is judged that the user's feedback information on the corresponding product is generally too much and the feedback time is relatively scattered. Therefore, it is necessary to perform data division and pre-processing. By establishing a data processing selection model, there is no need to analyze the feedback time concentration of different types of product usage information to determine whether it is necessary to divide and pre-process the product usage information fed back by users. Instead, it is automatically determined whether to divide and pre-process the product usage information based on the model output results, reducing the workload of the management system data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0042] Figure 1 It is a structural diagram of an intelligent management system for medical product sales data of the present invention;

[0043] Figure 2 It is a flow chart of a medical product sales data intelligent management method of the present invention. DETAILED DESCRIPTION

[0044] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0045] Combine the following Figure 1-Figure 2 The present invention is further described with reference to the accompanying drawings and specific embodiments.

[0046] Example 1: Figure 1 As shown, this embodiment provides an intelligent management system for medical product sales data, the system includes: a product data acquisition module, a product data storage management module, a data processing automatic judgment module and sales data transmission data;

[0047] Collect medical product type information and historical sales data of medical products through the product data collection module;

[0048] Retrieve historical sales data of medical products through the product data storage management module, analyze the time concentration of user feedback on the use of medical products, select preliminary storage management measures for the use information based on the concentration, analyze the data management results after preliminary storage management, and perform secondary storage management based on the data management results;

[0049] A data processing selection model for automatically judging whether to trigger a data storage management operation is established through a data processing automatic judgment module, and an automatic storage management operation is performed when the triggering condition is met;

[0050] After receiving the product data retrieval request, the sales data transmission module transmits the data to the requesting terminal;

[0051] The output end of the product data acquisition module is connected to the input end of the product data storage management module, the output end of the product data storage management module is connected to the input end of the data processing automatic judgment module, and the output end of the data processing automatic judgment module is connected to the input end of the sales data transmission module.

[0052] The product data collection module includes a product type collection unit, a usage data collection unit, and an after-sales information collection unit;

[0053] The product type collection unit is used to collect information on the types of medical products that have been sold in the past;

[0054] The usage data collection unit is used to collect the start time information of the user's previous use of different types of medical products;

[0055] The after-sales information collection unit is used to collect the time when users feedback product usage information after starting to use the product. The feedback product usage information refers to the user's feedback on the use experience or use effect information after using a certain product, for example: the user's feedback on the use experience or use effect information of dental implants after using them;

[0056] The output end of the product type acquisition unit is connected to the input end of the data acquisition unit.

[0057] The product data storage management module includes a feedback concentration analysis unit, a division preprocessing unit, a preprocessing data dispersion judgment unit and a data integration processing unit;

[0058] The feedback concentration analysis unit is used to analyze the feedback time concentration of different types of product usage information;

[0059] The division preprocessing unit is used to set a concentration threshold, compare the concentration of the feedback time and the concentration threshold: if the concentration of the feedback time does not exceed the concentration threshold, the product usage information initially stored in the same location by default is divided and preprocessed: the product usage information is divided according to the feedback time, and the product usage information divided into the same group is stored in the same location. The division preprocessing operation is a preliminary storage management measure; if the concentration of the feedback time exceeds the concentration threshold, no division preprocessing is performed;

[0060] The pre-processing data dispersion judgment unit is used to analyze the time dispersion index of the product usage information stored in different locations after the division pre-processing, and the time dispersion index obtained by the analysis is the data management result after taking preliminary storage management measures;

[0061] The data integration processing unit is used to set a dispersion index threshold and compare the time dispersion index with the threshold: if the time dispersion index exceeds the threshold, the data located in different storage locations after preprocessing are reintegrated, and the product usage information in the same group after the integration is stored in the same location; if the time dispersion index does not exceed the threshold, the data integration is not performed, and the process of selecting whether to reintegrate the data and storing the product usage information in the same group after the integration in the same location is a secondary storage management process;

[0062] The input end of the feedback concentration analysis unit is connected to the output ends of the usage data collection unit and the after-sales information collection unit, the output end of the feedback concentration analysis unit is connected to the input end of the division preprocessing unit, the output end of the division preprocessing unit is connected to the input end of the preprocessing data dispersion judgment unit, and the output end of the preprocessing data dispersion judgment unit is connected to the input end of the data integration processing unit.

