Medical device inventory supervision method and system based on data analysis

Abstract

The present invention discloses a medical device inventory supervision method and system based on data analysis, which relates to the field of medical device technology. It includes retrieving the historical inventory information and historical usage information of each medical device, predicting the inventory demand trend, and automatically marking the inventory status; generating an inventory status transfer matrix based on the inventory status marking results and combining the inventory demand trend prediction results, and dynamically updating the inventory status transfer probability; calculating the inventory transfer impact rate based on the historical usage information, the inventory status transfer probability and the supplier delivery cycle corresponding to each historical usage information; obtaining the inventory inspection time point in the current medical device inventory management plan, combining the inventory demand trend prediction results, calculating the risk value of each inspection time point, and outputting prompt information. The present invention not only improves the accuracy of inventory forecasting, but also can effectively cope with demand fluctuations and supply chain uncertainties.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a method and system for monitoring medical device inventory based on data analysis. Background Art

[0002] As a key link in the operation of modern medical institutions, medical device inventory management has always been a key research topic in the field of medical information technology. With the rapid development of medical technology and the increasing complexity of medical needs, traditional inventory management methods can no longer meet the current medical institutions' needs for accurate and efficient inventory control.

[0003] However, existing inventory management methods based on data analysis still have some shortcomings. For example, existing methods often use static forecasting models, which are difficult to adapt to the dynamic changes in medical needs and the impact of emergencies. In addition, the dynamic characteristics of inventory status transfer and the impact of supply chain factors on inventory management are not considered enough, resulting in limited accuracy and practicality of forecast results. Summary of the invention

[0004] In view of the problems of insufficient prediction accuracy and poor dynamic adaptability in the existing medical device inventory management technology based on data analysis, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to improve the scientificity and efficiency of medical device inventory management.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a medical device inventory supervision method based on data analysis, which includes retrieving historical inventory information and historical usage information of each medical device, predicting the inventory demand trend based on the historical inventory information and the historical usage information, and automatically marking the inventory status; generating an inventory status transfer matrix based on the inventory status marking results and the inventory demand trend prediction results, and dynamically updating the inventory status transfer probability; calculating the inventory transfer impact rate based on the historical usage information, the inventory status transfer probability and the supplier delivery cycle corresponding to each historical usage information; obtaining the inventory inspection time points in the current medical device inventory management plan, calculating the risk value of each inspection time point in combination with the inventory demand trend prediction results, and outputting prompt information.

[0008] As a preferred solution of the medical device inventory supervision method based on data analysis described in the present invention, the retrieving historical inventory information and historical usage information of each medical device includes: collecting historical inventory data in the medical device inventory management system, and aggregating the historical inventory data of various medical devices; according to the aggregated data, retrieving the historical inventory information and historical usage information corresponding to each medical device in the medical device inventory management system; according to the historical usage information, sorting out the historical inventory information, and pairing the historical inventory information at each time point with the corresponding historical usage information.

[0009] As a preferred solution of the medical device inventory supervision method based on data analysis described in the present invention, the generation of the inventory status transfer matrix includes the following steps: counting the inventory status marking results of all historical time points, and generating a set of historical time points, recorded as {Tn1, Tn2, ..., Tn m}, where Tn1, Tn2, ..., Tn m They are represented as the 1st, 2nd, ..., mth historical time points respectively; the excess state, normal state and insufficient state in the inventory state are marked as ST1, ST2 and ST3 respectively; according to the set of historical time points, the inventory state and the historical time point are mapped into a corresponding relationship of time → inventory state, where one historical time point corresponds to one inventory state; if the inventory state transfer occurs between two adjacent historical time points, it is marked as an inventory state transfer event, where the inventory state transfer is represented as ST i →ST j , among which, ST i 、ST j ∈{ST1,ST2,ST3}; identify all inventory state transfer events and generate the inventory state transfer matrix:

[0010]

[0011] P i1 +P i2 +P i3 =1

[0012] Among them, P is the inventory state transfer matrix, and the element P in the matrix ij Indicates that from state ST i Transfer to ST State j probability.

