Ophthalmologic nursing medicine stock management method and system

Abstract

The invention relates to the field of drug inventory data prediction, in particular to an ophthalmic nursing drug inventory management method and system. The method comprises the following steps: obtaining a data fluctuation degree according to data distribution of each data neighborhood in a drug consumption data sequence of a target drug, and obtaining a trend term weight coefficient in combination with data distribution in the sequence; obtaining a seasonal item weight coefficient according to data cycle characteristics in the sequence; according to the data fluctuation degree and the data dispersion degree in the sequence, obtaining a residual item weight coefficient; adjusting the initial prediction trend item, the initial prediction season item and the initial prediction residual item to obtain final drug use prediction data of the target drug use at each sampling moment in the future period; and carrying out inventory management on the ophthalmic nursing medicines according to the final prediction data of the medicines. According to the method, the intensity performance of each ophthalmology nursing medicine in different decomposition items is considered, an accurate prediction result is obtained, and the ophthalmology nursing medicine can be subjected to accurate inventory management.

Description

Technical Field

[0001] The present invention relates to the field of drug inventory data forecasting, and in particular to an ophthalmic care drug inventory management method and system. Background Art

[0002] Ophthalmic care medications mainly refer to various drugs for eye diseases, such as antibiotics, anti-inflammatory drugs, and intraocular pressure-lowering drugs. These drugs play an important role in the prevention, treatment, and control of ophthalmic diseases. Since ophthalmic care medications have certain requirements for packaging forms and storage conditions, in order to ensure the safety and effectiveness of drugs, medical institutions need to conduct reasonable and accurate inventory management, thereby improving medical quality and reducing waste and idle resources. In the prior art, inventory management of ophthalmic care medications is usually based on a medical supplies inventory management system (such as a HIS system). This system sets upper and lower limits for inventory to monitor inventory levels, and uses time series decomposition methods to predict future drug consumption in order to achieve early warning effects.

[0003] Prior art time series decomposition methods decompose the time series consumption data of different types of ophthalmic care medications to obtain trend terms, seasonal terms, and residual terms. These decomposition terms are then forecasted and combined to obtain future drug consumption forecasts. However, in practice, conventional time series decomposition methods fail to account for the diverse uses, methods, dosages, and patient populations of each ophthalmic care medication. This results in varying strengths for each ophthalmic care medication across different decomposition terms, leading to significant discrepancies between forecast results and actual conditions, making it impossible to accurately manage ophthalmic care medication inventory. Summary of the Invention

[0004] In order to solve the technical problem that in actual situations, conventional time series decomposition methods do not take into account the different uses, usage, dosages, and target patient groups of each ophthalmic care medication, resulting in different strength performance of each ophthalmic care medication in different decomposition items, causing a large difference between the prediction results and the actual situation, and thus making it impossible to accurately manage the inventory of ophthalmic care medications, the purpose of the present invention is to provide an ophthalmic care medication inventory management method and system. The technical solutions adopted are as follows:

[0005] A method for managing ophthalmic care medication inventory, the method comprising:

[0006] Obtain drug consumption data for each ophthalmic care drug at each sampling time during a preset period;

[0007] The drug consumption data of any one ophthalmic care drug within a preset period are selected to form a drug consumption data sequence for the target drug; based on the data distribution characteristics of each drug consumption data within a preset neighborhood range in the drug consumption data sequence, the data fluctuation degree of each drug consumption data is obtained; based on the data distribution characteristics and data fluctuation degree of all drug consumption data in the drug consumption data sequence, the trend item weight coefficient of the drug consumption data sequence is obtained; based on the data period characteristics of all drug consumption data in the drug consumption data sequence, the seasonal item weight coefficient of the drug consumption data sequence is obtained; based on the data fluctuation degree and data discrete characteristics of all drug consumption data in the drug consumption data sequence, the residual item weight coefficient of the drug consumption data sequence is obtained;

[0008] Obtaining, based on the data distribution of the drug consumption data in the drug consumption data sequence, an initial predicted trend item, an initial predicted season item, and an initial predicted residual item of the target drug at each sampling moment in the future period; adjusting the initial predicted trend item, the initial predicted season item, and the initial predicted residual item using the trend item weight coefficient, the seasonal item weight coefficient, and the residual item weight coefficient to obtain final predicted data of the target drug usage at each sampling moment in the future period;

[0009] Inventory management of ophthalmic care medications is performed based on the final medication forecast data.