[0063] The data processing automatic judgment module includes a medical product number creation unit, a processing necessary analysis unit, a processing selection model establishment unit and a data automatic processing unit;

[0064] The medical product number creation unit is used to create numbers for different types of medical products;

[0065] The processing necessity analysis unit is used to count the number of times and time that the usage information of medical products with different numbers is divided and preprocessed, and analyze the necessity index of dividing and preprocessing the usage information of different types of medical products and the frequency of processing of the usage information based on the statistical results;

[0066] The processing selection model building unit is used to map the processing frequency and necessary index of the product usage information to form data points in a two-dimensional coordinate system, and to establish a data processing selection model after performing linear fitting on the data points;

[0067] The data automatic processing unit is used to substitute the frequency of processing of the current product usage information into the data processing selection model to predict the necessary index for pre-processing the current product usage information, set the necessary index threshold, and if the predicted necessary index exceeds the necessary index threshold, automatically pre-process the product usage information when the user who uses the current product subsequently feeds back the product usage information;

[0068] The input end of the medical product number creation unit is connected to the output end of the data integration processing unit, the output end of the medical product number creation unit is connected to the input end of the processing necessary analysis unit, the output end of the processing necessary analysis unit is connected to the input end of the processing selection model establishment unit, and the output end of the processing selection model establishment unit is connected to the input end of the data automatic processing unit.

[0069] The sales data transmission module includes a data retrieval request receiving unit and a product data transmission unit;

[0070] The data retrieval request receiving unit is used to receive a request for retrieval of user feedback information of any product sent by a terminal where a relevant person is located, and the terminal where the relevant person is located is the request terminal;

[0071] The product data transmission unit is used for, after receiving a request, if the corresponding product usage information fed back by all users is stored in the same location, retrieving all the product usage information stored in the corresponding location and transmitting it to the terminal where the relevant personnel are located; if the corresponding product usage information fed back by the users is stored in different locations, retrieving and transmitting the product usage information stored in different locations in a random order to the terminal where the relevant personnel are located, and at the same time transmitting the average interval duration information between the time when the product usage information stored in the corresponding location is fed back and the time when the user starts to use the product to the terminal where the relevant personnel are located, the average interval duration is obtained by summing the interval duration between the time when the product usage information stored in the corresponding location is fed back and the time when the user starts to use the product and then dividing it by the number of users, for example: a certain location stores the corresponding product usage information fed back by 3 users, and the interval durations between the time when the 3 users fed back the information and the time when the corresponding users start to use the product are a1, a2 and a3 respectively, then the average interval duration is (a1+a2+a3) / 3;

[0072] The input end of the data retrieval request receiving unit is connected to the output end of the data automatic processing unit, and the output end of the data retrieval request receiving unit is connected to the input end of the product data transmission unit.

[0073] Example 2: Figure 2 As shown, this embodiment provides a medical product sales data intelligent management method, which is implemented based on the data management system in the embodiment and specifically includes the following steps:

[0074] S100: Collecting medical product type information and historical sales data of medical products: Collecting information on the types of medical products that have been sold in the past, collecting information on the start time of users using different types of medical products in the past, obtaining a random start time of a medical product being used once as T, and a total of n users starting to use the corresponding medical product at time T. After collecting n users starting to use the corresponding medical product at time T, the time set of feedback of the use information of the corresponding medical product is {t 1 , t 2 ,…,t i ,…t m}, t i represents the time at which the i-th item of usage information is fed back, m represents the number of usage information items fed back after the corresponding medical product is used at time T, and each piece of usage information fed back by the user is recorded as one piece of usage information. Since a user may feed back usage information more than once after starting to use the product, there are m items in the above set instead of n items. Feedback data on usage information of different types of medical products after they are started to be used at different times are collected, and the usage information is fed back by users who use the corresponding products;

[0075] S200: Retrieving historical sales data of medical products, analyzing the time concentration of user feedback on the use of medical products, selecting preliminary storage management measures for the use information based on the concentration, analyzing the data management results after the preliminary storage management, and performing secondary storage management based on the data management results;

[0076] The process of analyzing concentration and taking preliminary storage management measures for usage information is as follows:

