Intelligent analysis and processing method and system for commercial data of ultra-large international hub airport

Abstract

The invention relates to an intelligent analysis and processing method and system for commercial data of an ultra-large international hub airport, and the method comprises the steps: obtaining short-time sales data from a commercial data platform, and outputting a short-time sales result through model analysis; acquiring operation data from the commercial data, outputting the position of the commercial resource in the matrix through model analysis, and outputting a periodic operation result by comparing the position change of the same commercial resource in the matrix and the change trend of each index in two adjacent specific periods; obtaining sales data and market attention data of the brand from the commercial data, and outputting a brand adaptation degree through model analysis; and performing rent value evaluation on the commercial resources, and outputting an optimal distribution scheme of the commercial resources in combination with a short-time sales result, a periodic operation result and a brand adaptation degree of the commercial resources. According to the method, the business data processing capability, decision-making efficiency and resource configuration optimization level of the airport are greatly improved, and the overall operation benefit of the airport is promoted to be remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing and analysis, and in particular to a method and system for intelligent analysis and processing of commercial data of a super-large international hub airport. Background Art

[0002] With the rapid development of the global aviation industry, super-large international hub airports have become key nodes connecting all parts of the world. These airports not only carry heavy air traffic tasks, but also involve complex business operations and management activities. In daily operations, these airports need to process massive amounts of business data, which comes from a variety of systems, such as point-of-sale systems, flight information systems, and customer relationship management systems. Each type of data contains rich business information and operational insights. For airport managers, effectively integrating and analyzing this data is an important basis for improving business operation efficiency, accurately allocating resources, and enhancing customer satisfaction.

[0003] However, traditional data processing and analysis methods are unable to cope with such large and complex data sets. The complexity of data source access, the inefficiency of data processing, and the inaccuracy of analysis results have become the main problems hindering the optimization of airport commercial operations. These problems not only affect the ability of real-time data analysis and decision support, but may also lead to delays and errors in commercial operation decisions, ultimately having a negative impact on the overall operational benefits of the airport. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method and system for intelligent analysis and processing of commercial data of a super-large international hub airport, which solves the technical problem that traditional data processing methods cannot meet the needs of super-large international hub airports for efficient integration and accurate analysis of massive and diversified data sources.

[0006] (II) Technical solution

[0007] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0008] In a first aspect, an embodiment of the present invention provides a method for intelligent analysis and processing of commercial data of a super-large international hub airport, comprising:

[0009] Automatically identify and access airport business data from different data sources, perform vertical and / or horizontal fusion processing on the accessed airport business data and integrate them into a data platform to form a business data middle platform;

[0010] Obtain short-term sales data of business resources within a specific time period from the business data center, analyze the short-term sales data through the pre-built business dynamic early warning model, and output the short-term sales results;

[0011] Obtain the business data of business resources in two adjacent specific periods from the business data center, input at least one indicator value in the business resource business data into a pre-built three-dimensional matrix model, output the position of the business resource in the matrix, and output the periodic business results by comparing the position change of the same business resource in the matrix and the change trend of each indicator in two adjacent specific periods;

[0012] Obtain sales data and market attention data of the brands corresponding to the commercial resources in a specific time period from the business data center, analyze the sales data and market attention data of the brands through the pre-built brand fit model, and output the brand fit;

[0013] The rental value of commercial resources is evaluated through at least one calculation model, and the optimal allocation plan for each commercial resource is output based on the short-term sales results, periodic operating results and brand adaptability of the commercial resources.

[0014] Optionally, the airport business data from different data sources are automatically identified and accessed, and the accessed airport business data is fused and processed including the vertical dimension and / or the horizontal dimension and integrated into a data platform to form a business data middle platform including:

[0015] Introducing pre-trained natural language processing models to automatically identify and parse business data from different data sources and automatically fill in access configuration forms;

[0016] For each data in the access configuration form, the preset connector is used to automatically access it according to the order in which it is filled in;

[0017] Pre-process the incoming commercial data, including data cleaning, deduplication, and format conversion;

[0018] Classify and organize the pre-processed business data in the vertical dimension according to the hierarchical relationship determined by business attributes and business logic;

[0019] Find a reference data dimension or a common data element for the pre-processed business data or the business data processed in the vertical dimension, and classify and sort in the horizontal dimension based on the reference data dimension or the common data element;

[0020] Integrate the data that have been fused vertically and / or horizontally onto a unified data platform to form a business data middle platform.

[0021] Optionally, the short-term sales data of business resources in a specific time period is obtained from the business data center, and the short-term sales data is analyzed by a pre-built business dynamic early warning model, and the output short-term sales results include:

[0022] Obtain short-term sales data of specific business resources within a specific time period from the business data center;

[0023] Build a dynamic early warning model for business operations, and use historical sales data obtained from the business data center to estimate parameters including dynamic mean, dynamic standard deviation, and adaptive coefficient;

[0024] According to the dynamic average, dynamic standard deviation and adaptive coefficient, the upper and lower rails are dynamically calculated as adaptive warning thresholds;

[0025] Input short-term sales data within a specific time period into the business dynamic early warning model to generate early warning information including business resources that trigger the early warning, indicators that trigger the early warning, early warning direction, magnitude of exceeding the threshold, and trigger time;

[0026] Generate short-term sales results through a visual interface or report based on early warning information;

[0027] As new short-term sales data are continuously input into the business dynamics early warning model, the store business dynamics early warning model is continuously iterated and updated.