[0010] Furthermore, the method for obtaining the data fluctuation degree includes:

[0011] Select any drug consumption data in the drug consumption data sequence as target drug data;

[0012] The data fluctuation degree is obtained according to the data fluctuation degree calculation formula, and the data fluctuation degree calculation formula is as follows:

[0013]

[0014] Where U0 represents the data value of the target drug data; μ represents the data fluctuation degree of the target drug data; K represents the amount of drug consumption data within the preset range of the target drug data; U k Represents the data value of the kth drug consumption data within the preset range of the target drug data; L k Indicates the time difference between the sampling time of the kth drug consumption data within the preset area and the sampling time of the target drug data.

[0015] Furthermore, the method for obtaining the trend item weight coefficient includes:

[0016] The trend item weight coefficient is obtained according to the trend item weight coefficient calculation formula, and the trend item weight coefficient calculation formula is as follows:

[0017]

[0018] Where, represents the trend item weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; U i+1 Represents the data value of the i+1th drug consumption data in the drug consumption data sequence; U i represents the data value of the i-th drug consumption data in the drug consumption data sequence; μ i+1 Indicates the data fluctuation degree of the i+1th drug consumption data in the drug consumption data sequence; μ i It represents the degree of data fluctuation of the i-th drug consumption data in the drug consumption data sequence; norm[] represents the normalization function.

[0019] Furthermore, the method for obtaining the seasonal item weight coefficient includes:

[0020] Divide the preset period into equal parts to obtain all medication data time periods;

[0021] The seasonal item weight coefficient is obtained according to the seasonal item weight coefficient calculation formula, and the seasonal item weight coefficient calculation formula is as follows:

[0022]

[0023] Where, represents the seasonal weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; U i Represents the data value of the i-th drug consumption data in the drug consumption data sequence; represents the data mean of all drug consumption data in the medication data time period where the i-th drug consumption data in the drug consumption data sequence is located; J represents the number of medication data time periods in the drug consumption data sequence; V j represents the data and value of drug consumption data in the jth medication data time period; It represents the mean value of drug consumption data for each year in the drug consumption data series; norm[] represents the normalization function.

[0024] Furthermore, the method for obtaining the residual term weight coefficient includes:

[0025] Obtaining a residual item of each drug consumption data in the drug consumption data sequence;

[0026] The residual term weight coefficient is obtained according to the residual term weight coefficient calculation formula, and the residual term weight coefficient calculation formula is as follows:

[0027]

[0028] Where, represents the residual weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; μ i represents the data fluctuation degree of the i-th drug consumption data in the drug consumption data sequence; σ(R) represents the variance of the residual term of all drug consumption data in the drug consumption data sequence;

[0029] σ(U) represents the data variance of the drug consumption data series; norm[] represents the normalization function.

[0030] Furthermore, based on the data distribution of the drug consumption data in the drug consumption data sequence, the initial prediction trend item, the initial prediction season item, and the initial prediction residual item of the ophthalmic care drugs corresponding to the drug consumption data sequence at each sampling moment in the future period are obtained, including:

[0031] A decomposition algorithm is applied to the drug consumption data sequence to obtain a trend item, a seasonal item, and a residual item for each drug consumption data in the drug consumption data sequence;

[0032] Fitting trend items of all drug consumption data to obtain a trend item data curve; obtaining an initial predicted trend item at each sampling moment in a future period based on the trend item data curve;

[0033] Obtain the target drug consumption seasonal cycle based on the seasonal items of all drug consumption data in the drug consumption data sequence, and use the mean of the seasonal items of drug consumption data of other consumption seasonal cycles at the same location in the future period as the initial predicted seasonal item for each sampling moment in the future period;

[0034] The moving average method is used for the residual items of each drug consumption data to obtain the initial predicted residual items at each sampling moment in the future period.

[0035] Furthermore, the method for obtaining the final medication prediction data includes:

[0036] The final medication prediction data is obtained according to the calculation formula of the final medication prediction data. The calculation formula of the final medication prediction data is as follows:

[0037]

[0038] Where U t T represents the final predicted data of medication at the t-th sampling moment in the future period; t represents the initial forecast trend item at the t-th sampling moment in the future period; S t represents the initial forecast seasonal term at the tth sampling moment in the future period; Rt represents the initial prediction residual term at the t-th sampling moment in the future period; Represents the trend item weight coefficient of the drug consumption data series; Represents the seasonal weight coefficient of the drug consumption data series; Represents the residual weight coefficient of the drug consumption data series.

[0039] Furthermore, inventory management of ophthalmic care medications is performed based on the final medication forecast data, including:

[0040] Obtain the remaining inventory of the target medication at the current sampling time;

[0041] Obtaining the maximum usage time of the target medication based on the remaining amount of the medication inventory and the final predicted medication usage data at each sampling moment in the future period;

[0042] When the longest usage time is not greater than a preset first threshold, it is considered that the target medication has a risk of insufficient inventory; when the longest usage time is not less than a preset second threshold, it is considered that the target medication has a risk of inventory backlog.

