Comprehensive management platform and method for display of medicine commodities in retail drugstores

Abstract

The invention belongs to the technical field of medicine retail, and discloses a retail pharmacy medicine commodity display integrated management platform and method, and the method comprises the steps: collecting illumination data and environment data; constructing a luminosity prediction model based on the illumination data; predicting each drug placement point of a shelf in the retail pharmacy based on the constructed luminosity prediction model to obtain predicted illumination intensity, analyzing the predicted illumination intensity by adopting a spline analysis method to obtain a hierarchical light intensity curve graph, and performing illumination influence evaluation on the hierarchical light intensity curve graph to obtain an illumination influence coefficient; carrying out point location storage analysis on the hierarchical light intensity curve graph to obtain a validity period variation; constructing a target optimization function, carrying out optimization solution on the target optimization function by adopting an improved genetic algorithm to obtain an optimal drug layout, and carrying out drug placement on the retail drugstore according to the optimal drug layout; and the efficiency and the accuracy of medicine display are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of pharmaceutical retail technology, and more particularly to a comprehensive management platform and method for displaying pharmaceutical products in retail pharmacies. Background Art

[0002] In the merchandise display management of retail pharmacies, the effective storage and display of medicines are crucial to the operation of pharmacies. However, the current management system has many shortcomings, resulting in poor display effects of medicines and even affecting the quality and sales of medicines. For example: the display locations of retail pharmacies are usually fixed and lack the ability to dynamically adjust according to changes in external natural light intensity. Changes in light intensity in different time periods have an important impact on factors such as the quality and expiration date of medicines, but most existing systems fail to effectively consider this, resulting in medicines being displayed for a long time under inappropriate lighting conditions, affecting their quality and safety; in most retail pharmacies, the expiration date management of medicines is usually based on static models, without considering the impact of environmental changes on the expiration date of medicines, resulting in the expiration date of medicines being greatly shortened, thereby affecting the inventory management and expiration date control of medicines; the existing medicine display layout mostly relies on static planning, and does not fully combine the quality requirements of medicines and multiple factors such as light for optimization, resulting in the degradation of medicine quality and reducing the overall layout efficiency of the pharmacy.

[0003] In view of this, the present invention proposes a comprehensive management platform and method for displaying pharmaceutical products in retail pharmacies to solve the above problems. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned objectives, the present invention provides the following technical solution: a comprehensive management platform for the display of pharmaceutical products in retail pharmacies, characterized by comprising:

[0005] Data acquisition module: collects lighting data and environmental data;

[0006] Lighting prediction module: Build an initial luminosity prediction model based on lighting data, and use the lighting data as sample data to train the initial luminosity prediction model to obtain a luminosity prediction model. Based on the constructed luminosity prediction model, predict the light intensity at each drug placement point on the retail pharmacy shelf to obtain the predicted light intensity. Use spline analysis to analyze the predicted light intensity and obtain a hierarchical light intensity curve. Perform a lighting impact assessment on the hierarchical light intensity curve to obtain the lighting impact coefficient.

[0007] Drug expiration date assessment module: performs point storage analysis on the hierarchical light intensity curve to obtain the change in validity period;

[0008] Display optimization module: Construct a drug adaptability evaluation function based on the light impact coefficient and the change in the expiration date, construct a target optimization function based on the drug adaptability evaluation function, use an improved genetic algorithm to optimize and solve the target optimization function, obtain the optimal drug layout, and place drugs in retail pharmacies based on the optimal drug layout.

[0009] Furthermore, the lighting data includes: lighting intensity, time scale and position coordinates; environmental data includes ambient temperature; the method of obtaining the lighting data includes: spatially three-dimensionalizing the retail pharmacy, selecting a shelf point in the retail pharmacy as the point center, constructing a standard three-dimensional coordinate system with the point center, and using the straight-line distance of other points from the point center in each direction dimension in the three-dimensional coordinate system as the coordinate value of each dimension, and all coordinate values constitute the point coordinates; based on the point coordinates, some points in the retail pharmacy are selected as sampling points, and distributed photometry sensors are set at the sampling points, and a collection time interval is preset. The distributed photometry sensor collects data every collection time interval to obtain light intensity and time scale.

[0010] Furthermore, the method of constructing the initial luminosity prediction model includes:

[0011] Calculate the variance of light intensity at the same time scale and construct a point evaluation function based on the variance. The formula of the point evaluation function is: ;in, Representative The lighting data and The light intensity correlation of the illumination data, Representative The lighting data and The distance of the illumination data, common distance evaluation functions are Euclidean distance function and Chebyshev distance function, represents the variance of light intensity at the time scale;

[0012] Preset group value , is an integer greater than zero, dividing the illumination data equally into Group data, each time The data group is used as the sample group, and the remaining data group is used as the prediction group to train the initial photometric prediction model. Each training uses the sample group as input to predict each data in the prediction group. Repeat times; the formula of the initial luminosity prediction model is:

[0013] ;in, Represents the forecast group The predicted light intensity of the light data, Representative sample group The light weight of the lighting data, Representative sample group The illumination data and the prediction group The light intensity correlation of the illumination data, Representative sample group The light intensity of the lighting data, represents the index of the sample group, Represents the size of the prediction group.