[0028] Optionally, a running average is:

[0029]

[0030] In the formula, x i is the sales data at time point i, ω i is the data point x i The weight of , n is the moving average window size;

[0031] The dynamic standard deviation is:

[0032]

[0033] The adaptive coefficient is;

[0034] α=f(h d );

[0035] In the formula, α is based on historical sales data h d The learned adaptive coefficient is used to adjust the sensitivity of the warning threshold;

[0036] The warning thresholds are:

[0037]

[0038] Where UT , L T They are the upper and lower rails respectively. If the value of the lower rail is lower than 0, the lower rail is L T =max(0, D A -α·D S ).

[0039] Optionally, the business data of the business resources in two adjacent specific periods are obtained from the business data center, at least one indicator value in the business resource business data is input into a pre-built three-dimensional matrix model, and the position of the business resource in the matrix is ​​output. By comparing the position change of the same business resource in the matrix and the change trend of each indicator in two adjacent specific periods, the output period operation results include:

[0040] Obtain the business data of business resources within two adjacent specific periods from the business data center;

[0041] According to the characteristics of business data, we select key evaluation dimensions including sales capability, profitability and customer satisfaction, build a three-dimensional matrix model, and determine the value range and scale of each dimension;

[0042] Selecting at least one key indicator from the business resource's operating data, and inputting the value of the selected indicator into the three-dimensional matrix model;

[0043] Compare the direction, distance and speed of the migration of the same business resource in the matrix in two adjacent specific periods;

[0044] For each key indicator, analyze the changing trend of growth rate and decline rate in two adjacent specific periods;

[0045] Based on the position migration of business resources in the matrix and the changing trends of indicators, the periodic operating results are output through a visual interface or report form.

[0046] Optionally, the three-dimensional matrix model is:

[0047] f(S,P,C)=(g1(S),g2(P),g3(C));

[0048] g1(S)=a·S 2 +b·S+c;

[0049] g2(P)=d·e (f·P) ;

[0050] g3(C)=g·log(h·C+i);

[0051] Where g1(S) is the position coordinate on the dimension of sales ability S, g2(P) is the position coordinate on the dimension of profitability P, g3(C) is the position coordinate on the dimension of customer satisfaction C, a, b and c are the coefficients of the polynomial, d and f are the parameters of the exponential function, and g, h and i are the parameters of the logarithmic function.

[0052] Optionally, the sales data and market attention data of the brand corresponding to the commercial resources in a specific time period are obtained from the commercial data middle platform, and the sales data and market attention data of the brand are analyzed by a pre-built brand suitability model, and the brand suitability output includes;

[0053] Obtain sales data and market attention data of brands corresponding to commercial resources in a specific time period from the commercial data center;

[0054] The acquired sales data is used to fill in the missing time series data through interpolation or extrapolation;

[0055] The market attention data is analyzed by sentiment analysis technology on external crawled text data to extract brand-related sentiment tendency features and add them to the market attention data;

[0056] Normalize the filled sales data and the market attention data added with new data and convert them to the same level;

[0057] Build a brand fit model;

[0058] The normalized data is input into the brand fit model for comparative analysis to calculate the fit score of each brand.

[0059] Optionally, normalize the formula:

[0060] X n =(XX min ) / (X max -X min );

[0061] In the formula, X is the original data, X min , X max are the minimum and maximum values ​​of the data, respectively, n is the normalized data;

[0062] The brand fit model is:

[0063]

[0064] In the formula, represents sales, M represents market attention, S represents max and M maxare the maximum values ​​of sales and market attention respectively. α, β, γ, δ, and ε are all model parameters, which are determined by regression analysis, machine learning algorithms or expert scoring.

[0065] Optionally, the rental value of commercial resources is evaluated through at least one calculation model, and the optimal allocation plan for each commercial resource is outputted in combination with the short-term sales results, periodic operation results and brand suitability of the commercial resources, including:

[0066] Obtain historical rental data of commercial resources from the commercial data center;

[0067] Select at least one calculation model suitable for current business resource evaluation from the calculation models including investment income method, rent comparison method and turnover reverse calculation method;

[0068] Based on the selected calculation model, the rental value of commercial resources is evaluated, and the rental value range of commercial resources is determined according to the evaluation results;

[0069] Based on the rental value range, short-term sales results, periodic operating results and brand suitability of commercial resources, formulate the optimal allocation plan for each commercial resource, including rental pricing, lease term and brand matching.

[0070] In a second aspect, an embodiment of the present invention provides a system for intelligent analysis and processing of commercial data of a super-large international hub airport, comprising:

[0071] A commercial data middle platform formation module is used to automatically identify and access airport commercial data from different data sources, perform fusion processing on the accessed airport commercial data including vertical dimensions and / or horizontal dimensions and integrate them into a data platform to form a commercial data middle platform;

[0072] The short-term sales result output module is used to obtain the short-term sales data of business resources within a specific time period from the business data center, analyze the short-term sales data through the pre-built business dynamic early warning model, and output the short-term sales results;

[0073] The cycle operation result output module is used to obtain the operation data of the business resources in two adjacent specific cycles from the business data center, input at least one indicator value in the operation data of the business resources into a pre-built three-dimensional matrix model, output the position of the business resources in the matrix, and output the cycle operation results by comparing the position change of the same business resource in the matrix and the change trend of each indicator in two adjacent specific cycles;

[0074] The brand fit output module is used to obtain the sales data and market attention data of the brand corresponding to the commercial resources in a specific time period from the commercial data center, analyze the sales data and market attention data of the brand through the pre-built brand fit model, and output the brand fit;

[0075] The commercial resource allocation module is used to evaluate the rental value of commercial resources through at least one measurement model, and output the optimal allocation plan for each commercial resource based on the short-term sales results, periodic operating results and brand adaptability of the commercial resources.

