An airport ground traffic passenger flow prediction method and system

Abstract

The application provides an airport ground traffic passenger flow prediction method and system, which is applied to the field of prediction data processing; the application forms a dynamic data pool by integrating flight dynamic data, traffic flow data, weather forecast data and passenger movement data, ensures that the system has a comprehensive and timely information base, enables the system to dynamically update and timely reflect the real-time conditions of the airport and its surroundings, improves the prediction accuracy and response speed, and simultaneously ensures that only when the data is complete and up-to-date, the subsequent prediction and scheduling operation is triggered, avoids inaccurate prediction caused by insufficient or outdated data, and improves the reliability of system decision-making.

Description

Technical Field

[0001] This invention relates to the field of predictive data processing, and in particular to a method and system for predicting passenger flow in airport ground transportation. Background Technology

[0002] Airport ground transportation refers to the traffic flow of passengers in the area surrounding the airport. It covers various modes of transportation used by passengers to depart from or arrive at the airport, such as taxis, private cars, public transportation (subway, bus), shuttle buses, and shared mobility services. This part of the traffic flow is an important part of the overall operation of the airport and a key node in the urban transportation system.

[0003] Currently, although passengers in the area surrounding the airport can choose different modes of transportation (such as taxis, public transportation, and private cars) to enter the airport, the behavioral patterns corresponding to different modes of transportation are complex. These diverse choices make it complicated to predict the usage of different modes of transportation. Traditional traffic flow prediction methods usually assume that passenger behavior is homogeneous and predictable, but in reality, passenger behavior is often highly heterogeneous and uncertain, which cannot meet passenger needs and makes it difficult to improve the overall efficiency of the transportation system. Summary of the Invention

[0004] This invention aims to solve the problem that traditional traffic flow forecasting methods cannot meet passenger demand and are difficult to improve the overall efficiency of the transportation system, and provides a method and system for forecasting passenger flow in airport ground transportation.

[0005] The present invention employs the following technical means to solve the technical problem:

[0006] This invention provides a method for predicting passenger flow in airport ground transportation, comprising:

[0007] Based on the real-time data pre-collected by the airport management terminal, the real-time data is integrated and summarized to form a dynamic data pool. The real-time data specifically includes flight dynamic data, traffic flow data, weather forecast data, and passenger movement data.

[0008] Determine whether the dynamic data pool can execute the data update strategy preset by the airport management terminal;

[0009] If possible, the pre-trained prediction model is applied to perform rolling predictions on the latest data in the dynamic data pool to obtain prediction results. Based on the prediction results, the traffic management measures preset by the airport management terminal are adaptively triggered to dynamically schedule traffic resources in the area surrounding the airport. The airport management terminal receives feedback information uploaded by passengers in the area surrounding the airport. Specifically, the traffic management measures include adjusting traffic light durations, optimizing vehicle scheduling, and issuing congestion warnings. The traffic resources specifically include increasing traffic capacity, opening additional lanes, and activating temporary parking lots.

[0010] Determine whether the feedback information matches the response event corresponding to the area surrounding the airport;

[0011] If so, the flight information pre-entered by the passenger to the airport management terminal is collected, the traffic data corresponding to the response event and the flight information are input into the prediction model, a dynamic recommended route required by the passenger is generated, and the dynamic recommended route is synchronized from the airport management terminal to the passenger's preset terminal. Specifically, the dynamic recommended route is the fastest transportation route obtained by combining available transportation tools within the airport.

[0012] Furthermore, before the step of receiving feedback information uploaded by passengers in the area surrounding the airport through the airport management terminal, the method further includes:

[0013] Based on the passenger information pre-entered into the airport management terminal, the passenger group corresponding to the passenger is identified, and the traffic influencing factors of the passenger group on the feedback information are collected. The passenger information specifically includes ticketing information, waiting area entry status and boarding record. The passenger group specifically includes business travelers, family travelers and leisure travelers. The traffic influencing factors specifically include travel time, distance and weather conditions.

[0014] Determine whether the traffic-influencing factors are within the preset average value of the airport management terminal;

[0015] If not, then based on the existing management strategy of the airport management terminal, a dedicated channel window for the passenger group is established. Based on the dedicated channel window, the passenger group is configured with the required service content. The dedicated channel window and the required service content are updated to the airport management terminal in real time. The airport management terminal displays the usage status of the dedicated channel window and the required service content to the passengers in real time. Specifically, the dedicated channel window includes a dedicated security check channel, a dedicated channel for family passengers, and a dedicated boarding gate.

