Big data analysis and management system for high-value medical consumables

Abstract

The invention discloses a big data analysis and management system for high-value medical consumables, relates to the technical field of medical information, and is used for solving the problem that clinical diagnosis and treatment and medical service quality are affected due to the fact that the accuracy of demand prediction is reduced and the high-value consumables are not transported in time. Establishing a data analysis model, obtaining a precision evaluation coefficient, comparing the precision evaluation coefficient with a precision threshold value, obtaining a precision comparison result of each prediction frequency in a current calculation period, counting precision evaluation coefficient values smaller than the precision threshold value, sorting according to the values, obtaining total fluctuation intensity, comparing the total fluctuation intensity with a fluctuation threshold value, and evaluating the prediction model according to the comparison result. Inventory consumable distribution prediction analysis is performed through the time sequence model to obtain consumable demand data, and inventory consumable distribution is dynamically adjusted according to the maximum value and the minimum value of a prediction result, so that the accuracy of inventory management and the quality of clinical diagnosis and treatment and medical service are improved.

Description

Technical Field

[0001] This invention relates to the field of medical information technology, and more specifically, to a big data analysis and management system for high-value medical consumables. Background Technology

[0002] With the deepening of my country's medical system reform and the implementation of the zero-markup policy for medical consumables, medical consumables, especially high-value consumables, are no longer profit centers but have instead become cost centers. This has brought enormous pressure and challenges to the control of hospital operating costs. At the same time, with the continuous development of advanced clinical diagnosis and treatment technologies, the proportion of high-value consumables used in high-level surgeries is increasing year by year, making the need for scientific, full-process, and refined management of high-value consumables by medical institutions more urgent.

[0003] The existing technology has the following shortcomings:

[0004] Currently, in the process of implementing a "zero-inventory" high-value consumables management system based on the SPD (Supply, Process, and Distribution) model, minimizing high-value consumables inventory and reducing inventory backlog and waste are achieved by optimizing management processes and improving management efficiency. However, when supply chain collaboration capabilities decline and the complexity of multi-party collaboration arises, the accuracy of demand forecasting inevitably decreases. Furthermore, insufficient capacity to manage unforeseen risks leads to delays in the transportation of high-value consumables, thus affecting clinical diagnosis and treatment and the quality of medical services. Therefore, a big data analysis and management system for high-value medical consumables is proposed.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a high-value medical consumables big data analysis and management system, which solves the problems mentioned in the background art by employing different product testing methods.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A big data analysis and management system for high-value medical consumables includes a parameter acquisition module, a parameter processing module, a precision analysis module, and an inventory adjustment module; the modules are interconnected by signals.

[0009] The parameter acquisition module is used to collect forecast error analysis information and demand fluctuation information; through data processing, it obtains the average absolute error rate, forecast coverage, demand volatility, and seasonality index and their consistency with actual demand changes, and sends them to the parameter processing module. At the same time, it collects inventory data and supplier information and sends them to the inventory adjustment module.

[0010] The parameter processing module is used to receive prediction error analysis information and demand fluctuation information, establish a data analysis model, obtain accuracy evaluation coefficients, and send them to the accuracy analysis module.

[0011] The accuracy analysis module is used to obtain the accuracy evaluation coefficients and compare them with the accuracy threshold to obtain the accuracy comparison results of each prediction number in the current calculation cycle. It counts the accuracy evaluation coefficients that are less than the accuracy threshold and sorts them by value. It obtains the total fluctuation intensity and compares it with the fluctuation threshold. Based on the comparison results, it evaluates the prediction model and sends the total fluctuation intensity to the inventory adjustment module.

[0012] The inventory adjustment module receives total fluctuation intensity, inventory data, and supplier information. Through data processing, it obtains inventory turnover rate, stockout rate, and average reliability of supply from all suppliers. It then uses a time series model to perform predictive analysis on inventory consumable allocation, further refining and optimizing the inventory consumable allocation plan.

