Business hall site selection evaluation method and device and medium

Abstract

The invention provides a business hall site selection evaluation method and device and a medium, and relates to the technical field of data analysis, and the method comprises the steps: respectively obtaining thermodynamic diagram data and telephone traffic data of a to-be-built business hall and a plurality of stock business halls, the thermodynamic diagram data and the telephone traffic data having distribution characteristics; according to the thermodynamic diagram data, the telephone traffic data and the distribution thereof, obtaining human traffic parameters and telephone traffic parameters of the to-be-built business hall and each stock business hall; obtaining TopK stock business halls closest to the business hall to be built according to the human traffic parameters and the telephone traffic parameters; and according to the average productivity of the TopK stock business halls, obtaining expected benefits of the to-be-built business hall. According to the method, the thermodynamic diagram data and the telephone traffic data are subjected to parameterized conversion, the approximation of the to-be-built business hall and the stock business hall is obtained according to the conversion parameters, and the expected income of the to-be-built business hall is evaluated according to the most approximate productivity average value of the stock business halls, so that the rationality of target site selection is evaluated.

Description

Technical Field

[0001] The present disclosure at least relates to the field of data analysis technology, and in particular to a business hall site selection evaluation method, a business hall site selection evaluation device, and a computer-readable storage medium. Background Art

[0002] The site selection of business halls is an issue that every telecommunications operator must face. Whether the site selection is reasonable and effective largely determines the success or failure of the subsequent operation of the business hall.

[0003] At present, due to the lack of special research and management optimization on the site selection of business halls, the site selection of business halls is still in a relatively extensive management stage, mainly using field surveys or market research. Overall, there is a lack of standardized site selection methods and quantitative evaluation models for site selection. It is highly subjective, time-consuming and labor-intensive, which may cause deviations in decision-making information and affect the quality of site selection. Summary of the invention

[0004] The technical problem to be solved by the present disclosure is to provide a method, device and medium for evaluating the site selection of a business hall in view of the above-mentioned shortcomings, so as to solve the problem of how to evaluate whether the target site selection of the business hall to be built is reasonable.

[0005] In a first aspect, the present disclosure provides a method for evaluating the location of a business hall.

[0006] include:

[0007] The heat map data and call volume data of the business hall to be built and multiple existing business halls are obtained respectively. Both the heat map data and call volume data have distribution characteristics.

[0008] According to the heat map data and traffic data and their distribution, obtain the passenger flow parameters and traffic volume parameters of the business hall to be built and each existing business hall;

[0009] According to the passenger flow parameters and call volume parameters, obtain the top K existing business halls that are most similar to the business hall to be built;

[0010] Based on the average production capacity of the TopK existing business halls, the expected benefits of the business halls to be built are obtained.

[0011] Furthermore, the heat map data and traffic volume data of the business hall to be built and the multiple existing business halls are obtained respectively, which are specifically included in the business hall site selection reference map system:

[0012] Mark the locations of the business halls to be built and all existing business halls, and mark the business districts radiated by the business halls to be built and each existing business hall with their respective locations as the center;

[0013] Import the heat map data and traffic data of each business district. The heat map data reflects the flow distribution of people in the corresponding business district. The traffic data is the traffic distribution data of base stations distributed in each business district obtained from the operator's device.

[0014] Furthermore, according to the heat map data and the traffic volume data and their distribution, the flow parameters and traffic volume parameters of the business hall to be built and each existing business hall are obtained, including:

[0015] Get the color distribution of the heat map data for each business district, and get the distribution time and location of different colors;

[0016] Obtain the traffic distribution of each business district based on traffic data, and obtain the distribution time and location of different traffic volumes;

[0017] The pedestrian flow parameters are obtained based on the color and its distribution time and location, and the traffic volume parameters are obtained based on the traffic volume and its distribution time and location.

[0018] Further, the human flow parameters are obtained according to the color and its distribution time and distribution location, and the traffic volume parameters are obtained according to the traffic volume and its distribution time and distribution location, specifically including:

[0019] Different traffic scores are assigned to different colors, and different traffic scores are assigned according to the volume of traffic;

[0020] Different weights and coefficients are given according to the distribution time and distribution location, including: giving different weights to the distribution time according to different time periods of the day, and giving different coefficients according to the business type of the distribution location combined with date characteristics;

[0021] The traffic score is obtained by multiplying the traffic score by the corresponding weight and coefficient, and the traffic score is obtained by multiplying the traffic score by the corresponding weight and coefficient.