[0043] Furthermore, based on the remaining amount of the drug inventory and the final predicted data of drug use at each sampling moment in the future period, the maximum usage time of the target drug is obtained, specifically:

[0044] Calculate the final predicted data and value of medication at all sampling moments in the future period as the first sum value;

[0045] When the remaining amount of the target medication in stock is not less than the first sum value, the length of the future period corresponding to the maximum value of the first sum value is taken as the maximum usage time of the target medication.

[0046] An ophthalmic care medication inventory management system, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the steps of the above-mentioned ophthalmic care medication inventory management method when executing the computer program.

[0047] The present invention has the following beneficial effects:

[0048] The present invention first obtains the drug consumption data of each ophthalmic care drug at each sampling moment in a preset period for subsequent inventory forecast analysis; since the consumption of ophthalmic care drugs is random in actual situations, conventional time series decomposition methods cannot make accurate predictions, so it is necessary to analyze the trend item, seasonal item and residual item after decomposition separately; since the sudden change in drug consumption cannot reflect the long-term characteristics of the consumption of such ophthalmic care drugs, it is necessary to reduce the influence of the sudden change in drug consumption in the trend item, so the data fluctuation degree of each drug consumption data is obtained; since the faster the change rate of drug consumption, the higher the weight of the trend item in the prediction result, and the drug consumption data with a large degree of data fluctuation cannot represent the long-term characteristics of drug consumption, so according to the data distribution characteristics and data fluctuation degree of the drug consumption data, the trend item weight coefficient is obtained; since ophthalmic care drugs are used in a certain period of time, the drug consumption data is used in a certain period of time, so ... The consumption of medicines is cyclical, so according to the data period characteristics of the medicine consumption data, the seasonal item weight coefficient is obtained; since medicine consumption often changes unpredictably and may cause the medicine consumption data to mutate, the degree of data fluctuation will affect the residual item of the medicine consumption data, and the residual item in the medicine consumption data may be affected by the trend item and the seasonal item, so according to the data fluctuation degree and data discrete characteristics of all medicine consumption data, the residual item weight coefficient is obtained; the existing technology is used to obtain the initial predicted trend item, initial predicted seasonal item and initial predicted residual item of the target medicine at each sampling moment in the future period; and the corresponding weights are adjusted by the trend item weight coefficient, the seasonal item weight coefficient and the residual item weight coefficient to obtain the final predicted data of the medicine at each sampling moment in the future period; the final predicted data of the medicine is used to manage the inventory of ophthalmic care medicines. The present invention takes into account the strength performance of each ophthalmic care medicine in different decomposition items, obtains relatively accurate prediction results, and thus can accurately manage the inventory of ophthalmic care medicines. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. 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 paying any creative work.

[0050] Figure 1 This is a flow chart of an ophthalmic care medication inventory management method provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0051] To further illustrate the technical means and effectiveness of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of an ophthalmic care medication inventory management method and system proposed by the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0052] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0053] The following describes in detail a method and system for managing ophthalmic care medication inventory provided by the present invention with reference to the accompanying drawings.

[0054] See also Figure 1 , which shows an ophthalmic care medication inventory management method provided by one embodiment of the present invention, the method comprising:

[0055] Step S1: Obtain the drug consumption data of each ophthalmic care drug at each sampling moment in a preset period.

[0056] The embodiments of the present invention are mainly applied to the inventory management scenario of ophthalmic care medications. In order to manage the inventory of ophthalmic care medications, it is necessary to obtain the maximum usage time of each ophthalmic care medication. Relevant personnel can adjust the inventory of each ophthalmic care medication by comparing the maximum usage days with the scheduled inventory update time of each ophthalmic care medication. In order to obtain the maximum usage time of each ophthalmic care medication, the drug consumption data of each ophthalmic care medication at each sampling moment is first obtained, and the future drug consumption is obtained through subsequent predictive analysis for judgment. The specific process is shown in the following embodiment.

[0057] In one embodiment of the present invention, the preset period is set to 5 years and the sampling time is set to 1 day. That is, the drug consumption data of each ophthalmic care medication on a daily basis over 5 years is obtained as the drug consumption data. It should be noted that in other embodiments of the present invention, the preset period and sampling time can be set arbitrarily and are not limited here.

