Shopping mall passenger flow volume prediction method based on historical data auxiliary scene analysis

Abstract

The invention discloses a shopping mall passenger flow volume prediction method based on historical data auxiliary scene analysis, relates to the technical field of passenger flow volume prediction, and solves the technical problems that the analysis of a single factor has one-sided analysis and the prediction result has deviation due to the fact that multi-dimensional factors cannot be integrated for analysis. The missing values in the passenger flow volume data and the weather data are reasonably estimated and filled according to the characteristics of the data and the surrounding information, so that the interference of wrong data on the prediction result is avoided, various influence factors of the shopping mall passenger flow volume are considered, and the prediction result is more accurate by deeply analyzing the complex relationship between the factors and the passenger flow volume. According to the method, a more comprehensive and accurate prediction model is constructed, when the relation between the influence factors and the passenger flow volume is analyzed, a scientific statistical analysis method is applied, for example, a multiple linear regression model is combined with mean absolute error evaluation and similar scenes in historical data are combined for passenger flow volume prediction, and the overall prediction accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of passenger flow prediction, and specifically to a shopping mall passenger flow prediction method based on historical data-assisted scenario analysis. Background Art

[0002] In today's commercial operation field, the accurate prediction of shopping mall passenger flow is of crucial significance for the effective management and operation decision-making of shopping malls. With the rapid development of information technology, a large amount of shopping mall operation data has been collected and stored, which provides the possibility for data-driven passenger flow prediction.

[0003] The patent with the publication number CN112465566A discloses a shopping mall passenger flow prediction method based on historical data-assisted scenario analysis, which predicts the passenger flow in combination with the distribution of the commercial areas around the shopping mall, takes into account the sources and flows of passenger flows, first divides the link area, scenario and time period, then calculates the historical data, considers the sources and destinations of passenger flows, and obtains the time-related coefficient of passenger flow; based on the event-related coefficient and the method of smooth prediction, comprehensively predicts the passenger flow of the day to be predicted, and the prediction is more accurate.

[0004] However, when some existing shopping mall passenger flow prediction methods are used, they often focus on a single factor for prediction, may only consider the simple trend of historical passenger flow data, and fail to fully integrate these multi-dimensional influencing factors, resulting in a large deviation between the prediction result and the actual situation. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a shopping mall passenger flow prediction method based on historical data-assisted scenario analysis, which solves the problem that the analysis of a single factor is one-sided and the prediction result is deviated due to the failure to integrate multi-dimensional factors for analysis.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A shopping mall passenger flow prediction method based on historical data-assisted scenario analysis, and this method specifically includes the following steps:

[0007] Step 1: Obtain the shopping mall information of the target shopping mall and the corresponding shopping mall historical data, where the shopping mall information includes the business hours of the shopping mall;

[0008] Step 2: Analyze the shopping mall passenger flow according to the obtained shopping mall information and shopping mall historical data, and perform segmented processing on the passenger flow corresponding to different time periods to obtain segmented time periods;

[0009] Step 3: Obtain the classified peak passenger flow segments and normal passenger flow segments, at the same time analyze the existing passenger flow influencing factors in combination with historical data, calculate the corresponding influence factors respectively, and then calculate the overall influence factor;

[0010] Step 4: Based on the obtained overall influence factor, predict the passenger flow of the target shopping mall. By analyzing the prediction time and combining the influencing factors corresponding to the time, comprehensively calculate the predicted passenger flow and generate prediction information.

[0011] Step 5: Display the obtained prediction information to the corresponding operator.

[0012] As a further solution of the present invention, the specific method for obtaining the segmented time periods in Step 2 is as follows:

[0013] Obtain the business hours of the shopping mall, divide the business hours to obtain unit time periods, then obtain the passenger flow of the shopping mall corresponding to the unit time periods according to the historical data of the shopping mall, and at the same time obtain the passenger flow of the shopping mall corresponding to different unit time periods in the business data, and so on to classify the unit time periods into peak passenger flow segments and normal passenger flow segments.

[0014] As a further solution of the present invention, the specific method for analyzing the passenger flow influencing factors in Step 3 is as follows:

[0015] Obtain the number of people corresponding to the peak passenger flow segment and the normal passenger flow segment, and calculate the total passenger flow corresponding to the target shopping mall. Then analyze different influencing factors in the historical data, and the influencing factors include weather data, date information, and promotional activities.