[0022] Furthermore, according to the passenger flow parameters and the call volume parameters, the top K existing business halls that are most similar to the business hall to be built are obtained, specifically including:

[0023] Set the K value of the K nearest neighbor classification algorithm;

[0024] The passenger flow parameters and call volume parameters of the business hall to be built and each existing business hall are respectively combined into their own feature sets;

[0025] Each feature set is input into the K nearest neighbor classification algorithm to calculate the feature distance between the business hall to be built and each existing business hall;

[0026] In order of feature distance from near to far, select the TopK existing business halls that are most similar to the business hall to be built.

[0027] Furthermore, based on the average production capacity of the TopK existing business halls, the expected benefits of the business halls to be built are obtained, including:

[0028] Get the production capacity data of the top K existing business halls within the preset time period;

[0029] Based on the TopK production capacity data, obtain the average production capacity of the TopK existing business halls;

[0030] Obtain the predicted revenue of the business hall to be built within a preset time period in the future based on the average production capacity;

[0031] Obtain the predicted cost of the business hall to be built within a preset time period in the future;

[0032] Based on the ratio of predicted revenue to predicted cost, the expected benefits of the business hall to be built can be obtained.

[0033] Further, wherein:

[0034] The production capacity data includes the monthly sales volume of J products of the TopK existing business halls in the last consecutive I months. ij , i∈[1,I], j∈[1,J];

[0035] Average production capacity B ij =∑ k (A ij ) / K, k∈[1,K];

[0036] Predicted return = ∑ ij B ij *Average user revenue value of each product category*Life cycle of each product category / I;

[0037] The predicted cost is the predicted average monthly cost of the business hall to be built, including the average monthly site rent, labor costs, and management expenses.

[0038] Furthermore, the method further comprises:

[0039] In response to the expected benefit being greater than a preset benefit threshold, determining to construct the business hall to be built;

[0040] In response to the determination of the construction of the business hall to be built, the model of the business hall to be built is designed with reference to the models of the TopK existing business halls. The model includes the appearance of the business hall, staff configuration, product configuration, preferential policies, and marketing activities.

[0041] In a second aspect, the present disclosure provides a business hall site selection evaluation device, the device comprising:

[0042] A data acquisition module is used to obtain heat map data and call volume data of the business hall to be built and multiple existing business halls respectively, and both the heat map data and call volume data have distribution characteristics;

[0043] A parameter acquisition module, connected to the data acquisition module, is used to obtain the flow parameters and traffic parameters of the business hall to be built and each existing business hall according to the heat map data and traffic data and their distribution;

[0044] The approximate analysis module is connected to the parameter acquisition module and is used to obtain the TopK existing business halls that are most similar to the business hall to be built based on the passenger flow parameters and the call volume parameters;

[0045] The benefit prediction module is connected to the approximate analysis module and is used to obtain the expected benefits of the business halls to be built based on the average production capacity of the TopK existing business halls.

[0046] In a third aspect, the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the business hall site selection evaluation method as described above is implemented.

[0047] The present invention provides a method, device and medium for evaluating the site selection of a business hall. By performing parameterized conversion on heat map data and traffic data, the approximation between the business hall to be built and the existing business hall is obtained according to the conversion parameters. The expected profit of the business hall to be built is evaluated according to the average production capacity of several most similar existing business halls, so as to evaluate the rationality of the target site selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a flow chart of a method for evaluating the location of a business hall according to an embodiment of the present disclosure;

[0049] Figure 2 It is a structural schematic diagram of a business hall site selection evaluation device according to an embodiment of the present disclosure;

[0050] Figure 3 It is a structural schematic diagram of a business hall site selection reference map system according to an embodiment of the present disclosure;

[0051] Figure 4 It is a schematic diagram of a business hall capacity constraint factor model according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the embodiments of the present disclosure will be further described in detail below with reference to the accompanying drawings.

[0053] It should be understood that the specific embodiments and drawings described herein are only used to explain the present disclosure rather than to limit the present disclosure.

[0054] It can be understood that, in the absence of conflict, the various embodiments of the present disclosure and the various features in the embodiments can be combined with each other.

[0055] It can be understood that, for the convenience of description, the drawings of the present disclosure only show parts related to the present disclosure, while parts irrelevant to the present disclosure are not shown in the drawings.

[0056] It can be understood that each module or unit involved in the embodiments of the present disclosure may correspond to only one entity structure, or may be composed of multiple entity structures, or multiple modules or units may be integrated into one entity structure.

[0057] It will be understood that, in the absence of conflict, the functions and steps marked in the flowcharts and block diagrams of the present disclosure may occur in an order different from that marked in the drawings.

[0058] It is understood that the flowcharts and block diagrams of the present disclosure illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the various embodiments of the present disclosure. Each box in the flowchart or block diagram may represent a module, unit, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented by a hardware-based device that implements the specified functions, or may be implemented by a combination of hardware and computer instructions.

