Shop lease pricing evaluation method for commercial complex multi-source passenger flow data

Abstract

The invention provides a shop lease pricing evaluation method for commercial complex multi-source passenger flow data, and relates to the technical field of commercial real estate big data analysis and asset evaluation, and the method comprises the steps: obtaining multi-source passenger flow data of a Wi-Fi probe, video monitoring, POS transaction and the like, and carrying out the cleaning and fusion; calculating a passenger flow comprehensive index reflecting passenger flow quantity and quality characteristics based on the preprocessed data; calculating a passenger flow volume conversion rate in combination with sales data, and calculating a lease value score through a specific formula by coupling the indexes; evaluating a lease price according to the score and based on a historical lease data model; and finally generating a lease suggestion report containing pricing details and a visual chart. According to the method, the pricing decision is converted from empirical to data driving, and the objectivity, accuracy and efficiency of an evaluation result are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of commercial real estate big data analysis and asset evaluation, in particular to a shop rental pricing evaluation method for multi-source passenger flow data of a commercial complex. BACKGROUND

[0002] The shop rental pricing of a commercial complex is the core of its operation management and revenue. Traditional pricing methods mostly rely on the static physical properties of the shops, such as location and area, and make decisions in combination with market research and manager experience. This method is highly subjective and lacks objective data support, making it difficult to accurately reflect the actual value and future revenue potential of the shops, which can easily lead to pricing deviating from the real market level, resulting in undervaluation of high-quality assets or overvaluation of poor assets.

[0003] With the development of information technology, some commercial bodies have begun to try to use passenger flow data to assist decision-making. However, current practices are mostly limited to a single passenger flow data source, such as using only POS machine transaction amounts or simple video passenger flow counting. POS data can only reflect the consumption behavior that has already occurred and cannot capture the potential customer value that has not generated transactions; simple passenger flow counting ignores the quality of passenger flow, such as dwell time, visit frequency, and passenger flow stability. This single-dimensional analysis cannot fully and stereoscopically depict the commercial appeal of a shop.

[0004] In addition, even if multiple sources of data are obtained, how to effectively clean, fuse, and extract core evaluation indicators from these heterogeneous data that can scientifically guide pricing is still a technical difficulty in the industry. Existing methods lack a unified index system that can comprehensively quantify the "quantity" and "quality" of passenger flow, and they have not been able to correlate passenger flow data with the final rental value through a non-linear, intelligent model, resulting in inaccurate evaluation results and an inability to significantly improve asset management efficiency and revenue levels. SUMMARY

[0005] To solve the technical problems of relying on static attributes and subjective experience, relying on single data source analysis, and lacking multi-source data fusion and intelligent evaluation model in the prior art, the present application provides a shop rental pricing evaluation method for multi-source passenger flow data of a commercial complex.

[0006] The technical solution provided by the present application is as follows:

[0007] The shop rental pricing evaluation method for multi-source passenger flow data of a commercial complex provided by the present application comprises:

[0008] S1: Obtain multi-source passenger flow data of multiple shops in a commercial complex, the multi-source passenger flow data comprising Wi-Fi probe data, video monitoring data, and POS machine transaction data;

[0009] S2: cleaning and preprocessing the multi-source passenger flow data to remove noise and outliers, and realizing data fusion;

[0010] S3: calculating a passenger flow comprehensive index of each shop based on the preprocessed data, the passenger flow comprehensive index being used to quantify the quantity and quality characteristics of passenger flow;

[0011] S4: calculating a rental value score of each shop based on the passenger flow comprehensive index, the rental value score being used to evaluate the rental value potential of the shop;

[0012] S5: evaluating the rental pricing of each shop according to the rental value score, wherein the rental pricing is positively correlated with the rental value score.

