A smart business district management method and system based on the Internet of Things

Abstract

The present invention discloses a smart business district management method and system based on the Internet of Things, which relates to the technical field of the Internet of Things. The steps of the management method include: collecting heat indicators of target business districts, transmitting the heat indicators to a database through the Internet of Things technology, and displaying comparison results; its technical points are: inputting the encrypted and stored heat indicators into a heat pointing calculation model, outputting the heat index of each business district, and comparing it with the benchmarked and dynamically set standard threshold to ensure the accuracy and effectiveness of the comparison results. When the load of the corresponding business district is reflected in a targeted manner, that is, when the heat index exceeds the standard, an alarm scheduling strategy is automatically triggered. From the perspective of the business district, a pre-established personnel scheduling plan is called out to effectively deal with congestion or safety accidents in the business district. It can also achieve personnel diversion from the perspective of customers by pushing traffic for other surrounding business districts to ensure the safety of personnel activities.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a smart business district management method and system based on the Internet of Things. Background Art

[0002] The Internet of Things refers to the real-time collection of information on any object or process that needs to be monitored, connected, and interacted with through various information sensors, radio frequency identification technology, global positioning systems, and other technologies, and the realization of ubiquitous connections between objects and objects, and objects and people through network access, to achieve intelligent perception, identification, and management of objects and processes. In the management of smart business districts, the Internet of Things technology plays an important role. For example, through the Internet of Things technology, smart business districts can monitor the activities of each target business district in real time and collect data such as advertising placement, coupon usage, and participation in special activities. This data provides decision-making support for business district managers.

[0003] The technical solution pointed out in the document of the existing application publication number CN115660768A, entitled A Management Method and System for a Smart Business District, includes: obtaining a first online product, which is a product purchased by a first user; obtaining a first traceability code for the first online product; obtaining a first offline sales point set for the first online product based on the traceability code of the first online product; obtaining a second online product, which is different from the first online product; obtaining a second traceability code for the second online product; obtaining a second offline sales point set for the second online product based on the second traceability code; inputting the first offline sales point set and the second offline sales point set into a neural network model to obtain first business district information; after sending the first business district information to the first user, if the first user does not purchase the first online product and / or the second online product in the first business district, promoting the first business district online based on the first online promotional information, thereby solving the technical problems of low vitality and low customer flow in the physical business district in the prior art, but failing to provide an effective solution to the problem of excessive customer flow;

[0004] In combination with the above documents and existing technologies, in the process of smart management of different business districts, such as different shopping malls, traditional methods also collect relevant data such as customer flow and consumption to evaluate the popularity or vitality of the corresponding business district. When the activity in the business district is too high, scheduling operations will be actively carried out. For example, the flow of people in business district A is too high. The video shows that the distance between people is less than 1m and they move slowly, which will cause congestion or even trampling. Although the economy of business district A is instantly boosted, the safety issues caused by excessive customer flow cannot be ignored. Maintenance and scheduling alone cannot guarantee that the corresponding business district will not be overloaded. Therefore, the safety of the operation of each or each type of business district needs to be given sufficient attention. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a smart business district management method and system based on the Internet of Things, which can solve the problems raised in the background technology by implementing a specific and effective management method.

[0007] (2) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A smart business district management method based on the Internet of Things, the steps are as follows:

[0010] Collect the heat index of the target business district and transmit it to the database through IoT technology;

[0011] Pre-process the heat index of each target business district, input it into the established heat index calculation model, output each heat index, compare each heat index with the corresponding set standard threshold, and display the comparison results;

[0012] When the comparison result shows that the standard is exceeded, the alarm scheduling strategy is triggered to manage the flow of people;

[0013] Monitor the activities of each target business district, screen out the target business districts that meet the agreed conditions and do not exceed the standards, and use them as the objects of the activity promotion mechanism. Use machine learning algorithms to evaluate the promotion effect and optimize the activity promotion mechanism in a feedback manner.

[0014] Furthermore, popularity indicators include traffic flow, pedestrian flow and consumption data; among them, consumption data represents: the consumption amount per unit time obtained in real time within the target business district.

[0015] Furthermore, the preprocessing process is to denoise, remove duplication and unify the format of the heat index.

