Method and device for measuring and calculating reasonable capacity of tobacco market based on multi-objective optimization
Through the tobacco market reasonable capacity calculation method based on multi-objective optimization, the problem of market capacity calculation in the existing technology depends on experience and lack of systematic models, and the multi-dimensional analysis and multi-objective equilibrium of the tobacco market are achieved, and more accurate and scientific market capacity calculation results are provided.
Patent Information
- Application Number
- CN202510436025.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-09
- Publication Date
- 2025-06-27
AI Technical Summary
In the existing tobacco market management, market capacity calculation relies on empirical methods, making it difficult to accurately grasp market dynamics, and there is a lack of systematic models to comprehensively consider multi-dimensional market characteristics.
The tobacco market reasonable capacity calculation method based on multi-objective optimization is adopted. By collecting and fusion of multi-source heterogeneous data, a multi-objective optimization model is constructed, and different weights are calculated and assigned to different weights to achieve multi-dimensional analysis and multi-objective equilibrium of the tobacco market reasonable capacity.
It has achieved more accurate and scientific calculations of the reasonable capacity of the tobacco market, provided scientific decision-making support, and helped tobacco market management better balance multiple dimensions such as monopoly management, consumer convenience, retailer profit and marketing management.
Smart Images
Figure CN120218992A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of tobacco market management, and particularly to a method and device for calculating the reasonable capacity of a tobacco market based on multi-objective optimization. Background Art
[0002] In traditional tobacco market management, the calculation of market capacity mainly relies on empirical methods, which have many limitations. For example, they often fail to accurately grasp market dynamics and are difficult to formulate a reasonable market capacity according to the actual situation. With the development of big data technology, tobacco market management is gradually shifting from experience-driven to data-driven. However, most existing studies focus on the internal data of the tobacco industry and lack a systematic model that can comprehensively consider multi-dimensional market characteristics. Therefore, how to improve the scientific nature of tobacco market management with the help of big data technology and more accurately calculate the reasonable capacity of the market has become an urgent problem to be solved. Summary of the Invention
[0003] In view of the above-mentioned defects or deficiencies in the prior art, the present invention provides a method and device for calculating the reasonable capacity of a tobacco market based on multi-objective optimization. On the basis of analyzing multi-dimensional market characteristics, a large amount of multi-source heterogeneous data inside and outside the tobacco industry is collected and fused, and a reasonable capacity calculation model based on multi-objective optimization is constructed, realizing multi-dimensional analysis and multi-objective balance of the reasonable capacity of the tobacco market.
[0004] One aspect of the present invention provides a method for calculating the reasonable capacity of a tobacco market based on multi-objective optimization, including: Collecting multi-source data including internal data and external data and performing preprocessing, where the internal data includes retailer information, online order sales and inventory data, and grid supervision data, and the external data includes population and economic data, POI data, AOI data, and grid feature data; Calculating measurement indicators according to the preprocessed multi-source data; Constructing a reasonable capacity calculation model based on multi-objective optimization according to the measurement indicators, and assigning different weights to the measurement indicators through importance ranking; Inputting the preprocessed multi-source data and the parameters of the measurement indicators into the reasonable capacity calculation model, and solving the reasonable capacity calculation model to obtain a reasonable capacity calculation result.
[0005] Further, the step of inputting the preprocessed multi-source data and the parameters of the measurement indicators into the reasonable capacity calculation model, and solving the reasonable capacity calculation model to obtain a reasonable capacity calculation result includes: Establishing an objective function of the reasonable capacity calculation model: , Among them , ; is the i-th measurement index of the j-th grid, is the reasonable measurement capacity of the j-th grid, n is the number of measurement indexes, is the expected value of the i-th measurement index of the j-th grid, is the weight of the i-th measurement index of the j-th grid, ; The reasonable capacity measurement result of each grid is obtained by solving according to the objective function .