[0016] Furthermore, the step of applying a pre-trained prediction model to perform rolling predictions on the latest data in the dynamic data pool and obtaining the prediction results further includes:

[0017] The difference between the predicted result and the actual value is calculated based on the preset error metric. The trend of the prediction error is identified by the difference value, and the prediction deviation for a specific time period is collected from the trend.

[0018] Determine whether the prediction deviation has a corresponding prediction weight, wherein the prediction weight specifically includes traffic weight, weather weight, and flow weight;

[0019] If so, the model parameters of the prediction model are adjusted according to the prediction deviation, corresponding weight conditions are added to the prediction model in real time, the interaction features between the latest data and the weight conditions are obtained, the joint impact parameters of the prediction results are generated based on the interaction features, the joint impact parameters are input into the prediction model, and the joint impact value of the joint impact parameters on the prediction results is revealed. Specifically, the weight conditions include sudden weather data, holidays and special events, traffic conditions and traffic accidents and flight delay information.

[0020] Furthermore, the step of generating the dynamically recommended route required by the passenger and synchronizing the dynamically recommended route from the airport management terminal to the passenger's preset terminal further includes:

[0021] Based on the passenger's preset transportation preferences on the airport management terminal, obtain the passenger's feedback on the dynamically recommended route;

[0022] Determine whether the feedback indicates agreement to apply the dynamically recommended route;

[0023] If not, then routes that do not include the travel preference in the dynamically recommended routes are blocked, the specific time difference between the route content and the best recommended route is generated, and the specific time difference is synchronized to the passenger's preset mobile device in real time.

[0024] Furthermore, the step of determining whether the dynamic data pool can execute the data update strategy preset by the airport management terminal further includes:

[0025] Based on the simulated load preset by the airport management terminal, the maximum number of concurrent requests that the dynamic data pool can accommodate is obtained;

[0026] Determine whether the dynamic data pool can operate stably for a preset period of time when it is at the maximum concurrent request volume;

[0027] If so, the performance parameters of the dynamic data pool are monitored in real time, and a preset rapid simulated traffic is input into the dynamic data pool. Abnormal information of the dynamic data pool under different loads is recorded. The performance parameters specifically include response time, throughput, and resource usage, and the abnormal information specifically includes timeouts, failed requests, and resource exhaustion.

[0028] Furthermore, the step of determining whether the feedback information matches the response event corresponding to the area surrounding the airport also includes:

[0029] Based on the mobile terminal pre-connected to the airport management terminal, the location information of the passenger in the area surrounding the airport is obtained;

[0030] Determine whether the location information is within a preset range of the response event;

[0031] If so, the response event will be synchronously published to the online devices pre-connected to the airport management terminal, and a corresponding processing time will be adaptively generated based on the response event. The online devices specifically include airport traffic equipment and airport passenger equipment.

[0032] Furthermore, the step of integrating and summarizing the real-time data pre-collected based on the airport management terminal to form a dynamic data pool includes:

[0033] Based on the traffic organization preset by the airport management terminal, the traffic organization is uniformly deployed on a pre-built cloud data platform, and real-time data uploaded by the traffic organization to the cloud data platform is collected. The traffic organization specifically includes traffic management departments, airport management departments, and weather forecasting agencies.

[0034] Determine whether the real-time data has the corresponding timestamp format;

[0035] If not, the obsolete data that does not have the timestamp format will be transferred to the information silo preset by the cloud data platform, and public access to the obsolete data in the information silo will be restricted according to the access permissions preset by the airport management terminal.

[0036] The present invention also provides an airport ground traffic passenger flow prediction system, comprising:

[0037] The integration module is used to integrate and summarize the real-time data pre-collected by the airport management terminal to form a dynamic data pool. The real-time data specifically includes flight dynamic data, traffic flow data, weather forecast data, and passenger movement data.

[0038] The judgment module is used to determine whether the dynamic data pool can execute the data update strategy preset by the airport management terminal;

[0039] The execution module is used to, if possible, apply a pre-trained prediction model to perform rolling predictions on the latest data in the dynamic data pool, obtain prediction results, and adaptively trigger traffic management measures preset by the airport management terminal based on the prediction results, dynamically dispatch traffic resources in the area surrounding the airport, and receive feedback information uploaded by passengers in the area surrounding the airport through the airport management terminal. The traffic management measures specifically include adjusting traffic light durations, optimizing vehicle scheduling, and issuing congestion warnings. The traffic resources specifically include increasing traffic capacity, opening additional lanes, and activating temporary parking lots.