[0013] In a preferred embodiment, a calculation period is determined first, and the average absolute error rate Ma is obtained by averaging the percentage of the deviation between the predicted value and the actual value within the period relative to the actual value. i Where i represents the i-th prediction within the period;

[0014] Successful coverage is defined as a predicted value being greater than or equal to the actual demand within a calculation period. The number of successful coverages within the calculation period is counted and compared to the total number of successful coverages to obtain the prediction coverage rate (Cr). i ;

[0015] The demand volatility Dv is calculated by calculating the average demand during the period and based on the magnitude of the change in demand relative to the average demand.

[0016] The overall average demand is obtained by dividing the total demand of all calculation cycles by the total number of calculation cycles. The seasonality index of the current calculation cycle is obtained by dividing the seasonality index by the ratio of the total demand of all calculation cycles and the total number of calculation cycles. The degree of agreement between the seasonality index and the actual demand change is obtained by dividing the seasonality index by the ratio of the seasonality index to the actual demand change.

[0017] In a preferred embodiment, the mean absolute error rate, forecast coverage, demand volatility, and the degree of agreement between the seasonality index and actual demand changes are obtained; a data analysis model is established; and the accuracy evaluation coefficient Ac is obtained. i The formula used is as follows:

[0018]

[0019] In the formula, Ac i For accuracy evaluation coefficients, as well as These are preset proportional coefficients for the mean absolute error rate, forecast coverage, demand volatility, and the degree of agreement between the seasonal index and actual demand changes, respectively. as well as All are greater than 0.

[0020] In a preferred embodiment, accuracy evaluation coefficients are obtained, and the accuracy evaluation coefficients are compared and analyzed with continuously iterated accuracy thresholds.

[0021] If the accuracy evaluation coefficient is greater than or equal to the accuracy threshold, the current prediction is marked as an accurate prediction and an accurate signal is generated.

[0022] If the accuracy evaluation coefficient is less than the accuracy threshold, the current prediction is marked as an error prediction and an error signal is generated.

[0023] The values ​​of all accuracy evaluation coefficients that are less than the accuracy threshold within the current calculation period are sorted from largest to smallest to smooth out occasional numerical jumps in the natural number of calculations. This allows for an objective view of the differences between values ​​to obtain the total fluctuation intensity.

[0024] Let Z be the total number of accuracy evaluation coefficients less than the accuracy threshold, then Ac g Let g be the accuracy evaluation coefficient that is less than the accuracy threshold, where g is the g-th accuracy evaluation coefficient that is less than the accuracy threshold; then according to the formula:

[0025] P=|Ac1-Ac2|+|Ac2-Ac3|+…+|Ac Z-1 -Ac Z |

[0026] Where P is the total fluctuation intensity, and it is compared with the fluctuation threshold.

[0027] In a preferred embodiment, if the total fluctuation intensity is greater than or equal to the fluctuation threshold, an alarm is issued, and the error in the prediction model is reported, and the weights of each parameter in the prediction model are readjusted.

[0028] If the total fluctuation intensity is less than the fluctuation threshold, then the current calculation period is marked.

[0029] In a preferred embodiment, the inventory turnover rate is obtained by statistically calculating the ratio of the number of times consumables are used within the inventory during the period to the actual average inventory. The deviation It is then obtained by subtracting it from the target inventory turnover rate. k Where k represents the kth category of high-value medical consumables;

[0030] The stockout rate Sr is obtained by calculating the proportion of times or periods when demand for various consumables cannot be met due to insufficient inventory relative to the total number of times or periods of demand. k ;

[0031] On-time delivery rate is calculated by dividing the number of orders delivered on time by the total number of orders. The average reliability of all suppliers' supply is calculated by dividing the on-time delivery rate of all suppliers by the total number of suppliers. k .

[0032] In a preferred embodiment, the time series model used is the ARIMAX model. The specific steps for resource allocation prediction analysis using the time series model are as follows:

[0033] Step A1: Obtain the data for prediction;

[0034] Step A2: Create an ARIMAX model;

[0035] Step A3: Use the maximum likelihood estimation (MLE) method to estimate the ARIMAX model parameters;

[0036] Step A4: Verify the effectiveness of the fitted model by checking the goodness of fit of the model using residual analysis.