[0059] It can be understood that the modules and units involved in the embodiments of the present disclosure can be implemented by software or hardware, for example, the modules and units can be located in a processor.

[0060] Embodiment 1:

[0061] like Figure 1 As shown, the present disclosure provides a method for evaluating the location of a business hall, the method comprising:

[0062] S1. respectively obtain heat map data and call volume data of the business hall to be built and multiple existing business halls, where both the heat map data and call volume data have distribution characteristics;

[0063] S2. Obtain the flow parameters and traffic parameters of the business hall to be built and each existing business hall according to the heat map data and traffic data and their distribution;

[0064] S3. Obtain the top K existing business halls that are most similar to the business hall to be built based on the passenger flow parameters and the call volume parameters;

[0065] S4. Based on the average production capacity of the TopK existing business halls, obtain the expected benefits of the business halls to be built.

[0066] In this embodiment, the method performs parameter conversion on the heat map data and the traffic data, obtains the approximation between the business hall to be built and the existing business hall according to the conversion parameters, and evaluates the expected revenue of the business hall to be built according to the average capacity of several most similar existing business halls, so as to evaluate the rationality of the target site selection. Figure 2 The device shown in the figure comprises a data acquisition module 1 for executing step S1, a parameter acquisition module 2 for executing step S2, an approximate analysis module 3 for executing step S3, and a benefit prediction module 4 for executing step S4. The device has the same technical effect as the method, and may be implemented in the form of a computer.

[0067] Specifically, in terms of scientific and quantitative business hall site selection, field surveys are often required to obtain information about the market environment, population, traffic conditions, site conditions, etc. around the site, and then summarize this information as a basis for decision-making. Since most of the above information is difficult to quantify or standardize, it is necessary to conduct a systematic study on the site selection of business halls to obtain a scientific site selection model for business halls based on big data information to help operators make decisions.

[0068] The location selection principles for business halls include the following: the principle of integrity; the principle of customer-centricity; the principle of matching supply and demand; the principle of efficiency; the principle of coverage; the principle of competitiveness; the principle of coordination; the principle of importance; and the principle of legality.

[0069] The current business hall site selection faces many challenges in terms of data infrastructure system architecture and data analysis, with defects such as poor scalability, low resource utilization, complex application deployment, high operating costs and high energy consumption.

[0070] In order to effectively reduce the investment risk in business hall site selection, it is urgent to conduct accurate big data risk assessment before construction and avoid the risk of investment failure. This puts forward new demands for the innovation of business hall site selection risk assessment model.

[0071] Normally, the site selection of business halls requires selecting the key features that most affect the probability of business hall profitability from massive data. However, due to the limitations of data complexity, real-time and authenticity, it is often difficult to quickly and accurately capture the failure risks in the business of business hall site selection. Therefore, it is urgent to innovate and establish a high-precision model that can reduce the risk of business hall site selection.

[0072] At present, the amount of data faced in selecting business hall locations is increasing, and how to effectively process this massive data has become a difficult problem. It is urgent to adopt distributed processing, cloud computing and other technologies to improve the processing capabilities of ultra-large-scale data such as PB (1PB equals 1024TB) and ZB (about 1 trillion GB). It is necessary to have a deep insight into the massive data, extract key information from the massive data, and improve the accuracy of decision-making.

[0073] This embodiment is based on the objective need to accurately evaluate the risk and benefit of the site selection of the business hall. It optimizes and analyzes the data from the aspects of site selection integrity, customer demand, supply and demand matching, optimal benefit, coverage, competitiveness, coordination, importance and legality, predicts the business model, asset operation, income and profit structure of the business hall, develops an artificial intelligence algorithm model for site selection of the business hall based on big data, and makes intelligent and accurate site selection risk prediction for the business hall based on big data and machine learning. It can discover patterns that are difficult for humans to detect based on the huge amount of data, and assist the business hall in making more intelligent site selection decisions.

[0074] This embodiment uses the latest AI (Artificial Intelligence) algorithm to improve the accuracy and efficiency of data processing, improve the analysis of data on products or services provided to consumers in business halls, and use big data for accurate site selection and marketing. This embodiment builds a big data analysis model, fully utilizes the characteristic advantages of big data, and develops relevant big data analysis software to provide reliable and real-time decision-making references for solving business hall site selection.

[0075] More specifically, this embodiment optimizes the site selection method of the telecom operator's business hall from three aspects:

[0076] 1) Optimize the spatial coverage model of the business hall. When superimposing the geographical coverage of the business hall with the business coverage, set different weights to refine the impact of factors such as retail environment, population, and business at different distances on the business hall;

[0077] 2) Optimize the business hall load regression model. Explore the use of quadratic functions or even higher-order functions to establish the model. Improve the model accuracy by continuously optimizing the model and establishing load models for different scenarios.