[0016] Analyze the weather data, collect the historical passenger flow data and the corresponding weather data, clean the collected data, use each element in the weather data as the horizontal axis and the passenger flow as the vertical axis to draw a scatter plot, and establish a multiple linear regression model. Specifically, y 1 = β 0 + β 1 x 1 + β 2 x 2 +… β n x n + θ, where y 1 is the weather passenger flow, x is the weather element, β 0 is the intercept, β 1 is the regression coefficient, and the specific value of the regression coefficient is set by the operator, and θ is the error term.

[0017] And so on, calculate the weather passenger flow corresponding to different weather data, and calculate the numerical mean value corresponding to the weather passenger flow. Calculate the difference between the calculated numerical mean value and the passenger flow y 0 corresponding to the normal situation, and at the same time calculate the difference between the calculated difference and the change value of the weather data.

[0018] Similarly to the above analysis, the impact factors corresponding to the date information and promotional activities are calculated and analyzed respectively, and are denoted as y 2 and y 3 , and the overall impact factor is calculated by synthesizing the obtained impact factors.

[0019] As a further solution of the present invention, the specific method for calculating the overall impact factor by synthesizing the impact factors obtained in step three is as follows:

[0020] Next, substitute the obtained different impact factors into the formula to calculate the overall impact factor Q, where Sx i is the standard deviation of the independent variable x i , and Sy is the standard deviation of the dependent variable.

[0021] As a further solution of the present invention, the specific method for predicting the passenger flow of the target shopping mall in step four is as follows:

[0022] Obtain prediction information and classify the prediction information. If the prediction information is a single prediction, a single prediction signal is generated. On the contrary, if the prediction information is a comprehensive prediction, a comprehensive prediction signal is generated. Then, analyze the single prediction signal and the comprehensive prediction signal respectively.

[0023] As a further solution of the present invention, the specific method for analyzing the single prediction signal in step five is as follows:

[0024] Obtain the prediction time, and at the same time obtain the time information corresponding to the prediction time, and obtain the weather data corresponding to the time information. Then, obtain the promotional information, and synthesize the obtained time information, weather data, and promotional information to judge the specific factors having an impact and calculate the overall impact factor;

[0025] Next, obtain h identical scenarios corresponding to the time information according to the historical data, and at the same time obtain the passenger flows corresponding to the h identical scenarios and denote them as Lh, and calculate the overall passenger flow mean and denote it as Lp. Take the calculated passenger flow mean Lp as the standard value, and at the same time perform a multiplication operation on the obtained overall impact factor and the standard value Lp to obtain the predicted passenger flow, and generate prediction information.

[0026] As a further solution of the present invention, the specific method for analyzing the comprehensive prediction signal in step four is as follows:

[0027] Similarly to the analysis method of the single prediction signal, analyze and calculate all the single prediction quantities, and at the same time sum up the calculated single prediction quantities to obtain the comprehensive predicted passenger flow, and generate prediction information.

[0028] As a further solution of the present invention, the passenger flow data is obtained from the access control system of the shopping mall, the monitoring device statistical software or manual counting records, while the weather data is obtained from the records of the local meteorological department or professional meteorological data service providers, and the weather data includes multiple meteorological elements such as temperature, humidity, whether it is raining / snowing, wind speed, air pressure, and sunshine duration.

[0029] The present invention provides a method for predicting the passenger flow of a shopping mall based on historical data-assisted scenario analysis. Compared with the prior art, it has the following beneficial effects:

[0030] In the present invention, for the missing values in the passenger flow data and weather data, reasonable estimation and filling are respectively carried out according to the characteristics of the data and surrounding information, avoiding the interference of incorrect data on the prediction results, considering various influencing factors of the passenger flow of the shopping mall, and by deeply analyzing the complex relationship between these factors and the passenger flow, a more comprehensive and accurate prediction model is constructed. When analyzing the relationship between the influencing factors and the passenger flow, scientific statistical analysis methods are used, such as the multiple linear regression model combined with the mean absolute error evaluation, etc., which can accurately determine the influencing factors of each factor, laying a solid foundation for constructing a high-quality prediction model, and can accurately calculate the overall influencing factor according to the specific date, weather data and promotional activity information corresponding to the prediction time, and combine the similar scenarios in the historical data to predict the passenger flow. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a diagram of the method steps of the present invention;

[0032] Figure 2 It is a flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0034] Example 1, please refer to Figure 1 and Figure 2 , the present application provides a method for predicting the passenger flow of a shopping mall based on historical data-assisted scenario analysis, and the method specifically includes the following steps:

[0035] Step 1: Obtain the shopping mall information of the target shopping mall and the corresponding historical data of the shopping mall, where the shopping mall information includes the business hours of the shopping mall.