[0014] Furthermore, the method of training the initial photometric prediction model includes:

[0015] Preset Group light weight combination, is an integer greater than zero, the optimal weight combination is initialized to empty, the weight fitness of the optimal weight combination is infinitesimal, and the light weights in the light weight combination satisfy the weight constraints: ; Bring each light weight combination into the initial luminosity prediction model, and evaluate the adaptability of each light weight combination using the sample group and the prediction group. The formula for evaluating the adaptability of each light weight combination is: ;in, represents the weighted fitness, Represents the group value, Represents the prediction group The predicted light intensity of the data, Represents the forecast group The light intensity of the data, Represents the amount of data in the prediction group, Representative A subset of samples, Representation training times, Representative The prediction group of the training is trained; based on the weight fitness of each group of light weight combinations, the light weight combination with the largest weight fitness is selected as the candidate combination, and the weight fitness of the candidate combination is compared with the weight fitness of the optimal weight combination. When the weight fitness of the candidate combination is greater than the weight fitness of the optimal weight combination, the candidate combination is used as the new optimal weight combination, the disturbance factor is preset, and each light weight combination is combined and updated based on the disturbance factor. The random selection algorithm is used to select the light weight combination. The weights are used as the exchange factors, and the position of the exchange factors is replaced; this is repeated until the optimal weight combination no longer changes, and the optimal weight combination at this time is output; the optimal weight combination is placed into the initial photometric prediction model to obtain the photometric prediction model.

[0016] Furthermore, the formula for evaluating the illumination impact of the hierarchical light intensity curve is:

[0017] ;in, The coordinates of the representative points are The light influence coefficient at Represents the total span of the time scale, The coordinates of the representative points are In the The light intensity at each time scale, Representative Time scale time weight.

[0018] Furthermore, the method of performing point storage analysis on the hierarchical light intensity curve graph includes:

[0019] A light-temperature fluctuation relationship model is constructed based on the hierarchical light intensity curve. The calculation formula of the light-temperature relationship model is: ;in, The coordinates of the representative points are The temperature fluctuation value at Represents the coordinate value of the first dimension in the three-dimensional coordinate system, Represents the coordinate value of the second dimension in the three-dimensional coordinate system, Represents the coordinate value of the third dimension in the three-dimensional coordinate system, represents the thermal conductivity, The coordinates of the points in the light intensity curve representing the level are The function value at , that is, the light intensity, Represents the time scale, represents the light attenuation coefficient, Represents the time of sunrise, representing the total time of a day; summing the ambient temperature and the temperature fluctuation value to obtain the point-corrected temperature; obtaining drug stability test data through drug stability testing, performing degradation kinetic parameter analysis based on the drug stability test data to generate drug degradation kinetic characteristic data; determining the activation energy and pre-exponential factor of a specific drug based on the drug degradation kinetic characteristic data to generate drug degradation reaction parameter data; performing degradation reaction order determination processing based on the drug degradation reaction parameter data to generate degradation reaction kinetic model data; performing absolute temperature conversion processing based on the point-corrected temperature data to generate point-absolute temperature data; performing Arrhenius equation calculation based on the point-absolute temperature data to generate point-degradation reaction rate constant data; designing an intelligent analysis module for temperature impact assessment based on the point-degradation reaction rate constant data using the Q10 rule to obtain a temperature impact assessment engine; transmitting the degradation reaction kinetic model data in real time to the temperature impact assessment engine for intelligent drug stability analysis to generate drug stability prediction data; constructing a shelf life change model based on the drug stability prediction data to generate a shelf life calculation model; calculating the shelf life difference between the standard storage temperature and the actual storage temperature based on the shelf life calculation model to generate a shelf life change.

[0020] Furthermore, the formula of the drug adaptability evaluation function is:

[0021] ; The coordinates of the representative drug are The positional fitness at The coordinates of the representative drug are The change in validity period, Represents the light intensity weight, represents the effective period weight;

[0022] The formula of the objective optimization function is: ;in, Represents the global fitness.

[0023] Furthermore, the method of optimizing and solving the target optimization function includes:

[0024] Preset Group drug layouts. Each group of drug layouts represents a set of arrangement of drug display positions. Each group of drug layouts is considered a chromosome. Each gene in the chromosome represents a display position in the drug layout. The optimal chromosome is initialized to be empty.

[0025] The global fitness of each chromosome is calculated using the target optimization function, and the global fitness is used as the layout fitness of the chromosome. Based on the layout fitness, the chromosomes are selected, crossed, and mutated to obtain variants.

[0026] Calculate the layout fitness of the variants, select the preferred variant with the largest fitness as the candidate individual, and when the layout fitness of the candidate individual is greater than the optimal chromosome, use the candidate individual as the new optimal chromosome and the variant as the new preferred chromosome. Repeat until the optimal chromosome no longer changes. The optimal chromosome at this time is output as the optimal drug layout, and the drugs are placed in retail pharmacies based on the optimal drug layout.

[0027] Furthermore, the method of selecting chromosomes includes:

[0028] A clustering algorithm is used to perform cluster analysis on the layout fitness of each chromosome to obtain chromosome clusters; a filtering threshold is preset, and the chromosome clusters whose cluster centers are greater than or equal to the filtering threshold are used as the preferred pool; a selection wheel is constructed based on the preferred pool, and the selection probability of each chromosome in the selection wheel is equal to the layout fitness of the chromosome divided by the sum of the layout fitness of all chromosomes in the preferred pool. chromosomes as the preferred chromosomes;

[0029] The crossover and mutation method includes: performing permutations and combinations on the preferred chromosomes to obtain a crossover combination, using the mean of the layout fitness of the preferred chromosomes as a crossover threshold, and constructing a crossover operator based on the crossover threshold. The formula of the crossover operator is: ;in, Represents the preferred chromosome The crossover probability of a chromosome is represents a fixed probability, which is set by those skilled in the art based on the actual situation. Represents the maximum layout fitness within the preferred chromosome, Represents the preferred chromosome The layout fitness of chromosomes, represents the crossover threshold; the chromosome in the crossover combination is used as the parent chromosome, a gene in the parent chromosome is randomly selected as the crossover starting point, a crossover operation is performed based on the crossover probability, the genes after the crossover node are exchanged, the crossover chromosome is obtained, the layout fitness of the crossover chromosome is calculated, and the crossover chromosome with a layout fitness greater than that of the parent chromosome is regarded as an excellent individual; a random selection algorithm is used to randomly select a gene in the excellent individual as the mutation sequence, and the order of the genes in the mutation sequence is changed to obtain a variant.