[0013] As a preferred solution of the medical device inventory supervision method based on data analysis of the present invention, the dynamic updating of the inventory state transition probability includes: if the predicted inventory demand trend is an upward trend, then increasing the transition weight from the normal state ST2 to the insufficient state ST3, that is, increasing P 23value, while reducing P 21 and P 23 If the predicted inventory demand trend is downward, increase the transfer weight from the normal state ST2 to the excess state ST1, that is, increase P 21 value, while reducing P 22 and P 23 The value of; According to the adjusted inventory state transfer matrix P ’ , calculate the inventory state transition probability P of any inventory state transition event ij , the specific calculation formula is as follows:

[0014] P ij =n e / n z

[0015] Among them, n e Indicates that from state ST i Transfer to ST State j The total number of events; n z Indicates status ST i Total number of occurrences.

[0016] As a preferred solution of the medical device inventory supervision method based on data analysis described in the present invention, the calculation of the inventory transfer impact rate includes: retrieving the supplier delivery cycle corresponding to each historical usage information in the medical device inventory management system, performing statistics on all supplier delivery cycles corresponding to any historical usage information, and generating a supplier delivery cycle set, recorded as DC = {DC1, DC2, ...DC i ,...,DC mz}, where mz is the number of delivery cycles; when the inventory status of two adjacent historical usages is transferred, the average delivery cycle corresponding to the two historical usages is retrieved respectively, and the historical usage U x The corresponding average delivery cycle is recorded as ADC x , the historical usage U x+1 The corresponding average delivery cycle is recorded as ADC x+1 ; Calculate historical usage U x and historical usage U x+1 The inventory transfer impact rate when the inventory status transfer occurs between the two periods, and the inventory transfer impact rate is recorded as Y x , the specific calculation formula is as follows:

[0017] Y x =|ADC x+1 -ADC x | / max(ADC x ,ADC x+1 )×w+|U x+1 -Ux | / max(U x ,U x+1 )×(1 - w)

[0018] where w is the weight coefficient of the change in the delivery cycle (0 < w < 1), which can be adjusted according to the actual situation. Among them, max(ADC x ,ADC x+1 ) is to select the maximum value from ADC x+1 and ADC x , and max(U x ,U x+1 ) is to select the maximum value from U x+1 and U x .

[0019] As a preferred solution of the medical device inventory supervision method based on data analysis according to the present invention, wherein: calculating the risk value at each inspection time point includes the following steps: obtaining the set of inventory inspection time points in the current medical device inventory management plan, denoted as {TD1, TD2,... TD i ,..., TD K}, where K is the number of inspection time points; for each inspection time point TD i , perform the following steps: obtaining the predicted demand quantity F i at this inspection time point; obtaining the current inventory state transition probability matrix; obtaining the inventory transfer impact rate Y x calculated most recently; calculating the risk value R i at the inspection time point TD i ; setting a risk threshold R th , for each inspection time point TD i , if the risk value R i ≥R th , then generate a risk warning message.

[0020] As a preferred solution of the medical device inventory supervision method based on data analysis according to the present invention, wherein: the calculation formula for the risk value R i at the inspection time point TD i is:

[0021]

[0022] where α2, β2, and γ2 are the corresponding weights; S k is the current inventory quantity, and max(P ij ) is the maximum probability value in the inventory state transition probability matrix.

[0023] In the second aspect, an embodiment of the present invention provides a medical device inventory supervision system based on data analysis, which includes a historical data retrieval module for obtaining historical data from an inventory database to provide basic information; an inventory demand forecasting module for forecasting inventory demand trends based on historical data and marking inventory status; an inventory status transfer matrix generation module for generating a transfer matrix and dynamically updating transfer probabilities based on inventory markings and demand trends; an inventory transfer impact rate calculation module for calculating the inventory transfer impact rate in combination with supplier delivery cycles and usage; and a risk calculation module for calculating inventory risks and outputting prompt information based on inventory inspection time points and trend forecasts.

[0024] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the medical device inventory management method based on data analysis as described in the first aspect of the present invention are implemented.

[0025] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the medical device inventory management method based on data analysis as described in the first aspect of the present invention are implemented.