[0078] 3) Optimize data, including replacing base station data with signaling data, obtaining data from operator systems, and increasing data types and sample sizes.

[0079] This embodiment constructs a business hall site selection evaluation model, conducts accurate risk assessment based on the characteristics of the business hall, and is supplemented by historical data verification; relying on big data with authenticity and real-time advantages, a business hall site selection mathematical model with self-learning and continuous optimization is constructed; in the entire life cycle of providing services to telecom users, better risk assessment decisions before business operations, real-time risk monitoring during the post-operation process, and review of site selection effects after profitability, and further optimization of risk control decisions. Evaluation algorithm models such as deep neural networks for business hall site selection based on big data can be established, and decision trees, logistic regression, machine learning and other algorithms can also be used to calculate quantitative analysis site selection scores, and the site selection evaluation results of business halls can be obtained after verification. A host computer configuration software system based on Libview can be developed to provide a human-computer interaction friendly interface for business hall site selection. A big data management library for business hall site selection is established to make full use of the characteristic advantages of big data and provide reliable and real-time decision-making references for solving business hall site selection.

[0080] In one embodiment, heat map data and traffic volume data of a business hall to be built and a plurality of existing business halls are obtained respectively, specifically including in a business hall site selection reference map system:

[0081] Mark the locations of the business halls to be built and all existing business halls, and mark the business districts radiated by the business halls to be built and each existing business hall with their respective locations as the center;

[0082] Import the heat map data and traffic data of each business district. The heat map data reflects the flow distribution of people in the corresponding business district. The traffic data is the traffic distribution data of base stations distributed in each business district obtained from the operator system.

[0083] In this embodiment, the business hall site selection reference map system is as follows Figure 2 As shown, the specific implementation method can be to use the Baidu Map API (Application Programming Interface) data interface tool to secondary develop and build a business hall site selection reference map system. The map system mainly displays the distribution of communication outlets, business halls of the three major operators, and large chain stores in the city's business districts, and displays the density of communication outlets in each business district through map visualization.

[0084] Baidu Map has Baidu heat map data, which is a kind of data that represents the density of people. It uses the depth of color to show the density of hot spots, and can quickly show the distribution trend of popular places and the coverage of popular areas. This embodiment uses Baidu heat map data to evaluate the target location of the business hall.

[0085] Using the site selection reference map system, site selection managers conduct preliminary screening of site selection plans reported by various districts, focusing on business districts with insufficient network coverage and strong communication demand, and locking in site selection targets. Figure 3 The business hall capacity constraint factor model shown in the figure has a great impact on the site selection, and the information such as the business district where the store is located and the location of the store in the business district has a great impact on the site selection, and the flow of people in the business district and the store entry rate are taken into consideration. Therefore, Baidu thermal data and base station traffic data are further used to analyze the flow of people in the locked business district site selection, providing the most important flow parameters for the site selection evaluation model.

[0086] At the same time, the operator's traffic volume big data is used to calculate the traffic volume parameters of the site selection through a quantitative algorithm. Specifically, the distribution of the operator's traffic volume within the locked business district site selection range can be obtained by combining information such as base station location and terminal positioning.

[0087] In one embodiment, based on the heat map data and the traffic data and their distribution, the flow parameters and traffic parameters of the business hall to be built and each existing business hall are obtained, specifically including:

[0088] Get the color distribution of the heat map data for each business district, and get the distribution time and location of different colors;

[0089] Obtain the traffic distribution of each business district based on traffic data, and obtain the distribution time and location of different traffic volumes;

[0090] The pedestrian flow parameters are obtained based on the color and its distribution time and location, and the traffic volume parameters are obtained based on the traffic volume and its distribution time and location.

[0091] In this embodiment, the purpose of this step is to parameterize the data, and the parameters take into account the time distribution and location distribution characteristics of the data to facilitate subsequent approximation analysis.

[0092] In one embodiment, obtaining a human flow parameter according to color and its distribution time and distribution location, and obtaining a traffic volume parameter according to traffic volume and its distribution time and distribution location specifically includes:

[0093] Different traffic scores are assigned to different colors, and different traffic scores are assigned according to the volume of traffic;

[0094] Different weights and coefficients are given according to the distribution time and distribution location, including: giving different weights to the distribution time according to different time periods of the day, and giving different coefficients according to the business type of the distribution location combined with date characteristics;

[0095] The traffic score is obtained by multiplying the traffic score by the corresponding weight and coefficient, and the traffic score is obtained by multiplying the traffic score by the corresponding weight and coefficient.