[0076] (III) Beneficial effects

[0077] The beneficial effects of the present invention are:

[0078] The present invention can automatically identify and access airport business data from different data sources, and integrate these data into a unified data platform through efficient data fusion and processing technology to form a business data middle platform. This innovative data processing architecture not only simplifies the data access and integration process, but also greatly improves the efficiency and accuracy of data processing.

[0079] On this basis, the present invention further uses advanced mathematical models and algorithms to conduct in-depth analysis of commercial data. Through the comprehensive use of multiple models such as business dynamic early warning model, three-dimensional matrix model and brand adaptation model, it can comprehensively and deeply mine the valuable information in commercial data and provide scientific and accurate decision support for airport commercial operations.

[0080] Therefore, the present invention will greatly enhance the airport's commercial data processing capabilities, decision-making efficiency and resource allocation optimization level, thereby promoting a significant improvement in the airport's overall operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 A schematic diagram of a process for intelligent analysis and processing of commercial data of a super-large international hub airport provided by an embodiment of the present invention;

[0082] Figure 2 A specific flow chart of step S1 of a method for intelligent analysis and processing of commercial data of a super-large international hub airport provided by an embodiment of the present invention;

[0083] Figure 3 A diagram showing a longitudinal dimension of a method for intelligent analysis and processing of commercial data of a super-large international hub airport provided by an embodiment of the present invention;

[0084] Figure 4 A display interface of a data center for a method of intelligent analysis and processing of commercial data of a super-large international hub airport provided by an embodiment of the present invention;

[0085] Figure 5 A specific flow chart of step S2 of a method for intelligent analysis and processing of commercial data of a super-large international hub airport provided by an embodiment of the present invention;

[0086] Figure 6 A specific flow chart of step S3 of a method for intelligent analysis and processing of commercial data of a super-large international hub airport provided by an embodiment of the present invention;

[0087] Figure 7 A specific flow chart of step S4 of a method for intelligent analysis and processing of commercial data of a super-large international hub airport provided by an embodiment of the present invention;

[0088] Figure 8 A specific flow chart of step S5 of a method for intelligent analysis and processing of commercial data of a super-large international hub airport provided by an embodiment of the present invention;

[0089] Fig. 9 A detailed information list diagram of rental value assessment of a super-large international hub airport commercial data intelligent analysis and processing method provided by an embodiment of the present invention;

[0090] Fig.10 A heat map and information summary table of the commercial resource rental value of a super-large international hub airport commercial data intelligent analysis and processing method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0091] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0092] like Figure 1As shown, an embodiment of the present invention proposes a method for intelligent analysis and processing of commercial data of a super-large international hub airport, including: automatically identifying and accessing airport commercial data from different data sources, performing fusion processing including vertical dimensions and / or horizontal dimensions on the accessed airport commercial data and integrating it onto a data platform to form a commercial data middle station; obtaining short-term sales data of commercial resources within a specific time period from the commercial data middle station, analyzing the short-term sales data through a pre-built business dynamic early warning model, and outputting short-term sales results; obtaining business data of commercial resources within two adjacent specific periods from the commercial data middle station, and inputting at least one indicator value in the business data of the commercial resources into In the pre-constructed three-dimensional matrix model, the position of the business resources in the matrix is ​​output, and the periodic operation results are output by comparing the position changes of the same business resources in the matrix and the changing trends of each indicator in two adjacent specific periods; the sales data and market attention data of the brand corresponding to the business resources in a specific time period are obtained from the business data middle platform, and the brand sales data and market attention data are analyzed through the pre-constructed brand fitness model to output the brand fitness; the rental value of business resources is evaluated through at least one measurement model, and the optimal allocation plan for each business resource is output based on the short-term sales results, periodic operation results and brand fitness of the business resources.

[0093] The present invention can automatically identify and access airport business data from different data sources, and integrate these data into a unified data platform through efficient data fusion and processing technology to form a business data middle platform. This innovative data processing architecture not only simplifies the data access and integration process, but also greatly improves the efficiency and accuracy of data processing.

[0094] On this basis, the present invention further uses advanced mathematical models and algorithms to conduct in-depth analysis of commercial data. Through the comprehensive use of multiple models such as business dynamic early warning model, three-dimensional matrix model and brand adaptation model, it can comprehensively and deeply mine the valuable information in commercial data and provide scientific and accurate decision support for airport commercial operations.

[0095] Therefore, the present invention will greatly enhance the airport's commercial data processing capabilities, decision-making efficiency and resource allocation optimization level, thereby promoting a significant improvement in the airport's overall operating efficiency.

[0096] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0097] Specifically, an embodiment of the present invention provides a method and system for intelligent analysis and processing of commercial data of a super-large international hub airport, which includes:

[0098] S1. Automatically identify and access airport business data from different data sources, perform fusion processing on the accessed airport business data including vertical dimensions and / or horizontal dimensions and integrate them into a data platform to form a business data middle platform.

[0099] Furthermore, if Figure 2 As shown, step S1 includes:

[0100] S11. Introduce a pre-trained natural language processing model to automatically identify and parse business data from different data sources and automatically fill in the access configuration form.

[0101] The present invention collects business data samples from different data sources, including CSV files, database export files, API responses, etc. The data is preprocessed, such as cleaning and formatting, for subsequent model training. Among them, the business data samples collected from different data sources include CSV files, database export files, API responses, etc.

