An intelligent management method for the production line capacity of pharmaceutical production

Abstract

The present invention relates to the technical field of intelligent capacity management, and specifically discloses an intelligent capacity management method for a production line for drug production, comprising the following steps: step S1: setting a historical data collection interval to obtain sales and corresponding incidence rates of all indications of the drug; establishing a basic relationship between sales and incidence rates through a regression model; step S2: after excluding abnormal year data, generating an incidence rate curve by year and performing periodic division; for each disease cycle, aggregating the curves of the same period of previous years to form a set, analyzing the year difference and the curve similarity through pairwise combination, calculating the year weight coefficient, and finally weighted fusion to generate a standard incidence rate curve; step S3: in a new disease cycle, real-time monitoring is performed to generate a current incidence rate curve, and the incidence rate of the next node is predicted by comparing it with the standard curve; combining the basic relationship, the predicted incidence rate is converted into a recommended output, and dynamic capacity planning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent production capacity management, and particularly to an intelligent production capacity management method for a production line used in pharmaceutical production. Background Art

[0002] Pharmaceutical production capacity management refers to the management and control of the production quantity and quality of pharmaceuticals by pharmaceutical enterprises during the production process; it mainly includes production planning and scheduling, equipment management, raw material management, quality management, and supply chain management, etc.

[0003] In the existing pharmaceutical production technology system, the production volume of pharmaceuticals is usually arranged according to a pre-set production volume standard. This relatively fixed and standardized production method can, to a certain extent, ensure the planning and stability of production, but there are significant limitations. Especially when facing a dynamically changing market environment and patient needs, its drawbacks become more prominent.

[0004] Specifically, this traditional production method lacks sufficient flexibility and is difficult to make timely and adaptive adjustments according to the current complex and changeable environment and the actual incidence of indications. In different regions, different seasons, or even different years, the incidence of various diseases may fluctuate due to various factors. For example, for some diseases with seasonal incidence characteristics, the number of patients may increase sharply during the high-incidence season, and the demand for related pharmaceuticals will also increase significantly accordingly; in some special cases, such as sudden public health events or local disease epidemics, the demand for specific pharmaceuticals may show an explosive growth. However, the production mode based on the pre-set production volume standard cannot keenly capture these changes and cannot quickly optimize the production volume.

[0005] For those pharmaceuticals with a short shelf life, the problems brought about by this situation are even more severe. On the one hand, if the market demand cannot be accurately estimated during the production process, resulting in overproduction, then when the pharmaceuticals cannot be fully sold and used within the validity period, a large number of pharmaceuticals will face the fate of expiration and invalidation. This not only means that the resources such as raw materials, manpower, and material resources invested in the production process are wasted without reason, but may also have a negative impact on the economic benefits of the enterprise. On the other hand, if the production volume is underestimated and lower than the actual market demand, then when patients urgently need the pharmaceutical for treatment, there may be a shortage of supply. In this case, patients with indications will not be able to obtain effective drug treatment in a timely manner, which may delay the condition and pose a serious threat to the health and even life safety of the patients. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent production capacity management method for a production line used in pharmaceutical production to solve the above technical problems.

[0007] The object of the present invention can be achieved by the following technical solutions:

[0008] An intelligent management method for the production capacity of a production line for drug production, comprising the following steps:

[0009] Step S1: Obtain the indications of the drug, set a historical data collection interval, obtain historical data, where the historical data includes the sales volume of the drug and the incidence rate of the indication; obtain the basic relationship between the sales volume and the indication according to the historical data;

[0010] Step S2: Obtain an incidence rate curve according to the historical data, perform a periodic analysis on the incidence rate curve to obtain several incidence cycles; obtain all the cycle curves corresponding to the incidence cycle, obtain a year weight coefficient according to all the cycle curves of the incidence cycle; and obtain the standard incidence rate curve of the incidence cycle according to all the cycle curves and the year weight coefficient;

[0011] Step S3: When entering a new incidence cycle, record the new incidence cycle as the current incidence cycle; generate a real-time incidence rate curve of the current incidence cycle in real time, and compare the standard incidence rate curve and the standard incidence rate curve in real time to obtain a predicted incidence rate; obtain the recommended production volume of the drug according to the predicted incidence rate and the basic relationship.