[0006] Furthermore, it also includes: Including the measurement index of the regulation coefficient , which is calculated by the following formula:
[0007]
[0008]
[0009] Among them, is the prediction result of the reasonable measurement capacity of the j-th grid by the m-th decision tree, is the h-th feature of the j-th grid, is the weight of the h-th feature of the j-th grid, is the number of features, is the number of decision trees, is the fitting capacity of the j-th grid, is the number of grid retailers in the j-th grid.
[0010] Furthermore, it also includes: Including the measurement index of the per capita license holding rate , which is calculated by the following formula:
[0011] Among them, is the number of grid retailers in the j-th grid, is the grid population number of the j-th grid.
[0012] Furthermore, it also includes: Including the measurement index of the average walking distance of consumers , which is calculated by the following formula:
[0013] Among them, is the grid area of the j-th grid, is the average walking distance from the residence of consumers within the j-th grid to the retail store.
[0014] Furthermore, it also includes: including the measurement index of the annual average sales volume of the grid , which is obtained by the following formula:
[0015] Among them, is the annual sales volume of the j-th grid, is the number of retail households in the j-th grid.
[0016] Furthermore, it also includes: including the measurement index of the average monthly gross profit per household in the grid , which is obtained by the following formula:
[0017] Among them, is the monthly retail price of the j-th grid, is the monthly wholesale price of the j-th grid, is the number of retail households in the j-th grid.
[0018] Furthermore, it also includes: including the measurement index of the average gross profit margin of the grid , which is obtained by the following formula: .
[0019] is the monthly retail price of the j-th grid, is the average monthly gross profit per household in the j-th grid, is the number of retail households in the j-th grid.
[0020] Furthermore, it also includes: Calculating the mean square error , the mean absolute error , the mean relative error :
[0021]
[0022]
[0023] Among them, is the mean square error, is the mean absolute error, is the mean relative error, and k is the number of grids, is the number of tobacco retailers in the j-th grid, is the calculated reasonable capacity of the j-th grid; According to the mean square error obtained by calculation , mean absolute error and mean relative error , evaluate the calculation result of the reasonable capacity.
[0024] On the other hand, the present invention also provides a device for calculating the reasonable capacity of the tobacco market based on multi-objective optimization, including: The first module is configured to collect multi-source data including internal data and external data and perform preprocessing. The internal data includes retailer information, network order-sale-inventory data, and grid supervision data, and the external data includes population and economic data, POI data, AOI data, and grid feature data; The second module is configured to calculate measurement indicators according to the preprocessed multi-source data; The third module is configured to construct a reasonable capacity measurement model based on multi-objective optimization according to the measurement indicators, and assign different weights to the measurement indicators through importance ranking; The fourth module is configured to input the preprocessed multi-source data and the parameters of the measurement indicators into the reasonable capacity measurement model, and solve the reasonable capacity measurement model to obtain the calculation result of the reasonable capacity.
[0025] The method and device for calculating the reasonable capacity of the tobacco market based on multi-objective optimization provided in this embodiment comprehensively consider multiple dimensions such as monopoly control, consumer convenience, retailer profitability, and marketing management. By analyzing and calculating six key indicators including the regulation coefficient, per capita license holding rate, average walking distance of consumers, average monthly gross profit per household in the grid, average gross profit margin in the grid, and average annual sales volume in the grid, a reasonable capacity measurement model based on multi-objective optimization is constructed, realizing multi-dimensional analysis and multi-objective balance of the reasonable capacity of the tobacco market, and being conducive to providing scientific decision-making support for tobacco market management. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, purposes, and advantages of the present application will become more obvious: Figure 1 is a schematic flowchart of a method for calculating the reasonable capacity of the tobacco market based on multi-objective optimization provided by an embodiment of the present application; Figure 2It is a comparison chart of the existing quantity and the calculated reasonable capacity of a method for calculating the reasonable capacity of the tobacco market based on multi-objective optimization provided by an embodiment of the present application; Figure 3 It is a QQ chart of the existing quantity and the calculated reasonable capacity of a method for calculating the reasonable capacity of the tobacco market based on multi-objective optimization provided by an embodiment of the present application; Figure 4 It is a schematic structural diagram of a device for calculating the reasonable capacity of the tobacco market based on multi-objective optimization provided by an embodiment of the present application. Detailed implementation manners
[0027] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0028] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "the" and "said" used in the embodiments of the present invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.