[0077] According to the formula E=(1 / m)+[1 / [(∑ m i=1 ((t i -T)-(∑ m i=1 (t i -T)) / m) 2 ) / m] 1 / 2] Calculate the time concentration E of the feedback of the use information of a random medical product after it is used randomly once, and set the concentration threshold to E ’ , compare E and E ’ :If E≤E ’ , the product usage information initially stored in the same location by default is divided and preprocessed: the m items of usage information are arranged in the order of the interval time between the feedback time and T from small to large, and the arranged m items of usage information are randomly divided into c groups, and a random grouping result is obtained. The average value set of the interval time between the feedback time and T of each group of usage information in group c is V={V 1 , V 2 , …, V j ,…V c},V j It represents the average time interval between the feedback time of the j-th group usage information and T in a random grouping result. According to the formula Q=(∑ c j=1 (V j -(∑ c j=1 (V j )) / c) 2 ) / cCalculate the goodness metric value Q of a random grouping result, calculate the goodness metric values ​​of different grouping results, select the grouping result with the largest goodness metric value as the optimal grouping result, and store the product usage information of the same group in the optimal grouping result in the same location; if E>E ’ , no partition preprocessing is performed;

[0078] Product usage information is stored in the memory, and different storage locations represent different storage addresses;

[0079] For example, the start time of a random medical product (a certain dental implant) being used once is T, and a total of 8 users are counted to start using the certain dental implant at time T. After the 8 users start using the certain dental implant at time T, the time set of the corresponding dental implant usage information being fed back is {t 1 , t 2 , t 3 , t 4 , t 5 , t 6 , t 7 , t 8 , t 9 , t 10}, after 8 users start using the corresponding dental implant at time T, the interval time set between the time when the corresponding dental implant usage information is fed back and T is {5, 1, 15, 30, 100, 50, 12, 10, 6, 120}, in days. The concentration of the time when the corresponding medical product usage information is fed back after the corresponding dental implant is randomly used is calculated to be E≈0.13, and the concentration threshold is set to E ’ =0.7, E <E ’ , the usage information of the corresponding dental implants initially stored in the same location by default is divided and preprocessed: the 10 items of usage information are arranged in the order of the interval time between the feedback time and T from small to large: the interval time is {1, 5, 6, 10, 12, 15, 30, 50, 100, 120} from small to large, and the 10 items of usage information after arrangement are randomly divided into 5 groups, and the results of a random grouping are obtained: the 5 groups of usage information are respectively {1, 5, 6}, {10, 12, 15}, {30, 50}, {100} and {120} according to the interval time between the feedback time and T, and the average value set of the interval time between the feedback time and T of each group of usage information in the 5 groups is V={V 1 , V 2 , V 3 , V 4 , V 5}={4, 12.3, 40, 100, 120}, calculate the goodness metric value Q≈2180 of the corresponding grouping result, calculate the goodness metric values ​​of different grouping results, select the grouping result with the largest goodness metric value as the optimal grouping result, and the optimal grouping result is: the 5 groups of usage information are respectively {1, 5, 6, 10, 12, 15}, {30}, {50}, {100} and {120} according to the interval time sets between the feedback time and T, and store the product usage information divided into the same group in the optimal grouping result in the same location, for example: the dental implant usage information with the feedback interval time of {1, 5, 6, 10, 12, 15} is stored in the same location;

[0080] The process of analyzing data management results and performing secondary storage management on data is as follows:

[0081] After the division preprocessing, the average interval time set between the time when the product usage information stored in different locations is fed back and T is obtained as K={K 1 , K 2 , …, K j , …K c}, K jDenote the average interval between the time when the product usage information stored at the j-th position is fed back and T. For example, after partitioning and preprocessing, the set of intervals between the time when the product usage information stored in a certain position is fed back and T is {1, 5, 6, 10, 12, 15}, then the average interval at this time is the average value 8.2 of the values in the corresponding set. According to W = [(∑ c j=1 (K j -(∑ c j=1 (K j )) / c) 2 ) / c] 1 / 2 Calculate the dispersion index W of the time when the product usage information stored at different positions after partitioning and preprocessing is fed back. Set the dispersion index threshold w, and compare W and w: If W > w, re-integrate the data located at different storage positions after partitioning and preprocessing: Arrange the c groups of usage information in ascending order of the average interval in the set K, and perform secondary grouping on the arranged c groups of usage information: A total of f groups are divided, f < c, where the usage information within each group among the c groups is not split, that is, the usage information originally in the same group is still in the same group after secondary grouping. Obtain a random secondary grouping result. The set of average values of the intervals between the time when the usage information in each of the f groups is fed back and T is v = {v 1 ,v 2 ,…,v e ,…v f}, v e represents the average value of the interval between the time when the usage information in the e-th group in a random secondary grouping result is fed back and T. According to the formula Q ’ =(∑ f e=1 (v e -(∑ f e=1 (v e )) / f) 2 ) / f calculate the goodness-of-fit measure value Q ’ of a random secondary grouping result. Calculate the goodness-of-fit measure values of different secondary grouping results, and select the secondary grouping result with the largest goodness-of-fit measure value as the optimal secondary grouping result. Store the usage information in the same group in the optimal secondary grouping result at the same position, that is, integrate and store the usage information of some groups among the original c groups of usage information at the same position; If W ≤ w, choose not to perform data integration processing;