[0036] Step 2: Analyze the passenger flow of the mall based on the obtained mall information and historical mall data, and perform segmented processing according to the passenger flow corresponding to different time periods to obtain segmented time periods.

[0037] Obtain the business hours of the mall, divide the business hours to get unit time periods, and here the unit time period is represented as one hour. For example, the business hours of a certain mall are from 9 am to 9 pm, a total of 12 hours. Divide this business hour into unit time periods of one hour each. That is to say, a series of consecutive unit time periods with a duration of 1 hour such as 9:00 - 10:00, 10:00 - 11:00, 11:00 - 12:00, etc. will be obtained. Then, according to the mall historical data, obtain the passenger flow of the mall corresponding to the unit time period, and the passenger flow of the mall is represented as the number of people entering the mall. For example, after data query and collation, it is found that within the past month of this mall, the average passenger flow during the unit time period of 9:00 - 10:00 is 50 people, the average passenger flow of 10:00 - 11:00 is 80 people, the average passenger flow of 11:00 - 12:00 is 120 people, and so on. At the same time, obtain the passenger flow of the mall corresponding to different unit time periods in the business data, and classify the unit time periods in this way to obtain peak passenger flow periods and normal passenger flow periods.

[0038] Through comprehensive analysis of the historical passenger flow data of this mall and combined with the actual operation situation, it is found that during the time periods when the average passenger flow is 150 people or more, the people in the mall are relatively dense and the shopping atmosphere is enthusiastic. These time periods can be classified as peak passenger flow periods; while the time periods with an average passenger flow lower than 150 people are classified as normal passenger flow periods. Suppose during the above-mentioned time periods, the average passenger flow during time periods such as 11:00 - 12:00, 14:00 - 15:00, 18:00 - 19:00, etc. all reach or exceed 150 people, then these time periods belong to peak passenger flow periods; while time periods with a passenger flow lower than 150 people such as 9:00 - 10:00, 10:00 - 11:00 belong to normal passenger flow periods.

[0039] Step 3: Obtain the classified peak passenger flow periods and normal passenger flow periods, and at the same time analyze the existing passenger flow influencing factors in combination with historical data, calculate the corresponding influence factors respectively, and then calculate the overall influence factor.

[0040] Obtain the number of people corresponding to the peak passenger flow periods and normal passenger flow periods, and calculate the total passenger flow of the target mall. Here, the average passenger flow is represented as the total passenger flow corresponding to the target mall during the business hours. Then, analyze different influencing factors in the historical data, and the influencing factors include weather data, date information, and promotional activities;

[0041] Analyze weather data, collect historical passenger flow data and corresponding weather data. The passenger flow data can be obtained from the mall's access control system, surveillance equipment statistics software, or manual counting records. The weather data can be obtained from the records of the local meteorological department or professional meteorological data service providers, including multiple meteorological elements such as temperature, humidity, whether it is raining / snowing, wind speed, air pressure, sunshine duration, etc. Clean the collected data, check whether the passenger flow data and weather data are complete. For missing data points, appropriate methods can be used to fill them. For missing values in the passenger flow data, if the passenger flow in adjacent time periods is relatively stable, the average value of the passenger flow in adjacent time periods can be used to fill it. For missing values in the weather data, they can be estimated according to meteorological laws or using data from other surrounding meteorological stations. Take each element in the weather data as the horizontal axis (such as temperature), and the passenger flow as the vertical axis to draw a scatter plot. Considering that the passenger flow may be affected by the combined influence of multiple weather elements, a multiple linear regression model can be established. Specifically, y 1 =β 0 +β 1 x 1 +β 2 x 2 +…β n x n +θ, where y 1 is the weather passenger flow (dependent variable), x is the weather element (such as temperature, as the independent variable), β 0 is the intercept, β 1 is the regression coefficient, and the specific value of the regression coefficient is set by the operator. θ is the error term, and the error term here is the mean absolute error, specifically expressed as the average value of the absolute value of the difference between the predicted value and the actual value;

[0042] For example, there are 5 samples, and the actual passenger flows are 100, 120, 110, 90, and 105 respectively, and the predicted passenger flows are 95, 115, 100, 90, and 100 respectively. For the first sample, the absolute value of the difference between the predicted value and the actual value is 5. Similarly, calculate the absolute values of the differences for the remaining samples to get 5, 10, 0, and 5. Then, according to the formula the mean absolute error is calculated to be 5.