[0096] In this embodiment, the specific method for analyzing and calculating the passenger flow parameters is as follows: using the heat map tool of Baidu Map, the heat map conditions around the site are obtained in real time in the constructed site reference map system, and the heat map conditions of the target site are quantified through a quantitative model.

[0097] The quantitative model is as follows: 1) According to the color depth of Baidu heat map, specific traffic scores are given to different colors: red to purple are given 7 to 1 points respectively; 2) Different evaluation weights are given to traffic scores at different times of the day: 30% weight in the morning, 50% weight in the afternoon, and 20% weight in the evening; 3) Different evaluation coefficients are given to weekend and weekday traffic scores according to the type of business district selected: street shops, shopping mall shops, and community shops have a coefficient of 1.2 on weekends and a coefficient of 1 on weekdays; shops in office buildings have a coefficient of 1 on weekends and a coefficient of 1.2 on weekdays, etc.

[0098] The calculation of the traffic volume parameter is similar to the above, and the traffic volume score and traffic volume score are calculated separately. The above is just an example of a calculation method. You can also directly form a three-dimensional array of the color value of the Baidu heat map and its distribution location and time. The traffic volume parameter is similar, or the above scores can be calculated as a total score, and so on.

[0099] In one implementation, according to the passenger flow parameter and the call traffic parameter, the top K existing business halls that are most similar to the business hall to be built are obtained, specifically including:

[0100] Set the K value of the K nearest neighbor classification algorithm;

[0101] The passenger flow parameters and call volume parameters of the business hall to be built and each existing business hall are respectively combined into their own feature sets;

[0102] Each feature set is input into the K nearest neighbor classification algorithm to calculate the feature distance between the business hall to be built and each existing business hall;

[0103] In order of feature distance from near to far, select the TopK existing business halls that are most similar to the business hall to be built.

[0104] In this embodiment, the calculated passenger flow parameters and call volume parameters are combined, and a data mining algorithm, K-Nearest Neighbor (KNN) classification algorithm, is used to build a site selection regression model to analyze whether the target site is suitable for building a hall.

[0105] The specific method is as follows: 1) Calculate the passenger flow score and call volume score of the target location and existing stock business halls respectively to form the address feature set of the target location and each hall {passenger flow parameter, call volume parameter}; 2) Calculate the characteristic distance (Euclidean distance) between the target location and all existing halls, sort them from near to far, and select the top 10% halls closest to the target location; 3) For the 10% of existing halls selected, count the average monthly production capacity data of various products in the past year, and calculate the average value, and use the average capacity of this group as the predicted capacity output of the target location.

[0106] In one implementation, the expected benefits of the business halls to be built are obtained based on the average production capacity of the TopK existing business halls, specifically including:

[0107] Get the production capacity data of the top K existing business halls within the preset time period;

[0108] Based on the TopK production capacity data, obtain the average production capacity of the TopK existing business halls;

[0109] Obtain the predicted revenue of the business hall to be built within a preset time period in the future based on the average production capacity;

[0110] Obtain the predicted cost of the business hall to be built within a preset time period in the future;

[0111] Based on the ratio of predicted revenue to predicted cost, the expected benefits of the business hall to be built can be obtained.

[0112] In this embodiment, the investment benefit of the selected site is evaluated based on the predicted production capacity of the target site and the rent of the selected site. The specific value of K can be set manually as needed, such as selecting according to the size of the data set. The cross-validation method can be used to select the K value, that is, cross-validating different sample sets to verify the variance of the K value, so as to obtain the best K value.

[0113] In one embodiment, wherein:

[0114] The production capacity data includes the monthly sales volume of J products of the TopK existing business halls in the last consecutive I months. ij , i∈[1,I], j∈[1,J];

[0115] Average production capacity B ij =∑ k (A ij ) / K, k∈[1,K];

[0116] Predicted return = ∑ ij B ij *Average user revenue value of each product category*Life cycle of each product category / I;

[0117] The predicted cost is the predicted average monthly cost of the business hall to be built, including the average monthly site rent, labor costs, and management expenses.

[0118] In this embodiment, the evaluation model is as follows: 1) Site selection benefit = (∑ product average monthly predicted capacity * product ARPU (average revenue per user) value * product life cycle - average monthly cost) / average monthly cost, where: average monthly cost mainly includes site rental, labor costs, property water and electricity and other management expenses; 2) Site selection with benefits greater than or equal to 50% can be determined to be suitable for building a hall in mathematical logic (specific judgment criteria can be set according to the actual situation of the user). The above is just an example of a calculation method. You can also predict the expected profit growth rate, development potential, contribution rate, etc. based on the operating time as a basis for decision-making.