[0040] The second judgment module is used to determine whether the feedback information matches the response event corresponding to the area surrounding the airport.

[0041] The second execution module is used to, if yes, collect the flight information pre-entered by the passenger to the airport management terminal, input the traffic data corresponding to the response event and the flight information into the prediction model, generate the dynamic recommended route required by the passenger, and synchronize the dynamic recommended route from the airport management terminal to the passenger's preset terminal, wherein the dynamic recommended route is specifically the fastest transportation route obtained by combining available transportation tools within the airport.

[0042] Furthermore, it also includes:

[0043] The data collection module is used to identify the passenger group corresponding to the passenger based on the passenger information pre-entered into the airport management terminal, and to collect the traffic influencing factors of the passenger group on the feedback information. The passenger information specifically includes ticketing information, waiting area entry status and boarding record. The passenger group specifically includes business travelers, family travelers and leisure travelers. The traffic influencing factors specifically include travel time, distance and weather conditions.

[0044] The third judgment module is used to determine whether the traffic influencing factors are within the preset average value of the airport management terminal;

[0045] The third execution module is used to, if not, establish a dedicated channel window for the passenger group based on the original management strategy of the airport management terminal, configure the required service content to the passenger group according to the dedicated channel window, update the dedicated channel window and the required service content to the airport management terminal in real time, and display the usage status of the dedicated channel window and the required service content to the passengers in real time through the airport management terminal. The dedicated channel window specifically includes a dedicated security check channel, a dedicated channel for family passengers, and a dedicated boarding gate.

[0046] Furthermore, the execution module also includes:

[0047] The acquisition unit is used to calculate the difference between the prediction result and the actual value based on a preset error metric, identify the trend of prediction error through the difference value, and collect the prediction deviation for a specific time period from the trend.

[0048] The judgment unit is used to determine whether the prediction deviation has a corresponding prediction weight, wherein the prediction weight specifically includes traffic weight, weather weight and flow weight;

[0049] An execution unit is configured to, if so, adjust the model parameters of the prediction model according to the prediction deviation, add corresponding weight conditions to the prediction model in real time, obtain the interaction features between the latest data and the weight conditions, generate joint impact parameters of the prediction result based on the interaction features, input the joint impact parameters into the prediction model, and reveal the joint impact value of the joint impact parameters on the prediction result. Specifically, the weight conditions include sudden weather data, holidays and special events, traffic conditions and traffic accidents and flight delay information.

[0050] This invention provides a method and system for predicting passenger flow in airport ground transportation, which has the following beneficial effects:

[0051] This invention integrates flight dynamics data, traffic flow data, weather forecast data, and passenger movement data to form a dynamic data pool. This ensures the system has a comprehensive and timely information foundation, enabling it to dynamically update and reflect the real-time conditions of the airport and its surroundings. This improves the accuracy of predictions and the speed of response. Furthermore, by determining whether the dynamic data pool can execute a preset data update strategy, it ensures that subsequent prediction and scheduling operations are only triggered when the data is sufficiently complete and up-to-date. This avoids inaccurate predictions due to insufficient or outdated data, improving the reliability of system decisions. Moreover, by applying a pre-trained prediction model to perform rolling predictions on the latest data, it can dynamically and continuously predict traffic conditions. This allows the system to adjust prediction results in real time and adaptively trigger traffic management measures, improving the flexibility and accuracy of traffic management. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating an embodiment of the airport ground traffic passenger flow prediction method of the present invention;

[0053] Figure 2 This is a structural block diagram of an embodiment of the airport ground traffic passenger flow prediction system of the present invention. Detailed Implementation

[0054] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. The realization of the purpose, functional features, and advantages of the invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings.

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Reference Appendix Figure 1 An airport ground traffic passenger flow prediction method according to an embodiment of the present invention includes:

[0057] S1: Based on the real-time data pre-collected by the airport management terminal, the real-time data is integrated and summarized to form a dynamic data pool, wherein the real-time data specifically includes flight dynamic data, traffic flow data, weather forecast data and passenger movement data;

[0058] S2: Determine whether the dynamic data pool can execute the data update strategy preset by the airport management terminal;

[0059] S3: If possible, the pre-trained prediction model is applied to perform rolling prediction on the latest data in the dynamic data pool to obtain the prediction results. Based on the prediction results, the traffic management measures preset by the airport management terminal are adaptively triggered to dynamically schedule traffic resources in the area surrounding the airport. The airport management terminal receives feedback information uploaded by passengers in the area surrounding the airport. The traffic management measures specifically include adjusting traffic light duration, optimizing vehicle scheduling, and issuing congestion warnings. The traffic resources specifically include increasing traffic capacity, opening additional lanes, and activating temporary parking lots.