[0029] It should be understood that although the terms first, second, third, etc. may be used to describe the acquisition modules in the embodiments of the present invention, these acquisition modules should not be limited to these terms. These terms are only used to distinguish the acquisition modules from each other.
[0030] Depending on the context, the word "if" as used herein can be interpreted as "when" or "while" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detected (stated condition or event)" can be interpreted as "when determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)".
[0031] It should be noted that the orientation words such as "upper", "lower", "left" and "right" described in the embodiments of the present invention are described from the angles shown in the drawings, and should not be construed as limiting the embodiments of the present invention. In addition, in the context, it should also be understood that when it is mentioned that an element is formed "on" or "under" another element, it can not only be directly formed "on" or "under" another element, but also be indirectly formed "on" or "under" another element through an intermediate element.
[0032] See Figure 1 , an embodiment of the present invention provides a method for calculating the reasonable capacity of the tobacco market based on multi-objective optimization, including: Step S101, collect multi-source data including internal data and external data and perform preprocessing. The internal data includes retailer information, online order sales and inventory data, and grid supervision data. The external data includes population and economic data, POI data, AOI data, and grid feature data; Specifically, collect the internal data of the tobacco market's monopoly, marketing, and logistics in the sample area through the internal tobacco own database and data center, including: retailer information, grid order sales and inventory data, and grid supervision data. Among them, the retailer information includes license number, business format, business address, longitude coordinate, latitude coordinate, and grade; the grid order sales and inventory data includes wholesale price, retail price, sales amount, sales quantity, order quantity, and order amount, and the statistical period is one year; the grid supervision data includes the quantity of real cigarettes flowing into the grid, the quantity of real cigarettes flowing out of the grid, and the number of retailers involved in the case in the grid, including three-year statistical data; Collect the external data of the sample area through external data platforms such as government information disclosure websites and map applications, including: population and economic data, POI (Point of Interest) data, AOI (Area of Interest) data, and grid feature data. Among them, the population and economic data includes GDP, per capita income, and permanent population. The POI data includes the geographical coordinates of transportation, accommodation, catering, shopping, and enterprises, etc. The AOI data includes the geographical coordinates of primary and secondary schools, government agencies, and medical and health institutions, etc. The grid feature data includes grid basic information, monthly average passenger flow of the grid, grid asset level distribution, and grid active business district ranking, etc.
[0033] To ensure the integrity, consistency, correctness, and maximum utilization rate of the data, perform preprocessing on the collected internal data and external data, including data cleaning, grid division, data integration, and feature engineering. The specific steps are as follows: Data cleaning: Delete duplicate records, null values, and outliers in the data to ensure the accuracy and consistency of the data; fill in missing data. Exemplarily, some retailer longitude and latitude information is missing, and the address data is converted into longitude and latitude coordinates through the API interface address coding of the map application to fill in the missing geographical location information; through coordinate transformation and projection calculation, unify the data spatial reference system; Grid division: Perform grid clustering and aggregation division according to the population and economic data and POI data. Optionally, further adjust the grid boundary according to the actual business situation to ensure that the grid division meets the actual operation requirements, form the final grid division result, and assign a name to each grid; Data integration: Through spatial join calculation, locate data such as the license numbers, longitude coordinates, latitude coordinates of retail households, the geographical coordinates of POI data, and the geographical boundaries of grids to the smallest unit grid, so as to calculate key indicators such as the number of retail households, sales volume, and the number of POIs in each grid based on the license numbers of retail households and grid names, and associate the internal data with external data. Exemplarily, associate the retail household information and grid order-sale-inventory data in the internal data with the population and economic data, POI data, etc. in the external data according to the license numbers of retail households and grid names to form a unified data set based on the smallest unit grid; Feature engineering: Perform One-Hot encoding on discrete features and convert them into a numerical form acceptable to the model; mine new features through feature interaction and combination to improve the feature expression ability; feature generation. Exemplarily, convert the coordinate system to the EPSG:2380 coordinate system, project and calculate the grid area based on the Shoelace algorithm to generate grid area feature data; perform feature selection through correlation analysis based on the Pearson correlation coefficient; Optionally, the data can also be personalized. Exemplarily, perform weighted processing on population data. The population statistics data of the statistical bureau reflects relatively fixed population, while the flow data reflects mobile population. When calculating population data, perform weighted processing based on the statistical population and flow data to obtain more accurate regional population data and ensure that the data processing results are more in line with the actual situation.