[0012] As a further scheme of the present invention: The setting process of the historical data collection interval includes:

[0013] Set a year threshold Y, where the setting range of the year threshold is Y ∈ [3, 5]; obtain the current year, and the historical data collection interval is the time period between the previous Y years and the current year;

[0014] Set a time interval threshold, where the setting range of the time interval threshold is [1, 7] days. On the historical data collection interval, select a node every time interval threshold and record it as a production node, then several production nodes are obtained.

[0015] As a further scheme of the present invention: The historical data includes the historical sales records of the drug and the historical incidence records of the indication. The historical sales records are the sales volumes of the drug at each production node within the historical data collection interval, and the historical incidence records are the incidence rates of the indication at each production node within the historical data collection interval.

[0016] As a further scheme of the present invention: The obtaining process of the basic relationship includes:

[0017] An initial regression model is established. The historical incidence records and historical sales records are input into the initial regression model, and the initial regression model is trained to obtain a final regression model. Through the final regression model, a basic relationship between the sales volume Sv and the incidence rate Ir is obtained as Sv = Ir × β reg ×β dose +β noi , where β reg is a regulatory factor, β dose is the drug dosage for a single treatment, and β noi is a noise factor.

[0018] As a further solution of the present invention: The obtaining process of the incidence rate curve includes:

[0019] The incidence rates at each production node within a year are obtained to get the incidence rate set {Ir1, Ir1,..., Ir num} within the year, where Ir num represents the incidence rate at the num-th production node within the year, and num represents the total number of production nodes within the year. According to the incidence rate sets of each year, the average incidence rate of the historical incidence records is obtained , where Ir i represents the incidence rate at the i-th production node in the r-th year, i ∈ [1, num] and i is a positive integer, r ∈ [1, Y] and r is a positive integer; and the standard deviation of the historical incidence records is obtained ;

[0020] According to the average incidence rate, the standard scores of each year are obtained , where Z r represents the standard score of the r-th year. A standard score threshold is set. If the standard score within a year exceeds the standard score threshold, it is recorded that the incidence rate in that year is abnormal; otherwise, the incidence rate is normal. The years with abnormal incidence rates in the historical incidence records are screened out and removed, and an incidence rate curve is generated according to the incidence rates at each production node within the remaining years.

[0021] As a further solution of the present invention: The process of performing periodic analysis on the incidence rate curve includes:

[0022] A periodic function model is established. The periodic function model is based on the sine function. The incidence rate curve is fitted into the periodic function model by the least squares method, and by adjusting the parameters of the periodic function model, the fitting error is minimized to obtain the periodic function expression of the incidence rate curve. According to the periodic function expression, the period of the incidence rate curve is obtained, and a year is divided into several incidence periods according to the period.

[0023] As a further solution of the present invention: the process of obtaining the year weight coefficient includes:

[0024] Obtain the cycle curves corresponding to each year of the disease onset cycle, and obtain the set {I1(y), I2(y),..., I m (y)}, where y is the number of the production node, and I m (y) represents the cycle curve of the mth year, and m is the total number of incidence curves; randomly combine each element in the set in pairs to obtain a number of combinations;

[0025] For any combination, denote the combination as [I w1 (y), I w2 (y)], where w1 ∈ [1, m], w2 ∈ [1, m], both w1 and w2 are positive integers, and w1 ≠ w2; then obtain the year difference value Ydv = |w1 - w2|, and obtain the curve difference value , where t1 is the number of the first production node in the disease onset cycle, t2 is the number of the last production node in the disease onset cycle, and y k represents the kth production node in the disease onset cycle, k ∈ [t1, t2] and k is a positive integer;

[0026] Obtain the year weight coefficient according to the year difference value and the curve difference value of each combination , where h is the total number of combinations, Cdv e represents the curve difference value of the e-th combination, and Ydv e represents the year difference value of the e-th combination, and λ is a preset correction coefficient.

[0027] As a further solution of the present invention: the process of obtaining the standard incidence curve includes:

[0028] Obtain the incidence standard value of each cycle curve of the disease onset cycle at the same production node according to the year weight coefficient , where w g represents the year corresponding to the incidence curve where the g-th cycle curve is located, w0 is the current year, and Ir g ´ represents the incidence rate of the g-th cycle curve at the production node; generate a standard incidence curve according to the numbers of each production node in the disease onset cycle and their corresponding incidence standard values.