[0013] Further, S301: extracting daily passenger flow time series data of each shop from the preprocessed data;

[0014] S302: calculating the mean, peak value and standard deviation of daily passenger flow;

[0015] S303: calculating the passenger flow comprehensive index using the following formula:

[0016]

[0017] wherein TCI represents the passenger flow comprehensive index, represents the mean of daily passenger flow, unit: person / day, P represents the peak value of daily passenger flow, unit: person / day, σ represents the standard deviation of daily passenger flow, unit: person / day, and T represents the monitoring days, unit: day.

[0018] Further, S401: obtaining sales data of each shop from the preprocessed data;

[0019] S402: calculating the passenger flow conversion rate CR, wherein CR is defined as the ratio of sales to passenger flow;

[0020] S403: calculating the rental value score using the following formula:

[0021]

[0022] wherein RVS represents the rental value score, TCI represents the passenger flow comprehensive index, CR represents the passenger flow conversion rate, and TC avg represents the mean of passenger flow comprehensive index of all shops.

[0023] Further, the multi-source passenger flow data in S1 further comprises mobile device Bluetooth data and social media check-in data.

[0024] Further, the preprocessing in S2 includes data normalization, missing value imputation and time alignment operation.

[0025] Further, the evaluation of the lease pricing in S5 includes mapping the lease value score to a preset rent range, and the mapping is based on a historical lease data model.

[0026] Further, the average of the daily traffic volume in S302 is calculated by using a moving average method, and the moving average window size is 7 days.

[0027] Further, the traffic conversion rate CR in S402 is calculated based on monthly sales and monthly traffic.

[0028] Further, S6: generating a lease recommendation report according to the evaluated lease pricing, the report including pricing details and recommended strategies.

[0029] Further, the lease recommendation report in S6 further includes a visual chart and a comparative analysis part for showing the lease value evaluation results of different shops.

[0030] The technical solution provided by the present application has at least the following beneficial effects:

[0031] (1) In the present application, by comprehensively collecting Wi-Fi probes, video monitoring, POS transactions and other multi-source heterogeneous traffic data, a comprehensive and three-dimensional traffic portrait is constructed, overcoming the limitations of single data source, providing unprecedented data breadth and depth for lease pricing, and laying the data foundation for objective evaluation;

[0032] (2) In the present application, by creating two core indicators of traffic comprehensive index (TCI) and lease value score (RVS), the number, stability and conversion efficiency of traffic are nonlinearly coupled for the first time, and the real value potential of the shop is quantified scientifically, realizing the leap from experience judgment to data intelligent decision-making;

[0033] (3) In the present application, by using a machine learning model trained based on historical data, the lease value score is automatically mapped to a specific lease pricing suggestion, so that the pricing process can learn market rules autonomously, dynamically adapt to market changes, and finally output a precise pricing scheme that meets market conditions and maximizes the revenue of the commercial complex. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0035] Figure 1 A flowchart of a business complex multi-source passenger flow data shop rental pricing evaluation method provided by an embodiment of the present application is shown in FIG. 1. DETAILED DESCRIPTION

[0036] The technical solutions in the present application will be described below with reference to the drawings.

[0037] In the embodiments of the present application, the words such as "example", "for example" are used to represent an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0038] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0039] In the embodiments of the present application, sometimes the subscript such as W1 can be mistakenly used in the form of non-subscript such as W1. When the distinction is not emphasized, the meanings expressed are consistent.

[0040] To make the technical problems, technical solutions and advantages of the present application clearer, specific embodiments will be described in detail below with reference to the drawings.

[0041] Reference is made to the accompanying drawings and specific embodiments described in the specification Figure 1 , a flowchart of a business complex multi-source passenger flow data shop rental pricing evaluation method provided by an embodiment of the present application is shown.

[0042] The embodiments of the present application provide a business complex multi-source passenger flow data shop rental pricing evaluation method, and the processing flow can include the following steps:

[0043] S1: Obtain multi-source passenger flow data of a plurality of shops in a business complex, and the multi-source passenger flow data includes Wi-Fi probe data, video monitoring data and POS machine transaction data.