[0026] The beneficial effects of the present invention are as follows: the present invention constructs a dynamic inventory demand forecasting model by introducing the concepts of inventory status transfer matrix and inventory transfer impact rate, thereby realizing accurate characterization and dynamic update of inventory status changes. At the same time, by calculating the risk value at the inspection time point, a quantitative basis is provided for inventory management decisions. It can be seen that the method of the present invention not only improves the accuracy of inventory forecasting, but also can effectively cope with demand fluctuations and supply chain uncertainties, and provides a new technical path for medical institutions to achieve refined and intelligent inventory management. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 Flowchart of the medical device inventory supervision method based on data analysis.

[0029] Figure 2 This is a structural diagram of the medical device inventory supervision system based on data analysis. DETAILED DESCRIPTION

[0030] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0031] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0032] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0033] Example 1

[0034] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a medical device inventory supervision method based on data analysis, including:

[0035] S1: Retrieve the historical inventory information and historical usage information of each medical device in the medical device inventory management system, predict the inventory demand trend based on the historical inventory information and historical usage information, and automatically mark the inventory status, wherein the inventory information includes the incoming inventory, current inventory and safety inventory.

[0036] According to the classification standards of the State Food and Drug Administration, medical devices are divided into Class I, Class II and Class III. Different categories of medical devices have different inventory management requirements; for example: Class I includes scalpels, stethoscopes, medical X-ray films, etc.; Class II includes medical suture needles, blood glucose meters, ultrasound diagnostic equipment, etc.; Class III includes implantable pacemakers, artificial heart valves, coronary artery stents, etc.

[0037] Specifically, historical inventory data is collected from the medical device inventory management system, and the historical inventory data of various medical devices are summarized. According to the summarized data, the historical inventory information and historical usage information corresponding to each medical device are retrieved from the medical device inventory management system, and the historical inventory information is sorted according to the historical usage information, and the historical inventory information at each time point is paired with the corresponding historical usage information.

[0038] Furthermore, the historical data of each medical device is used to predict the inventory demand trend. The specific formula is as follows:

[0039] F t+u =α1×D t+β1×U t +γ1

[0040] Among them, F tu is the demand predicted at time t+u, D t is the inventory at time point t, U t is the usage at time point t, α1 is the weight coefficient of historical inventory, β1 is the weight coefficient of historical usage, and γ1 is the intercept term, which represents the impact of other uncontrollable factors (such as seasonal demand fluctuations).

[0041] Among them, the values ​​of α1 and β1 may be different for different types of medical devices. For example, for disposable medical consumables that are used frequently, the weight coefficient of historical usage may be higher; for medical devices used seasonally (such as nebulizers during the flu season), the weight of historical inventory may fluctuate with seasonal changes.

[0042] Obtain the historical inventory information of any medical device at a specific time point and the predicted demand corresponding to the time point, and mark the inventory status at the time point, where the inventory status includes excess status, normal status and insufficient status; in the historical inventory information at the time point, if the current inventory is greater than 1.5 times of the predicted demand and greater than the safety stock, then the time point is marked as an excess state; if the current inventory is between 0.8 times and 1.5 times of the predicted demand and is not less than the safety stock, then the time point is marked as a normal state; if the current inventory is less than 0.8 times of the predicted demand or less than the safety stock, then the time point is marked as an insufficient state.

[0043] S2: Based on the inventory status marking results and the inventory demand trend forecast results, the inventory status transfer matrix is ​​generated and the inventory status transfer probability is dynamically updated.

[0044] Preferably, the inventory status marking results of all historical time points are counted, and a set of historical time points is generated, which is recorded as {Tn1, Tn2, ..., Tn m}, where Tn1, Tn2, ..., Tn m They are represented as the 1st, 2nd, ..., mth historical time points respectively; the excess state, normal state and insufficient state are marked as ST1, ST2 and ST3 respectively, and according to the set of historical time points, the inventory state and the historical time point are mapped into a corresponding relationship of time → inventory state, wherein one historical time point corresponds to one inventory state.

[0045] According to the mapping relationship, the inventory status transfer events are sorted out. If the inventory status transfer occurs at two adjacent historical time points, the event is marked as an inventory status transfer event, where the inventory status transfer is represented by ST i →STj , among which, ST i 、ST j ∈{ST1,ST2,ST3}; identify all inventory state transfer events and generate the inventory state transfer matrix:

[0046]

[0047] P i1 +P i2 +P i3 =1

[0048] Among them, P is the inventory state transfer matrix, and the element P in the matrix ij Indicates that from state ST i Transfer to ST State j The probability of the inventory status changing from one state to another at different times.