[0029] As a further solution of the present invention: the process of obtaining the recommended yield includes:

[0030] Real-time obtain the difference value between the standard incidence curve and the real-time incidence curve , where G(y p ) represents the expression of the standard incidence curve, where G´(yp ) represents the expression of the real-time incidence rate curve, y p represents the p-th production node, t p represents the number of the latest production node;

[0031] Set the difference threshold. If the difference value is less than or equal to the difference threshold, the predicted incidence rate is the standard incidence rate of the next production node; if the difference value exceeds the difference threshold, the predicted incidence rate PIr = G´(t p +1)+Dv, where G´(t p +1) represents the incidence rate of the next production node in the real-time incidence rate curve;

[0032] The recommended output , where PIr s represents the predicted incidence rate of the s-th indication of the drug at the next production node, and M is the total number of indications of the drug.

[0033] Advantages of the present invention:

[0034] The present invention directly correlates the incidence rate with the drug sales volume through a regression model, combines the dosage and noise correction, accurately quantifies the real demand, avoids the deviation of traditional empirical prediction, reduces the risk of overproduction or insufficient supply; and based on the comparison between the real-time incidence rate curve and the standard curve, quickly corrects the prediction deviation, adapts to the sudden fluctuations of diseases (such as the outbreak of infectious diseases), and realizes the dynamic matching of output and demand; the generation of the standard incidence rate curve is through eliminating abnormal years, fusing historical cycle curves (weighting the year weight coefficient), extracting the periodic law of diseases (such as seasonal influenza), providing a stable benchmark for production planning, reducing the interference of random fluctuations, and supporting long-term production capacity allocation; by analyzing the year difference and curve difference, different weights are given to the data of different years, enhancing the representativeness and anti-interference ability of the standard curve, so as to maximize the value of historical data; by predicting the incidence rate of the next production node, adjusting the output in advance, avoiding the waste of inventory caused by the sudden drop in demand, and reducing the inventory backlog; real-time monitoring of the incidence trend, increasing the production capacity before the demand surges, ensuring the accessibility of drugs (such as the rapid increase in production of antiviral drugs at the beginning of the epidemic), and preventing supply shortages; through accurate prediction, optimizing the enterprise's production resources (such as raw material procurement, production line scheduling), reducing the overall operation cost, and at the same time enhancing the social emergency support ability; the present invention replaces the extensive production with data-driven, and through the closed-loop logic of "modeling to quantify demand - extracting periodic laws - real-time dynamic adjustment", realizes the scientific, agile and low-cost control of drug production, and has both commercial benefits and public health value. Description of the drawings

[0035] The present invention will be further described below with reference to the drawings.

[0036] Figure 1 It is a schematic flow chart of an intelligent management method for the production line capacity in the production of pharmaceuticals according to the present invention. Specific embodiments

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0038] Please refer to Figure 1 As shown, the present invention is an intelligent management method for the production line capacity in the production of pharmaceuticals, including the following steps:

[0039] Step S1: Set a historical data collection interval, and equally select a number of production nodes within the historical data collection interval; obtain all indications of the drug, obtain the historical sales records of the drug and the historical incidence records of the indications. The historical sales records are the sales volumes of the drug at each production node within the historical data collection interval, and the historical incidence records are the incidence rates of the indications at each production node within the historical data collection interval;

[0040] Establish an initial regression model, input the historical incidence records and historical sales records into the initial regression model, train the initial regression model to obtain a final regression model; through the final regression model, obtain the basic relationship between the sales volume Sv and the incidence rate Ir as Sv = Ir × β reg ×β dose +β noi , where β reg is a regulation factor, β dose is the dosage of the drug for a single treatment, β noi is a noise factor, and the units of the regulation factor and the noise factor are the same as the unit of the sales volume;