[0016] Furthermore, in the heat direction calculation model, the formula used is:

[0017] ;

[0018] Where, Represents the popularity index of the corresponding target business district, where i represents the number of the corresponding target business district, fc, pr, and cq represent vehicle flow, pedestrian flow, and consumption data, respectively.

[0019] Furthermore, the process of setting the standard threshold is as follows:

[0020] Obtain the scale of the corresponding target business district;

[0021] Build a rule engine and set the following formula to generate standard thresholds :

[0022] ;

[0023] In the formula, k and b are constants, both greater than 0, and ms represents the scale of the corresponding target business district.

[0024] Furthermore, the comparison results include:

[0025] If the heat index of the corresponding target business district exceeds the corresponding set standard threshold, it means that the standard is exceeded;

[0026] If the heat index of the corresponding target business district does not exceed the corresponding set standard threshold, it means that it has not exceeded the standard.

[0027] Furthermore, the triggered alarm dispatching strategy process is: issuing an early warning signal and calling out a pre-made complete personnel dispatching plan, including the number of dispatched personnel, dispatching time and dispatching location.

[0028] Furthermore, the agreed conditions are as follows:

[0029] Condition 1: There is overlap in activity periods;

[0030] Condition 2: With the exceeding business district as the center, other business districts within the set radius do not exceed the standard.

[0031] Furthermore, the content of the promotion mechanism for the activities implemented is as follows:

[0032] Develop additional promotional measures for the selected target business districts, including increasing the amount of advertising;

[0033] The process of using machine learning algorithms to evaluate promotion effectiveness is as follows:

[0034] Collect advertising data: record ad impressions, clicks, and conversions;

[0035] Select features: The number of ad impressions, click-through rate, and conversion rate are selected as features; click-through rate = clicks / impressions, and conversion rate = conversions / clicks;

[0036] Build a regression model: Using selected features as input and sales during the advertising period as output, build a regression model to predict sales. Use historical data to train the model and adjust model parameters in real time to fit the data and predict promotion effectiveness.

[0037] Evaluate promotion effectiveness: Use the trained regression model to predict sales under the current promotion mechanism and compare it with the expected target. If the predicted sales do not exceed the expected target, optimize the promotion mechanism in a feedback manner.

[0038] Furthermore, in the feedback optimization promotion mechanism, the feedback adjustment process is as follows:

[0039] Non-dimensionalize the forecasted sales and expected targets;

[0040] Assuming the additional advertising volume is ΔA, the new advertising volume A_new = A_base + ΔA; where A_base is the additional promotion volume set for the selected target business district, that is, the initial advertising volume;

[0041] Construct the following formula to estimate the additional advertising volume ΔA:

[0042] ;

[0043] Where S_target, S_pred, S_base, and S_initial represent the expected target, the predicted sales under the current promotion, the actual sales under the A_base delivery volume, and the sales before the delivery, respectively.

[0044] A smart business district management system based on the Internet of Things, the system comprising:

[0045] The data collection module collects the popularity index of the target business district and transmits the popularity index to the database through the Internet of Things technology;

[0046] The information processing module pre-processes the popularity indicators of each target business district, inputs the established popularity pointing calculation model, outputs each popularity index, compares each popularity index with the corresponding set standard threshold, and displays the comparison results;

[0047] The scheduling strategy module triggers an alarm scheduling strategy to manage the flow of people when the comparison result shows that the standard is exceeded;

[0048] The promotion optimization module monitors the activities of each target business district, selects the target business districts that meet the agreed conditions and do not exceed the standards, and uses them as the objects for executing the activity promotion mechanism. It also uses machine learning algorithms to evaluate the promotion effect and optimize the activity promotion mechanism in a feedback manner.

[0049] (3) Beneficial effects

[0050] The present invention provides a smart business district management method and system based on the Internet of Things, which has the following beneficial effects:

[0051] (1) This solution ensures the comprehensiveness of data by capturing dynamic information such as vehicle flow and pedestrian flow in real time and connecting it with the consumption data interface to obtain consumption data. At the same time, the blockchain technology is introduced to encrypt and store the data, forming an unchangeable data chain, which effectively guarantees the security and non-tamperability of the data.