[0049] It should be noted that for Class I medical devices (such as gauze, bandages, etc.), due to the lower risk, the state transition standards can be appropriately relaxed. For example, the range of normal state can be expanded to 0.7 to 1.6 times the predicted demand.

[0050] For Class II and Class III medical devices (such as ventilators, ECG monitors, etc.), due to the higher risks, the inventory status should be more strictly controlled, and the normal status range can be narrowed to 0.9 to 1.3 times the predicted demand.

[0051] For high-value medical devices (such as CT machines, MRI machines, etc.), additional states can be introduced to indicate that the inventory level is close to but has not yet reached the safety stock level, which can provide earlier warning of possible inventory shortages.

[0052] Combined with the inventory demand trend forecast results, the inventory status transfer matrix is ​​adjusted. The specific adjustment strategies are as follows:

[0053] If the predicted inventory demand trend is on the rise, increase the transfer weight from the normal state ST2 to the insufficient state ST3, that is, increase P 23 value, while reducing P 21 and P 23 The value of is adjusted as follows:

[0054] P' 23 =P 23 +Δp

[0055] P' 21 =P 21 -Δp / 2

[0056] P' 22 =P22 -Δp / 2

[0057] If the predicted inventory demand trend is downward, increase the transfer weight from the normal state ST2 to the excess state ST1, that is, increase P 21 value, while reducing P 22 and P 23 The value of is adjusted as follows:

[0058] P' 21 =P 21 +Δp

[0059] P' 22 =P 22 -Δp / 2

[0060] P' 23 =P 23 -Δp / 2

[0061] Among them, Δp is the adjusted weight increment, which can be set according to the actual predicted demand amplitude.

[0062] According to the adjusted inventory state transition matrix P', calculate the inventory state transition probability P of any inventory state transition event ij , the specific calculation formula is as follows:

[0063] P ij =n e / n z

[0064] Among them, n e Indicates that from state ST i Transfer to ST State j The total number of events; n z Indicates status ST i Total number of occurrences.

[0065] Since most of the existing technologies are based on static threshold settings and preset inventory warning lines, they are unable to respond to dynamic changes in inventory status in real time. The present invention generates an inventory status transfer matrix, which can accurately capture transition events between inventory statuses when the inventory status changes. The system is made more adaptable by dynamically adjusting the matrix.

[0066] Furthermore, it should be noted that the present invention closely combines inventory status management with demand forecasting by adjusting the transfer weights between different states according to the rising or falling trend of demand. For example, when demand is on the rise, the system automatically increases the transfer weight from the normal state to the insufficient state, and warns of insufficient inventory in advance. This adaptive adjustment makes inventory management more forward-looking, and can respond before actual demand changes, effectively reducing the risk of excessive or insufficient inventory.

[0067] S3: Calculate the inventory transfer impact rate based on the historical usage information, the inventory status transfer probability, and the supplier delivery cycle corresponding to each historical usage information retrieved from the medical device inventory management system.

[0068] S3.1: Retrieve the supplier delivery cycle corresponding to each historical usage information from the medical device inventory management system, count all the supplier delivery cycles corresponding to any one historical usage information, and generate a set of supplier delivery cycles, denoted as DC = {DC1, DC2,... DC i ,..., DC mz}, where mz is the number of delivery cycle quantities.

[0069] S3.2: Calculate the average delivery cycle corresponding to each historical usage information.

[0070] S3.3: When an inventory status transfer occurs between two adjacent historical usages, retrieve the average delivery cycles corresponding to these two historical usages respectively. Denote the average delivery cycle corresponding to the historical usage U x as ADC x , and denote the average delivery cycle corresponding to the historical usage U x+1 as ADC x+1 ;

[0071] Calculate the inventory transfer impact rate when an inventory status transfer occurs between the historical usage U x and the historical usage U x+1 . Denote the inventory transfer impact rate as Y x . The specific calculation formula is as follows:

[0072] Y x = |ADC x+1 - ADC x | / max(ADC x , ADC x+1 ) × w + |U x+1 - U x | / max(U x , U x+1 ) × (1 - w)

[0073] where w is the weight coefficient of the delivery cycle change (0 < w < 1), which can be adjusted according to the actual situation. Among them, max(ADC x , ADC x+1 ) is to select the maximum value from ADC x+1 and ADC x , and max(U x , U x+1 ) is to select the maximum value from U x+1 and U xSelect the maximum value among .