[0041] It should be noted that a time period is set as the data collection range to ensure that the data covers sufficient historical information; several time points (production nodes) are equally selected within the collection interval for recording the drug sales volume and the incidence rate of indications; the historical sales records (sales volume) of the drug and the historical incidence records (incidence rate) of the indications are collected to ensure that the data is aligned in the time dimension; the regression analysis method is adopted, with the incidence rate (Ir) as the independent variable and the drug sales volume (Sv) as the dependent variable, to establish the mathematical relationship between the two; the model is trained through the historical incidence records and sales records to optimize the model parameters so that it can accurately fit the historical data; after the training is completed, the relationship formula between the sales volume and the incidence rate is obtained, where the adjustment factor is used to calibrate the final regression model to adjust the influence degree of the adjusted incidence rate on the sales volume, β dose dose represents the dose of the drug for each indication, and the noise factor is used to correct the random errors or uncontrollable factors in the model, such as the influence of the market share; the model is trained through historical data to reveal the potential law between the incidence rate and the sales volume;

[0042] In a preferred embodiment of the present invention, a year threshold Y is set, and the setting range of the year threshold is Y ∈ [3, 5]; the current year is obtained, and the historical data collection interval is the time period between the previous Y years and the current year;

[0043] In a preferred embodiment of the present invention, the selection process of the production nodes includes:

[0044] A time interval threshold is set, and the setting range of the time interval threshold is [1, 7] days. On the historical data collection interval, a node is selected every time interval threshold and recorded as a production node, and then several production nodes are obtained;

[0045] It should be noted that the historical data of 3 to 5 years can not only reflect the recent trend but also avoid the interference of too old data on the model; this time range usually contains enough data points to capture the periodic or trend changes of the incidence rate and the sales volume; in addition, the time interval of 1 to 7 days can ensure the data density while avoiding the data being too sparse or redundant; adjust the time interval according to actual needs. For example, a shorter interval (1 day) is suitable for high-frequency data collection scenarios (such as drugs for acute infectious diseases); a longer interval (7 days) is suitable for low-frequency data collection scenarios (such as drugs for chronic diseases); through a reasonable time range and interval, ensure that the collected data is representative and timely;

[0046] In a preferred embodiment of the present invention, the obtaining process of the incidence rate:

[0047] Obtain the medical records of a medical institution, where the medical records include all patients and their diseases; according to the medical records, obtain the total number of patients, and obtain the number of patients with the disease as the indication, denoted as the number of indication patients, and obtain the incidence rate Ir = n / N, where n is the number of indication patients and N is the total number;

[0048] Step S2: According to the historical incidence records, screen and eliminate the years with abnormal incidence rates in the historical incidence records, and generate an incidence rate curve based on the incidence rates at each production node in the remaining years; perform a periodic analysis on the incidence rate curves of each year, divide a year into several incidence cycles, and denote the curve segment corresponding to the incidence cycle on the incidence rate curve as a cycle curve;

[0049] For any incidence cycle, obtain the cycle curves corresponding to the incidence cycle in each year, and obtain the set {I1(y), I2(y),..., I m (y)}, where y is the number of the production node, and I m (y) represents the cycle curve of the m-th year, and m is the total number of incidence rate curves; randomly combine each pair of elements in the set to obtain several combinations, obtain the year difference value and the curve difference value of each combination; and obtain the year weight coefficient according to the year difference value and the curve difference value of each combination; according to the cycle curves corresponding to the incidence cycle in each year and the year weight coefficient, obtain the standard incidence rate curve of the incidence cycle;

[0050] It should be noted that by eliminating the years with abnormal incidence rates, the interference of abnormal data on model training is avoided, ensuring the accuracy and reliability of the data; after eliminating the outliers, the incidence rate curve can better reflect the real disease trend, providing a high-quality data basis for subsequent analysis; dividing a year into several incidence cycles can capture the seasonal and periodic characteristics of the disease (such as the high incidence of influenza in winter); by extracting the curve segments of each incidence cycle, the change law of the disease in different time periods can be analyzed in detail, supporting more accurate prediction; over time, the disease epidemic pattern, environmental factors (such as climate, population density) and medical conditions may change; recent data can more accurately reflect the current disease epidemic trend and drug demand; therefore, by randomly combining each pair, analyzing the year difference value and the curve difference value, different weights are assigned to the data of different years to reflect their contribution degrees; combining the year weight coefficient, a standard incidence rate curve is generated to eliminate random fluctuations and highlight the core trend of the disease;