[0052] (2) This solution inputs the encrypted and stored heat index into the heat pointing calculation model, outputs the heat index of each business district, and compares it with the benchmark and dynamically set standard threshold to ensure the accuracy and effectiveness of the comparison result. When the load of the corresponding business district is reflected in a targeted manner, that is, when the heat index exceeds the standard, the alarm scheduling strategy is automatically triggered. From the perspective of the business district, the pre-established personnel scheduling plan is called up to effectively deal with congestion or safety accidents in the business district. It can also promote traffic for other surrounding business districts and realize personnel diversion from the perspective of customers to ensure the safety of personnel activities;

[0053] This solution enables real-time monitoring, intelligent assessment, and rapid response to business district popularity, providing a basis for scientific decision-making for business district managers, effectively improving the operational efficiency and safety of the business district, and reducing risks caused by excessive popularity.

[0054] (3) This solution can select target business districts that meet the agreed conditions and do not exceed the standards as promotion targets, and use machine learning algorithms to conduct in-depth evaluation of the promotion effect. When the predicted sales volume does not reach the expected target, the feedback optimization mechanism is used to estimate and increase the advertising volume. This not only improves the accuracy and efficiency of the promotion activities, but also achieves scientific adjustment of the advertising volume and optimizes resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flowchart of the overall steps of a smart business district management method based on the Internet of Things in the present invention;

[0056] Figure 2 This is a modular schematic diagram of a smart business district management system based on the Internet of Things in the present invention. DETAILED DESCRIPTION

[0057] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0058] Example 1:

[0059] See also Figure 1This embodiment provides a smart business district management method based on the Internet of Things. This smart business district management method integrates the Internet of Things, machine learning, algorithm optimization and blockchain technology to achieve unified and targeted management of multiple business districts in the same area. This ensures that the original load of each business district will not exceed the standard and maintain a certain stability. It can also disperse and guide the flow of people during activities, thereby reducing the probability of safety accidents and reflecting the effectiveness of business district management.

[0060] S1. Collect the popularity index of the target business district and transmit it to the database through the Internet of Things technology. The blockchain technology is introduced in the process of collecting and transmitting the popularity index to ensure the security and immutability of the data.

[0061] The target business district mentioned in this application may refer to:

[0062] Shopping malls, commercial clusters or business districts can be selected according to the specific situation;

[0063] Among them, popularity indicators include traffic flow, pedestrian flow and consumption data;

[0064] Sensor network: Various sensors, such as cameras, infrared sensors, and RFID readers, are deployed in each target business district to capture dynamic information on vehicle and pedestrian traffic. These sensors are connected to the database in the data center via IoT technology to ensure real-time data transmission. Consumption data interface: This interface connects with the POS systems and mobile payment platforms in each target business district to obtain consumer consumption data, including the types of goods purchased, the amount spent, and the time of consumption. The consumption data is the amount spent per unit time, collected in real time within the target business district.

[0065] Blockchain technology guarantee:

[0066] In the process of data collection or collection, blockchain technology is introduced to ensure the security and immutability of data. Each piece of data is encrypted at the time of generation and stored on the blockchain, forming an unchangeable data chain. In this way, even if the data is intercepted during transmission, it cannot be tampered with or deleted, ensuring the authenticity and integrity of the data.

[0067] In addition, data quality control can also be included: in addition to ensuring data security, data quality must also be controlled; by setting data verification rules, the collected data (i.e., the popularity index of the target business district) is verified in real time, abnormal data is eliminated, and data accuracy is ensured; data visualization and monitoring: data visualization functions can also be provided in the data center, using charts, dashboards, etc. to display the traffic flow, pedestrian flow, and consumption data of each target business district in real time, providing managers with intuitive decision-making basis;

[0068] For example:

[0069] For business district A, the deployed cameras and infrared sensors capture vehicle and pedestrian traffic in real time, while consumption data is obtained through the POS system. This data is encrypted and stored on the blockchain as it is generated, ensuring data security and non-tamperability. Managers can view the pedestrian and consumption situation in business district A in real time through the data visualization interface, providing a basis for subsequent decision-making.

[0070] By adopting the above technical solution, we can achieve the technical effect of collecting and analyzing the popularity index of the target business district in real time, safely and accurately, and solve the problems of insufficient security, easy data tampering and data quality control in the data collection process.