[0102] Next, preprocess the data, such as cleaning and formatting, for subsequent model training, and input the preprocessed data into the fine-tuned model. The model automatically identifies the structure and content of the data, such as table column names, data types, etc. The model further parses the data and extracts key information, such as store name, transaction amount, date, etc. Next, based on the key information parsed by the model, automatically fill in the data access configuration form. The form content includes data source type, access method, data format, field mapping, etc. Ensure that all necessary information is filled in correctly for subsequent data access and analysis.

[0103] S12. The data in the access configuration form are automatically accessed using a preset connector in the order in which they are filled in.

[0104] S13. Perform pre-processing on the received commercial data, including data cleaning, deduplication, and format conversion.

[0105] S14. Classify and organize the pre-processed business data in the vertical dimension according to the hierarchical relationship determined by business attributes and business logic.

[0106] Further, analyze the preprocessed business data to understand the business attributes and logic behind it. According to the business logic, divide the data into different vertical dimensions, refer to Figure 3, such as stores, orders, products, etc. Within each vertical dimension, further subdivide the data type to ensure that the hierarchical relationship of the data is clear. Set clear rules and standards for each data type to ensure the consistency and accuracy of the data.

[0107] S15. Find a reference data dimension or a common data element for the pre-processed business data or the business data processed in the vertical dimension, and perform horizontal dimension classification and arrangement based on the reference data dimension or the common data element.

[0108] In this step, common benchmark data dimensions or elements are identified from the pre-processed or longitudinally processed data. Based on the benchmark dimensions, the data is horizontally classified. For example, all time-related data can be classified into one category, and all location-related data can be classified into another category. In each horizontal classification, the data type is further sorted and improved to ensure the integrity and consistency of the data. Design data linkage, that is, establish the relationship between different types of data to achieve vertical interconnection and interaction of data.

[0109] S16. Integrate the data processed by longitudinal and / or transverse latitude fusion into a unified data platform to form Figure 4 The business data middle platform is shown. The business data middle platform is the core and foundation of the solution of the present invention. It is an integrated platform that integrates business data management, governance, and application. It aims to provide stable and efficient data support for business operations and management. Its core is to integrate all data resources and data applications into one platform to achieve centralized management and application of business data, thereby avoiding the duplication of various data centers, reducing resource waste and duplication of work, and improving data management efficiency and quality.

[0110] S2. Obtain the short-term sales data of business resources within a specific time period from the business data center, analyze the short-term sales data through the pre-built business dynamics early warning model, and output the short-term sales results.

[0111] Furthermore, if Figure 5 As shown, step S2 includes:

[0112] S21. Obtain short-term sales data of specific business resources within a specific time period from the business data middle platform; wherein, short-term sales data includes but is not limited to key operating indicators such as sales volume, number of sales, average order value, sales volume and sales efficiency per square meter.

[0113] S22. Build a dynamic early warning model for operations, and use the historical sales data obtained from the business data center to estimate parameters including dynamic mean, dynamic standard deviation, and adaptive coefficient.

[0114] S23. According to the dynamic average value, the dynamic standard deviation and the adaptive coefficient, the upper track line and the lower track line are dynamically calculated as the adaptive warning threshold.

[0115] In a specific embodiment, the dynamic average value is:

[0116]

[0117] In the formula, x i is the sales data at time point i, ω i is the data point x i The weight of , n is the moving average window size;

[0118] The dynamic standard deviation is:

[0119]

[0120] The adaptive coefficient is;

[0121] α=f(h d );

[0122] In the formula, α is based on historical sales data h d The learned adaptive coefficient is used to adjust the sensitivity of the warning threshold;

[0123] The warning thresholds are:

[0124]

[0125] Where U T , L T They are the upper and lower rails respectively. If the value of the lower rail is lower than 0, the lower rail is L T =max(0, D A -α·D S ).

[0126] Furthermore, the warning threshold meets the following requirements: middle line = 7-day moving average, upper line = middle line + 2 times standard deviation, lower line = middle line - 2 times standard deviation. If the lower line is lower than 0, the lower line = middle line - 1 times standard deviation; if the lower line is still lower than 0, the lower line = 0.

[0127] S24. Input the short-term sales data within a specific time period into the business dynamic early warning model to generate early warning information including the business resources that trigger the early warning, the indicators that trigger the early warning, the early warning direction, the magnitude of exceeding the threshold, and the triggering time. Real-time monitoring of the key business indicators of each store. Once the early warning is triggered online or offline, the system will respond immediately, such as changing the color of the indicator line (from green to red).

[0128] S25. Generate short-term sales results through a visual interface or report based on the early warning information;

[0129] S26. With the continuous input of new short-term sales data into the business dynamics early warning model, the store business dynamics early warning model is continuously iterated and updated.

[0130] S3. Obtain the operating data of the business resources in two adjacent specific periods from the business data center, input at least one indicator value in the operating data of the business resources into a pre-built three-dimensional matrix model, output the position of the business resources in the matrix, and output the periodic operating results by comparing the position changes of the same business resource in the matrix and the changing trends of each indicator in two adjacent specific periods.

[0131] Furthermore, if Figure 6 As shown, step S3 includes:

[0132] S31. Obtain the business data of business resources in two adjacent specific periods from the business data center. Determine two adjacent analysis periods, such as the previous period and the current period, or the same period last year and this year.

[0133] S32. According to the characteristics of business data, select key evaluation dimensions including sales ability, profitability and customer satisfaction, build a three-dimensional matrix model, and determine the value range and scale of each dimension.