[0037] Step A5: Use the fitted model to predict the future inventory consumable allocation plan, and take the maximum value of the prediction result as the highest point of consumable usage in the future calculation cycle.

[0038] In a preferred embodiment, the exogenous variables in the ARIMAX model include total volatility, inventory turnover rate, stockout rate, and average reliability of supply from all suppliers. The ARIMAX model formula is as follows:

[0039]

[0040] In the formula, y t The consumable requirements for the current computing cycle are represented by α, where α is a constant term. Let θ be the i-th order autoregressive parameter, p be the order of the autoregressive term, and θ be the autoregressive parameter. j Let y be the parameter of the j-th moving average. t-i It is the predicted target value with a lag of i periods, where q is the order of the moving average term, ∈ t-j It is a white noise term with a lag of j periods, ∈ t It is a white noise term, representing random error, β k It is an exogenous variable X t-k The coefficient, m is the lag order of the exogenous variable, X t-k It is an exogenous variable lagged by k periods.

[0041] In a preferred embodiment, in step A3, α, θ1, β1, and β2 are obtained by the maximum likelihood estimation method.

[0042] The technical effects and advantages of this invention are as follows:

[0043] 1. This invention establishes a data analysis model by collecting prediction error analysis information and demand fluctuation information, obtains accuracy evaluation coefficients, compares them with accuracy thresholds, obtains the accuracy comparison results of each prediction count within the current calculation cycle, counts the accuracy evaluation coefficients less than the accuracy threshold, sorts them by value, obtains the total fluctuation intensity, compares it with the fluctuation threshold, and evaluates the prediction model based on the comparison results. This ensures the accuracy and precision of the prediction model, while setting an adjustment range for the prediction model to optimize it, improving the reliability of high-value consumables management and reducing operating costs.

[0044] 2. This invention obtains total fluctuation intensity, inventory turnover rate, stockout rate, and average reliability of supply from all suppliers. It then uses a time series model to predict and analyze the allocation of consumable inventory, obtaining consumable demand data. Based on the maximum and minimum values ​​of the predicted results, the allocation of consumable inventory is dynamically adjusted, thereby refining and optimizing the allocation scheme. This improves the accuracy of inventory management and the quality of clinical diagnosis and medical services. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of a module of a high-value medical consumables big data analysis and management system according to the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] This invention discloses a big data analysis and management system for high-value medical consumables, which is based on the SPD model for "zero inventory" management of high-value consumables. By conducting comprehensive testing of the SPD platform and configuring a prediction model according to the hospital's consumable usage, the system predicts the demand for high-value consumables. The prediction accuracy is ensured by comprehensively analyzing the prediction error and volatility.

[0048] Example 1

[0049] This invention discloses a big data analysis and management system for high-value medical consumables, such as... Figure 1 As shown, it includes a parameter acquisition module, a parameter processing module, a precision analysis module, and an inventory adjustment module; the modules are interconnected by signals.

[0050] The parameter acquisition module is used to collect forecast error analysis information and demand fluctuation information. Through data processing, it obtains the average absolute error rate, forecast coverage, demand volatility, and seasonality index and their consistency with actual demand changes, and sends them to the parameter processing module. At the same time, it collects inventory data and supplier information and sends them to the inventory adjustment module.

[0051] The forecast error analysis information includes the average absolute error rate and forecast coverage; the demand fluctuation information includes the demand volatility and the degree of agreement between the seasonal index and actual demand changes.

[0052] The calculation period for the following parameters will be determined first, i.e., for all predictions, it will be set as "1 week", "1 month" or "1 year", etc. The specific setting period will be implemented by the researchers based on the hospital's management needs or the frequency of use of consumables, and will not be limited here.

[0053] The mean absolute error rate (MAR) is a metric that measures the accuracy of a forecasting model over a given period. It represents the percentage deviation between the predicted and actual values ​​within that period, calculated by averaging the predicted and actual values. The formula used is:

[0054]

[0055] In the formula, A i For the actual demand during the i-th cycle, F i For the i-th forecast demand within the period, Ma i Let be the average absolute error rate, where i is the i-th prediction within the period, and n is the total number of predictions within the period.