[0044] The multi-source passenger flow data is acquired, which is the data basis for the implementation of the application. In actual operation, the original data can be collected through the Internet of Things sensing equipment deployed in the public area of the commercial complex and the entrances and exits of each shop. The Wi-Fi probe data is collected through multiple wireless access points (APs) distributed in the complex, which can capture the MAC address, signal strength and connection time of the mobile device with Wi-Fi function turned on, thereby reversing the user's trajectory and residence time. The video monitoring data uses the existing security camera network, and processes the shooting picture through computer vision algorithm (such as target detection and tracking technology), and counts the passenger flow entering and exiting each shop area. The POS transaction data is obtained through the unified cash register system of the commercial body or the interface docking with the ERP system of each merchant, which contains detailed transaction time, amount and shop information. These heterogeneous data streams are preliminarily associated through time stamp and space label, and are prepared for subsequent fusion analysis.

[0045] In a possible implementation, the multi-source passenger flow data in S1 further includes mobile device Bluetooth data and social media check-in data.

[0046] In order to construct a more comprehensive and three-dimensional passenger flow portrait, the multi-source data system relied on by the application further incorporates two new data sources on the basis of Wi-Fi probe, video monitoring and POS transaction data. One is mobile device Bluetooth data, which captures the signals of Bluetooth-enabled devices through the deployment of Bluetooth beacons (Beacon). The positioning accuracy is usually higher than that of Wi-Fi, which can be used to analyze the accurate movement path and hot spot area of users in the shop. The second is social media check-in data, which obtains the active check-in records of users at specific location points in the commercial complex by applying for authorized access to the API interface of mainstream social media platforms. This kind of data is a high-quality active location signal, which not only enriches the passenger flow count, but more importantly provides emotional dimension information such as user's interest points and brand preferences, providing additional basis for evaluating the attractiveness and brand effect of the shop.

[0047] S2: cleaning and preprocessing of multi-source passenger flow data to remove noise and outliers, and realizing data fusion.

[0048] Data cleaning and preprocessing aims to transform the original multi-source heterogeneous data into high-quality, analyzable data for modeling. The process first performs data cleaning, which filters out invalid probe signals with weak signal strength and short residence time for Wi-Fi probe data, and identifies and excludes noise data such as staff mobile devices; for video data, it corrects the counting errors caused by camera angle, light changes, etc., and removes abnormal peaks. Subsequently, missing value processing is performed, which can use methods such as linear interpolation or filling based on historical data to repair short-term data missing caused by device failure or signal interruption. Finally, data fusion and time alignment are performed to unify data from different sources with different sampling frequencies to the same time granularity (e.g., one hour as a period), and align based on the unified time axis to form a complete data cube covering passenger flow, residence time, consumption behavior, etc.

[0049] In one possible implementation, the preprocessing in S2 includes data normalization, missing value interpolation, and time alignment operations.

[0050] The data preprocessing stage includes a series of standardization operations to ensure data quality and consistency. Data normalization processing is mainly for data fields with different dimensions, such as converting Wi-Fi signal strength, passenger flow count, sales, etc. to a unified numerical interval or distribution through Min-Max scaling or Z-score standardization, etc. to eliminate the influence of dimensions and lay a foundation for subsequent fusion and comparison. Missing value interpolation is for the inevitable intermittent problem in data collection process, according to the characteristics and missing mode of the data, using strategies such as forward / backward filling, linear interpolation or filling based on average values of similar shops to ensure the continuity of data sequence. Time alignment operation uses high-precision NTP time server to ensure synchronization of all collection terminal clocks, and resamples and aligns all data streams according to a unified time axis (e.g., Unix timestamp) to finally synthesize a time series data with multiple synchronized dimensions.

[0051] S3: Based on the preprocessed data, calculate the passenger flow comprehensive index of each shop, which is used to quantify the quantity and quality characteristics of passenger flow.