[0043] And so on, calculate the weather passenger flows corresponding to different weather data, and calculate the numerical mean of the weather passenger flows. Calculate the difference between the calculated numerical mean and the passenger flow y 0 Here, the normal situation is set by the operator himself. For example, when the temperature is 20°C and there is no rain, the corresponding passenger flow in the normal situation is also the mean value. The difference calculation is the mean of the weather passenger flows - the passenger flow y 0Obtain the corresponding difference, and at the same time calculate the calculated difference with the change value of weather data. The specific calculation method is to divide the difference by the change value of weather data. For example, if the calculated difference is 30 and the corresponding change value of weather data is 3, the obtained impact factor is 10;

[0044] For example, collect the mall passenger flow data for each unit time period (such as every hour) under different weather conditions (such as sunny, cloudy, rainy, snowy, and different temperature ranges, etc.) in the past year. Assume that in the weather condition of temperature between 15 - 18°C and no precipitation, the passenger flow data in the morning of the past year is as follows: [80, 85, 78, 82, 75, 88, 90, 83, 79, 86]. Calculate the numerical mean of these specific weather passenger flow data. Taking the above data as an example, the calculated mean is 82.6. Here, the normal situation is set as the weather scenario of temperature 20°C and no rain. Similarly, calculate the passenger flow mean under this normal situation through historical data, assume it is 90, that is, calculate the difference between the weather passenger flow mean and the normal situation passenger flow 82.6 - 90 = -7.4. The temperature changes from the normal 20°C to the current analyzed average temperature 16.5°C (the middle value of 15 - 18°C), and the temperature change value is 3.5. Divide the difference by the weather data change value to get -2.11.

[0045] In the same analysis method as above, calculate and analyze the impact factors corresponding to the date information and promotional activities respectively, and denote them as y 2 and y 3 Then, substitute the obtained different impact factors into the formula Calculate the overall impact factor Q, where Sx i is the standard deviation of the independent variable x i and Sy is the standard deviation of the dependent variable (passenger flow).

[0046] Step 4: Predict the passenger flow of the target mall based on the obtained overall impact factor. Through analyzing the prediction time and combining the influencing factors corresponding to the time, calculate the predicted passenger flow comprehensively and generate prediction information.

[0047] Obtain the prediction information, and here the prediction information specifically refers to the corresponding prediction time, such as predicting the passenger flow of the target mall three days later, and classify the prediction information. If the prediction information is a single prediction, generate a single prediction signal. On the contrary, if the prediction information is a comprehensive prediction, generate a comprehensive prediction signal. And here, the single prediction means predicting the passenger flow of only one day, while the comprehensive prediction means predicting for multiple days. Then analyze the single prediction signal and the comprehensive prediction signal respectively;

[0048] Analyze a single prediction signal to obtain the prediction time, and at the same time obtain the time information corresponding to the prediction time. Here, the time information specifically represents the corresponding date, and the current time point is used as the prediction starting point. Then obtain the weather data corresponding to the time information. The weather data is obtained from the records of the local meteorological department or a professional meteorological data service provider. The prediction time of the specific weather data is within seven days to ensure accurate values. Next, obtain the promotion information. Integrate the obtained time information, weather data, and promotion information to determine the specific influencing factors and calculate the overall influence factor. Here, the overall influence factor is calculated by calculating the specific influencing factors. For example, if the weather and date have an impact, then substitute the corresponding influence factors during the calculation to calculate the overall influence factor;

[0049] Suppose the current time point is December 15, 2024, and the prediction time is December 18, 2024 (the day after tomorrow). Then the corresponding date information is December 18. This day is Thursday. The obtained weather data shows that the weather on December 18 is sunny, and the temperature is 10 - 15°C. At the same time, assume that it is found that some merchants are having a "full reduction" promotion activity on that day;

[0050] Through historical data analysis, it can be seen that the passenger flow on Thursday is usually slightly lower than that on weekends. Let the date influence factor of Thursday relative to the average passenger flow be -0.2 (here, the influence factor is a relative value obtained through statistical methods such as regression analysis of a large amount of historical data, indicating the increasing or decreasing trend of the passenger flow);

[0051] For the weather data, according to past data research, it is found that when the temperature is 10 - 15°C and it is sunny, the passenger flow will increase slightly compared to normal weather. Let the weather influence factor be 0.1. In terms of promotion activities, past experience shows that similar "full reduction" promotion activities can increase the passenger flow by about 30%. Converted to an influence factor, it is 0.3. Since the date, weather, and promotion information all have an impact on the passenger flow, calculate the overall influence factor: 1 + (-0.2) + 0.1 + 0.3 = 1.2.