[0119] In one embodiment, the method further comprises:

[0120] In response to the expected benefit being greater than a preset benefit threshold, determining to construct the business hall to be built;

[0121] In response to the determination of the construction of the business hall to be built, the model of the business hall to be built is designed with reference to the models of the TopK existing business halls. The model includes the appearance of the business hall, staff configuration, product configuration, preferential policies, and marketing activities.

[0122] In this embodiment, combined with Figure 3 As shown in Figure 1, factors affecting the production capacity of business halls also include: store appearance, salesperson's ability to attract customers, product factors, preferential policies, marketing capabilities, etc. In order to make the target business hall reach a scale similar to that of the TopK existing business halls, the model of the target business hall can be designed with reference to the model of the TopK existing business halls.

[0123] In addition, in order to meet the challenges of large-scale data, we should study the decision to strengthen data processing capabilities, improve data processing speed by introducing technologies such as high-performance computing and in-memory computing, and use technologies such as artificial intelligence and machine learning to achieve automated data processing and intelligent analysis.

[0124] The above scheme is implemented by combining Python and Libview technology to develop the corresponding business hall site selection intelligent evaluation software. Based on the algorithm provided above, combined with the call of the pedestrian flow distribution data in the map heat map and the traffic distribution data obtained from the operator, the site selection of the business hall to be built is intelligently evaluated, and the location and construction method of the new business hall are scientifically selected.

[0125] The key technical points of this embodiment include:

[0126] 1) Model: Based on the data quantification model of Baidu heat map and operator traffic volume, users can conveniently and efficiently quantify the flow of people and traffic volume of the target site, and provide a site selection model that can predict cost-effectiveness.

[0127] 2) Algorithm: Based on big data with multiple time scales, high accuracy, strong real-time performance and strong reliability, we develop an artificial intelligence evaluation algorithm for business hall site selection based on deep reinforcement learning.

[0128] 3) Software: Make full use of the characteristics and advantages of big data to develop a software system as the upper computer configuration, provide reliable and real-time decision-making reference for solving the site selection of business halls, and provide technical support for the development of data-driven telecommunications industry.

[0129] 4) Technology: The intelligent evaluation modeling, solution and optimization evaluation system for business hall site selection is developed by combining Python and Libview. The developed system has functions such as autonomous learning and real-time adjustment, which can meet the real-time evaluation needs of business hall site selection.

[0130] The site selection method provided in this embodiment is based on big data information, and proposes a parameter quantification model based on Baidu heat map data and operator traffic data, and based on the two key parameters of traffic and traffic volume, a quantitative prediction of the target site selection benefits is made based on the data mining clustering regression algorithm and the benefit evaluation model. It not only fully considers the situation of users across the entire network, but also comprehensively considers the cost-effectiveness of site selection, which can help the scientific site selection of business halls, meet the real-time evaluation needs of business hall site selection, save the time and manpower costs of communication operators in site selection, and improve customer perception.

[0131] Embodiment 2:

[0132] like Figure 2 As shown, the present disclosure provides a business hall site selection evaluation device, the device comprising:

[0133] Data acquisition module 1, used to respectively acquire heat map data and call volume data of the business hall to be built and multiple existing business halls, where both the heat map data and call volume data have distribution characteristics;

[0134] Parameter acquisition module 2, connected to data acquisition module 1, is used to obtain the flow parameters and traffic parameters of the business hall to be built and each existing business hall according to the heat map data and traffic data and their distribution;

[0135] The approximate analysis module 3 is connected to the parameter acquisition module 2 and is used to obtain the TopK existing business halls that are most similar to the business hall to be built according to the passenger flow parameters and the traffic volume parameters;

[0136] The benefit prediction module 4 is connected to the approximate analysis module 3 and is used to obtain the expected benefit of the business hall to be built according to the average production capacity of the TopK existing business halls.

[0137] In one embodiment, the data acquisition module 1 specifically includes:

[0138] Business hall location reference map system, and

[0139] The marking unit is used to mark the locations of the business halls to be built and all the existing business halls in the business hall site selection reference map system, and to mark the business districts radiated by the business halls to be built and each existing business hall with their respective locations as the center;

[0140] The import unit is connected to the marking unit and is used to import the heat map data and traffic data of each business district into the business hall site selection reference map system. The heat map data reflects the distribution of pedestrian flow in the corresponding business district, and the traffic data is the traffic distribution data of base stations distributed in each business district obtained from the operator system.

[0141] In one implementation, the parameter acquisition module 2 specifically includes:

[0142] A color distribution acquisition unit is used to acquire the color distribution of the heat map data of each business district, and to acquire the distribution time and distribution location of different colors;

[0143] A traffic volume distribution acquisition unit, used to acquire the traffic volume distribution of each business district according to the traffic volume data, and acquire the distribution time and distribution location of different traffic volumes;

[0144] The parameterization unit is connected to the color distribution acquisition unit and the traffic volume distribution acquisition unit, and is used to obtain the human flow parameters according to the color and its distribution time and distribution location, and to obtain the traffic volume parameters according to the traffic volume and its distribution time and distribution location.