[0056] The logic for obtaining the predicted coverage rate is as follows: If the predicted value is greater than or equal to the actual demand for medical consumables within the calculation period, it is considered that the predicted demand covers the actual demand, and this is counted as one successful coverage. The number of successful coverages within the calculation period is counted and compared with the total number of successful coverages to obtain the predicted coverage rate (Cr). i .

[0057] The prediction coverage rate is not a fixed value, but varies depending on the number of predictions within the calculation period.

[0058] It should be noted that the calculation period for the predicted coverage rate is the same as the calculation period for the mean absolute error rate.

[0059] The logic for obtaining demand volatility is to acquire actual demand data within the calculation period and then calculate the average demand during that period using the following formula:

[0060]

[0061] In the formula, A i The actual demand during the period is for the i-th time. This represents the average demand during the period.

[0062] Through formula Calculate the deviation between the actual demand for each forecast and the average demand, i.e., demand fluctuation.

[0063] Demand volatility is expressed as the magnitude of change in quantity demanded relative to average quantity demanded, and is based on the following formula:

[0064]

[0065] The demand volatility Dv is obtained.

[0066] The seasonality index and its correlation with actual demand changes refer to the degree of change in demand within a calculation period (such as a month or season) relative to the annual average demand. The logic behind this is to collect demand data over many years, classify it by calculation period, calculate the average demand for each calculation period within each year, and then use the total demand of all calculation periods and the total number of calculation periods to obtain the overall average demand. The ratio of this average demand to the average demand of the current calculation period is then used to obtain the seasonality index for the current calculation period.

[0067] Actual demand change refers to the change in actual demand during the calculation period as the number of forecasts increases, and the ratio of this change to the seasonality index yields the degree of agreement between the seasonality index and actual demand change, Si.

[0068] The parameter processing module receives prediction error analysis information and demand fluctuation information, establishes a data analysis model, obtains accuracy evaluation coefficients, and sends them to the accuracy analysis module.

[0069] To obtain the mean absolute error rate, forecast coverage, demand volatility, and the degree of agreement between the seasonality index and actual demand changes, a data analysis model is established to obtain the accuracy evaluation coefficient Ac. i The formula used is as follows:

[0070]

[0071] In the formula, Ac i For accuracy evaluation coefficients, as well as These are preset proportional coefficients for the mean absolute error rate, forecast coverage, demand volatility, and the degree of agreement between the seasonal index and actual demand changes, respectively. as well as All are greater than 0.

[0072] Among them, the mean absolute error rate, forecast coverage, demand volatility, and the degree of agreement between the seasonality index and actual demand changes are all data-driven representations of long-term forecast deviations and accuracy assessment coefficients.

[0073] As can be seen from the above formula, the greater the mean absolute error rate and the greater the demand volatility, the greater the difficulty of forecasting, which leads to lower accuracy of the forecasting model and a lower accuracy evaluation coefficient. Conversely, the greater the consistency between the forecast coverage and the seasonality index and the actual demand changes, the more accurately the forecasting model can cover the actual demand, the higher the model accuracy, and the higher the accuracy evaluation coefficient.

[0074] The accuracy analysis module is used to obtain the accuracy evaluation coefficients and compare them with the accuracy threshold to obtain the accuracy comparison results of each prediction number in the current calculation cycle. It counts the accuracy evaluation coefficients that are less than the accuracy threshold and sorts them by value. It obtains the total fluctuation intensity and compares it with the fluctuation threshold. Based on the comparison results, it evaluates the prediction model and sends the total fluctuation intensity to the inventory adjustment module.

[0075] The logic for obtaining the accuracy threshold is to collect a set of distributed samples from the historical prediction accuracy analysis database, then divide the dataset into a training set and a test set, set evaluation metrics and clustering algorithms, train the model on the training set in each iteration of cross-validation, evaluate the model performance on the test set, and then adjust the accuracy threshold based on the performance of the validation set. Therefore, the accuracy threshold is constantly iterated and updated.