[0052] The calculation of the passenger flow comprehensive index aims to go beyond simple passenger flow counting and construct a core index that can comprehensively reflect the "quantity" and "quality" of passenger flow. The step takes the fusion data cube generated in the previous step as input. First, the daily passenger flow data is extracted from the pre-processed time series data. Then, the daily passenger flow data is analyzed to extract its statistical characteristics. These characteristics include not only the concentration tendency index reflecting the size of passenger flow, but also the dispersion degree index revealing the stability and fluctuation of passenger flow. Finally, through a comprehensive calculation model, these multi-dimensional characteristics are aggregated into a single, quantifiable index value. The higher the index value, the higher the potential commercial value of the store, as it not only has a large passenger flow, but also has a relatively stable passenger flow distribution.

[0053] In one possible implementation, S3 specifically includes:

[0054] S301: Extracting the daily passenger flow time series data of each store from the pre-processed data;

[0055] S302: Calculating the mean, peak, and standard deviation of daily passenger flow;

[0056] S303: Calculating the passenger flow comprehensive index using the following formula:

[0057]

[0058] where TCI represents the passenger flow comprehensive index, represents the mean of daily passenger flow, with units of people / day, P represents the peak of daily passenger flow, with units of people / day, σ represents the standard deviation of daily passenger flow, with units of people / day, and T represents the monitoring period, with units of days.

[0059] In calculating the passenger flow comprehensive index (TCI), a multi-step statistical and calculation method is used to achieve a more detailed quantification of passenger flow. First, from the pre-processed and fused data cube, the passenger flow time series of each store is cut and extracted according to natural days. Then, in-depth statistical analysis is performed on the daily passenger flow sequence: the arithmetic mean is calculated to reflect the average daily level of passenger flow; the maximum value in the sequence is identified as the peak of daily passenger flow to capture the maximum carrying potential of passenger flow; and the standard deviation of the sequence is calculated to quantify the fluctuation of daily passenger flow around the mean, i.e., the stability of passenger flow. Finally, through a comprehensive formula, the above indicators are combined with the duration of the monitoring period. The design concept of this formula is that the product of the average passenger flow and the peak passenger flow constitutes the basis of passenger flow size, dividing it by the standard deviation means that high stability (low fluctuation) will amplify the basic value, and multiplying by the natural logarithm of the monitoring period introduces the gain of the time dimension, so that the index derived from long-term monitoring is more representative, resulting in a more scientific and robust passenger flow comprehensive index.

[0060] It should be noted that +1 here is to ensure that the value of the logarithmic term ln(T+1) is also 0 at the initial stage of monitoring (e.g. T=0), so that the calculation of the entire passenger flow comprehensive index (TCI) remains mathematically valid from the starting day of monitoring, avoiding undefined cases.

[0061] In a possible implementation, the average daily passenger flow in S302 is calculated using a moving average method, and the moving average window size is 7 days.

[0062] When calculating the average daily passenger flow, in order to better reflect the recent passenger flow trend and smooth accidental fluctuations, instead of simply using arithmetic mean, a moving average method is used. Specifically, the system slides on the daily passenger flow time series with a window length of 7 days (i.e. a natural week). For each day in the sequence, its moving average is calculated by taking the arithmetic mean of the passenger flow data of that day and the previous six days. This method can effectively filter out abnormal fluctuations caused by accidental weather, single-day events, etc., and better reveal the stable trend and periodic changes (such as weekend effect) of passenger flow in the medium and short term, making the passenger flow comprehensive index (TCI) calculated based on this average more representative and predictable.

[0063] S4: Based on the passenger flow comprehensive index, calculate the rental value score of each shop, and the rental value score is used to evaluate the rental value potential of the shop.