[0030] A comprehensive management method for displaying pharmaceutical products in a retail pharmacy, comprising:

[0031] S1, collect lighting data and environmental data;

[0032] S2. Construct an initial luminosity prediction model based on the illumination data, and train the initial luminosity prediction model using the illumination data as sample data to obtain a luminosity prediction model; predict each drug placement point on a shelf in a retail pharmacy based on the constructed luminosity prediction model to obtain a predicted light intensity; analyze the predicted light intensity using a spline analysis method to obtain a hierarchical light intensity curve; and perform a light impact assessment on the hierarchical light intensity curve to obtain a light impact coefficient;

[0033] S3. Perform point storage analysis on the hierarchical light intensity curve to obtain the effective period change;

[0034] S4. Construct a drug adaptability evaluation function based on the illumination influence coefficient and the change in the expiration date, construct a target optimization function based on the drug adaptability evaluation function, use an improved genetic algorithm to optimize and solve the target optimization function, obtain the optimal drug layout, and place drugs in retail pharmacies based on the optimal drug layout.

[0035] The technical effects and advantages of the integrated management platform and method for displaying pharmaceutical products in retail pharmacies provided by the present invention are as follows:

[0036] The present invention realizes accurate light intensity prediction for the placement of medicines by constructing a photometric prediction model. The model combines multiple factors such as light intensity, time scale, and position coordinates to provide high-precision light intensity prediction for the medicine display environment, ensuring that the medicines are stored under appropriate light intensity, thereby extending the shelf life of the medicines and preventing damage caused by excessive or insufficient light. By performing point storage analysis on the hierarchical light intensity curve chart, considering the relationship between the speed of medicine aging and temperature, the shelf life of the medicines is dynamically evaluated and the display layout is optimized. This evaluation mechanism not only helps retailers avoid the occurrence of expired medicines, but also reduces medicine waste and improves the efficiency of medicine use. The target optimization function is optimized and solved through an improved genetic algorithm to optimize the display position of the medicines so that the medicines meet the optimal light and temperature conditions, which greatly improves the operating efficiency and medicine management level of retail pharmacies, reduces the risk and error of human intervention, and ensures the optimal environment for medicine storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a schematic diagram of a comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to the present invention;

[0038] Figure 2 This is a schematic diagram of a comprehensive management method for displaying pharmaceutical products in retail pharmacies according to the present invention. DETAILED DESCRIPTION

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0040] Example 1;

[0041] See also Figure 1 As shown, the embodiment of the present invention provides a comprehensive management platform for displaying pharmaceutical products in retail pharmacies, including:

[0042] Data acquisition module: collects lighting data and environmental data;

[0043] Lighting prediction module: Build an initial luminosity prediction model based on lighting data, and use the lighting data as sample data to train the initial luminosity prediction model to obtain a luminosity prediction model. Based on the constructed luminosity prediction model, predict the light intensity at each drug placement point on the retail pharmacy shelf to obtain the predicted light intensity. Use spline analysis to analyze the predicted light intensity and obtain a hierarchical light intensity curve. Perform a lighting impact assessment on the hierarchical light intensity curve to obtain the lighting impact coefficient.

[0044] Drug expiration date assessment module: performs point storage analysis on the hierarchical light intensity curve to obtain the change in validity period;

[0045] Display Optimization Module: This module constructs a drug adaptability evaluation function based on the light impact coefficient and the change in the expiration date. It also constructs a target optimization function based on the drug adaptability evaluation function. This target optimization function is optimized and solved using an improved genetic algorithm to obtain the optimal drug layout. This optimal drug layout is then used to display drugs in retail pharmacies.

[0046] Each module is connected via wired and / or wireless means to achieve data transmission between modules;

[0047] Lighting data includes: light intensity, time scale and location coordinates; environmental data includes ambient temperature; methods for obtaining light data include: spatially three-dimensionalizing the retail pharmacy, selecting a shelf point in the retail pharmacy as the point center, constructing a standard three-dimensional coordinate system with the point center, and using the straight-line distance from other points to the point center in each direction dimension in the three-dimensional coordinate system as the coordinate value of each dimension, and all coordinate values constitute the point coordinates; based on the point coordinates, some points in the retail pharmacy are selected as sampling points, and distributed photometry sensors are set at the sampling points, and the collection time interval is preset. The distributed photometry sensor collects data every collection time interval to obtain light intensity and time scale; the ambient temperature is obtained through a temperature sensor.

[0048] In retail environments, the shelf life of medications is affected by light intensity, especially for sensitive medications. Because light intensity varies dynamically across time and space, traditional methods cannot track and analyze its changes in real time. The light estimation module accurately fits light variation trends, dynamically tracks light changes, quantifies the impact of light on medications, and optimizes light management, thereby reducing the risk of medication expiration due to inappropriate lighting. Specifically:

[0049] Calculate the variance of light intensity at the same time scale and construct a point evaluation function based on the variance. The formula of the point evaluation function is: ;in, Representative The lighting data and The light intensity correlation of the illumination data, Representative The lighting data and The distance of the illumination data, common distance evaluation functions are Euclidean distance function and Chebyshev distance function, represents the variance of light intensity at the time scale;

[0050] Preset group value , is an integer greater than zero, dividing the illumination data equally into Group data, each time The data group is used as the sample group, and the remaining data group is used as the prediction group to train the initial photometric prediction model. Each training uses the sample group as input to predict each data in the prediction group. Repeat times; the formula of the initial luminosity prediction model is:

[0051] ;in, Represents the forecast group The predicted light intensity of the light data, Representative sample group The light weight of the lighting data, Representative sample group The illumination data and the prediction group The light intensity correlation of the illumination data, Representative sample group The light intensity of the lighting data, represents the index of the sample group, represents the size of the prediction group;

[0052] Preset The light weight combination contains the light weights of all lighting data and meets the weight conditions. The sum of all light weights is 1. is an integer greater than zero, the optimal weight combination is initialized to empty, and the weight fitness of the optimal weight combination is infinitesimal; each light weight combination is brought into the initial luminosity prediction model, and the adaptability of each light weight combination is evaluated with the sample group and the prediction group. The formula for evaluating the adaptability of each light weight combination is: ;in, represents the weighted fitness, Represents the group value, Represents the prediction group The predicted light intensity of the data, Represents the forecast group The light intensity of the data, Represents the amount of data in the prediction group, Representation training times, Representative The prediction group of the training is trained; based on the weight fitness of each group of light weight combinations, the light weight combination with the largest weight fitness is selected as the candidate combination, and the weight fitness of the candidate combination is compared with the weight fitness of the optimal weight combination. When the weight fitness of the candidate combination is greater than the weight fitness of the optimal weight combination, the candidate combination is used as the new optimal weight combination, the disturbance factor is preset, and each light weight combination is combined and updated based on the disturbance factor. The random selection algorithm is used to select the light weight combination. The weights are used as the exchange factors, and the exchange factors are replaced; this process is repeated until the optimal weight combination no longer changes, and the optimal weight combination at this time is output; the optimal weight combination is placed into the initial luminosity prediction model to obtain the luminosity prediction model; the luminosity prediction model can better simulate the illumination distribution in the actual environment, avoids the neglect of illumination changes in traditional methods and the jump or discontinuity phenomenon in illumination prediction, and ensures the robustness of the model in practical applications.

[0053] Based on the constructed luminosity prediction model, each drug placement point on the shelf in the retail pharmacy is predicted to obtain the predicted light intensity. The spline analysis method is used to perform spline analysis on the predicted light intensity and the collected light intensity on each shelf layer in the retail pharmacy to obtain a hierarchical light intensity curve. It should be noted that the collected light intensity is the light intensity in the light data;

[0054] The point illumination impact assessment of the hierarchical light intensity map is performed based on the time scale. The formula for the point illumination impact assessment of the hierarchical light intensity map is: ;in, The coordinates of the representative points are The light influence coefficient at Represents the total span of the time scale, The coordinates of the representative points are In the The light intensity at each time scale, Representative The time weight of each time scale represents the weight of the impact of light on the drug in different time periods; by calculating the light impact coefficient, it can accurately reflect the cumulative effect of light on the shelf life of the drug over a long period of time. The impact of light on the drug in different time periods is different. This method not only takes into account the light intensity, but also fully considers the impact of the time dimension on the drug, making the prediction more in line with the actual situation, thereby improving the accuracy of the shelf life prediction.

[0055] The shelf life of drugs is affected by both light and temperature. Traditional methods, which consider only one factor, cannot accurately reflect the coupled relationship between the two. Furthermore, the effect of temperature on drug aging is nonlinear, and traditional methods often fail to fully consider the impact of temperature changes on the chemical reaction rates of drugs. The drug expiration date assessment module can simultaneously consider the interaction between light and temperature, improving the accuracy of drug expiration date prediction. This module comprehensively considers multiple environmental variables and predicts drug expiration dates using precise mathematical models. Compared with traditional single-factor models, it can better adapt to complex real-world environments. Specifically:

[0056] A light-temperature fluctuation relationship model is constructed based on the hierarchical light intensity curve. The calculation formula of the light-temperature relationship model is: ;in, The coordinates of the representative points are The temperature fluctuation value at Represents the coordinate value of the first dimension in the three-dimensional coordinate system, Represents the coordinate value of the second dimension in the three-dimensional coordinate system, Represents the coordinate value of the third dimension in the three-dimensional coordinate system, represents the thermal conductivity, The coordinates of the points in the light intensity curve representing the level are The function value at , that is, the light intensity, Represents the time scale, represents the light attenuation coefficient, Represents the time of sunrise, Represents the total time of a day; sums the ambient temperature and the temperature fluctuation value to obtain the point-corrected temperature; obtains drug stability test data through drug stability test, performs degradation kinetic parameter analysis based on the drug stability test data, and generates drug degradation kinetic characteristic data; determines the activation energy and pre-exponential factor of a specific drug based on the drug degradation kinetic characteristic data, and generates drug degradation reaction parameter data; performs degradation reaction order determination processing based on the drug degradation reaction parameter data, and generates degradation reaction kinetic model data; performs absolute temperature conversion processing based on the point-corrected temperature data to generate point absolute temperature data; performs Arrhenius equation calculation based on the point absolute temperature data to generate point degradation reaction rate constant data; performs Q10 method based on the point degradation reaction rate constant data Then, an intelligent analysis module for temperature impact assessment is designed to obtain a temperature impact assessment engine; the degradation reaction kinetics model data is transmitted to the temperature impact assessment engine in real time for intelligent analysis of drug stability to generate drug stability prediction data; a shelf life change model is constructed based on the drug stability prediction data to generate a shelf life calculation model; the shelf life difference between the standard storage temperature and the actual storage temperature is calculated based on the shelf life calculation model to generate the shelf life change; by predicting the shelf life change of drugs under different environments, timely warnings are provided for drug management in actual operations, ensuring that drugs are always in the best storage environment, reducing drug waste caused by adverse environmental conditions, greatly improving the refinement of drug management, and providing a scientific basis for drug inventory management;