[0082] S300: Establish a data processing selection model for automatically determining whether to trigger a data storage management operation, and perform an automatic storage management operation when the triggering condition is met: Create numbers for different types of medical products, each number corresponds to a type of medical product, and the number of times the usage information of medical products with different numbers is divided and preprocessed is A={A 1 , A 2 , …A p , …A g}, the number of times in set A is greater than 2, g represents the total number, according to H p =A p / (∑ g p=1 (A p )) Obtain the necessary index H for the pre-processing of the usage information of a randomly numbered medical product p , the necessary index set is {H 1 , H 2 , …H p , …H g}, the time interval set for the collection of the usage information of a randomly numbered medical product to be divided and preprocessed is U={U 1 , U 2 , …U d , …U r}, r+1 represents the number of times the usage information of a randomly numbered medical product is divided and preprocessed, U d The frequency of processing of the usage information of a random medical product with a number L is obtained by representing the time interval between the time when the usage information of a random medical product with a number is divided and preprocessed for the d+1th time and the time when the usage information of the random medical product with a number is divided and preprocessed for the dth time. p , L p =(∑ r d=1 U d ) / r, the time intervals for collecting the usage information of medical products with different numbers and dividing them for preprocessing are obtained, and the frequency set of the usage information of medical products with different numbers being processed is obtained as {L 1 , L 2 , …L p , …L g}, map the frequency and necessary index to a two-dimensional coordinate system to form data points {(L 1 , H 1 ), L 2 , H 2 ),…(L p , H p ),…(L g , H g)}, after performing a straight line fitting on the data points, a data processing selection model is established: y=β1*x+β2, where * represents a multiplication sign, β1 and β2 are fitting coefficients, and the frequency of processing of the user feedback usage information of the corresponding type of product currently sold after being used in the past is obtained as J, and J is substituted into the model: Let x=J, and the necessary index for pre-processing the usage information of the current product is predicted to be: β1*J+β2, and the necessary index threshold is set to G. If β1*J+β2>G, β1*J+β2>G means that the trigger condition is met, and the product usage information is automatically pre-processed after the user who uses the current product subsequently feedbacks the product usage information. The automatic pre-processing operation mode is the same as the operation mode in step S200;

[0083] S400: after receiving the product data retrieval request, transmitting the data to the requesting terminal: receiving the request for retrieving the user feedback information of any product sent by the terminal where the relevant personnel is located, after receiving the request, if the corresponding product usage information fed back by all users is stored in the same location, all the product usage information stored in the corresponding locations will be retrieved and transmitted to the terminal where the relevant personnel is located; if the corresponding product usage information fed back by the users is stored in different locations, the product usage information stored in different locations will be retrieved and transmitted to the terminal where the relevant personnel is located in a random order, and at the same time, the average interval time between the time when the product usage information stored in the corresponding location is fed back and the time when the user starts using the product will be transmitted to the terminal where the relevant personnel is located.