[0145] In one embodiment, the parameterization unit specifically includes:

[0146] The scoring unit is used to assign different traffic scores to different colors and assign different traffic scores according to the traffic volume;

[0147] A weight coefficient unit is used to give different weights and coefficients according to the distribution time and distribution location, including: giving different weights to the distribution time according to different time periods of the day, and giving different coefficients according to the business type of the distribution location combined with date characteristics;

[0148] The score calculation unit is connected to the score unit and the weight coefficient unit, and is used to multiply the traffic score by the corresponding weight and coefficient to obtain the traffic score, and multiply the traffic score by the corresponding weight and coefficient to obtain the traffic score.

[0149] In one embodiment, the approximate analysis module 3 specifically includes:

[0150] An initial value unit, used to set the K value of the K nearest neighbor classification algorithm;

[0151] A feature set unit, used to respectively combine the passenger flow parameters and call traffic parameters of the business hall to be built and each existing business hall into their own feature sets;

[0152] An approximate calculation unit, connected to the initial value unit and the feature set unit, is used to input each feature set into a K nearest neighbor classification algorithm to calculate the feature distance between the business hall to be built and each existing business hall;

[0153] The approximate sorting unit is connected to the approximate calculation unit and is used to select the top K existing business halls that are most similar to the business hall to be built in the order of feature distance from near to far.

[0154] In one embodiment, the benefit prediction module 4 specifically includes:

[0155] The capacity statistics unit is used to obtain the capacity data of the top K existing business halls within a preset period of time;

[0156] The capacity average unit is connected to the capacity statistics unit and is used to obtain the average capacity of the top K existing business halls based on the top K capacity data;

[0157] The revenue forecasting unit is connected to the capacity averaging unit and is used to obtain the forecasted revenue of the business hall to be built within a preset time period in the future according to the average capacity;

[0158] Cost prediction unit, to obtain the predicted cost of the business hall to be built within a preset time period in the future;

[0159] The benefit calculation unit is connected with the revenue prediction unit and the cost estimation unit, and is used to obtain the expected benefit of the business hall to be built according to the ratio of the predicted revenue to the predicted cost.

[0160] In one embodiment, wherein:

[0161] The production capacity data includes the monthly sales volume of J products of the TopK existing business halls in the last consecutive I months. ij , i∈[1,I], j∈[1,J];

[0162] Average production capacity B ij =∑ k (A ij ) / K, k∈[1,K];

[0163] Predicted return = ∑ ij B ij*Average user revenue value of each product category*Life cycle of each product category / I;

[0164] The predicted cost is the predicted average monthly cost of the business hall to be built, including the average monthly site rent, labor costs, and management expenses.

[0165] In one embodiment, the device further comprises:

[0166] A construction decision unit, connected to the benefit prediction module 4, for determining to construct the business hall to be built in response to the expected benefit being greater than a preset benefit threshold;

[0167] The model decision unit is connected to the construction decision unit, and is used to respond to the determination of the construction of the business hall to be built, and design the model of the business hall to be built with reference to the models of the TopK existing business halls. The model includes the appearance of the business hall, staff configuration, product configuration, preferential policies, and marketing activities.

[0168] Embodiment 3:

[0169] Embodiment 3 of the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the method for evaluating the location of a business hall as described in Embodiment 1 or the device for evaluating the location of a business hall as described in Embodiment 2 is implemented.

[0170] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program elements or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0171] In addition, the present disclosure may also provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the business hall site selection evaluation method as described in Example 1. The computer device may be the business hall site selection evaluation device as described in Example 2.

[0172] The memory is connected to the processor, the memory may be a flash memory or a read-only memory or other memory, and the processor may be a central processing unit or a single-chip microcomputer.

[0173] Embodiments 1-3 of the present disclosure provide a business hall site selection evaluation method, a business hall site selection evaluation device, and a computer-readable storage medium. By performing parameterized conversion on heat map data and traffic data, the approximation between the business hall to be built and the existing business hall is obtained according to the conversion parameters. The expected profit of the business hall to be built is evaluated based on the average production capacity of several most similar existing business halls, so as to evaluate the rationality of the target site selection.

[0174] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A method for evaluating the location of a business hall, characterized in that: The method comprises: The heat map data and call volume data of the business hall to be built and multiple existing business halls are obtained respectively. Both the heat map data and call volume data have distribution characteristics. According to the heat map data and traffic data and their distribution, obtain the passenger flow parameters and traffic volume parameters of the business hall to be built and each existing business hall; According to the passenger flow parameters and call volume parameters, obtain the top K existing business halls that are most similar to the business hall to be built; Based on the average production capacity of the TopK existing business halls, the expected benefits of the business halls to be built are obtained.