[0057] The drug stability test was conducted in accordance with the ICHQ1A (R2) guidelines, with three conditions designed: a long-term test (25°C ± 2°C / 60% RH ± 5% RH), an intermediate test (30°C ± 2°C / 65% RH ± 5% RH), and an accelerated test (40°C ± 2°C / 75% RH ± 5% RH). At the same time, a high-temperature test condition of 50°C ± 2°C / 75% RH ± 5% RH was added to improve the accuracy of the model prediction. At the above four temperature points, samples were taken for testing at 0, 1, 3, and 6 months, and the changes in the content of the active ingredient were recorded. The content of the active ingredient was determined by high-performance liquid chromatography (HPLC), and the precision of the test method was RSD < 1.0%, and the accuracy was within the range of 98.0%-102.0%, ensuring the reliability of the test data. Based on the obtained content-time data, the degradation rate constants at different temperatures were calculated. (i.e., degradation kinetic characteristic data). For example, the degradation rate constants of a certain drug at 25°C, 30°C, 40°C and 50°C are 0.0015, 0.0032, 0.0128 and 0.0423, respectively, indicating that the drug is significantly sensitive to temperature increases.

[0058] Determine the activation energy and pre-exponential factor of a specific drug based on the drug degradation kinetics data, and draw right The relationship diagram is obtained by least squares linear regression to obtain the regression equation:

[0059] ;in, represents the correlation coefficient, and the pre-exponential factor represents the pre-exponential factor. represents the activation energy. For the above drug example, the correlation coefficient is 0.9987. Extract the parameters from the regression equation: is 25.6341, which means the pre-exponential factor is 1.36×10¹¹; The value of the activation energy is -8653.7, which means the activation energy is 72.0. These parameters reflect the energy barrier and molecular collision frequency of the degradation reaction of the drug molecule, and are key inputs for subsequent stability prediction. The degradation reaction order is determined based on the drug degradation reaction parameter data, and the drug active ingredient content-time data are fitted with zero-order, first-order, and second-order reaction kinetic models respectively. Taking the above-mentioned drug as an example, the R² of the zero-order model fitting is 0.9532, the R² of the first-order model fitting is 0.9987, and the R² of the second-order model fitting is 0.9645. Since the first-order model has the highest goodness of fit, it is determined that the degradation reaction of the drug follows the first-order kinetic model: Ct=C0·e^(-kT·t). In addition, by calculating the residual analysis of the model prediction value and the measured value at different temperatures, the root mean square error (RMSE) of the residual is 0.42%, which further verifies the applicability of the first-order kinetic model.

[0060] Based on the point-corrected temperature data, absolute temperature conversion is performed, combined with the temperature fluctuation value calculated by the light temperature fluctuation model to obtain the point-corrected temperature. The point-corrected temperature is converted to Kelvin temperature to obtain the point-absolute temperature data. For example, if the ambient temperature at a point is 27.5°C and the calculated temperature fluctuation value is 1.2°C, the point-corrected temperature is 28.7°C, corresponding to an absolute temperature of 301.85K.

[0061] According to the absolute temperature data of the point, the Arrhenius equation is calculated. The activation energy Ea, pre-exponential factor A and absolute temperature TK of the drug are substituted into the Arrhenius equation: k=A·e^(-Ea / (R·TK)), and the degradation reaction rate constant k of the drug at the point is calculated. Taking the temperature of 301.85K as an example, the calculated degradation reaction rate constant is Through the calculation of the Arrhenius equation, a quantitative relationship between temperature and degradation rate was established, providing a theoretical basis for the prediction of drug stability.

[0062] The Q10 rule is used to assess temperature impacts based on point-by-point degradation reaction rate constant data. This method calculates the increase in the drug degradation rate, Q10, for every 10°C increase in temperature. Based on the Arrhenius equation, Q10 can be expressed as: Q10 = e^((Ea / R)·(10 / (T·(T+10)))). For the example drug mentioned above, Q10 = 2.13 at a temperature of 25°C, indicating that the degradation rate increases approximately 2.13-fold for every 10°C increase in temperature. A temperature impact assessment engine was developed. Inputs include the drug's activation energy, pre-exponential factor, baseline temperature, and actual temperature. Outputs include the degradation rate ratio and the percentage change in shelf life, providing a quantitative temperature impact assessment for drug storage management.

[0063] The degradation reaction kinetic model data is transmitted in real time to the temperature impact assessment engine for intelligent analysis of drug stability. Based on the first-order reaction kinetic model: Ct=C0·e^(-k·t), the curve of the change of the drug active ingredient content over time is calculated. The lower limit of the drug active ingredient content is set to 90%, that is, Ct / C0=0.9, and the calculated shelf life of the drug is t=-ln(0.9) / k=0.1054 / k. Taking the above example drug at the standard storage temperature For example, the calculated validity period is t=0.1054 / 0.0015=70.3 months, or about 5.9 years.

[0064] Based on drug stability prediction data, a shelf life change model was constructed. Based on the Arrhenius equation, the relationship between temperature and shelf life was derived: t_exp = 0.1054·A^(-1)·e^(Ea / (R·T)). For the example drug described above, the shelf life calculation model is: t_exp = 7.75×10^(-13)·e^(8653.7 / T). This model can be used to calculate the shelf life of a drug at any temperature. For example, the shelf life at 30°C is 32.9 months, and at 40°C it is 8.2 months.

[0065] The shelf life calculation model calculates the difference between the standard storage temperature and the actual storage temperature, generating a shelf life change. For example, using a standard storage temperature of 25°C and an actual storage temperature of 28°C, the calculated shelf lives are 70.3 months and 43.5 months, respectively. The shelf life change is Δt_exp = 70.3 - 43.5 = 26.8 months. This indicates that the shelf life of a drug stored at 28°C is approximately 38.1% shorter than that at 25°C. This result intuitively demonstrates the significant impact of elevated temperatures on drug stability and provides a quantitative basis for controlling drug storage conditions.