[0076] In this invention, clustering algorithm is a type of unsupervised learning algorithm used to standardize all feature data to ensure that each feature is compared on the same scale. K initial centroids are randomly selected, and each data point is assigned to the cluster containing the nearest centroid according to Euclidean distance. The new centroid of each cluster is calculated and used as the centroid of the next iteration. The iteration is repeated until the centroid no longer changes or the change is less than a set threshold, and the final accuracy threshold is determined.

[0077] Obtain the accuracy evaluation coefficients and compare and analyze them with the continuously iterative accuracy thresholds.

[0078] If the accuracy evaluation coefficient is greater than or equal to the accuracy threshold, the prediction is marked as an accurate prediction and an accurate signal is generated.

[0079] If the accuracy evaluation coefficient is less than the accuracy threshold, the current prediction is marked as an error prediction and an error signal is generated.

[0080] The values ​​of all accuracy evaluation coefficients that are less than the accuracy threshold within the current calculation period are sorted from largest to smallest to smooth out occasional numerical jumps in the natural number of calculations. This allows for an objective view of the differences between values ​​to obtain the total fluctuation intensity.

[0081] Let Z be the total number of accuracy evaluation coefficients less than the accuracy threshold, then Ac g Let g be the accuracy evaluation coefficient that is less than the accuracy threshold, where g is the g-th accuracy evaluation coefficient that is less than the accuracy threshold; then according to the formula:

[0082] P=|Ac1-Ac2|+|Ac2-Ac3|+…+|Ac Z-1 -Ac Z |

[0083] Where P is the total fluctuation intensity, and its fluctuation threshold is compared.

[0084] If the total volatility intensity is greater than or equal to the volatility threshold, an alarm will be issued, and the error in the prediction model will be reported. The weights of each parameter in the prediction model will be readjusted, for example, by adjusting the parameter weights by 20% to adjust the prediction accuracy.

[0085] If the total fluctuation intensity is less than the fluctuation threshold, the current calculation period is marked for use in setting the subsequent accuracy threshold and the weights for calculating the bias assessment parameters of the prediction model.

[0086] The fluctuation threshold is obtained through a comprehensive analysis of historical calculation cycles and current application parameters.

[0087] This invention establishes a data analysis model by collecting prediction error analysis information and demand fluctuation information, obtains accuracy evaluation coefficients, compares them with accuracy thresholds, obtains the accuracy comparison results of each prediction count within the current calculation cycle, counts the accuracy evaluation coefficients that are less than the accuracy threshold, sorts them by value, obtains the total fluctuation intensity, and compares it with the fluctuation threshold. Based on the comparison results, the prediction model is evaluated, thereby ensuring the accuracy and precision of the prediction model. At the same time, the prediction model is optimized by setting an adjustment range, improving the reliability of high-value consumables management and reducing operating costs.

[0088] Example 2

[0089] In Embodiment 1 of this invention, a key example is given of collecting prediction error analysis information and demand fluctuation information, establishing a data analysis model, obtaining accuracy evaluation coefficients, comparing them with accuracy thresholds, obtaining accuracy comparison results for each prediction count within the current calculation cycle, statistically analyzing accuracy evaluation coefficients less than the accuracy threshold, sorting them by value, obtaining the total fluctuation intensity, comparing it with the fluctuation threshold, and evaluating the operational strategy of the prediction model based on the comparison results. However, Embodiment 1 only improves the accuracy of the prediction model by optimizing the prediction model for the current calculation cycle, but does not consider the dynamic management problem of high-value medical consumables inventory under extreme peak conditions, which traditional supply and demand balancing cannot account for. Obviously, how to manage inventory to both avoid the problem of clearing out inventory due to excessively low inventory and curb the waste of resources due to excessively high inventory is a challenge. Embodiment 2 of this invention further refines the above-mentioned issues.

[0090] The inventory adjustment module receives total fluctuation intensity, inventory data, and supplier information. Through data processing, it obtains inventory turnover rate, stockout rate, and average reliability of supply from all suppliers. It then uses a time series model to perform predictive analysis on inventory consumable allocation, further refining and optimizing the inventory consumable allocation plan.

[0091] The inventory data includes inventory turnover rate and stockout rate, while the supplier information includes the average reliability of supply from all suppliers.