[0071] Specifically, the solution deploys a sensor network within the target business district to capture dynamic information such as vehicle and pedestrian traffic in real time, and connects it to a consumer data interface to obtain consumption data, ensuring the comprehensiveness of the data. At the same time, blockchain technology is introduced to encrypt and store the data, forming an unchangeable data chain, effectively ensuring data security and non-tampering. Furthermore, by setting data verification rules, the collected data is verified in real time, eliminating abnormal data and ensuring data accuracy. Finally, the data center provides data visualization capabilities, allowing managers to intuitively and real-timely understand the popularity of each target business district, providing a strong basis for subsequent decision-making.

[0072] This technical solution not only improves the efficiency of data collection and analysis, but also enhances the credibility and practicality of the data, providing strong technical support for business district management and operations.

[0073] S2. Preprocess the popularity index of each target business district, input it into the established popularity index calculation model, output each popularity index, compare each popularity index with the corresponding set standard threshold, and display the comparison results;

[0074] The preprocessing process or content is as follows:

[0075] De-noising, de-duplication, and formatting of popularity indicators ensure data accuracy and consistency. Furthermore, data from different sensors is integrated to form a complete business district data view, specifically matching the popularity indicator data view of the target business district.

[0076] The formula running in the heat pointing calculation model is as follows:

[0077] ;

[0078] Where, represents the popularity index of the corresponding target business district, where i represents the number of the corresponding target business district, i = any of the business districts A, B, C, or D, and fc, pr, and cq represent the traffic flow, pedestrian flow, and consumption data, respectively;

[0079] Logic description:

[0080] The above formula is composed of logarithmic function, rational function and exponential function. The logarithmic function can handle a wide range of data, and its growth rate will gradually slow down for larger traffic flow values, which is consistent with the influence of traffic flow on popularity in actual situations; adding 1 is to avoid the logarithmic function being undefined when fc=0; the rational function maps the pedestrian flow pr to a value between 0 and 1, and as pr increases, its growth rate first increases rapidly and then slows down. The square operation is to enhance the influence of pedestrian flow on popularity, so that when the pedestrian flow is large, its contribution to popularity is still significant; the exponential function can grow rapidly and is suitable for representing the impact of consumption data on popularity, because an increase in consumption amount can often quickly increase the popularity of a business district; dividing by 10,000 is to adjust the scale of the consumption data so that the growth rate of the exponential function is within a reasonable range; subtracting 1 is to make the contribution of the exponential function 0 when cq=0;

[0081] Sample data example:

[0082] Assuming that the traffic volume fc in business district A is 500 vehicles / h, the pedestrian flow pr is 200 people / h, and the consumption data cq is 5000 yuan / h, then:

[0083] ;

[0084] ;

[0085] ;

[0086] Therefore, the popularity index of business district A is:

[0087] (keep two decimal places);

[0088] The corresponding process of setting the standard threshold is as follows:

[0089] Get the size of the corresponding target business district, that is, the actual area occupied;

[0090] Build a rule engine and set the following formula to generate standard thresholds :

[0091] ;

[0092] In the formula, k and b are both constants, both greater than 0, and ms represents the scale of the corresponding target business district;

[0093] Logic description:

[0094] The square root function is used because the actual impact of area on the heat index may not be linear. As the area increases, the growth rate of the heat index may gradually slow down. The square root function can capture this nonlinear relationship. Dividing by 100 is a scaling factor used to reduce the impact of area to a reasonable range, so that the heat index remains within a smaller range. k is the slope constant, which determines the extent of the area's influence on the heat index. b is the intercept constant, indicating that even when the area is 0, there is a basic heat index threshold. k and b can be set and adjusted according to actual conditions.

[0095] Sample data example:

[0096] Assuming we set k=2 and b=4, and the size of business district A is 1000 square meters, the formula becomes:

[0097] ;

[0098] Therefore, the standard threshold corresponding to business district A is:

[0099] ;

[0100] Comparing the popularity index of business district A with the corresponding standard threshold, since 4.24 is less than 4.63, the popularity of business district A has not exceeded the standard and is still within the tolerance range of business district A.