[0084] Finally, it should be noted that the above description is only a preferred example of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A medical product sales data intelligent management method, characterized by: The following steps are involved: S100: Collect medical product type information and historical sales data of medical products; S200: Retrieving historical sales data of medical products, analyzing the time concentration of user feedback on the use of medical products, selecting preliminary storage management measures for the use information based on the concentration, analyzing the data management results after the preliminary storage management, and performing secondary storage management based on the data management results; S300: Establishing a data processing selection model for automatically determining whether to trigger a data storage management operation, and performing an automatic storage management operation when a trigger condition is met; S400: after receiving the product data retrieval request, transmitting the data to the requesting terminal; In S100: information on the types of medical products sold in the past is collected, and information on the start time of users using different types of medical products in the past is collected. The start time of a random medical product being used once is obtained as T. A total of n users start using the corresponding medical product at time T. After the collection of n users starting to use the corresponding medical product at time T, the time set of the feedback of the use information of the corresponding medical product is {t1, t2, ..., t i ,…t m }, t i represents the time at which the i-th item of usage information is fed back, and m represents the number of items of usage information fed back after the corresponding medical product is used at time T, and the data on the feedback of usage information of different types of medical products after they are started to be used at different times are collected; In S200: According to the formula E=(1 / m)+[1 / [(∑ m i=1 ((t i -T)-(∑ m i=1 (t i -T)) / m) 2 ) / m] 1 / 2 ] Calculate the time concentration E of the feedback of the use information of a random medical product after it is used randomly once, and set the concentration threshold to E ’ , compare E and E ’ :If E≤E ’ , the product usage information initially stored in the same location by default is divided and preprocessed: the m items of usage information are arranged in the order of the interval time between the feedback time and T from small to large, and the arranged m items of usage information are randomly divided into c groups, and a random grouping result is obtained. The average value set of the interval time between the feedback time and T of each group of usage information in group c is V={V1, V2, …, V j ,…V c },V j It represents the average time interval between the feedback time of the j-th group usage information and T in a random grouping result. According to the formula Q=(∑ c j=1 (V j -(∑ c j=1 (V j )) / c) 2 ) / cCalculate the goodness metric value Q of a random grouping result, calculate the goodness metric values ​​of different grouping results, select the grouping result with the largest goodness metric value as the optimal grouping result, and store the product usage information of the same group in the optimal grouping result in the same location; if E>E ’ , no partitioning preprocessing is performed.

2. The medical product sales data intelligent management method according to claim 1, characterized in that: After performing the partitioning preprocessing, the set of average interval times between the times when the product usage information stored at different positions is fed back and T is obtained as K = {K1, K2, …, K j , … K c}, where K j represents the average interval time between the time when the product usage information stored at the j-th position is fed back and T. According to W = [(∑ c j=1 (K j - (∑ c j=1 (K j )) / c) 2 ) / c] 1 / 2 calculate the time dispersion index W of the product usage information stored at different positions after the partitioning preprocessing. Set the dispersion index threshold w, and compare W and w: If W > w, re-integrate the data located at different storage positions after the partitioning preprocessing: Arrange the c groups of usage information in ascending order of the average interval time in the set K, and perform secondary grouping on the arranged c groups of usage information: A total of f groups are divided, where f < c. In a randomly obtained secondary grouping result, the set of average values of the interval times between the times when the usage information in each of the f groups is fed back and T is v = {v1, v2, …, v e , … v f}, where v e represents the average value of the interval time between the time when the usage information in the e-th group in a randomly obtained secondary grouping result is fed back and T. According to the formula Q ’ = (∑ f e=1 (v e - (∑ f e=1 (v e )) / f) 2 ) / f calculate the goodness-of-fit measure value Q ’ of a randomly obtained secondary grouping result. Calculate the goodness-of-fit measure values of different secondary grouping results, and select the secondary grouping result with the largest goodness-of-fit measure value as the optimal secondary grouping result. Store the usage information in the same group in the optimal secondary grouping result at the same position; If W≤w, choose not to perform data integration processing.

3. The medical product sales data intelligent management method according to claim 2, characterized in that: In S300: create numbers for different types of medical products, each number corresponds to a type of medical product, and the number of times the usage information of medical products with different numbers is divided and pre-processed is A={A1, A2, ...A p , …A g }, g represents the total number of numbers, according to H p =A p / (∑ g p=1 (A p )) Obtain the necessary index H for the pre-processing of the usage information of a randomly numbered medical product p , the necessary index set is {H1, H2, ...H p , …H g }, the usage information of a randomly numbered medical product is collected and the time interval set for preprocessing is divided into U={U1, U2, …U d , …U r }, r+1 represents the number of times the usage information of a randomly numbered medical product is divided and preprocessed, U d The frequency of processing of the usage information of a random medical product with a number L is obtained by representing the time interval between the time when the usage information of a random medical product with a number is divided and preprocessed for the d+1th time and the time when the usage information of the random medical product with a number is divided and preprocessed for the dth time. p , L p =(∑ r d=1 U d ) / r, and the processing frequency set of the medical product usage information with different numbers is obtained as {L1, L2, ...L p , …L g }, map the frequency and necessary index into a two-dimensional coordinate system to form data points {(L1, H1), (L2, H2), … (L p , H p ),…(L g , H g )}, after performing a straight line fitting on the data points, a data processing selection model is established: y=β1*x+β2, where β1 and β2 are fitting coefficients. The frequency of processing of the usage information fed back by users of the corresponding type of product currently sold in the past is obtained as J. Substitute J into the model: Let x=J, and predict the necessary index for pre-processing the usage information of the current product as: β1*J+β2. Set the necessary index threshold to G. If β1*J+β2>G, the product usage information will be automatically divided and pre-processed after the users who subsequently use the current product feed back the product usage information.