[0134] In a specific embodiment, the three-dimensional matrix model is:

[0135] f(S,P,C)=(g1(S),g2(P),g3(C));

[0136] g1(S)=a·S 2 +b·S+c;

[0137] g2(P)=d·e (f·P) ;

[0138] g3(C)=g·log(h·C+i);

[0139] In the formula, g1(S) is the position coordinate on the sales ability dimension S, g2(P) is the position coordinate on the profitability dimension P, and g3(C) is the position coordinate on the customer satisfaction dimension C.

[0140] It is important to understand that a, b, and c are the coefficients of the polynomial, which can be determined by fitting the data. For example, you can choose a = 0.01, b = 0.5, c = 10, so that when S = 0, g1(S) = 10; when S = 100, g1(S) = 110, assuming that this is a reasonable range on the x-axis.

[0141] d and f are the parameters of the exponential function. Assuming that the value range of profitability P is also between 0 and 100, d = 1 and f = 0.01 can be selected. In this way, when P = 0, g2(P) = 1; when P = 100, g2(P) ≈ 2.72, assuming that this is a reasonable range on the y-axis.

[0142] g, h, and i are the parameters of the logarithmic function. For customer satisfaction C, a logarithmic function can be used to map to the position coordinates on the z-axis, and g, h, and i are the parameters of the logarithmic function. Assuming that the value range of customer satisfaction C is between 0 and 100, g = 10, h = 0.1, and i = 1 can be selected. In this way, when C = 0, g3(C) = 0 (it is necessary to deal with the domain problem of the logarithmic function, and it is assumed here that C will not be 0); when C = 100, g3(C) = 10*log(11)≈10.45, assuming that this is a reasonable range on the z-axis.

[0143] In addition, for the logarithmic function g3(C), there is a domain problem when C=0. In practical applications, the case of log(0) can be avoided by adding a small positive number, or other methods can be used to handle this special case.

[0144] S33. Select at least one key indicator from the business resource's operating data, and input the value of the selected indicator into the three-dimensional matrix model. Generally speaking, nine key composite indicators are selected, and these indicators should cover multiple aspects such as sales capability, operational efficiency, financial status, and customer satisfaction.

[0145] S34. Compare the direction, distance and speed of the migration of the position of the same business resource in the matrix in two adjacent specific periods.

[0146] Compare the positions of the same business resource in the matrix in two adjacent specific periods to observe whether it has migrated. Record the direction, distance and speed of migration to reflect the changes in the business resource's operating efficiency.

[0147] The migration direction can be determined by calculating the difference between the position vectors in two cycles. For example, the position of business resource A in the first cycle is (X1, Y1, Z1) and the position in the second cycle is (X2, Y2, Z2). Then the migration vector is (ΔX, ΔY, ΔZ) = (X2-X1, Y2-Y1, Z2-Z1).

[0148] The migration distance can be determined by calculating the modulus of the migration vector, i.e. The migration speed can be defined as the ratio of the migration distance to the time interval between two cycles.

[0149] S35. For each key indicator, analyze the changing trend of the growth rate and decline rate in two adjacent specific periods.

[0150] For each key indicator, analyze its changing trend in two adjacent specific periods. Calculate the growth rate, decline rate or other relevant statistics of the indicator to quantify the change of the indicator.

[0151] For example, if the sales of business resource A increases from S1 in the first cycle to S2 in the second cycle, the sales growth rate is (S2-S1) / S1*100%. Similarly, the trend of other indicators such as profit rate growth rate can be calculated.

[0152] S36. Based on the position migration of business resources in the matrix and the changing trends of indicators, the periodic operating results are output through a visual interface or report form.

[0153] S4. Obtain the sales data and market attention data of the brand corresponding to the commercial resources in a specific time period from the business data middle platform, analyze the brand sales data and market attention data through the pre-built brand fitness model, and output the brand fitness.

[0154] Furthermore, if Figure 7 As shown, step S4 includes:

[0155] S41. Obtain sales data and market attention data of the brand corresponding to the commercial resources within a specific time period from the commercial data middle platform.

[0156] Selectively extract brand sales data corresponding to commercial resources within a specific time period from the business data center. This step needs to ensure the accuracy and completeness of the data while filtering out any invalid or erroneous data points.

[0157] At the same time, obtain the market attention data of these brands in the same time period from the business data platform or other relevant data sources. Market attention data can be collected through various channels, including but not limited to social media interactions, search engine query volume, news report frequency, etc.

[0158] S42. Fill in the missing time series data with the acquired sales data through interpolation or extrapolation.

[0159] S43. Use sentiment analysis technology to perform sentiment analysis on externally crawled text data to extract brand-related sentiment tendency features and add them to the market attention data.

[0160] S44. Normalize the filled sales data and the market attention data with the new data added, and convert them to the same level.

[0161] Among them, the normalization formula is:

[0162] X n=(XX min ) / (X max -X min );

[0163] In the formula, X is the original data, X min , X max are the minimum and maximum values ​​of the data, respectively, n is the normalized data.

[0164] S45. Establish a brand fit model.

[0165] The brand fit model is:

[0166]

[0167] In the formula, represents sales, M represents market attention, S represents max and M max are the maximum values ​​of sales and market attention respectively. α, β, γ, δ, and ε are all model parameters, which are determined by regression analysis, machine learning algorithms or expert scoring.

[0168] Part I is the weighted average of sales and market attention, where α determines the relative importance of the two. An interaction term is introduced to account for the synergistic effect between sales and market attention. This interaction term increases the brand fit score when both are high.

[0169] S46. Input the normalized data into the brand fit model for comparative analysis and calculate the fit score of each brand.