[0051] It is understandable that the standard incidence curve, as a benchmark, can be used for real-time comparison and calibration to improve the accuracy of prediction; the methods of periodic analysis and generation of the standard curve are applicable to various diseases (such as seasonal infectious diseases, chronic diseases), and have wide application value; based on the standard incidence curve, enterprises can adjust the drug production in advance to avoid under-supply or overstocking;

[0052] In a preferred embodiment of the present invention, the process of generating the incidence curve includes:

[0053] Number each production node within a year, use the number of the production node as the abscissa and the incidence rate as the ordinate to establish a coordinate system; convert the number of each production node and its corresponding incidence rate into coordinate points at the corresponding positions on the coordinate system; and connect the coordinate points with a smooth curve, and record the curve as the incidence curve;

[0054] In a preferred embodiment of the present invention, the process of screening out the years with abnormal incidence rates in the historical incidence records includes:

[0055] Obtain the incidence rates at each production node within a year to obtain the incidence rate set {Ir1, Ir1,..., Ir num} within the year, where Ir num represents the incidence rate at the num-th production node within the year, and num represents the total number of production nodes within the year; according to the incidence rate sets of each year, obtain the average incidence rate of the historical incidence records, where Ir i represents the incidence rate at the i-th production node in the r-th year, i ∈ [1, num] and i is a positive integer, r ∈ [1, Y] and r is a positive integer; and obtain the standard deviation of the historical incidence records; according to the average incidence rate, obtain the standard score of each year, where Z r represents the standard score of the r-th year; set a standard score threshold, if the standard score within the year exceeds the standard score threshold, then record that the incidence rate of that year is abnormal; otherwise, the incidence rate is not abnormal;

[0056] In a preferred embodiment of the present invention, the process of performing periodic analysis on the incidence curves of each year includes:

[0057] Establish a periodic function model, the periodic function model is based on the sine function, fit the incidence curve into the periodic function model by the least squares method, and by adjusting the parameters of the periodic function model, minimize the fitting error to obtain the periodic function expression of the incidence curve; according to the periodic function expression, obtain the period of the incidence curve, and divide a year into several incidence periods according to the period;

[0058] In a preferred embodiment of the present invention, the process of obtaining the year difference value and the curve difference value includes:

[0059] For any combination, denote the combination as [I w1 (y), I w2 (y)], where w1 ∈ [1, m], w2 ∈ [1, m], both w1 and w2 are positive integers, and w1 ≠ w2; then obtain the year difference value Ydv = |w1 - w2|, and obtain the curve difference value , where t1 is the number of the first production node within the disease onset cycle, t2 is the number of the last production node within the disease onset cycle, y k represents the k-th production node within the disease onset cycle, k ∈ [t1, t2] and k is a positive integer;

[0060] In a preferred embodiment of the present invention, the process of obtaining the year weight coefficient includes:

[0061] According to the year difference value and the curve difference value of each combination, obtain the year weight coefficient , where h is the total number of combinations, Cdv e represents the curve difference value of the e-th combination, Ydv e represents the year difference value of the e-th combination, and λ is a preset correction coefficient;

[0062] In a preferred embodiment of the present invention, the process of obtaining the standard incidence rate curve of the disease onset cycle includes:

[0063] According to the year weight coefficient, obtain the standard incidence rate value at the same production node of each cycle curve of the disease onset cycle , where w g represents the year corresponding to the incidence rate curve where the g-th cycle curve is located, w0 is the current year, Ir g ´ represents the incidence rate of the g-th cycle curve at the production node; generate a standard incidence rate curve according to the numbers of each production node within the disease onset cycle and their corresponding standard incidence rate values;

[0064] Step S3: When entering a new disease onset cycle, denote the new disease onset cycle as the current disease onset cycle, monitor the incidence rate at each production node of the current disease onset cycle in real time, generate a real-time incidence rate curve, and obtain the standard incidence rate curve of the current disease onset cycle; compare the standard incidence rate curve and the real-time incidence rate curve in real time, predict the incidence rate of the next production node, denoted as the predicted incidence rate; according to the predicted incidence rate of each indication and the basic relationship, obtain the recommended production volume of the drug at the next production node;