[0034] Step S102, calculate measurement indicators according to the preprocessed multi-source data; Specifically, first, calculate the grid portrait index based on the preprocessed multi-source data for data analysis. The grid portrait index mainly includes population index, order index, economic and social index, sales index, sales volume index, and spacing index. Among them, the population index is obtained by dividing the comprehensive population weighted by the statistical population and pedestrian flow data by the area of the grid region to obtain the grid population density, and then taking the ratio of each grid population density to the maximum population density in the sample region as the population index of each grid. The population index is used to reflect the population density ranking and comprehensive population passenger flow of each grid in the sample region. The order index is obtained by statistically converting the order quantity and order amount of retail households in the grid, and is used to reflect the consumption demand and purchasing power in the grid. The economic and social index is synthesized by analyzing social and economic indicators such as the income level, consumption ability, and education level of residents in the grid, and is used to reflect the economic development level and the quality of life of residents in the grid. The sales index is closely related to the sales volume index. The sales index is obtained by summing up the sales amounts in each grid, and the sales volume index is obtained by counting the sales quantities in each grid by the box and converting them into an index, which is used to measure the market consumption ability of each grid. The spacing index is based on the KD tree (k-dimensional tree, a tree structure for k-dimensional space data). According to the longitude and latitude coordinates of retail households, the distance to the nearest neighbor retail household is calculated, and then the average grid spacing is calculated. Finally, it is converted into an index according to the maximum value. This index is used to measure the distribution density and service scope of retail households in the grid. Analyze the grid portrait index. For the dimensions of population index, order index, sales volume index, sales index, and economic and social index, when the grid index is too high, the grid capacity should be increased; when the grid index is too low, the grid capacity should be decreased. For the spacing index, when the grid index is too high, the grid capacity should be decreased; when the grid index is too low, the grid capacity should be increased, so as to achieve a certain degree of balance between grids.
[0035] Secondly, comprehensively considering business requirements and data availability, starting from the four perspectives of monopoly control, consumer convenience, retailer profitability, and marketing management, six measurement indicators are constructed based on machine learning, statistical analysis, professional field customization, etc. Data analysis is carried out on each grid in the sample region according to the six measurement indicators calculated from the preprocessed multi-source feature data. The six measurement indicators include: regulation coefficient, per capita license holding rate, average walking distance of consumers, monthly average gross profit per household in the grid, average gross profit margin in the grid, and average annual sales volume in the grid. Exemplarily, divide a sample region into k grids. Based on the random forest algorithm, analyze and fit each grid through multiple feature data such as the number of retail households, population, and number of POIs in the grid. Calculate the regulation coefficient measurement indicator of the jth grid according to the following formula , :
[0036]
[0037]
[0038] Among them, is the predicted result of the measured reasonable capacity of the m-th decision tree for the j-th grid, is the h-th feature of the j-th grid, is the weight of the h-th feature of the j-th grid, is the number of features, is the number of decision trees, is the fitted capacity of the j-th grid, is the number of grid retail households in the j-th grid; the measurement index of the regulation coefficient is the ratio of the existing number of grids, that is, the number of grid retail households to the fitted capacity of the grid, and the floating range of the adjusted capacity is determined by adjusting the measurement index of the regulation coefficient; The measurement index of the per capita license holding rate is calculated according to the following formula :
[0039] Among them, is the number of grid retail households in the j-th grid, is the grid population in the j-th grid; the measurement index of the per capita license holding rate is used to evaluate and adjust the layout and number of retail households through statistical analysis of the number of retail households and the population, so as to optimize the allocation of market resources; The measurement index of the average walking distance of consumers is calculated according to the following formula :