[0058] Step S2: The drug consumption data of any one ophthalmic care drug within a preset period are selected to form a drug consumption data sequence for the target drug; the data fluctuation degree of each drug consumption data is obtained according to the data distribution characteristics within a preset neighborhood range of each drug consumption data in the drug consumption data sequence; the trend item weight coefficient of the drug consumption data sequence is obtained according to the data distribution characteristics and data fluctuation degree of all drug consumption data in the drug consumption data sequence; the seasonal item weight coefficient of the drug consumption data sequence is obtained according to the data period characteristics of all drug consumption data in the drug consumption data sequence; the residual item weight coefficient of the drug consumption data sequence is obtained according to the data fluctuation degree and data discrete characteristics of all drug consumption data in the drug consumption data sequence.

[0059] In the process of decomposing the drug consumption data series, due to the different consumption characteristics of different types of ophthalmic care drugs, for example, spring conjunctivitis is a highly prevalent ophthalmic disease in spring, and the consumption of anti-allergic drugs and corticosteroid eye drops required for treatment increases significantly in spring. However, in actual situations, there may be sudden events such as large-scale screening activities, air pollution, and chemical leaks, which lead to a sudden increase in the number of patients with ophthalmic diseases, and thus a sudden increase in the consumption of certain types of ophthalmic care drugs. The consumption of ophthalmic care drugs is random, and conventional time series decomposition methods cannot make accurate predictions. Therefore, it is necessary to analyze the trend term, seasonal term, and residual term after decomposition separately to reduce the impact of the randomness of drug consumption. Since the sudden change in drug consumption cannot reflect the long-term characteristics of the consumption of such ophthalmic care drugs, it is necessary to reduce the impact of the sudden change in drug consumption in the trend term. Therefore, in the embodiment of the present invention, first, according to the data distribution characteristics within the preset neighborhood range of each drug consumption data in the drug consumption data series, the data fluctuation degree of each drug consumption data is obtained.

[0060] In one embodiment of the present invention, the STL decomposition algorithm is used to obtain the trend term, seasonal term, and residual term in the drug consumption data in the drug consumption data sequence. It should be noted that in other embodiments of the present invention, the ARIMA model can also be used to decompose the drug consumption data in the drug consumption data sequence, which is not limited here.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining the degree of data fluctuation includes:

[0062] Select any drug consumption data in the drug consumption data sequence as target drug data;

[0063] In one embodiment of the present invention, the preset neighborhood range is set to 5 drug consumption data on each side of the target drug data. It should be noted that in other embodiments of the present invention, the preset neighborhood range can be set arbitrarily and is not limited here.

[0064] The data fluctuation degree is obtained according to the data fluctuation degree calculation formula, which is as follows:

[0065]

[0066] Where U0 represents the data value of the target drug data; μ represents the data fluctuation degree of the target drug data; K represents the amount of drug consumption data within the preset range of the target drug data; U k Represents the data value of the kth drug consumption data within the preset range of the target drug data; L k Indicates the time difference between the sampling time of the kth drug consumption data within the preset area and the sampling time of the target drug data.

[0067] In the data fluctuation degree calculation formula, the difference between the target drug data and the k-th drug consumption data within the preset neighborhood is calculated. The larger the difference, the higher the prominence of the target drug data within the preset neighborhood. At this time, if the time difference L between the sampling time of the k-th drug consumption data within the preset area and the sampling time of the target drug data k The smaller the value, the greater the change of target drug data in a shorter period of time, which means that the target drug data cannot be used as a trend item. As the difference weight between the target drug data and the k-th drug consumption data within the preset neighborhood, The larger the value is, the greater the data fluctuation of the target drug data will be.

[0068] The trend term reflects the long-term variation in ophthalmic care medication consumption. Generally, the faster the change in medication consumption, the higher the weight of the trend term in the prediction result. However, according to the above process, the degree of data fluctuation of each medication consumption data in the medication consumption data sequence is obtained. Medication consumption data with a large degree of data fluctuation cannot represent the long-term characteristics of medication consumption. Therefore, the weight of medication consumption data with a large degree of data fluctuation is reduced in the trend term of the medication consumption data sequence. Therefore, in this embodiment of the present invention, the trend term weight coefficient of the medication consumption data sequence is obtained based on the data distribution characteristics and data fluctuation degree of all medication consumption data in the medication consumption data sequence.

[0069] Preferably, in one embodiment of the present invention, the method for obtaining the trend item weight coefficient includes:

[0070] Obtain the trend item weight coefficient according to the trend item weight coefficient calculation formula. The trend item weight coefficient calculation formula is as follows:

[0071]

[0072] Where, represents the trend item weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; U i+1 Represents the data value of the i+1th drug consumption data in the drug consumption data sequence; U i represents the data value of the i-th drug consumption data in the drug consumption data sequence; μ i+1 Indicates the data fluctuation degree of the i+1th drug consumption data in the drug consumption data sequence; μ i It represents the degree of data fluctuation of the i-th drug consumption data in the drug consumption data sequence; norm[] represents the normalization function.