[0060] S4: Determine whether the feedback information matches the response event corresponding to the area surrounding the airport;

[0061] S5: If so, collect the flight information pre-entered by the passenger to the airport management terminal, input the traffic data corresponding to the response event and the flight information into the prediction model, generate the dynamic recommended route required by the passenger, and synchronize the dynamic recommended route from the airport management terminal to the passenger's preset terminal. Specifically, the dynamic recommended route is the fastest transportation route obtained by combining available transportation tools within the airport.

[0062] In this embodiment, the system integrates and summarizes real-time data pre-collected by the airport management terminal, including flight dynamics data, traffic flow data, weather forecast data, and passenger movement data, to form a dynamic data pool. The system then determines whether this dynamic data pool can execute the data update strategy pre-set by the airport management terminal, and proceeds accordingly. For example, if the system determines that the dynamic data pool cannot execute the pre-set data update strategy, it assumes that the processing capacity or storage capacity of the dynamic data pool may have reached its limit, preventing the system from performing new data update operations. Based on the actual situation, the system will suggest that the airport management team increase the system's processing capacity or storage capacity. This can be achieved by expanding server resources or optimizing the database structure, compressing data, or cleaning up outdated or low-value data to free up storage space and improve system processing efficiency. Furthermore, data update operations can be performed in batches to avoid overloading the system by processing large amounts of data at once. For example, when the system determines that the dynamic data pool can execute the data update strategy pre-set by the airport management terminal, it will consider the dynamic data pool ready for a new data update operation. The system will then apply a pre-trained prediction model to perform rolling predictions on the latest data in the dynamic data pool. After obtaining the prediction results, it will adaptively trigger pre-set traffic management measures in the airport management terminal based on these prediction results. These traffic management measures specifically include adjusting traffic lights... The system optimizes vehicle scheduling and issues congestion warnings, dynamically allocating traffic resources around the airport. These resources include increasing transport capacity, opening additional lanes, and activating temporary parking lots. The system receives feedback from passengers in the surrounding area via airport management terminals. By applying a pre-trained predictive model, the system performs rolling predictions based on the latest data in a dynamic data pool. It can capture real-time information such as traffic flow, flight status, and weather changes, enabling more accurate predictions of future traffic conditions and faster responses. This reduces prediction errors and improves the overall system's responsiveness. Furthermore, based on the prediction results, the system adaptively triggers traffic management measures, such as adjusting traffic light durations, optimizing vehicle scheduling, and issuing congestion warnings. The system dynamically adjusts traffic conditions around the airport to reduce congestion, improve traffic flow, and enhance the travel experience for passengers. Guided by forecasts, the system can dynamically allocate traffic resources, such as increasing capacity, opening additional lanes, and activating temporary parking lots. This effectively addresses peak passenger flow or emergencies, maximizes the use of available resources, optimizes traffic efficiency, and ensures the orderly operation of traffic in and around the airport. Furthermore, by receiving and processing passenger feedback through airport management terminals, the system allows passengers' actual experiences to directly influence traffic management decisions. This feedback mechanism not only increases passenger participation and satisfaction but also helps managers understand passenger needs more accurately, thereby optimizing subsequent traffic management strategies.The system then determines whether these feedback messages match the corresponding response events in the airport's surrounding area, and executes the appropriate steps accordingly. For example, if the system determines that the feedback information uploaded by a passenger in the airport's surrounding area does not match the corresponding response events, the system assumes that the passenger's feedback may contain errors or misunderstandings, preventing the system from matching it with the actual event. The system will verify the accuracy of the feedback information by cross-validating other data sources (such as traffic surveillance videos and other passenger feedback). If a discrepancy is found, the system will prompt the passenger to resubmit the feedback or automatically adjust the feedback information. In cases where the feedback is inaccurate, the system will send a request to the passenger for further confirmation or more detailed information to improve the accuracy of the feedback. Simultaneously, based on the new feedback information, the system expands its response event database, incorporating previously uncovered events into the system's processing scope so that similar feedback can be identified and matched in the future. Mismatched feedback information is categorized and recorded for future system improvements and expansions. Based on this, the system ensures its gradual improvement, covering more possible events. Upon receiving feedback, it performs data integrity checks to ensure correct data format and complete content. If errors are found, the system requests the passenger to resubmit the feedback. By checking system logs, the system analyzes potential problems in the transmission process, such as network latency and data packet loss, and repairs and optimizes them to ensure that feedback information can be accurately transmitted and interpreted. For example, when the system determines that the feedback information uploaded by the passenger in the airport's surrounding area matches the corresponding response event in the airport's surrounding area, the system assumes that the response event corresponding to the feedback information has indeed occurred in the airport's surrounding area. The system collects the flight information pre-entered by the passenger to the airport management terminal, inputs the traffic data and flight information corresponding to the response event into the prediction model, and generates a dynamically recommended route for the passenger. The dynamically recommended route is the fastest transportation route obtained by combining available transportation tools within the airport, and this dynamically recommended route is synchronized from the airport management terminal to the passenger's pre-installed mobile terminal.By collecting passenger flight information and combining it with actual response events, the system can generate dynamically recommended routes that best meet the needs of each passenger. This personalized service provides the optimal travel plan based on real-time conditions, reducing passenger waiting time and inconvenience at and around the airport, and improving the smart flight display experience within the airport. Simultaneously, based on current traffic conditions and passenger flight times, it generates the fastest transportation route, ensuring passengers reach their destinations in the shortest possible time. Through adjustments to dynamically recommended routes, it effectively avoids congested sections or delays, maximizing passenger travel efficiency. Furthermore, by collecting and analyzing feedback information in real time, it can quickly adjust passenger travel routes after confirming the occurrence of a response event. This dynamic response mechanism can quickly provide passengers with alternative solutions in the event of emergencies (such as traffic accidents, temporary road closures, etc.), avoiding unnecessary delays or inconvenience. By providing passengers with personalized dynamically recommended routes, the system can more rationally allocate available transportation within the airport. This precise allocation not only reduces unnecessary resource consumption but also optimizes the utilization efficiency of airport transportation resources, avoiding traffic congestion or service deficiencies caused by improper resource allocation.