[0052] Next, obtain h identical scenarios corresponding to the time information according to historical data. The specific value of h is set by the operator. Here, the value of h is 5. At the same time, obtain the passenger flows corresponding to the h identical scenarios and record them as Lh, and calculate the overall average passenger flow and record it as Lp. Take the calculated average passenger flow Lp as the standard value. At the same time, multiply the obtained overall influence factor by the standard value Lp to obtain the predicted passenger flow and generate prediction information;

[0053] From the historical customer flow database of the shopping mall, extract the customer flow data of every Thursday in the past five years with similar weather conditions (temperature between 10-15℃ and sunny) and similar promotional activities ("full-reduction" promotion or equivalent promotional efforts). Assume that the operator sets the data for 5 identical scenarios, the customer flows corresponding to these five scenarios are 1200 people, 1300 people, 1250 people, 1180 people, and 1320 people, respectively. Then calculate the mean of these customer flows Lp=1250, and multiply the mean of the customer flows by the overall influencing factor to obtain the predicted customer flow of 1250×1.2=1500.

[0054] The comprehensive prediction signal is analyzed, and similarly, the single prediction signal is analyzed to obtain single prediction information, and then the single prediction information is summed to obtain the predicted passenger flow, and the prediction information is generated at the same time.

[0055] For example, the predicted passenger flow tomorrow (December 16, Wednesday) is 1,300 people, the passenger flow the day after tomorrow (December 17, Thursday) is 1,500 people, and the passenger flow the day after tomorrow (December 18, Friday) is 1,400 people. Further summing up the predicted passenger flow is 4,200.

[0056] Step 5: Display the obtained prediction information to the corresponding operator.

[0057] Embodiment 2, as Embodiment 2 of the present invention, is implemented on the basis of Embodiment 1, and differs from Embodiment 1 in that:

[0058] The value of the weather data in step 4 is different. In the first embodiment, the value is seven days, but the specific value needs to be determined in combination with the actual situation in the corresponding time period. For example, if the prediction is about the change in the flow of customers in the mall in the next month, since the accuracy of the weather forecast will decrease over time, the value of the weather data for the longer term (such as half a month later) may only refer to the forecast information within three days to ensure that the data has a high degree of credibility and reference value.

[0059] For near-term forecasts (e.g. within a week), five days of weather data can be used as a reference to more comprehensively consider the potential impact of weather factors on passenger flow.

[0060] For another example, when predicting the customer flow of a shopping mall during a specific holiday peak season (such as the Spring Festival holiday), since the Spring Festival holiday is relatively short and weather changes have a greater impact on travel and shopping intentions, the weather data value range can be set to the length of the holiday. For example, for a seven-day Spring Festival holiday, seven days of weather data will be used to accurately analyze the correlation effect between weather fluctuations and customer flow during that period, thereby providing more targeted data support for the mall's operational decisions during special periods.

[0061] Example 3 focuses on combining the implementation processes of Example 1 and Example 2 for implementation.

[0062] Some of the data in the above formula are numerically calculated by removing their dimensions, and the content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0063] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A shopping mall passenger flow prediction method based on historical data assisted scenario analysis, characterized in that: The method specifically comprises the following steps: Step 1: Obtain the mall information of the target mall and the corresponding mall historical data, wherein the mall information includes the mall's business hours; Step 2: Analyze the shopping mall customer flow according to the obtained shopping mall information and shopping mall historical data, and segment the customer flow corresponding to different time periods to obtain segmented time periods; Step 3: Obtain the peak passenger flow segments and normal passenger flow segments obtained by classification, analyze the existing passenger flow influencing factors in combination with historical data, calculate the corresponding influencing factors respectively, and then calculate the overall influencing factor; Step 4: Predict the target shopping mall’s passenger flow based on the overall influencing factors obtained, analyze the predicted time, and calculate the predicted passenger flow in combination with the influencing factors of the corresponding time to generate prediction information; Step 5: Display the obtained prediction information to the corresponding operator.