[0040] Among them, is the grid area of the j-th grid, is the average walking distance from the residence of consumers in the j-th grid to the retail store; the measurement index of the average walking distance of consumers calculates the average walking distance through geospatial analysis of the locations of retail stores and the residences of consumers in the grid, and is used to evaluate the rationality and optimization degree of the layout of retail stores; The measurement index of the annual average sales volume of the grid is calculated according to the following formula :
[0041] Among them, is the annual sales volume of the j-th grid, and is the number of grid retail households in the j-th grid; the measurement index of the annual average sales volume of the grid is used to reflect the overall consumption capacity of the market and the activity degree of the retail business, and is an important quantitative index for measuring market potential and evaluating economic contributions; The average monthly gross profit per household and the average gross profit rate of the grid are calculated by statistically analyzing the monthly retail price and the monthly wholesale price of the retail households in the grid. The difference between the two is used as the total gross profit amount, which is divided by the number of retail households to obtain the average monthly gross profit per household in the grid. The average gross profit rate is calculated based on the gross profit rate formula. The measurement index of the average monthly gross profit per household in the grid is calculated according to the following formula :
[0042] where is the monthly retail price of the j-th grid, is the monthly wholesale price of the j-th grid, and is the number of retail households in the j-th grid; The measurement index of the average gross profit rate of the grid is calculated according to the following formula :
[0043] where is the monthly retail price of the j-th grid, is the average monthly gross profit per household in the j-th grid, is the number of retail households in the j-th grid; Analyze the measurement indicators. For the measurement indicators of the average number of licenses per person and the average walking distance of consumers, when the grid indicators are too high, the grid capacity should be reduced; when the grid indicators are too low, the grid capacity should be increased. For the measurement indicators of the average gross profit rate of the grid, the average annual sales volume of the grid, and the average monthly gross profit per household in the grid, when the grid indicators are too high, the grid capacity should be increased; when the grid indicators are too low, the grid capacity should be reduced, so as to achieve a certain degree of balance among the grids.
[0044] Step S103: Construct a reasonable capacity measurement model based on multi-objective optimization according to the measurement indicators, and assign different weights to the measurement indicators through importance ranking; Specifically, comprehensively consider six measurement indicators: the regulation coefficient, the average number of licenses per person, the average walking distance of consumers, the average monthly gross profit per household in the grid, the average gross profit rate of the grid, and the average annual sales volume of the grid. Construct a reasonable capacity measurement model based on multi-objective optimization, and find the optimal solution that satisfies all objectives by setting different objective functions and constraint conditions; Establish the objective function of the reasonable capacity measurement model: , where , ; where is the i-th measurement indicator of the j-th grid, is the reasonable capacity measurement of the j-th grid, n is the number of measurement indicators, is the expected value of the i-th measurement indicator of the j-th grid, is the weight of the i-th measurement indicator of the j-th grid, ; For the input parameters and data of different grids, conflicts between indicators easily lead to no feasible solutions, and there are actually different importance levels among the indicators. Through importance ranking, different weights are assigned to the measurement indicators according to different rankings, and finally a weighted solution is obtained to obtain the reasonable capacity measurement results for each grid. ; , is a constraint condition, that is, the reasonable capacities measured for all grids need to be greater than or equal to zero, and the calculation results of the measurement indicators need to be within a predetermined expected range.
[0045] Step S104, input the preprocessed multi-source data and the parameters of the measurement indicators into the reasonable capacity measurement model, and solve the reasonable capacity measurement model to obtain the reasonable capacity measurement results.