[0170] S5. Evaluate the rental value of commercial resources through at least one calculation model, combine the short-term sales results, periodic operating results and brand adaptability of commercial resources, and output the optimal allocation plan for each commercial resource.

[0171] Furthermore, if Figure 8 As shown, step S5 includes:

[0172] S51. Obtain historical rental data of commercial resources from the commercial data center.

[0173] S52. Select at least one calculation model suitable for the current business resource evaluation from the calculation models including the investment income method, the rental comparison method and the turnover reverse calculation method.

[0174] In a specific embodiment, the rent comparison method is: find cases that are similar to the target commercial resources and have recent leasing transactions. Compare the similarities and differences between the cases and the target commercial resources in terms of geographical location, area, facilities, etc. Estimate the rental value of the target commercial resources based on the rental level and difference adjustment factor of similar cases; the investment income method is: predict the future income of the commercial resources, considering factors such as rental income, expense expenditure, and capitalization rate. Use an appropriate discount rate to discount future income to present value to derive the rental value of the commercial resources. And, the turnover reverse method: This is a method of reversely inferring the rental value of a commercial resource through its turnover. First, it is necessary to determine the turnover data of the target commercial resource, which is obtained through financial statements, sales records or market research. Next, analyze the potential relationship between turnover and rent, which may require considering factors such as industry averages, profit margins of commercial resources, and market competition.

[0175] Conduct a market-based rental value assessment of commercial resources within a certain period of time (1 month, 1 quarter, 6 months, 1 year) to provide decision-making references for resource optimization and adjustment and determining the base price for investment promotion. Fig. 9 As shown in Figure 1, it provides two presentation modes: intuitive visualization of resource rental value and detailed information list. Fig.10 , the rental value of commercial resources is evaluated through three models and compared with the actual rental level of the resources.

[0176] S53. Based on the selected calculation model, conduct rental value assessment on the commercial resources, and determine the rental value range of the commercial resources based on the assessment results.

[0177] S54. Based on the rental value range of commercial resources, short-term sales results, periodic operating results and brand adaptability of commercial resources, formulate the best allocation plan for each commercial resource, including rental pricing, lease term and brand matching.

[0178] In a specific embodiment, a multi-objective genetic algorithm is used to output the optimal allocation scheme:

[0179] (1) Randomly generate an initial population, where each individual represents a possible allocation scheme, including information on rental pricing, lease term, and brand pairing. Set the algorithm's parameters such as the number of iterations, crossover rate, and mutation rate.

[0180] (2) For each individual (allocation scheme), calculate its fitness value. The fitness function can comprehensively consider multiple objectives such as rental income, brand fitness and resource utilization efficiency. Rental income is calculated by multiplying the rental price and the lease term. Brand fitness can be obtained through market research or expert scoring. Resource utilization efficiency can be evaluated by the utilization of commercial resources (such as area utilization, sales, etc.).

[0181] (3) Use the non-dominated sorting and crowding comparison methods in multi-objective optimization to select excellent individuals to enter the next generation. Non-dominated sorting is used to determine the superiority relationship between individuals, while crowding comparison is used to maintain the diversity of the population.

[0182] (4) Randomly select two individuals to perform a crossover operation to generate new individuals. The crossover operation can be achieved by exchanging some genes (such as rental pricing, lease term, etc.). Perform a mutation operation on the newly generated individuals to increase the diversity of the population. The mutation operation can randomly change the values ​​of certain genes.

[0183] (5) When the preset number of iterations is reached or other termination conditions are met, the algorithm stops and outputs the optimal allocation plan, that is, the individual with the highest fitness.

[0184] In addition, an embodiment of the present invention provides a super-large international hub airport business data intelligent analysis and processing system, including:

[0185] The commercial data middle platform forms a module, which is used to automatically identify and access airport commercial data from different data sources, perform fusion processing on the accessed airport commercial data including vertical dimensions and / or horizontal dimensions and integrate them into a data platform to form a commercial data middle platform.

[0186] The short-term sales result output module is used to obtain the short-term sales data of business resources within a specific time period from the business data center, analyze the short-term sales data through the pre-built business dynamics early warning model, and output the short-term sales results.

[0187] The periodic operating result output module is used to obtain the operating data of business resources in two adjacent specific periods from the business data center, input at least one indicator value in the operating data of the business resources into a pre-built three-dimensional matrix model, output the position of the business resources in the matrix, and output the periodic operating results by comparing the position changes of the same business resource in the matrix and the change trend of each indicator in two adjacent specific periods.

[0188] The brand suitability output module is used to obtain the sales data and market attention data of the brand corresponding to the commercial resources in a specific time period from the commercial data center, analyze the brand's sales data and market attention data through the pre-built brand suitability model, and output the brand suitability.

[0189] The commercial resource allocation module is used to evaluate the rental value of commercial resources through at least one measurement model, and output the optimal allocation plan for each commercial resource based on the short-term sales results, periodic operating results and brand adaptability of the commercial resources.

[0190] Furthermore, an embodiment of the present invention provides a computer-readable medium having computer executable instructions stored thereon, and when the executable instructions are executed by a processor, the above-mentioned super-large international hub airport commercial data intelligent analysis and processing method is implemented. This computer-readable medium is equipped with a complete set of instructions, which can guide the processor to complete complex airport commercial data intelligent analysis tasks, thereby significantly improving the operating efficiency of super-large international hub airports.