4. A medical product sales data intelligent management system, applied to a medical product sales data intelligent management method as claimed in claim 1, characterized in that: The system includes: a product data collection module, a product data storage management module, a data processing automatic judgment module and a sales data transmission module; Collecting medical product type information and historical sales data of medical products through the product data collection module; Retrieving historical sales data of medical products through the product data storage management module, analyzing the time concentration of user feedback on the use of medical products, selecting preliminary storage management measures for the use information based on the concentration, analyzing the data management results after the preliminary storage management, and performing secondary storage management based on the data management results; A data processing selection model for automatically judging whether to trigger a data storage management operation is established through the data processing automatic judgment module, and an automatic storage management operation is performed when the triggering condition is met; After receiving the product data retrieval request, the sales data transmission module transmits the data to the requesting terminal.

5. The medical product sales data intelligent management system according to claim 4, characterized in that: The product data collection module includes a product type collection unit, a usage data collection unit and an after-sales information collection unit; The product type collection unit is used to collect information on the types of medical products sold in the past; The usage data collection unit is used to collect the start time information of the user's previous use of different types of medical products; The after-sales information collection unit is used to collect the time when the user previously provided product usage information after starting to use the product.

6. The medical product sales data intelligent management system according to claim 5, characterized in that: The product data storage management module includes a feedback concentration analysis unit, a division preprocessing unit, a preprocessing data dispersion judgment unit and a data integration processing unit; The feedback concentration analysis unit is used to analyze the feedback time concentration of different types of product usage information; The division preprocessing unit is used to set a concentration threshold, compare the feedback time concentration and the concentration threshold: if the feedback time concentration does not exceed the concentration threshold, perform division preprocessing on the product usage information initially stored in the same location by default: divide the product usage information according to the feedback time, and store the product usage information divided into the same group in the same location; If the feedback time concentration exceeds the concentration threshold, no division preprocessing is performed; The pre-processed data dispersion judgment unit is used to analyze the time dispersion index of the product usage information stored in different locations fed back after the division pre-processing; The data integration processing unit is used to set a dispersion index threshold, compare the time dispersion index and the threshold: if the time dispersion index exceeds the threshold, the data located in different storage locations after preprocessing are re-integrated, and the product usage information in the same group after integration is stored in the same location; If the time dispersion index does not exceed the threshold, no data integration processing is performed.

7. The medical product sales data intelligent management system according to claim 6, characterized in that: The data processing automatic judgment module includes a medical product number creation unit, a processing necessity analysis unit, a processing selection model establishment unit and a data automatic processing unit; The medical product number creation unit is used to create numbers for different types of medical products; The processing necessity analysis unit is used to count the number of times and time that the usage information of medical products with different numbers is divided and preprocessed, and analyze the necessity index of dividing and preprocessing the usage information of different types of medical products and the frequency of processing of the usage information based on the statistical results; The processing selection model building unit is used to map the processing frequency and necessary index of the product usage information to form data points in a two-dimensional coordinate system, and to establish a data processing selection model after performing linear fitting on the data points; The data automatic processing unit is used to substitute the frequency of processing of the current product usage information into the data processing selection model to predict the necessary index for dividing and preprocessing the current product usage information, and set a necessary index threshold. If the predicted necessary index exceeds the necessary index threshold, the product usage information is automatically divided and preprocessed when users who subsequently use the current product feedback the product usage information.

8. The medical product sales data intelligent management system according to claim 7, characterized in that: The sales data transmission module includes a data retrieval request receiving unit and a product data transmission unit; The data retrieval request receiving unit is used to receive a request for retrieving user feedback information of any product sent by a terminal where a relevant person is located; The product data transmission unit is used to retrieve all the product usage information stored in the corresponding location and transmit it to the terminal where the relevant personnel are located after receiving the request if the corresponding product usage information fed back by all users is stored in the same location; If the corresponding product usage information fed back by the user is stored in different locations, the product usage information stored in different locations will be retrieved and transmitted to the terminals of the relevant personnel in a random order. At the same time, the average interval duration between the time when the product usage information stored in the corresponding location is fed back and the time when the user starts using the product will be transmitted to the terminals of the relevant personnel.

Images