[0046] Specifically, input the preprocessed multi-source data and the parameters of the measurement indicators into the reasonable capacity measurement model. Preferably, in this embodiment, the Python language is used for coding to solve and optimize the analysis. Considering the grid portrait index and the measurement indicators, different objective functions and constraint conditions are set, and multi-dimensional analysis and multi-objective balance of the reasonable capacity of the tobacco market are realized through collaborative optimization to find the optimal solution that meets all objectives.
[0047] In addition, for the case where there are persistent conflict solutions for possible constraints, it is necessary to adjust the constraint range during the solution process. The specific steps are as follows: Solve under each measurement indicator and constraint condition. If there is a persistent conflict solution for the constraint, retain all global constraints and remove the constraint of the current measurement indicator to solve; if a feasible solution is obtained, adjust back to the original constraint range. If there is still no feasible solution, continue to combine and remove non-global constraints until the first combined solution is found following the principle of minimum removal, then adjust back to the original constraint range and continue to obtain the weighted solution.
[0048] Exemplarily, referring to Table 1 and Figure 2 , the reasonable capacity measurement is carried out for a certain area through the reasonable capacity measurement model based on multi-objective optimization provided in this embodiment. The comparison chart of the measured reasonable capacities of all grids in this area and the existing number of grid retailers is as Figure 2 shown. Compared with the existing number, there are 18 grids with an increase in the measured reasonable capacity, 10 grids with no change, and 12 grids with a decrease. Among them, the trend of the grid capacity change is generally consistent with the analysis results of the grid portrait index and the measurement indicators. Individual grids are affected by the importance level between the index or indicators, resulting in some minor differences; Table 1 Existing Quantity and Measured Reasonable Capacity Table of a Certain Area Grid Name Existing Quantity Calculated Reasonable Capacity Trend Change Grid 1 42 42 Unchanged Grid 2 52 51 Decrease Grid 3 65 63 Increase Grid 4 19 21 Increase Grid 5 17 17 Unchanged Grid 6 51 51 Unchanged Grid 7 192 186 Increase Grid 8 60 65 Increase Grid 9 91 98 Increase Grid 10 38 42 Increase Grid 11 120 128 Increase Grid 12 41 43 Increase Grid 13 89 93 Increase Grid 14 77 77 Unchanged Grid 15 82 86 Increase Grid 16 105 110 Increase Grid 17 27 27 Unchanged Grid 18 89 89 Unchanged Grid 19 57 57 Unchanged Grid 20 112 112 Unchanged Grid 21 165 168 Decrease Grid 22 35 34 Increase Grid 23 17 19 Decrease Grid 24 135 134 Increase Grid 25 63 70 Decrease Grid 26 129 116 Increase Grid 27 140 141 Decrease Grid 28 188 171 Decrease Grid 29 222 211 Increase Grid 30 9 9 Unchanged Grid 31 35 38 Increase Grid 32 46 46 Unchanged Grid 33 81 84 Decrease Grid 34 55 54 Increase Grid 35 74 79 Decrease Grid 36 77 76 Decrease Grid 37 114 111 Decrease Grid 38 86 81 Decrease Grid 39 21 20 Decrease Grid 40 43 40 Increase Meanwhile, the current number of retailers in a region also reflects the consumption capacity and market size of that area. Only when there is market demand will new retailers be added to the region. Therefore, to a certain extent, the current number of regional retailers is reasonable, and the calculated reasonable capacity result should not differ significantly from the current number. The mean squared error MSE, mean absolute error MAE, and mean relative error MRE between the current number of regional retailers and the calculated reasonable capacity are statistically calculated to evaluate the calculated reasonable capacity result. The calculation formulas are as follows:
[0049]
[0050]
[0051] Among them, k is the number of grids, is the number of retailers in the j-th grid, i.e., the current number, is the calculated reasonable capacity of the j-th grid; through calculation, the result is rounded to two decimal places. Exemplarily, the MAE of this region is 4.28, the MSE is 46.43, and the MRE is 7.28%, indicating that the difference between the calculated reasonable capacity and the current number in this region is small.