[0066] Traditional drug display location optimization is often static and cannot respond to multiple factors such as the ambient light impact and drug expiration date in a timely manner. It is difficult to handle complex constraints and dynamic changes, which leads to poor layout optimization and unsatisfactory optimization results. The display optimization module comprehensively considers the impact of light and expiration date changes, optimizes the drug display location through a comprehensive scoring function, flexibly adapts to actual changes, avoids the limitations of single-factor optimization, makes the display layout more reasonable, and improves the shelf life of drugs. Specifically:

[0067] The drug adaptability evaluation function is constructed based on the light impact coefficient and the change in shelf life. The calculation formula of the drug adaptability evaluation function is: ; The coordinates of the representative drug are The positional fitness at The coordinates of the representative drug are The change in validity period, Represents the light intensity weight, Represents the effective period weight; the target optimization function is constructed based on the drug adaptability evaluation function. The formula of the target optimization function is:

[0068] ;in, Represents global fitness; the drug adaptability evaluation function comprehensively evaluates the optimization potential of each location by integrating the impact of light and expiration date prediction, providing a comprehensive basis for drug display, improving the scientific nature of the display layout, flexibly adjusting the importance of different factors, and further improving the accuracy of optimization.

[0069] Preset Group drug layouts. Each group of drug layouts represents a group of drug display positions. Each group of drug layouts is considered a chromosome. Each gene in the chromosome represents a column position in the drug layout. The optimal chromosome is initialized to be empty.

[0070] The target optimization function is used to calculate the global fitness of each chromosome, and the global fitness is used as the layout fitness of the chromosome;

[0071] A clustering algorithm is used to perform cluster analysis on the layout fitness of each chromosome to obtain chromosome clustering. Common clustering algorithms include K-Means clustering algorithm and hierarchical clustering algorithm. A filtering threshold is preset, and the chromosome clusters whose cluster centers are greater than or equal to the filtering threshold are used as the preferred pool. A selection wheel is constructed based on the preferred pool. The selection probability of each chromosome in the selection wheel is equal to the layout fitness of the chromosome divided by the sum of the layout fitness of all chromosomes in the preferred pool. chromosomes as the preferred chromosomes;

[0072] The preferred chromosomes are arranged and combined to obtain crossover combinations. The mean value of the layout fitness of the preferred chromosomes is used as the crossover threshold. The crossover operator is constructed based on the crossover threshold. The formula of the crossover operator is:

[0073] ;in, Represents the preferred chromosome The crossover probability of a chromosome is represents a fixed probability, which is set by those skilled in the art based on the actual situation. Represents the maximum layout fitness within the preferred chromosome, Represents the preferred chromosome The layout fitness of chromosomes, represents the crossover threshold; the chromosome in the crossover combination is used as the parent chromosome, a gene in the parent chromosome is randomly selected as the crossover starting point, a crossover operation is performed based on the crossover probability, the genes after the crossover node are exchanged to obtain the crossover chromosome, the layout fitness of the crossover chromosome is calculated, and the crossover chromosome with a layout fitness greater than that of the parent chromosome is regarded as a good individual; a random selection algorithm is used to randomly select a gene in the good individual as the mutation sequence, and the order of the genes in the mutation sequence is changed to obtain a variant;

[0074] Calculate the layout fitness of the variants, select the preferred variant with the largest fitness as the candidate individual, and when the layout fitness of the candidate individual is greater than the optimal chromosome, use the candidate individual as the new optimal chromosome and the variant as the new preferred chromosome. Repeat until the optimal chromosome no longer changes. The optimal chromosome at this time is output as the optimal drug layout, and the drugs are placed in retail pharmacies based on the optimal drug layout.

[0075] This embodiment realizes accurate light intensity prediction of the placement of medicines by constructing a photometric prediction model. The model combines multiple factors such as light intensity, time scale, and position coordinates to provide high-precision light intensity prediction for the medicine display environment, ensuring that the medicines are stored under appropriate light intensity, thereby extending the shelf life of the medicines and preventing damage caused by excessive or insufficient light. By performing point storage analysis on the hierarchical light intensity curve chart, taking into account the relationship between the speed of medicine aging and temperature, the shelf life of the medicines is dynamically evaluated and the display layout is optimized. This evaluation mechanism not only helps retailers avoid the occurrence of expired medicines, but also reduces medicine waste and improves the efficiency of medicine use. The target optimization function is optimized and solved through an improved genetic algorithm to optimize the display position of the medicines so that the medicines meet the optimal light and temperature conditions, which greatly improves the operating efficiency and medicine management level of retail pharmacies, reduces the risk and error of human intervention, and ensures the optimal environment for medicine storage.

[0076] Example 2;

[0077] See also Figure 2 As shown, for parts not described in detail in this embodiment, please refer to the description of Example 1. A method for comprehensive management of pharmaceutical product displays in retail pharmacies is provided, including:

[0078] S1, collect lighting data and environmental data;

[0079] S2. Construct an initial luminosity prediction model based on the illumination data, and train the initial luminosity prediction model using the illumination data as sample data to obtain a luminosity prediction model; predict each drug placement point on a shelf in a retail pharmacy based on the constructed luminosity prediction model to obtain a predicted light intensity; analyze the predicted light intensity using a spline analysis method to obtain a hierarchical light intensity curve; and perform a light impact assessment on the hierarchical light intensity curve to obtain a light impact coefficient;

[0080] S3. Perform point storage analysis on the hierarchical light intensity curve to obtain the effective period change;

[0081] S4. Construct a drug adaptability evaluation function based on the illumination influence coefficient and the change in the expiration date, construct a target optimization function based on the drug adaptability evaluation function, use an improved genetic algorithm to optimize and solve the target optimization function, obtain the optimal drug layout, and place drugs in retail pharmacies based on the optimal drug layout.