[0101] The comparison results include:

[0102] If the heat index of the target business district exceeds the corresponding set standard threshold, it means that the standard is exceeded, which means that the probability of serious congestion or safety accidents in the business district has greatly increased, and timely safety dispatch or handling is required;

[0103] If the heat index of the corresponding target business district does not exceed the corresponding set standard threshold, it means that it has not exceeded the standard, which means that the business district has the ability to handle it on its own. The probability of serious congestion or safety accidents exists, but it is not high, and the business district can handle it on its own.

[0104] S3. When the comparison result shows that the value exceeds the standard, the alarm scheduling strategy is triggered;

[0105] If the comparison result shows that the standard is not exceeded, no response action will be taken;

[0106] The triggered alarm scheduling strategy process or content is as follows:

[0107] The system issues an early warning signal (for example, a flashing indicator light in the data center) and calls up a pre-defined personnel dispatch plan, including the number of dispatched personnel, dispatch time, and dispatch location. The specific dispatch plan is set based on the historical situation of the corresponding business district. For example, if the mall has historically exceeded the standard, two teams of three people each will be dispatched to the crowded dining area at the time of the early warning.

[0108] In addition, in extreme situations, such as severe congestion or safety accidents in the business district, the emergency plan is activated and more urgent measures are taken to respond; for example, adding temporary exits and limiting the number of people entering. Strategy optimization and adjustment: Based on historical data and real-time conditions, the alarm scheduling strategy is continuously optimized and adjusted (the specific adjustment is determined by the success of the historical scheduling strategy) to improve management efficiency and response capabilities. The system can learn from the successful experiences and lessons in historical data to provide a reference for future scheduling strategies.

[0109] By adopting the above technical solutions, we achieved the technical effect of accurately evaluating the popularity of the target business district and intelligently scheduling it, solving the difficult problems of business district popularity monitoring and emergency response.

[0110] Specifically, the solution first pre-processes the heat index to ensure data accuracy and consistency, then inputs it into the heat-directed calculation model, outputs the heat index of each business district, and compares it with the benchmark and dynamically set standard threshold to ensure the accuracy and effectiveness of the comparison results. When the load of the corresponding business district is specifically reflected, that is, the heat index exceeds the standard, the system automatically triggers the alarm scheduling strategy. From the perspective of the business district, it calls up the pre-established personnel scheduling plan and even activates the emergency plan in extreme cases to effectively deal with congestion or safety accidents in the business district. At the same time, it can also promote traffic to other surrounding business districts, realize personnel diversion from the perspective of customers, and ensure the safety of personnel activities.

[0111] This solution enables real-time monitoring, intelligent assessment, and rapid response to business district popularity, providing a basis for scientific decision-making for business district managers, effectively improving the operational efficiency and safety of the business district, and reducing the risks caused by excessive popularity.

[0112] S4. Monitor the activity status of each target business district in real time, select target business districts that meet the agreed conditions and do not exceed the standards, and use them as the targets for implementing the activity promotion mechanism. Use machine learning algorithms to evaluate the promotion effect and optimize the activity promotion mechanism in a feedback-based manner;

[0113] The agreed conditions are as follows:

[0114] Condition 1: There is overlap in activity periods;

[0115] Condition 2: With the exceeding commercial district as the center, other commercial districts within the set radius do not exceed the standard;

[0116] It should be noted that the event period refers to the period during which the business district holds the event;

[0117] For example, Business District A needs to hold an event during the National Day holiday, which runs from October 1st to 3rd. Business Districts B and C also hold events during the National Day holiday, from October 1st to 2nd and October 3rd to 8th, respectively. Therefore, Business Districts A, B, and C all have overlapping event periods, meeting Condition 1.

[0118] If business district A is the one that exceeds the standard, then set the radius to the square of the radius of business district A. If business district B is the only other business district within the set radius, then we only need to determine whether business district B exceeds the standard. If business district B does not exceed the standard, then condition 2 is met.