[0074] S4: Obtain the inventory inspection time points in the current medical device inventory management plan, combine the inventory demand trend forecast results, inventory status transfer probability and inventory transfer impact rate, calculate the risk value of each inspection time point, output prompt information, and send it to the inventory management personnel.

[0075] Preferably, a set of inventory inspection time points in the current medical device inventory management plan is obtained, which is recorded as {TD1, TD2, ... TD i ,...,TD K}, where K is the number of inspection time points.

[0076] For each inspection time point TD i , perform the following steps:

[0077] Get the predicted demand F at the inspection time point i ;

[0078] Get the current inventory status transition probability matrix;

[0079] Get the most recently calculated inventory transfer impact rate Y x ;

[0080] Calculate the inspection time point TD i The risk value R i , the specific calculation formula is as follows:

[0081]

[0082] Among them, α2, β2 and γ2 are the corresponding weights respectively; S k is the current inventory, max(P ij ) is the maximum probability value in the inventory state transition probability matrix.

[0083] Set the risk threshold R th , for each inspection time point TD i , if its risk value R i ≥R th , then risk warning information is generated.

[0084] Based on risk warning information, inventory management suggestions are automatically generated, including but not limited to: adjusting order quantity, changing order frequency, updating safety stock levels, and finding alternative suppliers.

[0085] The risk warning information and inventory management suggestions are integrated into prompt information and sent to inventory management personnel. The prompt information includes the inspection time point, risk value, main factors leading to high risk (demand trend, state transition probability, inventory transfer impact rate) and inventory management suggestions.

[0086] Furthermore, this embodiment also provides a medical device inventory monitoring system based on data analysis, including:

[0087] The historical data retrieval module is used to obtain historical data from the inventory database and provide basic information. The inventory demand forecasting module is used to forecast the inventory demand trend based on historical data and mark the inventory status. The inventory status transfer matrix generation module is used to generate the transfer matrix and dynamically update the transfer probability based on the inventory mark and demand trend. The inventory transfer impact rate calculation module is used to calculate the inventory transfer impact rate based on the supplier's delivery cycle and usage. The risk calculation module is used to calculate the inventory risk and output prompt information based on the inventory inspection time point and trend forecast.

[0088] This embodiment also provides a computer device, which is suitable for the case of a medical device inventory supervision method based on data analysis, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the medical device inventory supervision method based on data analysis proposed in the above embodiment.

[0089] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0090] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for medical device inventory supervision based on data analysis as proposed in the above embodiment is implemented.

[0091] In summary, the present invention constructs a dynamic inventory demand forecasting model by introducing the concepts of inventory status transfer matrix and inventory transfer impact rate, thereby realizing accurate characterization and dynamic update of inventory status changes. At the same time, by calculating the risk value at the inspection time point, a quantitative basis is provided for inventory management decisions. It can be seen that the method of the present invention not only improves the accuracy of inventory forecasting, but also can effectively cope with demand fluctuations and supply chain uncertainties, and provides a new technical path for medical institutions to achieve refined and intelligent inventory management.

[0092] Example 2

[0093] Referring to Table 1 and Table 2, the second embodiment of the present invention provides a medical device inventory supervision method based on data analysis. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0094] Specifically, three typical medical devices in the cardiology department of a tertiary hospital were selected as research objects: disposable cardiac catheters (Class I), portable ECG monitors (Class II) and coronary artery stents (Class III). The experimental period is from May 1, 2023 to April 30, 2024, a total of 365 days.

[0095] The experimental process is as follows:

[0096] First, the historical inventory and usage data of the three medical devices in the past two years (2021-2022) are extracted from the hospital's medical device inventory management system, and the data is preprocessed and summarized by month. Then, the prediction algorithm of the present invention is used to predict the demand for each month. Among them, the initial values ​​of α1 and β1 are set to 0.4 and 0.5 respectively, and γ1 is set to 10% of the average monthly demand of each device in 2022.