[0064] The calculation of the rental value score is a key step to directly link passenger flow data with commercial value. This score aims to more comprehensively evaluate the rental value potential of the shop, and its calculation is not only based on the passenger flow comprehensive index reflecting passenger flow, but also deeply integrates the conversion ability index reflecting the operating efficiency of the shop. Specifically, the system will obtain the sales data of each shop from the preprocessed data, and calculate a conversion rate index based on this, which measures the success rate of the shop in converting potential passenger flow into actual consumption. Then, a comprehensive index is coupled with the conversion rate index, and a specific function model is used to amplify the advantages of high-value shops. This model ensures that the score grows nonlinearly, i.e. for high-quality shops with high passenger flow comprehensive index and conversion rate, the score will increase more, thus producing more significant differentiation in pricing evaluation.

[0065] In a possible implementation, S4 specifically includes:

[0066] S401: Obtain sales data of each shop from preprocessed data;

[0067] S402: Calculate the passenger flow conversion rate CR, where CR is defined as the ratio of sales to passenger flow;

[0068] S403: Calculate the rental value score using the following formula:

[0069]

[0070] Where RVS represents Rental Value Rating, TCI represents Passenger Traffic Index, CR represents Passenger Traffic Conversion Rate, and TC represents Passenger Traffic Conversion Rate. avg This represents the average of the overall customer traffic index for all shops.

[0071] To more accurately couple customer traffic value with commercial conversion efficiency when calculating the Rental Value Score (RVS), a key financial indicator, Customer Traffic Conversion Rate (CR), is introduced. This indicator is obtained by retrieving the total monthly sales and total monthly customer traffic from preprocessed data and dividing the two. Its physical meaning is the average sales generated per visitor, directly reflecting the store's operational efficiency. Subsequently, the previously obtained Comprehensive Customer Traffic Index (TCI) and Conversion Rate (CR) are geometrically averaged (the square root of the product). This ensures that the score is influenced by a balanced approach of customer traffic base and conversion efficiency. To further highlight the value of top-tier stores, the calculation model also introduces an exponential amplification factor. This factor uses the average TCI of all stores (TC_avg) as a benchmark and exponentially calculates the ratio of the current store's TCI to this average. This means that the higher a store's TCI is than the average, the more significant the additive effect obtained through the exponential function, thus allowing the final rental value score to non-linearly reflect its superior rental value potential.

[0072] In one possible implementation, the customer traffic conversion rate CR in S402 is calculated based on monthly sales and monthly customer traffic.

[0073] To ensure the robustness and commercial applicability of the Customer Conversion Rate (CR) metric, the sales revenue and customer traffic data used in its calculation are calculated on a monthly basis. In practice, the system extracts the total sales revenue and total customer traffic for each shop within the past full calendar month from the pre-processed data. Then, the monthly sales revenue is divided by the total monthly customer traffic to obtain the monthly customer conversion rate (CR = monthly sales revenue / monthly customer traffic). Using a monthly period avoids the significant fluctuations associated with daily calculations (such as the difference between weekdays and weekends) and the potential incompleteness of the period in weekly calculations, resulting in a more stable conversion rate metric that better reflects the long-term profitability of shops, which is then used for subsequent rental value scoring calculations.

[0074] S5: Evaluate the rental price of each shop based on the rental value score, where the rental price is positively correlated with the rental value score.

[0075] The rental pricing assessment is the process of generating a final rental plan based on the aforementioned quantitative scoring. This step establishes a mapping relationship between rental value scores and rental prices. This relationship is pre-set to be positively correlated, meaning that the higher the score of a shop, the higher its assessed rental price per unit area should be. To achieve the conversion from score to specific rental price, the system calls a pricing model based on historical data. This model is trained using machine learning or statistical regression methods on the final transaction rents of a large number of shops in history and their corresponding multiple indicators (including the calculated rental value score). During the assessment, the rental value score of the shop to be assessed is input into this trained model, which then outputs a suggested rental price range or specific value that conforms to market rules and the overall pricing strategy of the commercial complex.