[0073] In the trend item weight coefficient calculation formula, the data difference between each two adjacent drug consumption data is calculated. The larger the data difference, the faster the target drug changes. At this time, if the data fluctuation degree of the two adjacent drug consumption data is smaller, it means that the two adjacent drug consumption data can better reflect the long-term change characteristics of the target drug consumption. Therefore, the reciprocal of the sum of the data fluctuation degrees of the two adjacent drug consumption data is taken as the data difference |U i+1 -U i The weight coefficient between | The larger the value is, the more obvious the trend of the two adjacent drug consumption data is. At this time, the product of the difference between each two adjacent drug consumption data and the weight coefficient is averaged. The larger the mean value is, the greater the weight of the trend item in the drug consumption data sequence of the target drug, that is, the greater the weight coefficient of the trend item in the drug consumption data sequence of the target drug.

[0074] In actual situations, the alternation of environmental climate will cause certain eye diseases to enter a high incidence period during a fixed period each year. At this time, the consumption of related eye care drugs will increase, and medical institutions will usually arrange more eye surgeries on weekends each week. At this time, the consumption of related eye care drugs will increase, reflecting that some drug consumption data in the drug consumption data sequence are periodic, that is, the seasonal item of the periodic drug consumption data is larger. Since the larger the seasonal item of the drug consumption data in the drug consumption data sequence, the greater the weight of the seasonal item in the prediction result, so in an embodiment of the present invention, the seasonal item weight coefficient of the drug consumption data sequence is obtained according to the data period characteristics of all drug consumption data in the drug consumption data sequence.

[0075] Preferably, in one embodiment of the present invention, the method for obtaining the seasonal item weight coefficient includes:

[0076] The preset period is evenly divided to obtain all medication data time periods. In one embodiment of the present invention, the medication data time period is set to 1 week. It should be noted that in other embodiments of the present invention, the medication data time period can be set arbitrarily and is not limited here.

[0077] The seasonal item weight coefficient is obtained according to the seasonal item weight coefficient calculation formula. The seasonal item weight coefficient calculation formula is as follows:

[0078]

[0079] Where, represents the seasonal weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; U i Represents the data value of the i-th drug consumption data in the drug consumption data sequence; represents the data mean of all drug consumption data in the medication data time period where the i-th drug consumption data in the drug consumption data sequence is located; J represents the number of medication data time periods in the drug consumption data sequence; V j represents the data and value of drug consumption data in the jth medication data time period; It represents the mean value of drug consumption data for each year in the drug consumption data series; norm[] represents the normalization function.

[0080] In the seasonal item weight coefficient calculation formula, the drug consumption data of different drug consumption data time periods in the drug consumption data sequence are analyzed. The smaller the difference between each drug consumption data and the data mean of all drug consumption data in the drug consumption data time period, the more likely the drug consumption data time period is to be the cycle of the drug consumption data sequence, that is, the stronger the periodicity of the drug consumption data sequence in the drug consumption data time period; since there are high incidence periods of certain ophthalmic diseases every year in actual situations, the consumption of related ophthalmic care drugs in this period will increase, so the data period characteristics of each drug consumption data time period in each year are analyzed. Among them, the greater the difference between the data and value in the drug consumption data time period and the mean of the drug consumption data in each year, the stronger the periodicity of the drug consumption data series in each year. At this time, the weight of the seasonal item in the drug consumption data sequence in the prediction result is greater, that is, the greater the seasonal item weight coefficient.

[0081] In reality, drug consumption often undergoes unpredictable changes, such as emergencies such as public health accidents. At this time, the drug consumption data will undergo sudden changes, that is, the data fluctuation degree of the drug consumption data affects the weight of the residual term in the prediction result. Because the residual term in the drug consumption data may be affected by trend terms and seasonal terms, it is necessary to remove the influence of trend terms and seasonal terms in the process of analyzing the weight of the residual term. Because there is no regularity in the residual term, in an embodiment of the present invention, the residual term weight coefficient of the drug consumption data sequence is obtained according to the data fluctuation degree and data discrete characteristics of all drug consumption data in the drug consumption data sequence.

[0082] Preferably, in one embodiment of the present invention, the method for obtaining the residual term weight coefficient includes:

[0083] The residual term of each drug consumption data in the drug consumption data sequence is obtained according to the STL decomposition algorithm.

[0084] The residual item weight coefficient is obtained according to the residual item weight coefficient calculation formula. The residual item weight coefficient calculation formula is as follows:

[0085]

[0086] Where, represents the residual weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; μ i It represents the degree of data fluctuation of the i-th drug consumption data in the drug consumption data sequence; σ(R) represents the variance of the residual consumption data; σ(U) represents the data variance of all drug consumption data in the drug consumption data sequence; norm[] represents the normalization function.