2. The method for predicting shopping mall passenger flow based on historical data-assisted scenario analysis according to claim 1 is characterized in that: The specific method of obtaining the segmented time period in step 2 is: Get the business hours of the mall, and divide the business hours into unit time periods. Then get the mall customer flow corresponding to the unit time period based on the mall's historical data. At the same time, get the mall customer flow corresponding to different unit time periods in the business data. Similarly, classify the unit time periods to get peak customer flow sections and normal customer flow sections.

3. The method for predicting shopping mall passenger flow based on historical data-assisted scenario analysis according to claim 1, characterized in that: The specific method of analyzing the factors affecting passenger flow in step 3 is as follows: Obtain the number of people corresponding to the peak passenger flow period and the normal passenger flow period, and calculate the total passenger flow corresponding to the target shopping mall, and then analyze the different influencing factors in the historical data, and the influencing factors include weather data, date information and promotional activities; Analyze weather data, collect historical passenger flow data and corresponding weather data, clean the collected data, draw a scatter plot with each element in the weather data as the horizontal axis and passenger flow as the vertical axis, and establish a multivariate linear regression model, specifically y1=β0+β1x1+β2x2+…β n x n +θ, where y1 is the weather passenger flow, x is the weather factor, β0 is the intercept, β1 is the regression coefficient, and the specific value of the regression coefficient is set by the operator, and θ is the error term; Similarly, the weather passenger flow corresponding to different weather data is calculated, and the numerical mean corresponding to the weather passenger flow is calculated. The calculated numerical mean is calculated with the passenger flow y0 corresponding to the normal situation, and the calculated difference is calculated with the weather data change value; Similarly to the above analysis, the impact factors corresponding to the date information and promotion activities are calculated and analyzed respectively, and recorded as y2 and y3, and the obtained impact factors are combined to calculate the overall impact factor.

4. The method for predicting shopping mall passenger flow based on historical data assisted scenario analysis according to claim 3 is characterized in that: The specific method of calculating the overall impact factor from the impact factor obtained in step 3 is: Then substitute the different impact factors obtained into the formula The overall impact factor Q is calculated, where Sx i is the independent variable x i Sy is the standard deviation of the dependent variable.

5. The method for predicting shopping mall passenger flow based on historical data assisted scenario analysis according to claim 1, characterized in that: The specific method of predicting the passenger flow of the target shopping mall in step 4 is: Obtain prediction information and classify the prediction information. If the prediction information is a single prediction, generate a single prediction signal. Otherwise, if the prediction information is a comprehensive prediction, generate a comprehensive prediction signal. Then analyze the single prediction signal and the comprehensive prediction signal respectively.

6. The method for predicting shopping mall passenger flow based on historical data-assisted scenario analysis according to claim 5, characterized in that: The specific method of analyzing a single prediction signal in step 5 is: Obtain the forecast time, the time information corresponding to the forecast time, and the weather data corresponding to the time information, then obtain the promotion information, comprehensively determine the specific factors that have an impact, and calculate the overall impact factor. Then, based on the historical data, h identical scenes corresponding to the time information are obtained, and the passenger flow corresponding to the h identical scenes is obtained, recorded as Lh, and the overall passenger flow average is calculated, recorded as Lp. The calculated passenger flow average Lp is used as the standard value, and the obtained overall influencing factor is multiplied by the standard value Lp to obtain the predicted passenger flow, and the prediction information is generated.

7. The method for predicting shopping mall passenger flow based on historical data-assisted scenario analysis according to claim 6, characterized in that: The specific method of analyzing the comprehensive prediction signal in step 4 is: Similarly, the analysis method of a single prediction signal is used to analyze and calculate all single prediction quantities, and the calculated single prediction quantities are summed up to obtain a comprehensive predicted passenger flow, and generate prediction information.

8. The method for predicting shopping mall passenger flow based on historical data-assisted scenario analysis according to claim 3 is characterized in that: The customer flow data is obtained from the shopping mall's access control system, monitoring equipment statistical software or manual counting records, while the weather data is obtained from the records of the local meteorological department or a professional meteorological data service provider, and the weather data includes multiple meteorological elements such as temperature, humidity, whether it is raining / snowing, wind speed, air pressure, and sunshine duration.

Images