2. The method according to claim 1, characterized in that The heat map data and call volume data of the business hall to be built and multiple existing business halls are obtained respectively, which are specifically included in the business hall site selection reference map system: Mark the locations of the business halls to be built and all existing business halls, and mark the business districts radiated by the business halls to be built and each existing business hall with their respective locations as the center; Import the heat map data and traffic data of each business district. The heat map data reflects the flow distribution of people in the corresponding business district. The traffic data is the traffic distribution data of base stations distributed in each business district obtained from the operator's device.

3. The method according to claim 2, characterized in that According to the heat map data and traffic data and their distribution, obtain the traffic parameters and traffic parameters of the business hall to be built and each existing business hall, including: Get the color distribution of the heat map data for each business district, and get the distribution time and location of different colors; Obtain the traffic distribution of each business district based on traffic data, and obtain the distribution time and location of different traffic volumes; The pedestrian flow parameters are obtained based on the color and its distribution time and location, and the traffic volume parameters are obtained based on the traffic volume and its distribution time and location.

4. The method according to claim 3, characterized in that According to the color and its distribution time and distribution location, the human flow parameters are obtained; according to the traffic volume and its distribution time and distribution location, the traffic volume parameters are obtained, including: Different traffic scores are assigned to different colors, and different traffic scores are assigned according to the volume of traffic; Different weights and coefficients are given according to the distribution time and distribution location, including: giving different weights to the distribution time according to different time periods of the day, and giving different coefficients according to the business type of the distribution location combined with date characteristics; The traffic score is obtained by multiplying the traffic score by the corresponding weight and coefficient, and the traffic score is obtained by multiplying the traffic score by the corresponding weight and coefficient.

5. The method according to any one of claims 1 to 4, characterized in that: According to the traffic parameters and call volume parameters, obtain the top K existing business halls that are most similar to the business hall to be built, including: Set the K value of the K nearest neighbor classification algorithm; The passenger flow parameters and call volume parameters of the business hall to be built and each existing business hall are respectively combined into their own feature sets; Each feature set is input into the K nearest neighbor classification algorithm to calculate the feature distance between the business hall to be built and each existing business hall; In order of feature distance from near to far, select the TopK existing business halls that are most similar to the business hall to be built.

6. The method according to any one of claims 1 to 4, characterized in that: Based on the average production capacity of the TopK existing business halls, the expected benefits of the business halls to be built are obtained, including: Get the production capacity data of the top K existing business halls within the preset time period; Based on the TopK production capacity data, obtain the average production capacity of the TopK existing business halls; Obtain the predicted revenue of the business hall to be built within a preset time period in the future based on the average production capacity; Obtain the predicted cost of the business hall to be built within a preset time period in the future; Based on the ratio of predicted revenue to predicted cost, the expected benefits of the business hall to be built can be obtained.

7. The method according to claim 6, characterized in that in: The production capacity data includes the monthly sales volume of J products of the TopK existing business halls in the last consecutive I months. ij , i∈[1,I], j∈[1,J]; Average production capacity B ij =∑ k (A ij ) / K, k∈[1,K]; Predicted return = ∑ ij B ij *Average user revenue value of each product category*Life cycle of each product category / I; The predicted cost is the predicted average monthly cost of the business hall to be built, including the average monthly site rent, labor costs, and management expenses.

8. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: In response to the expected benefit being greater than a preset benefit threshold, determining to construct the business hall to be built; In response to the determination of the construction of the business hall to be built, the model of the business hall to be built is designed with reference to the models of the TopK existing business halls. The model includes the appearance of the business hall, staff configuration, product configuration, preferential policies, and marketing activities.

9. A business hall site selection evaluation device, characterized in that: The device comprises: A data acquisition module is used to obtain heat map data and call volume data of the business hall to be built and multiple existing business halls respectively, and both the heat map data and call volume data have distribution characteristics; A parameter acquisition module, connected to the data acquisition module, is used to obtain the flow parameters and traffic parameters of the business hall to be built and each existing business hall according to the heat map data and traffic data and their distribution; The approximate analysis module is connected to the parameter acquisition module and is used to obtain the TopK existing business halls that are most similar to the business hall to be built based on the passenger flow parameters and the call volume parameters; The benefit prediction module is connected to the approximate analysis module and is used to obtain the expected benefits of the business halls to be built based on the average production capacity of the TopK existing business halls.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the business hall site selection evaluation method as described in any one of claims 1 to 8 is implemented.

Images