[0087] In the residual weight coefficient calculation formula, the mean of the data fluctuation degree of all drug consumption data in the drug consumption data series is The larger it is, the more drug consumption data that have mutated, the lower the predictability of the drug consumption data series, and the higher the weight of the residual term in the prediction result; the larger the ratio of the variance of the residual term of the drug consumption data to the data variance of all drug consumption data in the drug consumption data series, the higher the data dispersion of the residual term after removing the influence of the trend term and the seasonal term, and the higher the weight of the residual term in the prediction result, that is, the larger the weight coefficient of the residual term in the drug consumption data series.

[0088] Step S3: According to the data distribution of the drug consumption data in the drug consumption data sequence, the initial predicted trend item, the initial predicted season item and the initial predicted residual item of the target medication at each sampling moment in the future period are obtained; the initial predicted trend item, the initial predicted season item and the initial predicted residual item are adjusted using the trend item weight coefficient, the season item weight coefficient and the residual item weight coefficient to obtain the final predicted data of the target medication in the future period.

[0089] Preferably, in one embodiment of the present invention, the method for obtaining the initial predicted trend item, the initial predicted season item, and the initial predicted residual item of the target medication in the future period includes:

[0090] The STL decomposition algorithm is used on the drug consumption data series to obtain the trend term, seasonal term and residual term of each drug consumption data in the drug consumption data series.

[0091] Since the trend items of all drug consumption data in the drug consumption data sequence have strong trend characteristics, the trend items of all drug consumption data are fitted to obtain the trend item data curve; based on the trend item data curve, the initial predicted trend item at each sampling moment in the future period is obtained.

[0092] The seasonal consumption cycle of the target medication is obtained by analyzing the periodicity of the drug consumption data in the drug consumption data sequence. According to prior art, since the seasonal terms of drug consumption data at the same location within each seasonal consumption cycle are essentially the same, the mean of the seasonal terms of drug consumption data at the same location within each seasonal consumption cycle is used as the initial predicted seasonal term for each sampling moment in the future period.

[0093] The moving average method is used for the residual items of each drug consumption data to obtain the initial predicted residual items at each sampling moment in the future period.

[0094] It should be noted that the least square method, periodicity and moving average methods are all technical means well known to those skilled in the art and will not be described in detail here.

[0095] After obtaining the initial predicted trend item, initial predicted season item and initial predicted residual item at each sampling moment in the future period, they need to be adjusted to obtain the final predicted data of medication at each sampling moment in the future period.

[0096] Preferably, in one embodiment of the present invention, the calculation formula for the final medication prediction data includes:

[0097]

[0098] Where U t T represents the final predicted data of medication at the t-th sampling moment in the future period; trepresents the initial forecast trend item at the t-th sampling moment in the future period; S t represents the initial forecast seasonal term at the tth sampling moment in the future period; R t represents the initial prediction residual term at the t-th sampling moment in the future period; Represents the trend item weight coefficient of the drug consumption data series; Represents the seasonal weight coefficient of the drug consumption data series; Represents the residual weight coefficient of the drug consumption data series.

[0099] In the calculation formula for the final prediction data of medication use, the data values of the initial prediction trend item, initial prediction season item and initial prediction residual item at each sampling moment in the future period are equal. Since the respective strengths of the trend item, season item and residual item are different in actual situations, the trend item, season item and residual item corresponding to the final prediction data of medication use are not the same. Therefore, the trend item weight coefficient, season item weight coefficient and residual item weight coefficient of the drug consumption data series are used to adjust the initial prediction trend item, initial prediction season item and initial prediction residual item at each sampling moment in the future period, and the ratio between each item weight coefficient and the sum of the trend item weight coefficient, season item weight coefficient and residual item weight coefficient is calculated to obtain and Since there are three decomposition items for the final prediction data of medication, it is necessary to expand the product between the initial prediction item and the corresponding weight coefficient ratio after obtaining it. and

[0100] Multiply by the value 3 to obtain the final predicted data of medication.

[0101] Step S4: Inventory management of ophthalmic care medications is performed based on the final medication forecast data.

[0102] Preferably, in one embodiment of the present invention, inventory management of ophthalmic care medications is performed based on final medication forecast data, including:

[0103] Get the remaining amount of target medication in stock at the current sampling time.

[0104] Calculate the final predicted data and value of medication at all sampling moments in the future period as the first sum value; when the remaining inventory of the target medication is not less than the first sum value, the length of the future period corresponding to the maximum value of the first sum value is taken as the longest usage time of the target medication.