[0097] Based on the forecast results and actual inventory data, the inventory status is marked for each month in 2023. Taking into account the characteristics of different categories of medical devices, different marking standards are set: the normal status range for disposable cardiac catheters is 0.7-1.6 times the forecast demand, for portable ECG monitors is 0.8-1.5 times, and for coronary stents is 0.9-1.3 times.

[0098] Based on the labeling results, an inventory state transition matrix was constructed. The initial matrix was generated based on the state transition data in 2022 and was dynamically updated during the experiment. When updating, the transition probability was adjusted according to the predicted demand trend.

[0099] When calculating the impact rate of inventory transfer, the supplier delivery cycle data of these three medical devices were collected. The average delivery cycle of disposable cardiac catheters is 7 days, portable ECG monitors is 21 days, and coronary stents is 14 days. The weight coefficient w of the delivery cycle change is set to 0.6.

[0100] The end of each month is set as the inventory inspection time point, and the risk value of each inspection time point is calculated. The risk threshold is set to 0.7. If this value is exceeded, a risk warning information is generated.

[0101] Some experimental results are shown in the following table:

[0102] Table 1 Medical equipment inventory data table

[0103]

[0104]

[0105] Table 2 Medical device inventory risk assessment table

[0106]

[0107]

[0108] From Table 1 and Table 2, it can be seen that by adopting differentiated inventory status marking standards for different categories of medical devices, this method effectively captures the inventory status changes of various types of medical devices. For example, the disposable cardiac catheter changed from the "normal" status in January to the "excess" status in February and March, reflecting the method's relatively loose inventory management strategy for Class I medical devices, and the probability of inventory status transition showed an upward trend over time, which reflects that this method can effectively capture the dynamic changes of inventory status.

[0109] By comparing the risk assessment results of the three types of medical devices, it can be seen that this method can implement differentiated management strategies for different categories of medical devices. For example, the risk value of Class III coronary stents is generally higher than that of Class I disposable cardiac catheters, reflecting stricter management requirements for high-risk medical devices.

[0110] It can be seen that the medical device inventory supervision method based on data analysis proposed in the present invention not only improves the accuracy of demand forecasting, but also realizes the accurate characterization of inventory status changes through the construction of dynamic inventory status marking and transfer matrix, making risk assessment more comprehensive.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A medical device inventory supervision method based on data analysis, characterized by: include, Retrieving historical inventory information and historical usage information of each medical device, predicting inventory demand trends based on the historical inventory information and the historical usage information, and automatically marking inventory status; According to the inventory status marking results and combined with the inventory demand trend forecast results, the inventory status transfer matrix is ​​generated, and the inventory status transfer probability is dynamically updated; Calculate the inventory transfer impact rate based on historical usage information, inventory status transfer probability, and supplier delivery cycle corresponding to each historical usage information; Obtain the inventory inspection time points in the current medical device inventory management plan, combine the inventory demand trend forecast results and calculate the inventory transfer impact rate, calculate the risk value of each inspection time point, and output prompt information; The dynamically updating the inventory status transfer probability includes: If the predicted inventory demand trend is on the rise, increase the transfer weight from the normal state ST2 to the insufficient state ST3, that is, increase P 23 value, while reducing P 21 and P 23 The value of If the predicted inventory demand trend is downward, increase the transfer weight from the normal state ST2 to the excess state ST1, that is, increase P 21 value, while reducing P 22 and P 23 The value of According to the adjusted inventory state transition matrix P', calculate the inventory state transition probability P of any inventory state transition event ij , the specific calculation formula is as follows: P ij =n e / n z Among them, n e Indicates that from state ST i Transfer to ST State j The total number of events; n z Indicates status ST i Total number of occurrences.

2. The medical device inventory supervision method based on data analysis according to claim 1, characterized in that: The retrieval of historical inventory information and historical usage information of each medical device includes: Collect historical inventory data from the medical device inventory management system and summarize the historical inventory data of various medical devices; According to the summary data, retrieve the historical inventory information and historical usage information corresponding to each medical device in the medical device inventory management system; According to the historical usage information, the historical inventory information is sorted out, and the historical inventory information at each time point is paired with the corresponding historical usage information.