[0065] It is understandable that during the current incidence period, incidence data is collected at production nodes to generate a real-time incidence curve, which reflects the latest disease epidemic trend; a standardized incidence curve generated based on historical data reflects the typical trend of the disease during a specific incidence period; the standard curve serves as a benchmark for prediction and calibration and is used to evaluate the deviation of real-time data; the real-time incidence curve is compared with the standard incidence curve to analyze the differences between the two, and based on the comparison results, the incidence rate at the next production node (predicted incidence rate) is predicted to provide a basis for production adjustment;

[0066] In a preferred embodiment of the present invention, the process of predicting the incidence rate includes:

[0067] Obtain the difference value between the standard incidence curve and the real-time incidence curve in real time , where G(y p ) represents the expression of the standard incidence curve, where G´(y p ) represents the expression of the real-time incidence curve, y p represents the p-th production node, and t p represents the number of the latest production node;

[0068] Set a difference threshold. If the difference value is less than or equal to the difference threshold, the predicted incidence rate is the incidence rate standard value of the next production node; if the difference value exceeds the difference threshold, the predicted incidence rate PIr = G´(t p + 1) + Dv is obtained, where G´(t p + 1) represents the incidence rate of the next production node on the real-time incidence curve;

[0069] It should be noted that in the actual scenario, there may be delays in drug production, distribution, and supply, resulting in some patients not being able to obtain drugs in a timely manner; these unmet demands will accumulate and be reflected as additional incidence rates or drug demands at subsequent time points; therefore, the difference value Dv needs to be added to ensure the needs of patients who have not obtained drugs in a timely manner are met;

[0070] In a preferred embodiment of the present invention, the process of obtaining the recommended production quantity includes:

[0071] The recommended production quantity , where PIr s represents the predicted incidence rate of the s-th indication of the drug at the next production node, and M is the total number of indications of the drug.

[0072] The above has described in detail an embodiment of the present invention, but the above content is only a preferred embodiment of the present invention and cannot be considered as defining the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. An intelligent management method for the production line capacity in pharmaceutical production, characterized in that, Including the following steps: Step S1: Obtain the indications of the drug, set the historical data collection interval, obtain historical data, where the historical data includes the sales volume of the drug and the incidence rate of the indications; obtain the basic relationship between the sales volume and the indications according to the historical data; Step S2: According to the historical data, obtain the incidence rate curve, conduct a periodic analysis on the incidence rate curve to obtain several incidence cycles; obtain all the periodic curves corresponding to the incidence cycles, obtain the year weight coefficient according to all the periodic curves of the incidence cycle; and obtain the standard incidence rate curve of the incidence cycle according to all the periodic curves and the year weight coefficient; Step S3: When entering a new incidence cycle, record the new incidence cycle as the current incidence cycle; generate the real-time incidence rate curve of the current incidence cycle in real time, and compare the standard incidence rate curve and the real-time incidence rate curve in real time to obtain the predicted incidence rate; obtain the recommended production volume of the drug according to the predicted incidence rate and the basic relationship; In step S1, the historical data includes the historical sales records of the drug and the historical incidence records of the indications. The historical sales records are the sales volumes of the drug at each production node within the historical data collection interval, and the historical incidence records are the incidence rates of the indications at each production node within the historical data collection interval; In step S1, the process of obtaining the basic relationship includes: Establish an initial regression model, input the historical incidence records and historical sales records into the initial regression model, and train the initial regression model to obtain a final regression model; through the final regression model, obtain the basic relationship between the sales volume Sv and the incidence rate Ir as Sv = Ir × β reg ×β dose +β noi , where β reg is a regulatory factor, β dose is the drug dosage for a single treatment, and β noi is a noise factor; In step S2, the process of obtaining the incidence rate curve includes: Obtain the incidence rates at each production node within the year to get the set of incidence rates within the year {Ir1, Ir1,..., Ir num}, where Ir num represents the incidence rate at the num-th production node within the year, and num represents the total number of production nodes within the year; according to the sets of incidence rates for each year, obtain the average incidence rate of the historical incidence records , where Ir i represents the incidence rate at the i-th production node in the r-th year, i ∈ [1, num] and i is a positive integer, r ∈ [1, Y] and r is a positive integer; and obtain the standard deviation of the historical incidence records ; Based on the mean incidence rate, obtain the standard scores for each year , where Z r represents the standard score for the r-th year; set a standard score threshold. If the standard score within a year exceeds the standard score threshold, it is recorded that the incidence rate in that year is abnormal; otherwise, the incidence rate is not abnormal; screen and exclude the years with abnormal incidence rates in the historical incidence records, and generate an incidence rate curve based on the incidence rates at each production node within the remaining years.