[0076] In one possible implementation, assessing rental pricing in S5 includes mapping rental value scores to a preset rental range, the mapping being based on a historical rental data model.

[0077] The process of mapping the calculated Rental Value Score (RVS) to specific rental pricing relies on a pre-built historical data-driven model. This model is either a machine learning-based regression model or a statistically fitted curve function. Its training data comes from historical rental data of commercial complexes over many years, including the final transaction rents (rent per unit area) of a large number of leased shops and various indicator data of these shops before the leases took effect (including calculated RVS, location, area, etc.). Through training, the model learns the complex non-linear mapping relationship between RVS and market-accepted rental prices. In practical applications, the system only needs to input the RVS value of the shop to be evaluated into this trained model, and the model will automatically output a reasonable suggested rental price or price range that conforms to market rules. This method ensures that the pricing assessment is not subjective speculation, but is supported by solid historical market data.

[0078] In one specific implementation, the machine learning regression model can be trained using algorithms such as Gradient Boosting Decision Tree (GBDT), Random Forest, or Support Vector Regression (SVR). The model training process includes: collecting a historical dataset where each sample contains features such as a shop's Rental Value Score (RVS), location category, and area size, along with the corresponding actual rent per unit area as a label; then dividing the dataset into training and test sets; using the training set to train the selected algorithm to minimize the error (e.g., mean squared error) between the predicted rent and the actual rent; and finally, using the test set to evaluate the model's generalization performance and determine the optimal hyperparameters through methods such as cross-validation.

[0079] In one possible implementation, the method provided in this embodiment further includes:

[0080] S6: Based on the assessed lease pricing, generate a lease recommendation report, which includes pricing details and recommended strategies.

[0081] After completing the rental pricing assessment for all shops, the system automatically generates a report. This step structures the core inputs of the assessment process, the calculation results of intermediate indicators, and the final pricing recommendations into a detailed rental recommendation report. The main content of the report typically includes, but is not limited to: each shop's number, location, area, calculated Comprehensive Traffic Index (TCI), Rental Value Score (RVS), and detailed pricing information such as the assessed recommended unit rent and total rent. Furthermore, the report generates strategic recommendations based on the assessment results. For example, for shops with extremely high scores, a competitive bidding process is recommended; for shops with scores that need improvement, it may be suggested to include rent-free periods or other incentives to attract tenants, thus providing direct data support and action guidelines for management's rental decisions.

[0082] In one possible implementation, the rental recommendation report in S6 also includes visualization charts and comparative analysis sections to display the rental value assessment results of different shops.

[0083] To enhance intuitiveness and decision-making efficiency, the generated leasing recommendation report integrates a wealth of visualization components and comparative analysis modules. The visualization charts include, but are not limited to: using bar charts or radar charts to compare key indicators (TCI, RVS, etc.) across different shops; using heat maps to visually display the distribution of rental prices on the floor plan of the commercial complex; and using line charts to show historical trends in customer traffic and sales for key shops. The comparative analysis section aims to provide deeper insights, such as grouping and comparing assessment results by business type (e.g., catering, retail, entertainment) to analyze the rental affordability and value contribution of different business types; or conducting year-on-year and month-on-month analyses to reveal the changing trends in the value of each shop. These visualizations and analyses enable report users to quickly grasp the overall situation, identify high-quality assets and potential problems, thereby making more scientific and efficient asset management decisions.

[0084] In terms of technical implementation, the rental recommendation report can be automatically generated in HTML format by the backend server using a template engine (such as Apache FreeMarker or Thymeleaf), or it can be generated as a PDF document by integrating a dedicated report generation tool (such as JasperReports). The visualization charts can be dynamically rendered from JSON format data into interactive charts by calling the API interfaces of frontend charting libraries (such as ECharts, D3.js, or Chart.js).