[0191] In summary, the embodiments of the present invention provide a method and system for intelligent analysis and processing of commercial data of a super-large international hub airport, which has multiple advantageous functions:

[0192] First, the present invention can automatically identify and access airport business data from different data sources, effectively solving the problem of complex data source access. After fusion processing and integration into a data platform, a business data middle platform is formed, thereby achieving unified management and efficient use of data. This step not only simplifies the data processing process, but also improves the efficiency and accuracy of data processing.

[0193] Secondly, by obtaining short-term sales data from the business data center and analyzing it using the business dynamic early warning model, the method and system can quickly output short-term sales results. This provides airport managers with timely sales feedback, helping them make quick and accurate business decisions.

[0194] In addition, the present invention can also obtain business data from the business data center, input the key indicator values ​​in the business data into the three-dimensional matrix model, and then output the position of the business resources in the matrix. By comparing the changes in the position of the same business resource in the matrix and the change trend of each indicator in two adjacent specific periods, the business status of the business resources can be clearly understood, and the cycle business results can be output accordingly. This function helps airport managers to fully grasp the operation of business resources and provide strong support for future resource allocation.

[0195] At the same time, the present invention can also analyze the brand's sales data and market attention data through the brand suitability model and output the brand suitability. This indicator helps airport managers understand the performance of each brand in the market and provide a scientific basis for brand cooperation and selection.

[0196] Finally, the rental value of commercial resources is evaluated through the calculation model, and combined with the short-term sales results, periodic operation results and brand adaptability of commercial resources, the method and system can output the best allocation plan for each commercial resource. This not only optimizes resource allocation and improves resource utilization efficiency, but also brings greater economic benefits to the airport.

[0197] Therefore, the present invention provides airport managers with comprehensive, accurate and timely operational information through intelligent data processing and analysis, which helps them make more informed decisions and thus improve the airport's commercial operating efficiency and overall competitiveness.

[0198] Since the system / device described in the above embodiments of the present invention is a system / device used to implement the method of the above embodiments of the present invention, a person skilled in the art can understand the specific structure and deformation of the system / device based on the method described in the above embodiments of the present invention, and thus will not be described in detail here. All systems / devices used in the method of the above embodiments of the present invention belong to the scope of protection of the present invention.

[0199] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0200] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions.

[0201] It should be noted that in the claims, any reference numerals placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by the same hardware. The use of the words first, second, third, etc., is for convenience of expression only and does not indicate any order. These words may be understood as part of the component name.

[0202] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0203] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0204] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.

Claims

1. A method for intelligent analysis and processing of commercial data of a super-large international hub airport, characterized in that: include: Automatically identify and access airport business data from different data sources, perform vertical and / or horizontal fusion processing on the accessed airport business data and integrate them into a data platform to form a business data middle platform; Obtain short-term sales data of business resources within a specific time period from the business data center, analyze the short-term sales data through the pre-built business dynamic early warning model, and output the short-term sales results; Obtain the business data of business resources in two adjacent specific periods from the business data center, input at least one indicator value in the business resource business data into a pre-built three-dimensional matrix model, output the position of the business resource in the matrix, and output the periodic business results by comparing the position change of the same business resource in the matrix and the change trend of each indicator in two adjacent specific periods; Obtain sales data and market attention data of the brands corresponding to the commercial resources in a specific time period from the business data center, analyze the sales data and market attention data of the brands through the pre-built brand fit model, and output the brand fit; The rental value of commercial resources is evaluated through at least one calculation model, and the optimal allocation plan for each commercial resource is output based on the short-term sales results, periodic operating results and brand adaptability of the commercial resources.

2. The method for intelligent analysis and processing of commercial data of a super-large international hub airport according to claim 1, characterized in that: Automatically identify and access airport business data from different data sources, integrate the accessed airport business data into a data platform including vertical and / or horizontal dimensions, and form a business data middle platform including: Introducing pre-trained natural language processing models to automatically identify and parse business data from different data sources and automatically fill in access configuration forms; For each data in the access configuration form, the preset connector is used to automatically access it according to the order in which it is filled in; Pre-process the incoming commercial data, including data cleaning, deduplication, and format conversion; Classify and organize the pre-processed business data in the vertical dimension according to the hierarchical relationship determined by business attributes and business logic; Find a reference data dimension or a common data element for the pre-processed business data or the business data processed in the vertical dimension, and classify and sort in the horizontal dimension based on the reference data dimension or the common data element; Integrate the data that have been fused vertically and / or horizontally onto a unified data platform to form a business data middle platform.

3. The method for intelligent analysis and processing of commercial data of a super-large international hub airport according to claim 1, characterized in that: Obtain short-term sales data of business resources within a specific time period from the business data center, analyze the short-term sales data through the pre-built business dynamic early warning model, and output short-term sales results including: Obtain short-term sales data of specific business resources within a specific time period from the business data center; Build a dynamic early warning model for business operations, and use historical sales data obtained from the business data center to estimate parameters including dynamic mean, dynamic standard deviation, and adaptive coefficient; According to the dynamic average, dynamic standard deviation and adaptive coefficient, the upper and lower rails are dynamically calculated as adaptive warning thresholds; Input short-term sales data within a specific time period into the business dynamic early warning model to generate early warning information including business resources that trigger the early warning, indicators that trigger the early warning, early warning direction, magnitude of exceeding the threshold, and trigger time; Generate short-term sales results through a visual interface or report based on early warning information; As new short-term sales data are continuously input into the business dynamics early warning model, the store business dynamics early warning model is continuously iterated and updated.