2. The intelligent management method for the production line capacity of drug production according to claim 1, wherein, In step S1, the process of setting the historical data collection interval includes: Set the year threshold Y, where the setting range of the year threshold is Y ∈ [3, 5]; obtain the current year, then the historical data collection interval is the time period between the previous Y years and the current year; Set the time interval threshold, where the setting range of the time interval threshold is [1, 7] days. On the historical data collection interval, select a node every time interval threshold and record it as a production node, then several production nodes are obtained.

3. An intelligent management method for the production line capacity of drug production according to claim 1, characterized in that, In step S2, the process of conducting a periodic analysis on the incidence rate curve includes: Establish a periodic function model. The periodic function model is based on the sine function. Fit the incidence rate curve into the periodic function model by the least squares method, and minimize the fitting error by adjusting the parameters of the periodic function model to obtain the periodic function expression of the incidence rate curve; obtain the period of the incidence rate curve according to the periodic function expression, and divide a year into several incidence cycles according to the period.

4. The intelligent management method for production line capacity in pharmaceutical production according to claim 1, characterized in that, In step S2, the process of obtaining the year weight coefficient includes: Obtain the periodic curves corresponding to each year of the onset cycle to obtain the set {I1(y), I2(y),..., I m (y)}, where y is the number of the production node, and I m (y) represents the periodic curve of the m-th year, and m is the total number of incidence curves; randomly combine each element in the set in pairs to obtain a number of combinations; For any combination, denote the combination as [I w1 (y), I w2 (y)], where w1 ∈ [1, m], w2 ∈ [1, m], both w1 and w2 are positive integers, and w1 ≠ w2; then obtain the year difference value Ydv = |w1 - w2|, and obtain the curve difference value , where t1 is the number of the first production node within the disease onset period, t2 is the number of the last production node within the disease onset period, y k represents the k-th production node within the disease onset period, k ∈ [t1, t2] and k is a positive integer; Obtain the year weight coefficient according to the year difference value and curve difference value of each combination , where h is the total number of combinations, and Cdv e represents the curve difference value of the e-th combination, and Ydv e represents the year difference value of the e-th combination, and λ is a preset correction coefficient.

5. The intelligent management method for the production line capacity of drug production according to claim 4, characterized in that, In step S2, the process of obtaining the standard incidence rate curve includes: According to the annual weight coefficient, obtain the incidence standard value of each cycle curve of the disease incidence cycle at the same production node , where w g represents the year corresponding to the incidence curve where the g-th cycle curve is located, w0 is the current year, and Ir g ´ represents the incidence rate of the g-th cycle curve at the production node; generate a standard incidence rate curve according to the numbers of each production node within the disease incidence cycle and their corresponding incidence standard values.

6. The intelligent management method for production line capacity in pharmaceutical production according to claim 4, characterized in that, In step S3, the process of obtaining the recommended production volume includes: Obtain the difference value between the standard incidence rate curve and the real-time incidence rate curve in real time , where G(y p ) represents the expression of the standard incidence rate curve, where G´(y p ) represents the expression of the real-time incidence rate curve, y p represents the p-th production node, and t p represents the number of the latest production node; Set a difference threshold. If the difference value is less than or equal to the difference threshold, the predicted incidence rate is the standard incidence rate of the next production node; if the difference value exceeds the difference threshold, the predicted incidence rate PIr = G´(t p +1)+Dv, where G´(t p +1) represents the incidence rate of the next production node in the real-time incidence rate curve; The recommended production volume , where PIr s represents the predicted incidence of the s-th indication of the drug at the next production node, and M is the total number of indications of the drug.

Images