[0119] The promotion mechanism implemented is as follows:

[0120] Develop additional promotional measures for the selected target business districts, including: increasing the amount of advertising;

[0121] In addition, it can also include issuing a certain amount of coupons and holding special events and performances, but this embodiment only uses the increase of a certain amount of advertising as the specific content of the event promotion; wherein, the system can automatically execute the promotion plan to ensure the smooth progress of the promotion activity;

[0122] Using machine learning algorithms to evaluate promotion effectiveness can achieve in-depth analysis of promotion activity data, thereby more accurately understanding the effectiveness of the activity and providing data support for future promotion activities;

[0123] The following are specific instructions and examples:

[0124] Application of machine learning in promotion effect evaluation:

[0125] Data collection and organization:

[0126] Data Collection: Collect various data related to promotional activities, including advertising data obtained after increasing a certain amount of advertising, including at least the number of impressions, click-through rate, and conversion rate; for the issuance of a certain amount of coupons, corresponding coupon usage data is collected, including at least the number of collections, the number of uses, and the amount of money used; for special events and performances, the corresponding special event participation data includes at least the number of participants, the number of interactions, and the sales during the event; Data Collation: Clean and organize the collected data, remove invalid data, duplicate data, and abnormal data, and ensure data quality and accuracy;

[0127] Feature selection and extraction:

[0128] Feature selection: Select relevant features based on the campaign's goals and evaluation requirements. For evaluating ad effectiveness, select impressions, click-through rate, and conversion rate as relevant features. For evaluating coupon usage, select features such as number of coupons received, number of coupons used, and amount of coupons used. The same principles apply to subsequent evaluations and will not be elaborated on here. Feature extraction: Use machine learning algorithms to extract more advanced features from raw data. For example, clustering algorithms can be used to divide similar consumer groups into different market segments, or dimensionality reduction algorithms can be used to reduce the dimensionality of high-dimensional data for better subsequent analysis.

[0129] Model building and training:

[0130] Model building: Select an appropriate machine learning model, such as a regression model (for predicting continuous values, such as sales and conversion rates), a classification model (for predicting discrete values, such as whether a user will use a coupon or participate in an event), or a clustering model (for discovering potential patterns in the data, such as consumer group segmentation). Model training: Use historical data to train the model and adjust the model parameters to enable the model to better fit the data and predict future results.

[0131] Effect evaluation and prediction:

[0132] Effectiveness evaluation: Use trained models to evaluate the effectiveness of current promotional activities. For example, regression models can be used to predict key indicators such as sales and conversion rates. Classification models can be used to predict whether users will use coupons or participate in special events. Prediction and optimization: Based on the evaluation results, future promotional activities can be predicted and optimized. For example, if a certain advertising channel has a low conversion rate, the advertising strategy can be adjusted to increase advertising in other channels. If a certain coupon has a low usage rate, the coupon amount or distribution method can be adjusted to increase usage.

[0133] In this embodiment, the following is a detailed description of increasing the quantity of advertisement delivery:

[0134] Collect advertising data: record the number of ad impressions, clicks, and conversions (e.g., the number of times users visited a store after viewing an ad) on various platforms (e.g., social media, search engines, local life service platforms, etc.);

[0135] Select features: such as ad impressions, click-through rate (clicks / impressions), and conversion rate (conversions / clicks); Feature extraction: Use clustering algorithms to divide similar consumer groups into different market segments to better understand the consumption preferences and response behaviors of different groups;

[0136] Build a regression model: Using features such as ad impressions, click-through rates, and conversion rates as inputs and sales during the campaign (i.e., the period during which the ads were run) as outputs, a regression model is constructed to predict the sales of the promotional campaign. Train the model: Using historical data (i.e., historical data corresponding to impressions, clicks, and conversions) to train the model and adjust its parameters to better fit the data and predict future results.

[0137] Evaluate promotion effectiveness: Use the trained regression model to predict the sales of the current promotion campaign, compare it with the expected target, and evaluate the overall effectiveness of the campaign. Forecast and optimize: Based on the evaluation results, predict and optimize future promotion activities. For example, if the conversion rate of a certain advertising channel is found to be low, the advertising strategy can be adjusted to increase the amount of advertising in other channels.

[0138] This approach allows the system to evaluate the effectiveness of promotional activities in real time, collect consumer feedback, and provide data support for future promotional activities. Machine learning algorithms can also uncover potential patterns in the data, such as consumer group segmentation and consumption preferences, providing a basis for more accurate marketing decisions.