4. The method for intelligent analysis and processing of commercial data of a super-large international hub airport according to claim 3, characterized in that: The moving average is: In the formula, x i is the sales data at time point i, ω i is the data point x i The weight of , n is the moving average window size; The dynamic standard deviation is: The adaptive coefficient is; α=f(h d ); In the formula, α is based on historical sales data h d The learned adaptive coefficient is used to adjust the sensitivity of the warning threshold; The warning thresholds are: Where U T , L T They are the upper and lower rails respectively. If the value of the lower rail is lower than 0, the lower rail is L T =max(0, D A -α·D S ).

5. The method for intelligent analysis and processing of commercial data of a super-large international hub airport according to claim 1, characterized in that: The business data of business resources in two adjacent specific periods are obtained from the business data center. At least one indicator value in the business resource business data is input into a pre-built three-dimensional matrix model, and the position of the business resource in the matrix is ​​output. By comparing the position change of the same business resource in the matrix and the change trend of each indicator in two adjacent specific periods, the output period business results include: Obtain the business data of business resources within two adjacent specific periods from the business data center; According to the characteristics of business data, we select key evaluation dimensions including sales capability, profitability and customer satisfaction, build a three-dimensional matrix model, and determine the value range and scale of each dimension; Selecting at least one key indicator from the business resource's operating data, and inputting the value of the selected indicator into the three-dimensional matrix model; Compare the direction, distance and speed of the migration of the same business resource in the matrix in two adjacent specific periods; For each key indicator, analyze the changing trend of growth rate and decline rate in two adjacent specific periods; Based on the position migration of business resources in the matrix and the changing trends of indicators, the periodic operating results are output through a visual interface or report form.

6. The method for intelligent analysis and processing of commercial data of a super-large international hub airport according to claim 5, characterized in that: The three-dimensional matrix model is: f(S,P,C)=(g1(S),g2(P),g3(C)); g1(S)=a·S 2 +b·S+c; g2(P)=d·e (f·P) ; g3(C)=g·log(h·C+i); Where g1(S) is the position coordinate on the dimension of sales ability S, g2(P) is the position coordinate on the dimension of profitability P, g3(C) is the position coordinate on the dimension of customer satisfaction C, a, b and c are the coefficients of the polynomial, d and f are the parameters of the exponential function, and g, h and i are the parameters of the logarithmic function.

7. The method for intelligent analysis and processing of commercial data of a super-large international hub airport according to claim 1, characterized in that: Obtain sales data and market attention data of the brands corresponding to the commercial resources in a specific time period from the business data center, analyze the sales data and market attention data of the brands through the pre-built brand fit model, and output the brand fit including; Obtain sales data and market attention data of brands corresponding to commercial resources in a specific time period from the commercial data center; The acquired sales data is used to fill in the missing time series data through interpolation or extrapolation; The market attention data is analyzed by sentiment analysis technology on external crawled text data to extract brand-related sentiment tendency features and add them to the market attention data; Normalize the filled sales data and the market attention data added with new data and convert them to the same level; Build a brand fit model; The normalized data is input into the brand fit model for comparative analysis to calculate the fit score of each brand.

8. The method for intelligent analysis and processing of commercial data of a super-large international hub airport according to claim 7, characterized in that: Normalization formula: X n =(X-X min ) / (X max -X min ); In the formula, X is the original data, X min , X max are the minimum and maximum values ​​of the data, respectively, n is the normalized data; The brand fit model is: In the formula, represents sales, M represents market attention, S represents max and M max are the maximum values ​​of sales and market attention respectively. α, β, γ, δ, and ε are all model parameters, which are determined by regression analysis, machine learning algorithms or expert scoring.

9. The method for intelligent analysis and processing of commercial data of a super-large international hub airport according to claim 1, characterized in that: The rental value of commercial resources is evaluated through at least one calculation model. The optimal allocation plan for each commercial resource is output based on the short-term sales results, periodic operating results and brand suitability of the commercial resources, including: Obtain historical rental data of commercial resources from the commercial data center; Select at least one calculation model suitable for current business resource evaluation from the calculation models including investment income method, rent comparison method and turnover reverse calculation method; Based on the selected calculation model, the rental value of commercial resources is evaluated, and the rental value range of commercial resources is determined according to the evaluation results; Based on the rental value range, short-term sales results, periodic operating results and brand suitability of commercial resources, formulate the optimal allocation plan for each commercial resource, including rental pricing, lease term and brand matching.

10. A super-large international hub airport business data intelligent analysis and processing system, characterized in that: include: A commercial data middle platform formation module is used to automatically identify and access airport commercial data from different data sources, perform fusion processing on the accessed airport commercial data including vertical dimensions and / or horizontal dimensions and integrate them into a data platform to form a commercial data middle platform; The short-term sales result output module is used to obtain the short-term sales data of business resources within a specific time period from the business data center, analyze the short-term sales data through the pre-built business dynamic early warning model, and output the short-term sales results; The cycle operation result output module is used to obtain the operation data of the business resources in two adjacent specific cycles from the business data center, input at least one indicator value in the operation data of the business resources into a pre-built three-dimensional matrix model, output the position of the business resources in the matrix, and output the cycle operation results by comparing the position change of the same business resource in the matrix and the change trend of each indicator in two adjacent specific cycles; The brand fit output module is used to obtain the sales data and market attention data of the brand corresponding to the commercial resources in a specific time period from the commercial data center, analyze the sales data and market attention data of the brand through the pre-built brand fit model, and output the brand fit; The commercial resource allocation module is used to evaluate the rental value of commercial resources through at least one measurement model, and output the optimal allocation plan for each commercial resource based on the short-term sales results, periodic operating results and brand adaptability of the commercial resources.

Images