[0105] When the longest usage time is not greater than the preset first threshold, it is considered that the target medication is at risk of insufficient inventory; when the longest usage time is not less than the preset second threshold, it is considered that the target medication is at risk of inventory backlog.

[0106] In one embodiment of the present invention, the inventory update time N and the shelf life M of the target medication are obtained, and the preset first threshold is set to 1.2N. That is, when the maximum usage time is not greater than 1.2N, it is considered that the target medication is at risk of insufficient inventory and may be out of supply; the preset second threshold is set to 0.8M. That is, when the maximum usage time is not less than 0.8M, it is considered that the target medication is at risk of inventory backlog, which may cause the target medication to expire and be wasted. It should be noted that in other embodiments of the present invention, the preset first threshold and the preset second threshold can be set arbitrarily and are not limited here.

[0107] In summary, the drug consumption data of each ophthalmic care drug at each sampling moment in a preset period is obtained; the drug consumption data of any one ophthalmic care drug in a preset period is selected to form a drug consumption data sequence of the target drug; according to the data distribution characteristics within the preset neighborhood range of each drug consumption data in the drug consumption data sequence, the data fluctuation degree of each drug consumption data is obtained; according to the data distribution characteristics and data fluctuation degree of all drug consumption data in the drug consumption data sequence, the trend item weight coefficient of the drug consumption data sequence is obtained; according to the data period characteristics of all drug consumption data in the drug consumption data sequence, the seasonal item weight coefficient of the drug consumption data sequence is obtained; based on According to the data fluctuation degree and data discrete characteristics of all drug consumption data in the drug consumption data sequence, the residual item weight coefficient of the drug consumption data sequence is obtained; according to the data distribution of the drug consumption data in the drug consumption data sequence, the initial forecast trend item, initial forecast season item and initial forecast residual item of the target medication at each sampling moment in the future period are obtained; the trend item weight coefficient, seasonal item weight coefficient and residual item weight coefficient are used to adjust the initial forecast trend item, initial forecast season item and initial forecast residual item to obtain the final forecast data of the target medication at each sampling moment in the future period; inventory management of ophthalmic care medications is carried out according to the final forecast data of medication.

[0108] An embodiment of the present invention also provides an ophthalmic care medication inventory management system, which includes a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, and the processor is used to run the corresponding computer program. When the computer program runs in the processor, it can implement the method described in steps S1-S4.

[0109] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0110] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for managing ophthalmic care medication inventory, characterized in that: The method comprises: Obtain drug consumption data for each ophthalmic care drug at each sampling time during a preset period; The drug consumption data of any one ophthalmic care drug within a preset period are selected to form a drug consumption data sequence for the target drug; based on the data distribution characteristics of each drug consumption data within a preset neighborhood range in the drug consumption data sequence, the data fluctuation degree of each drug consumption data is obtained; based on the data distribution characteristics and data fluctuation degree of all drug consumption data in the drug consumption data sequence, the trend item weight coefficient of the drug consumption data sequence is obtained; based on the data period characteristics of all drug consumption data in the drug consumption data sequence, the seasonal item weight coefficient of the drug consumption data sequence is obtained; based on the data fluctuation degree and data discrete characteristics of all drug consumption data in the drug consumption data sequence, the residual item weight coefficient of the drug consumption data sequence is obtained; Obtaining, based on the data distribution of the drug consumption data in the drug consumption data sequence, an initial predicted trend item, an initial predicted season item, and an initial predicted residual item of the target drug at each sampling moment in the future period; adjusting the initial predicted trend item, the initial predicted season item, and the initial predicted residual item using the trend item weight coefficient, the seasonal item weight coefficient, and the residual item weight coefficient to obtain final predicted data of the target drug usage at each sampling moment in the future period; Inventory management of ophthalmic care medications is performed based on the final medication forecast data.

2. The ophthalmic care medication inventory management method according to claim 1, characterized in that: The method for obtaining the data fluctuation degree includes: Select any drug consumption data in the drug consumption data sequence as target drug data; The data fluctuation degree is obtained according to the data fluctuation degree calculation formula, and the data fluctuation degree calculation formula is as follows: Where U0 represents the data value of the target drug data; μ represents the data fluctuation degree of the target drug data; K represents the amount of drug consumption data within the preset range of the target drug data; U k Represents the data value of the kth drug consumption data within the preset range of the target drug data; L k Indicates the time difference between the sampling time of the kth drug consumption data within the preset area and the sampling time of the target drug data.