[0082] Example 3;

[0083] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the above-mentioned method for comprehensive management of pharmaceutical product displays in retail pharmacies is implemented.

[0084] Since the electronic device described in this embodiment is an electronic device used to implement the method for comprehensive management of pharmaceutical product displays in a retail pharmacy according to the embodiment of this application, those skilled in the art will be able to understand the specific implementation and various variations of the electronic device of this embodiment based on the method for comprehensive management of pharmaceutical product displays in a retail pharmacy according to the embodiment of this application. Therefore, how the electronic device implements the method in the embodiment of this application will not be described in detail here. As long as those skilled in the art can implement the electronic device used in the method for comprehensive management of pharmaceutical product displays in a retail pharmacy according to the embodiment of this application, it falls within the scope of protection of this application.

[0085] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0086] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that for users of ordinary skill in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies, characterized by: include: Data acquisition module: collects lighting data and environmental data; Lighting prediction module: Build an initial luminosity prediction model based on lighting data, and use the lighting data as sample data to train the initial luminosity prediction model to obtain a luminosity prediction model. Based on the constructed luminosity prediction model, predict the light intensity at each drug placement point on the retail pharmacy shelf to obtain the predicted light intensity. Use spline analysis to analyze the predicted light intensity and obtain a hierarchical light intensity curve. Perform a lighting impact assessment on the hierarchical light intensity curve to obtain the lighting impact coefficient. Drug expiration date assessment module: performs point storage analysis on the hierarchical light intensity curve to obtain the change in validity period; Display optimization module: Construct a drug adaptability evaluation function based on the light impact coefficient and the change in the expiration date, construct a target optimization function based on the drug adaptability evaluation function, use an improved genetic algorithm to optimize and solve the target optimization function, obtain the optimal drug layout, and place drugs in retail pharmacies based on the optimal drug layout.

2. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to claim 1, characterized in that: The illumination data includes: illumination intensity, time scale and position coordinates; the environmental data includes ambient temperature; the method for obtaining the illumination data includes: spatially three-dimensionalizing the retail pharmacy, selecting a shelf point in the retail pharmacy as the point center, constructing a standard three-dimensional coordinate system with the point center, and using the straight-line distance from other points to the point center in each direction dimension in the three-dimensional coordinate system as the coordinate value of each dimension, and all coordinate values constitute the point coordinates; based on the point coordinates, some points in the retail pharmacy are selected as sampling points, and distributed photometric sensors are set at the sampling points, and a collection time interval is preset. The distributed photometric sensor collects data every collection time interval to obtain light intensity and time scale.

3. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to claim 2, characterized in that: The method of constructing the initial luminosity prediction model includes: Calculate the variance of light intensity at the same time scale and construct a point evaluation function based on the variance. The formula of the point evaluation function is: ;in, Representative The lighting data and The light intensity correlation of the illumination data, Representative The lighting data and The distance of the illumination data, common distance evaluation functions are Euclidean distance function and Chebyshev distance function, represents the variance of light intensity at the time scale; Preset group value , is an integer greater than zero, dividing the illumination data equally into Group data, each time The data group is used as the sample group, and the remaining data group is used as the prediction group to train the initial photometric prediction model. Each training uses the sample group as input to predict each data in the prediction group. Repeat times; the formula of the initial luminosity prediction model is: ;in, Represents the forecast group The predicted light intensity of the light data, Representative sample group The light weight of the lighting data, Representative sample group The illumination data and the prediction group The light intensity correlation of the illumination data, Representative sample group The light intensity of the lighting data, represents the index of the sample group, Represents the size of the prediction group.

4. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to claim 3, characterized in that: The method of training the initial photometric prediction model includes: Preset Group light weight combination, is an integer greater than zero, the optimal weight combination is initialized to empty, the weight fitness of the optimal weight combination is infinitesimal, and the light weights in the light weight combination satisfy the weight constraints: ; Bring each light weight combination into the initial luminosity prediction model, and evaluate the adaptability of each light weight combination using the sample group and the prediction group. The formula for evaluating the adaptability of each light weight combination is: ;in, represents the weighted fitness, Represents the group value, Represents the prediction group The predicted light intensity of the data, Represents the forecast group The light intensity of the data, Represents the amount of data in the prediction group, Representative A subset of samples, Representation training times, Representative The prediction group of the training is trained; based on the weight fitness of each group of light weight combinations, the light weight combination with the largest weight fitness is selected as the candidate combination, and the weight fitness of the candidate combination is compared with the weight fitness of the optimal weight combination. When the weight fitness of the candidate combination is greater than the weight fitness of the optimal weight combination, the candidate combination is used as the new optimal weight combination, the disturbance factor is preset, and each light weight combination is combined and updated based on the disturbance factor. The random selection algorithm is used to select the light weight combination. The weights are used as the exchange factors, and the position of the exchange factors is replaced; this is repeated until the optimal weight combination no longer changes, and the optimal weight combination at this time is output; the optimal weight combination is placed into the initial photometric prediction model to obtain the photometric prediction model.

5. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to claim 4, characterized in that: The formula for evaluating the illumination impact of the hierarchical light intensity curve is: ;in, The coordinates of the representative points are The light influence coefficient at Represents the total span of the time scale, The coordinates of the representative points are In the The light intensity at each time scale, Representative Time scale time weight.

6. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to claim 5, characterized in that: The method of performing point storage analysis on the hierarchical light intensity curve diagram includes: A light-temperature fluctuation relationship model is constructed based on the hierarchical light intensity curve. The calculation formula of the light-temperature relationship model is: ;in, The coordinates of the representative points are The temperature fluctuation value at Represents the coordinate value of the first dimension in the three-dimensional coordinate system, Represents the coordinate value of the second dimension in the three-dimensional coordinate system, Represents the coordinate value of the third dimension in the three-dimensional coordinate system, represents the thermal conductivity, The coordinates of the points in the light intensity curve representing the level are The function value at , that is, the light intensity, Represents the time scale, represents the light attenuation coefficient, Represents the time of sunrise, representing the total time of a day; summing the ambient temperature and the temperature fluctuation value to obtain the point-corrected temperature; obtaining drug stability test data through drug stability testing, performing degradation kinetic parameter analysis based on the drug stability test data to generate drug degradation kinetic characteristic data; determining the activation energy and pre-exponential factor of a specific drug based on the drug degradation kinetic characteristic data to generate drug degradation reaction parameter data; performing degradation reaction order determination processing based on the drug degradation reaction parameter data to generate degradation reaction kinetic model data; performing absolute temperature conversion processing based on the point-corrected temperature data to generate point-absolute temperature data; performing Arrhenius equation calculation based on the point-absolute temperature data to generate point-degradation reaction rate constant data; designing an intelligent analysis module for temperature impact assessment based on the point-degradation reaction rate constant data using the Q10 rule to obtain a temperature impact assessment engine; transmitting the degradation reaction kinetic model data in real time to the temperature impact assessment engine for intelligent drug stability analysis to generate drug stability prediction data; constructing a shelf life change model based on the drug stability prediction data to generate a shelf life calculation model; calculating the shelf life difference between the standard storage temperature and the actual storage temperature based on the shelf life calculation model to generate a shelf life change.

7. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to claim 6, characterized in that: The formula of the drug adaptability evaluation function is: ; The coordinates of the representative drug are The positional fitness at The coordinates of the representative drug are The change in validity period, Represents the light intensity weight, represents the effective period weight; The formula of the objective optimization function is: ;in, Represents the global fitness.

8. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to claim 7, characterized in that: The method of optimizing and solving the target optimization function includes: Preset Group drug layouts. Each group of drug layouts represents a set of arrangement of drug display positions. Each group of drug layouts is considered a chromosome. Each gene in the chromosome represents a display position in the drug layout. The optimal chromosome is initialized to be empty. The global fitness of each chromosome is calculated using the target optimization function, and the global fitness is used as the layout fitness of the chromosome. Based on the layout fitness, the chromosomes are selected, crossed, and mutated to obtain variants. Calculate the layout fitness of the variants, select the preferred variant with the largest fitness as the candidate individual, and when the layout fitness of the candidate individual is greater than the optimal chromosome, use the candidate individual as the new optimal chromosome and the variant as the new preferred chromosome. Repeat until the optimal chromosome no longer changes. The optimal chromosome at this time is output as the optimal drug layout, and the drugs are placed in retail pharmacies based on the optimal drug layout.

9. A comprehensive management platform for displaying pharmaceutical products in retail pharmacies according to claim 8, characterized in that: The method of selecting chromosomes includes: A clustering algorithm is used to perform cluster analysis on the layout fitness of each chromosome to obtain chromosome clusters; a filtering threshold is preset, and the chromosome clusters whose cluster centers are greater than or equal to the filtering threshold are used as the preferred pool; a selection wheel is constructed based on the preferred pool, and the selection probability of each chromosome in the selection wheel is equal to the layout fitness of the chromosome divided by the sum of the layout fitness of all chromosomes in the preferred pool. chromosomes as the preferred chromosomes; The crossover and mutation method includes: performing permutations and combinations on the preferred chromosomes to obtain a crossover combination, using the mean of the layout fitness of the preferred chromosomes as a crossover threshold, and constructing a crossover operator based on the crossover threshold. The formula of the crossover operator is: ;in, Represents the preferred chromosome The crossover probability of a chromosome is represents a fixed probability, which is set by those skilled in the art based on the actual situation. Represents the maximum layout fitness within the preferred chromosome, Represents the preferred chromosome The layout fitness of chromosomes, represents the crossover threshold; the chromosome in the crossover combination is used as the parent chromosome, a gene in the parent chromosome is randomly selected as the crossover starting point, a crossover operation is performed based on the crossover probability, the genes after the crossover node are exchanged, the crossover chromosome is obtained, the layout fitness of the crossover chromosome is calculated, and the crossover chromosome with a layout fitness greater than that of the parent chromosome is regarded as an excellent individual; a random selection algorithm is used to randomly select a gene in the excellent individual as the mutation sequence, and the order of the genes in the mutation sequence is changed to obtain a variant.

10. A method for integrated management of pharmaceutical product displays in retail pharmacies, which is implemented based on a platform for integrated management of pharmaceutical product displays in retail pharmacies according to any one of claims 1 to 9, characterized in that: include: S1, collect lighting data and environmental data; S2. Construct an initial luminosity prediction model based on the illumination data, and train the initial luminosity prediction model using the illumination data as sample data to obtain a luminosity prediction model; predict each drug placement point on a shelf in a retail pharmacy based on the constructed luminosity prediction model to obtain a predicted light intensity; analyze the predicted light intensity using a spline analysis method to obtain a hierarchical light intensity curve; and perform a light impact assessment on the hierarchical light intensity curve to obtain a light impact coefficient; S3. Perform point storage analysis on the hierarchical light intensity curve to obtain the effective period change; S4. Construct a drug adaptability evaluation function based on the illumination influence coefficient and the change in the expiration date, construct a target optimization function based on the drug adaptability evaluation function, use an improved genetic algorithm to optimize and solve the target optimization function, obtain the optimal drug layout, and place drugs in retail pharmacies based on the optimal drug layout.

Images