[0092] Inventory turnover deviation refers to the number of times various high-value medical consumables are used and replenished within the calculation period. It is obtained by statistically analyzing the ratio of the number of times consumables are used within the inventory to the actual average inventory level during the calculation period, then subtracting this ratio from the target inventory turnover rate to obtain the inventory turnover deviation (It). k Where k represents the kth category of high-value medical consumables.

[0093] It should be noted that the target inventory turnover rate was set by the researchers based on the operational needs of the medical institution and historical usage data, and will not be limited here.

[0094] The logic for obtaining the stockout rate is to calculate the proportion of times or periods during which various consumables cannot meet demand due to insufficient inventory relative to the total number of demand times or periods within the calculation period, thus obtaining the stockout rate Sr. k .

[0095] The average reliability of supply from all suppliers is obtained by averaging the on-time delivery rates of all suppliers. The on-time delivery rate is calculated as the ratio of the number of orders delivered on time to the total number of orders. The average reliability of supply from all suppliers (Dr) is then calculated by dividing the on-time delivery rates of all suppliers by the total number of suppliers. k .

[0096] It should be noted that the time series model used in this embodiment is the ARIMAX model. Refining and optimizing the inventory consumable allocation scheme means that, under the condition that the total fluctuation intensity is greater than or equal to the fluctuation threshold, further analysis and adjustment are made to make the inventory consumable allocation scheme more scientific and reasonable, thereby achieving control over the energy consumption of LED panels.

[0097] Furthermore, the specific steps for resource allocation forecasting analysis using time series models are as follows:

[0098] Step A1: Obtain the data for prediction;

[0099] Step A2: Create an ARIMAX model;

[0100] Step A3: Use the maximum likelihood estimation (MLE) method to estimate the ARIMAX model parameters;

[0101] Step A4: Verify the effectiveness of the fitted model by checking the goodness of fit of the model using residual analysis.

[0102] Step A5: Use the fitted model to predict the future inventory consumable allocation plan, and take the maximum value of the prediction result as the highest point of consumable usage in the future calculation cycle.

[0103] Specifically, the forecasting data includes total volatility intensity, inventory turnover rate, stockout rate, average reliability of supply from all suppliers, and consumable demand data; among them, the total volatility intensity corresponding to the calculation period has been exemplified in Example 1 and will not be repeated here.

[0104] Consumable demand data refers to historical consumable demand data, which serves as the main variable in the time series.

[0105] Furthermore, the basic form of the ARIMAX model is:

[0106]

[0107] In the formula, y t The consumable requirements for the current computing cycle are represented by α, where α is a constant term. Let θ be the i-th order autoregressive parameter, p be the order of the autoregressive term, and θ be the autoregressive parameter. j Let y be the parameter of the j-th moving average. t-i It is the predicted target value with a lag of i periods, where q is the order of the moving average term, ∈ t-j It is a white noise term with a lag of j periods, ∈ t It is a white noise term, representing random error, β k It is an exogenous variable X t-k The coefficient, m is the lag order of the exogenous variable, X t-k It is an exogenous variable lagged by k periods.

[0108] It should be noted that the exogenous variable section can integrate the impact of other relevant variables (such as total volatility, inventory turnover rate, stockout rate, and average reliability of supply from all suppliers) on the demand for consumables.

[0109] It should be noted that in step A3, α, θ1, β1, and β2 are calculated using the maximum likelihood estimation (MLE) method. The specific steps are as follows:

[0110] Error term ∈ t Follows a normal distribution N(0,σ) 2 If ), then the likelihood function is:

[0111]

[0112] Taking the logarithm of the likelihood function yields the log-likelihood function:

[0113]

[0114] The parameter estimates α are obtained by maximizing the log-likelihood function. θ1, β1, β2.

[0115] By predicting the maximum and minimum values ​​of the results, the allocation strategy for inventory consumables in future calculation cycles is dynamically adjusted.

[0116] For different categories of consumables, the prediction results were obtained by incorporating them into a time series model.