3. The ophthalmic care medication inventory management method according to claim 1, characterized in that: The method for obtaining the trend item weight coefficient includes: The trend item weight coefficient is obtained according to the trend item weight coefficient calculation formula, and the trend item weight coefficient calculation formula is as follows: Where, represents the trend item weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; U i+1 Represents the data value of the i+1th drug consumption data in the drug consumption data sequence; U i represents the data value of the i-th drug consumption data in the drug consumption data sequence; μ i+1 Indicates the data fluctuation degree of the i+1th drug consumption data in the drug consumption data sequence; μ i It represents the degree of data fluctuation of the i-th drug consumption data in the drug consumption data sequence; norm[] represents the normalization function.

4. The ophthalmic care medication inventory management method according to claim 1, characterized in that: The method for obtaining the seasonal item weight coefficient includes: Divide the preset period into equal parts to obtain all medication data time periods; The seasonal item weight coefficient is obtained according to the seasonal item weight coefficient calculation formula, and the seasonal item weight coefficient calculation formula is as follows: Where, represents the seasonal weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; U i Represents the data value of the i-th drug consumption data in the drug consumption data sequence; represents the data mean of all drug consumption data in the medication data time period where the i-th drug consumption data in the drug consumption data sequence is located; J represents the number of medication data time periods in the drug consumption data sequence; V j represents the data and value of drug consumption data in the jth medication data time period; It represents the mean value of drug consumption data for each year in the drug consumption data series; norm[] represents the normalization function.

5. The ophthalmic care medication inventory management method according to claim 1, characterized in that: The method for obtaining the residual term weight coefficient includes: Obtaining a residual item of each drug consumption data in the drug consumption data sequence; The residual term weight coefficient is obtained according to the residual term weight coefficient calculation formula, and the residual term weight coefficient calculation formula is as follows: Where, represents the residual weight coefficient of the drug consumption data series; I represents the number of drug consumption data in the drug consumption data series; μ i It represents the degree of data fluctuation of the i-th drug consumption data in the drug consumption data sequence; σ(R) represents the variance of the residual terms of all drug consumption data in the drug consumption data sequence; σ(U) represents the data variance of the drug consumption data sequence; norm[] represents the normalization function.

6. The ophthalmic care medication inventory management method according to claim 1, characterized in that: According to the data distribution of the drug consumption data in the drug consumption data sequence, the initial forecast trend item, initial forecast seasonal item and initial forecast residual item of the ophthalmic care drugs corresponding to the drug consumption data sequence at each sampling moment in the future period are obtained, including: A decomposition algorithm is applied to the drug consumption data sequence to obtain a trend item, a seasonal item, and a residual item for each drug consumption data in the drug consumption data sequence; Fitting trend items of all drug consumption data to obtain a trend item data curve; obtaining an initial predicted trend item at each sampling moment in a future period based on the trend item data curve; Obtain the target drug consumption seasonal cycle based on the seasonal items of all drug consumption data in the drug consumption data sequence, and use the mean of the seasonal items of drug consumption data of other consumption seasonal cycles at the same location in the future period as the initial predicted seasonal item for each sampling moment in the future period; The moving average method is used for the residual items of each drug consumption data to obtain the initial predicted residual items at each sampling moment in the future period.

7. The ophthalmic care medication inventory management method according to claim 1, characterized in that: The method for obtaining the final medication prediction data includes: The final medication prediction data is obtained according to the calculation formula of the final medication prediction data. The calculation formula of the final medication prediction data is as follows: Where U t T represents the final predicted data of medication at the t-th sampling moment in the future period; t represents the initial forecast trend item at the t-th sampling moment in the future period; S t represents the initial forecast seasonal term at the tth sampling moment in the future period; R t represents the initial prediction residual term at the t-th sampling moment in the future period; Represents the trend item weight coefficient of the drug consumption data series; Represents the seasonal weight coefficient of the drug consumption data series; Represents the residual weight coefficient of the drug consumption data series.

8. The ophthalmic care medication inventory management method according to claim 1, characterized in that: Inventory management of ophthalmic care medications is performed based on the final medication forecast data, including: Obtain the remaining inventory of the target medication at the current sampling time; Obtaining the maximum usage time of the target medication based on the remaining amount of the medication inventory and the final predicted medication usage data at each sampling moment in the future period; When the longest usage time is not greater than a preset first threshold, it is considered that the target medication has a risk of insufficient inventory; when the longest usage time is not less than a preset second threshold, it is considered that the target medication has a risk of inventory backlog.

9. The ophthalmic care medication inventory management method according to claim 8, characterized in that: Based on the remaining amount of the drug inventory and the final predicted data of drug use at each sampling time in the future period, the maximum usage time of the target drug is obtained, specifically: Calculate the final predicted data and value of medication at all sampling moments in the future period as the first sum value; When the remaining amount of the target medication in stock is not less than the first sum value, the length of the future period corresponding to the maximum value of the first sum value is taken as the maximum usage time of the target medication.

10. An ophthalmic care medication inventory management system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images