[0139] When using machine learning algorithms to evaluate promotion effectiveness, the trained regression model is used to predict the sales of the current promotion activity and compare it with the expected target. If the predicted sales do not reach or exceed the expected target, the promotion mechanism is optimized in a feedback manner. The specific feedback adjustment process is as follows:

[0140] Calculations are performed based on dimensionless sales forecasts and expected targets;

[0141] Assuming the additional advertising volume is ΔA, the new advertising volume A_new = A_base + ΔA; where A_base is the additional promotion volume for the selected target business district, including: increasing a certain amount of advertising, that is, the initial advertising volume;

[0142] Since there is a linear relationship between sales and advertising volume, that is: S∝A;

[0143] Then, construct the following formula to estimate the additional advertising volume ΔA:

[0144] ;

[0145] Where S_target, S_pred, S_base, and S_initial represent the target (i.e., the expected sales), the predicted sales under the current promotion, the actual sales under the A_base delivery volume (obtained from existing historical data), and the sales before the launch, respectively.

[0146] Logically, if the predicted sales are lower than expected, ΔA is positive, indicating that more advertising is needed. Estimate the amount of advertising needed to increase sales by one dollar and use that as your baseline efficiency;

[0147] For example:

[0148] Assume S_target=1000, S_pred=800, A_base=500, S_base=700 (sales under A_base delivery) and S_initial (sales before delivery);

[0149] but, ;

[0150] Therefore, it is necessary to increase the advertising volume by approximately 167,000 times on top of the original increase in the fixed amount of advertising. Mille (thousand impressions): Since the amount of advertising can be very large, the advertising industry often uses mille as a unit for ease of calculation.

[0151] Also included: secondary forecasting and channel changes;

[0152] Secondary forecast: Use the new advertising volume A_new to make a secondary sales forecast;

[0153] Compare and adjust: If the second-forecast sales still do not meet the target, consider changing the delivery channel or further adjusting the delivery strategy, usually by changing the channel where the advertisement is delivered.

[0154] By adopting the above technical solutions, we have achieved the technical effect of real-time monitoring and optimization of the activity promotion mechanism, and solved the problem of evaluating and adjusting the promotion effect of business district activities;

[0155] This solution can select target business districts that meet agreed conditions and do not exceed the standards as promotion targets, and use machine learning algorithms to conduct in-depth evaluations of promotion effectiveness. By collecting, organizing, and analyzing advertising data, a regression model is constructed to predict sales, compare them with expected targets, and evaluate campaign effectiveness in real time. If the predicted sales do not meet expectations, the solution uses a feedback optimization mechanism to estimate and increase advertising volume, making secondary forecasts and adjustments until the expected target is achieved.

[0156] This process not only improves the accuracy and efficiency of promotional activities, but also achieves scientific adjustment of advertising volume and optimizes resource allocation. At the same time, the application of machine learning algorithms discovers potential patterns in the data, providing a basis for more accurate marketing decisions. This solution effectively improves the promotion effect of business district activities, reduces promotion costs, and provides business district managers with scientific and efficient promotion strategy formulation and adjustment tools.

[0157] Example 2:

[0158] See also Figure 2 Based on Example 1, this embodiment further provides a smart business district management system based on the Internet of Things, which includes:

[0159] The data collection module collects the popularity index of the target business district and transmits the popularity index to the database through the Internet of Things technology;

[0160] The information processing module pre-processes the popularity indicators of each target business district, inputs the established popularity pointing calculation model, outputs each popularity index, compares each popularity index with the corresponding set standard threshold, and displays the comparison results;

[0161] The scheduling strategy module triggers an alarm scheduling strategy to manage the flow of people when the comparison result shows that the standard is exceeded;

[0162] The promotion optimization module monitors the activities of each target business district, selects the target business districts that meet the agreed conditions and do not exceed the standards, and uses them as the objects for executing the activity promotion mechanism. It also uses machine learning algorithms to evaluate the promotion effect and optimize the activity promotion mechanism in a feedback manner.