[0117] The following is an example of this embodiment:

[0118] Maximum value of the prediction result This represents the peak demand for consumables in the future computing cycle. A peak-period allocation strategy is formulated during the predicted peak demand period (i.e., close to or equal to the peak demand). At the specified time point, all suppliers are ranked by their on-time delivery rate from highest to lowest, and the supplier with the highest on-time delivery rate is given a 20% increase in supply to ensure supply and demand balance, avoid inventory shortages leading to clearance issues, and achieve "zero inventory" management of high-value consumables based on the SPD model.

[0119] Based on the lowest value of the prediction results Or, during periods of low demand, formulate a low-demand allocation strategy, targeting inventory consumables during periods of low demand (i.e., close to or below the current level). (At the specified time point), all suppliers are ranked according to their on-time delivery rate from smallest to largest, and the supplier with the lowest on-time delivery rate is reduced by 15% in terms of supply volume, in order to ensure supply and demand balance and curb the waste of medical resources.

[0120] This invention obtains total fluctuation intensity, inventory turnover rate, stockout rate, and average reliability of supply from all suppliers. It then uses a time series model to predict and analyze the allocation of consumable inventory, obtaining consumable demand data. Based on the maximum and minimum values ​​of the predicted results, the allocation of consumable inventory is dynamically adjusted, thereby refining and optimizing the allocation scheme. This improves the accuracy of inventory management and the quality of clinical diagnosis and medical services.

[0121] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0122] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0123] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0124] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0125] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0126] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0127] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0128] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0129] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0130] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A big data analysis and management system for high-value medical consumables, characterized in that: It includes a parameter acquisition module, a parameter processing module, an accuracy analysis module, and an inventory adjustment module; signal connections between the modules; The parameter acquisition module is used to collect forecast error analysis information and demand fluctuation information; through data processing, it obtains the average absolute error rate, forecast coverage, demand volatility, and seasonality index and their consistency with actual demand changes, and sends them to the parameter processing module. At the same time, it collects inventory data and supplier information and sends them to the inventory adjustment module. The parameter processing module is used to receive prediction error analysis information and demand fluctuation information, establish a data analysis model, obtain accuracy evaluation coefficients, and send them to the accuracy analysis module. The accuracy analysis module is used to obtain the accuracy evaluation coefficients and compare them with the accuracy threshold to obtain the accuracy comparison results of each prediction number in the current calculation cycle. It counts the accuracy evaluation coefficients that are less than the accuracy threshold and sorts them by value. It obtains the total fluctuation intensity and compares it with the fluctuation threshold. Based on the comparison results, it evaluates the prediction model and sends the total fluctuation intensity to the inventory adjustment module. The inventory adjustment module receives total fluctuation intensity, inventory data, and supplier information. Through data processing, it obtains inventory turnover rate, stockout rate, and average reliability of supply from all suppliers. It then uses a time series model to perform predictive analysis on inventory consumable allocation, further refining and optimizing the inventory consumable allocation plan.

2. The high-value medical consumables big data analysis and management system according to claim 1, characterized in that: First, determine the calculation period. Then, calculate the mean absolute error rate (Ma) by averaging the percentage difference between the predicted and actual values ​​within that period relative to the actual values. i Where i represents the i-th prediction within the period; Successful coverage is defined as a predicted value being greater than or equal to the actual demand within a calculation period. The number of successful coverages within the calculation period is counted and compared to the total number of successful coverages to obtain the prediction coverage rate (Cr). i ; The demand volatility Dv is calculated by calculating the average demand during the period and based on the magnitude of the change in demand relative to the average demand. The overall average demand is obtained by dividing the total demand of all calculation cycles by the total number of calculation cycles. The seasonality index of the current calculation cycle is obtained by dividing the seasonality index by the ratio of the total demand of all calculation cycles and the total number of calculation cycles. The degree of agreement between the seasonality index and the actual demand change is obtained by dividing the seasonality index by the ratio of the seasonality index to the actual demand change.

3. The high-value medical consumables big data analysis and management system according to claim 2, characterized in that: To obtain the mean absolute error rate, forecast coverage, demand volatility, and the degree of agreement between the seasonality index and actual demand changes, a data analysis model is established to obtain the accuracy evaluation coefficient Ac. i The formula used is as follows: In the formula, Ac i For accuracy evaluation coefficients, as well as These are preset proportional coefficients for the mean absolute error rate, forecast coverage, demand volatility, and the degree of agreement between the seasonal index and actual demand changes, respectively. as well as All are greater than 0.