[0052] Furthermore, the Shapiro-Wilk normality test is performed on the two sets of data of the current number and the calculated reasonable capacity, and a QQ plot (Quantile-Quantile Plot) is drawn as Figure 3 shown, and the test results are shown in the following table: Table 2 Shapiro-Wilk Normality Test Table Statistic P - value Existing Quantity 0.9218103885650635 0.008783689700067043 Calculated Reasonable Capacity 0.9363895058631897 0.02616889216005802 As can be seen from the above table, the P-values of the two sets of test results are less than the significance level of 0.05, and the null hypothesis is rejected, that is, both the current number and the calculated reasonable capacity do not follow a normal distribution. Therefore, the Wilcoxon rank sum test is further used to evaluate whether there is a significant difference between the two sets of data. Finally, the obtained test result P-value is 0.4271, which is greater than the significance level of 0.05, indicating that there is no significant difference between the two sets of data.
[0053] In summary, the reasonable capacity calculation model provided in this embodiment can comprehensively balance multiple indicators to calculate the reasonable capacity of the tobacco market, and the calculation result conforms to the actual situation and there is no significant difference from the current number.
[0054] A method for calculating the reasonable capacity of the tobacco market based on multi-objective optimization provided in this embodiment comprehensively considers multiple dimensions such as monopoly control, consumer convenience, retailer profitability, and marketing management. By analyzing and calculating six key indicators, namely the regulation coefficient, the per capita license holding rate, the average walking distance of consumers, the average monthly gross profit per household in the grid, the average gross profit rate of the grid, and the average annual sales volume of the grid, a reasonable capacity calculation model based on multi-objective optimization is constructed, realizing multi-dimensional analysis and multi-objective balance of the reasonable capacity of the tobacco market, which is beneficial to providing scientific and comprehensive decision-making support for tobacco market management. Moreover, the model provided in this embodiment is convenient for continuously introducing and integrating more real-time data and new data sources to continuously enhance the dynamic response ability and accuracy of the model, and can continuously introduce more social and economic factors and market behavior data to continuously enrich the analysis dimensions of the model.
[0055] Reference Figure 4 , another embodiment of the present invention provides a device 200 for calculating the reasonable capacity of the tobacco market based on multi-objective optimization, including: The first module 201 is configured to collect multi-source data including internal data and external data and perform preprocessing. The internal data includes retailer information, online order sales and inventory data, and grid supervision data, and the external data includes population and economic data, POI data, AOI data, and grid feature data; The second module 202 is configured to calculate measurement indicators according to the preprocessed multi-source data; The third module 203 is configured to construct a reasonable capacity calculation model based on multi-objective optimization according to the measurement indicators and assign different weights to the measurement indicators through importance ranking; The fourth module 204 is configured to input the preprocessed multi-source data and the parameters of the measurement indicators into the reasonable capacity calculation model, and solve the reasonable capacity calculation model to obtain a reasonable capacity calculation result.
[0056] It should be noted that the device 200 for calculating the reasonable capacity of the tobacco market based on multi-objective optimization provided in this embodiment corresponds to the technical solutions that can be used to execute the method embodiments. Its implementation principle and technical effects are similar to those of the method, and will not be elaborated here.
[0057] The above description is only a preferred embodiment of the present invention. Those skilled in the art should understand that the scope of disclosure involved in the present invention is not limited to the technical solutions formed by the specific combination of the above technical features, but also covers other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features with similar functions disclosed in the present invention.
Claims
1. A method for calculating the reasonable capacity of tobacco market based on multi-objective optimization, characterized in that: include: Collect and pre-process multi-source data including internal data and external data, wherein the internal data includes retail information, online ordering and sales data, and grid supervision data, and the external data includes population and economic data, POI data, AOI data, and grid feature data; Calculate the measurement index according to the preprocessed multi-source data; According to the measurement indicators, a reasonable capacity measurement model based on multi-objective optimization is constructed, and different weights are assigned to the measurement indicators by ranking them according to their importance; The preprocessed multi-source data and the parameters of the calculation index are input into the reasonable capacity calculation model, and the reasonable capacity calculation model is solved to obtain a reasonable capacity calculation result.