[0163] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0164] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0165] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A smart business district management method based on the Internet of Things, characterized in that: Here are the steps: Collect the heat index of the target business district and transmit it to the database through IoT technology; Pre-process the heat index of each target business district, input it into the established heat index calculation model, output each heat index, compare each heat index with the corresponding set standard threshold, and display the comparison results; In the heat direction calculation model, the formula is: ； Where, Represents the popularity index of the corresponding target business district, where i represents the number of the corresponding target business district, fc, pr, and cq represent vehicle flow, pedestrian flow, and consumption data, respectively. The process of setting the corresponding standard threshold is as follows: obtain the scale of the corresponding target business district; build a rule engine and set the following formula to generate the standard threshold : ； In the formula, k and b are constants, both greater than 0, and ms represents the size of the corresponding target business district. When the comparison result shows that the standard is exceeded, the alarm scheduling strategy is triggered to manage the flow of people; Monitor the activities of each target business district, screen out target business districts that meet the agreed conditions and do not exceed the standards, and use them as the targets for implementing the activity promotion mechanism. Use machine learning algorithms to evaluate the promotion effect and optimize the activity promotion mechanism in a feedback manner. The agreed conditions are as follows: Condition 1: There is overlap in the activity period; Condition 2: The other business districts within the set radius of the business district that exceeds the standards do not exceed the standards; In the feedback optimization promotion mechanism, the feedback adjustment process is as follows: Non-dimensionalize the forecasted sales and expected targets; Assuming the additional advertising volume is ΔA, the new advertising volume A_new = A_base + ΔA; where A_base is the additional promotion volume set for the selected target business district, that is, the initial advertising volume; Construct the following formula to estimate the additional advertising volume ΔA: ； In the formula, S_target, S_pred, S_base and S_initial represent the expected target, the predicted sales under the current promotion activity, the actual sales under the A_base delivery volume and the sales before the delivery respectively; the content of the implemented activity promotion mechanism is as follows: formulate additional promotion volume for the selected target business district, including increasing the quantitative advertising delivery; the process of using machine learning algorithms to evaluate the promotion effect is as follows: collect advertising delivery data: record the number of ad impressions, clicks and conversions; select features: use the number of ad impressions, click-through rate and conversion rate as selected features; among which click-through rate = clicks / impressions, conversion rate = conversions / clicks; build a regression model: use the selected features as input and the sales during the advertising period as output to build a regression model to predict sales, and use historical data to train the model. By adjusting the model parameters in real time, the regression model can fit the data and predict the promotion effect; evaluate the promotion effect: use the trained regression model to predict the sales under the current activity promotion mechanism and compare it with the expected target. When the predicted sales do not exceed the expected target, the activity promotion mechanism is optimized in a feedback manner.

2. The method for managing a smart business district based on the Internet of Things according to claim 1, wherein: Popularity indicators include traffic flow, pedestrian flow and consumption data; among them, consumption data represents: the consumption amount per unit time obtained in real time within the target business district.

3. The method for managing a smart business district based on the Internet of Things according to claim 1, wherein: The preprocessing process is to denoise, remove duplicates and unify the format of the heat index.

4. The method for managing a smart business district based on the Internet of Things according to claim 1, wherein: The comparison results include: If the heat index of the corresponding target business district exceeds the corresponding set standard threshold, it means that the standard is exceeded; If the heat index of the corresponding target business district does not exceed the corresponding set standard threshold, it means that it has not exceeded the standard; The triggered alarm dispatch strategy process is: issuing an early warning signal and calling out a pre-made complete personnel dispatch plan, including the number of dispatched personnel, dispatch time and dispatch location.

5. A smart business district management system based on the Internet of Things, used to implement the management method according to any one of claims 1 to 4, characterized in that: The system includes: The data collection module collects the popularity index of the target business district and transmits the popularity index to the database through the Internet of Things technology; The information processing module pre-processes the popularity indicators of each target business district, inputs the established popularity pointing calculation model, outputs each popularity index, compares each popularity index with the corresponding set standard threshold, and displays the comparison results; The scheduling strategy module triggers an alarm scheduling strategy to manage the flow of people when the comparison result shows that the standard is exceeded; The promotion optimization module monitors the activities of each target business district, selects the target business districts that meet the agreed conditions and do not exceed the standards, and uses them as the objects for executing the activity promotion mechanism. It also uses machine learning algorithms to evaluate the promotion effect and optimize the activity promotion mechanism in a feedback manner.

Images