4. The high-value medical consumables big data analysis and management system according to claim 3, characterized in that: Obtain the accuracy evaluation coefficients and compare and analyze them with the continuously iterative accuracy thresholds; If the accuracy evaluation coefficient is greater than or equal to the accuracy threshold, the current prediction is marked as an accurate prediction and an accurate signal is generated. If the accuracy evaluation coefficient is less than the accuracy threshold, the current prediction is marked as an error prediction and an error signal is generated. The values ​​of all accuracy evaluation coefficients that are less than the accuracy threshold within the current calculation period are sorted from largest to smallest to smooth out occasional numerical jumps in the natural number of calculations. This allows for an objective view of the differences between values ​​to obtain the total fluctuation intensity. Let Z be the total number of accuracy evaluation coefficients less than the accuracy threshold, then Ac g Let g be the accuracy evaluation coefficient that is less than the accuracy threshold, where g is the g-th accuracy evaluation coefficient that is less than the accuracy threshold; then according to the formula: P=|Ac1-Ac2|+|Ac2-Ac3|+…+|Ac Z-1 -Ac Z | Where P is the total fluctuation intensity, and it is compared with the fluctuation threshold.

5. The high-value medical consumables big data analysis and management system according to claim 4, characterized in that: If the total fluctuation intensity is greater than or equal to the fluctuation threshold, an alarm will be issued, and the error in the prediction model will be reported, and the weights of each parameter in the prediction model will be readjusted. If the total fluctuation intensity is less than the fluctuation threshold, then the current calculation period is marked.

6. The high-value medical consumables big data analysis and management system according to claim 1, characterized in that: Inventory turnover rate is obtained by statistically calculating the ratio of the number of times consumables are used within the inventory during the period to the actual average inventory. The deviation It is then obtained by subtracting it from the target inventory turnover rate. k Where k represents the kth category of high-value medical consumables; The stockout rate Sr is obtained by calculating the proportion of times or periods when demand for various consumables cannot be met due to insufficient inventory relative to the total number of times or periods of demand. k ; On-time delivery rate is calculated by dividing the number of orders delivered on time by the total number of orders. The average reliability of all suppliers' supply is calculated by dividing the on-time delivery rate of all suppliers by the total number of suppliers. k .

7. The high-value medical consumables big data analysis and management system according to claim 6, characterized in that: The time series model uses the ARIMAX model. The specific steps for resource allocation forecasting analysis using the time series model are as follows: Step A1: Obtain the data for prediction; Step A2: Create an ARIMAX model; Step A3: Use the maximum likelihood estimation (MLE) method to estimate the ARIMAX model parameters; Step A4: Verify the effectiveness of the fitted model by checking the goodness of fit of the model using residual analysis. Step A5: Use the fitted model to predict the future inventory consumable allocation plan, and take the maximum value of the prediction result as the highest point of consumable usage in the future calculation cycle.

8. The high-value medical consumables big data analysis and management system according to claim 7, characterized in that: The exogenous variables in the ARIMAX model include total volatility, inventory turnover, stockout rate, and average reliability of supply from all suppliers. The ARIMAX model formula is: In the formula, y t The consumable requirements for the current computing cycle are represented by α, which is a constant term. Let θ be the i-th order autoregressive parameter, p be the order of the autoregressive term, and θ be the autoregressive parameter. j Let y be the parameter of the j-th moving average. t-i It is the predicted target value with a lag of i periods, where q is the order of the moving average term, ∈ t-j It is a white noise term with a lag of j periods, ∈ t It is a white noise term, representing random error, β k It is an exogenous variable X t-k The coefficient, m is the lag order of the exogenous variable, X t-k It is an exogenous variable lagged by k periods.

9. The high-value medical consumables big data analysis and management system according to claim 8, characterized in that: In step A3, α, θ1, β1, and β2 are obtained by the maximum likelihood estimation method.

Images