[0085] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0086] (1) In this invention, by comprehensively collecting multi-source heterogeneous customer flow data such as Wi-Fi probes, video surveillance, and POS transactions, a comprehensive and three-dimensional customer flow profile is constructed, overcoming the limitations of a single data source, providing unprecedented data breadth and depth for rental pricing, and laying the data foundation for objective evaluation;

[0087] (2) In this invention, by creating two core indicators, the Comprehensive Customer Flow Index (TCI) and the Rental Value Score (RVS), the multi-dimensional characteristics of customer flow, such as quantity, stability, and conversion efficiency, are coupled nonlinearly for the first time, which scientifically quantifies the real value potential of shops and realizes a leap from experience-based judgment to data-driven intelligent decision-making.

[0088] (3) In this invention, the rental value score is automatically mapped to a specific rental pricing suggestion by a machine learning model trained based on historical data, so that the pricing process can learn market rules autonomously, dynamically adapt to market changes, and finally output an accurate pricing scheme that conforms to market conditions and maximizes the revenue of the commercial complex.

[0089] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0090] The following points need to be explained:

[0091] (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0092] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the present invention; that is, these drawings are not drawn to actual scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element, or there may be intermediate elements.

[0093] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0094] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex, characterized in that, include: S1: Obtain multi-source customer flow data from multiple shops within the commercial complex, including Wi-Fi probe data, video surveillance data, and POS transaction data; S2: Clean and preprocess the multi-source passenger flow data to remove noise and outliers, and achieve data fusion; S3: Based on the preprocessed data, calculate the comprehensive customer flow index for each shop, which is used to quantify the quantity and quality characteristics of customer flow; S4: Based on the comprehensive passenger flow index, calculate the rental value score for each shop, which is used to assess the rental value potential of the shop; S5: Based on the rental value score, assess the rental price of each shop, where the rental price is positively correlated with the rental value score.

2. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 1, characterized in that, S3 specifically includes: S301: Extract the daily customer flow time series data for each shop from the preprocessed data; S302: Calculate the average, peak, and standard deviation of daily passenger flow; S303: Calculate the overall passenger flow index using the following formula: TCI stands for Comprehensive Passenger Flow Index. σ represents the average daily passenger flow in person / day, P represents the peak daily passenger flow in person / day, σ represents the standard deviation of daily passenger flow in person / day, and T represents the number of monitoring days in days.

3. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 1, characterized in that, S4 specifically includes: S401: Obtain the sales data for each store from the preprocessed data; S402: Calculate the customer traffic conversion rate CR, where CR is defined as the ratio of sales revenue to customer traffic. S403: Calculate the rental value score using the following formula: Where RVS represents Rental Value Rating, TCI represents Passenger Traffic Index, CR represents Passenger Traffic Conversion Rate, and TC represents Passenger Traffic Conversion Rate. avg This represents the average of the overall customer traffic index for all shops.

4. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 1, characterized in that, include: The multi-source passenger flow data in S1 also includes mobile device Bluetooth data and social media check-in data.

5. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 1, characterized in that, include: The preprocessing in S2 includes data normalization, missing value imputation, and time alignment operations.

6. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 1, characterized in that, include: The S5 process for assessing rental pricing includes mapping rental value scores to a preset rental range, with the mapping based on a historical rental data model.

7. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 2, characterized in that, include: In step S302, the average daily passenger flow is calculated using the moving average method, with a moving average window size of 7 days.

8. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 3, characterized in that, include: In S402, the customer traffic conversion rate CR is calculated based on monthly sales revenue and monthly customer traffic.

9. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 1, characterized in that, Also includes: S6: Based on the assessed lease pricing, generate a lease recommendation report, which includes pricing details and recommended strategies.

10. The method for evaluating shop rental pricing based on multi-source customer flow data in a commercial complex according to claim 9, characterized in that, include: The rental recommendation report in S6 also includes visualization charts and comparative analysis sections to display the rental value assessment results of different shops.

Images