2. The method for calculating the reasonable capacity of tobacco market based on multi-objective optimization according to claim 1 is characterized in that: The step of inputting the pre-processed multi-source data and the parameters of the calculation index into the reasonable capacity calculation model, and solving the reasonable capacity calculation model to obtain a reasonable capacity calculation result comprises: The objective function of establishing a reasonable capacity estimation model is: ， in , ; is the i-th measurement indicator of the j-th grid, is the calculated reasonable capacity of the jth grid, n is the number of calculated indicators, is the expected value of the i-th measurement indicator of the j-th grid, is the weight of the i-th measurement indicator of the j-th grid, ; According to the objective function, the reasonable capacity calculation result of each grid is obtained. .
3. The method for calculating the reasonable capacity of the tobacco market based on multi-objective optimization according to claim 2 is characterized in that: Including control coefficient calculation indicators , calculated by the following formula: in, is the prediction result of the mth decision tree for the reasonable capacity of the jth grid, is the h-th feature of the j-th grid, is the weight of the h-th feature of the j-th grid, is the feature number, is the number of decision trees, is the fitting capacity of the j-th grid, is the number of grid retailers in the j-th grid.
4. The method for calculating the reasonable capacity of the tobacco market based on multi-objective optimization according to claim 3 is characterized by: Including per capita certification rate calculation indicators , calculated by the following formula: in, is the number of grid retailers in the jth grid, is the grid population of the j-th grid.
5. The method for calculating the reasonable capacity of the tobacco market based on multi-objective optimization according to claim 4 is characterized in that: Including the average walking distance measurement index for consumers , calculated by the following formula: in, is the grid area of the jth grid, is the average walking distance from the consumer’s residence to the retail store in the j-th grid.
6. The method for calculating the reasonable capacity of tobacco market based on multi-objective optimization according to claim 5 is characterized in that: Also includes: Including grid annual average sales volume calculation indicators , calculated by the following formula: in, is the annual sales volume of the j-th grid, is the number of grid retailers in the j-th grid.
7. The method for calculating the reasonable capacity of tobacco market based on multi-objective optimization according to claim 6 is characterized in that: Also includes: Including grid monthly average gross profit calculation indicators , calculated by the following formula: in, is the monthly retail price of the j-th grid, is the monthly wholesale price of the j-th grid, is the number of grid retailers in the j-th grid.
8. The method for calculating the reasonable capacity of tobacco market based on multi-objective optimization according to claim 7 is characterized in that: Also includes: Including grid average gross profit margin calculation indicators , calculated by the following formula: is the monthly retail price of the j-th grid, is the monthly average gross profit per household of the j-th grid, is the number of grid retailers in the j-th grid.
9. The method for calculating the reasonable capacity of tobacco market based on multi-objective optimization according to claim 1, characterized in that: Also includes: The mean square error is calculated by , mean absolute error and the mean relative error : in, is the mean square error, is the mean absolute error, is the average relative error, k is the number of grids, is the number of grid retailers in the jth grid, Calculate the reasonable capacity of the jth grid; The mean square error obtained by calculation , mean absolute error and the mean relative error , evaluate the reasonable capacity calculation results.
10. A device for calculating the reasonable capacity of tobacco market based on multi-objective optimization, characterized in that: include: The first module is configured to collect and pre-process multi-source data including internal data and external data, wherein the internal data includes retail information, online order and sales data, and grid supervision data, and the external data includes population and economic data, POI data, AOI data, and grid feature data; A second module is configured to calculate the measurement index according to the preprocessed multi-source data; The third module is configured to construct a reasonable capacity measurement model based on multi-objective optimization according to the measurement indicators, and assign different weights to the measurement indicators by ranking them according to their importance; The fourth module is configured to input the pre-processed multi-source data and the parameters of the measurement index into the reasonable capacity measurement model, and solve the reasonable capacity measurement model to obtain a reasonable capacity measurement result.