3. The medical device inventory supervision method based on data analysis according to claim 2 is characterized in that: Generating the inventory status transfer matrix comprises the following steps: Count the inventory status marking results of all historical time points and generate a set of historical time points, recorded as {Tn1,Tn2,...,Tn m }, where Tn1, Tn2, ..., Tn m Represented as the 1st, 2nd, ..., mth historical time points respectively; The excess state, normal state and insufficient state in the inventory state are marked as ST1, ST2 and ST3 respectively; According to the set of historical time points, the inventory status and the historical time points are mapped into a corresponding relationship of time → inventory status, wherein one historical time point corresponds to one inventory status; If the inventory status transfer occurs at two adjacent historical time points, it is marked as an inventory status transfer event, where the inventory status transfer is represented by ST i →ST j , among which, ST i 、ST j ∈{ST1,ST2,ST3}; Identify all inventory status transfer events and generate an inventory status transfer matrix: Among them, P is the inventory state transfer matrix, and the element P in the matrix ij Indicates that from state ST i Transfer to ST State j probability.

4. The medical device inventory supervision method based on data analysis as claimed in claim 3 is characterized by: The calculation of the inventory transfer impact rate includes: In the medical device inventory management system, the supplier delivery cycle corresponding to each historical usage information is retrieved, all supplier delivery cycles corresponding to any historical usage information are counted, and a supplier delivery cycle set is generated, which is recorded as DC = {DC1, DC2, ... DC i ,...,DC mz }, where mz is the number of delivery cycles; When the inventory status of two adjacent historical usages changes, the average delivery cycle corresponding to the two historical usages is retrieved and the historical usage U x The corresponding average delivery cycle is recorded as ADC x , the historical usage U x+1 The corresponding average delivery cycle is recorded as ADC x+1 ; Calculate historical usage U x and historical usage U x+1 The inventory transfer impact rate when the inventory status transfer occurs between the two periods, and the inventory transfer impact rate is recorded as Y x , the specific calculation formula is as follows: Y x =|ADC x+1 -ADC x | / max(ADC x ,ADC x+1 )×w+|U x+1 -U x | / max(U x ,U x+1 )×(1-w) Among them, w is the weight coefficient of the delivery cycle change (0 < w < 1), which can be adjusted according to the actual situation. Among them, max(ADC x , ADC x+1 ) is to select the maximum value from ADC x+1 and ADC x , and max(U x , U x+1 ) is to select the maximum value from U x+1 and U x .

5. The method for medical device inventory supervision based on data analysis according to claim 4, characterized in that: Calculating the risk value at each inspection time point comprises the following steps: Get the inventory inspection time point set in the current medical device inventory management plan, recorded as {TD1, TD2, ... TD i ,...,TD K }, where K is the number of inspection time points; For each inspection time point TD i , perform the following steps: Get the predicted demand F at the inspection time point i ; Get the current inventory status transition probability matrix; Get the most recently calculated inventory transfer impact rate Y x ; Calculate the inspection time point TD i The risk value R i ; Set the risk threshold R th , for each inspection time point TD i , if the risk value R i ≥R th , then risk warning information is generated.

6. The method for medical device inventory supervision based on data analysis according to claim 5, characterized in that: The inspection time point TD i The risk value R i The calculation formula is: Among them, α2, β2 and γ2 are the corresponding weights respectively; S k is the current inventory, max(P ij ) is the maximum probability value in the inventory state transition probability matrix.

7. A medical device inventory monitoring system based on data analysis, based on the medical device inventory monitoring method based on data analysis according to any one of claims 1 to 6, characterized in that: Also includes, The historical data retrieval module is used to obtain historical data from the inventory database and provide basic information; Inventory demand forecasting module, used to forecast inventory demand trends based on historical data and mark inventory status; Inventory status transition matrix generation module, used to generate the transition matrix and dynamically update the transition probability according to the inventory mark and demand trend; Inventory transfer impact rate calculation module, used to calculate the inventory transfer impact rate based on supplier delivery cycle and usage; The risk calculation module is used to calculate inventory risks and output prompt information based on inventory inspection time points and trend forecasts.

Images