[0063] It should be noted that the specific example of the training process for the prediction model is as follows:

[0064] Taking the prediction of traffic congestion around the airport as an example, the entire process is explained in detail. First, the system obtains historical data from the past two years from the airport management system, traffic monitoring system, and weather forecast center. This data includes flight dynamic information (such as flight take-off and landing times, delays, etc.), traffic flow data (such as the number of vehicles on roads around the airport, average vehicle speed, etc.), weather data (such as temperature, precipitation, wind speed, etc.), and passenger movement data (such as the mode of transportation used by passengers, travel time, etc.). This data provides the foundation for model training.

[0065] The system then cleans and standardizes the collected raw data to ensure data quality. Data cleaning includes removing noisy data, filling in missing values, and handling outliers. Standardization transforms data from different scales to the same range, enabling the model to understand it correctly. Subsequently, the system extracts more meaningful features from the raw data through feature engineering. For example, it creates a "flight density" feature to represent the density of flights at an airport within a certain time period, and it can also create a "weather index" to comprehensively assess the impact of weather on transportation.

[0066] In terms of model selection, the system uses the LSTM (Long Short-Term Memory) model because the LSTM model is very powerful in processing time series data and can effectively capture the long-term and short-term time dependencies in the data. The system uses the traffic flow, flight density and weather index of the past 6 hours as input to predict the traffic flow of the next hour; for example, if the system has historical data from 10 am to 3 pm, the system will use this data to predict the traffic situation from 3 pm to 4 pm.

[0067] During the model training phase, the system divides the preprocessed data into a training set and a validation set. 80% of the data is used for training, and 20% of the data is used for validation. The training set is used to adjust the model's parameters so that the model can predict future traffic flow as accurately as possible. In this process, the LSTM model iterates repeatedly to minimize the error between the prediction results and the actual traffic flow, and gradually optimizes the model's performance.

[0068] After the model is trained, the system evaluates the model on the validation set. The system selects mean squared error (MSE) as the evaluation metric because it can effectively measure the difference between the model's predicted value and the actual value. If the model performs well on the validation set and the error is small, it means that the model's prediction accuracy is high and it can predict future traffic flow more accurately.

[0069] However, if the model is found to be unsatisfactory after evaluation, the system will optimize and tune the model. Optimization methods include adjusting the number of layers, neurons, and learning rate of the LSTM network, or reselecting features, or even introducing new data sources to improve the model's performance. The system will also combine other models (such as random forest or XGBoost) to enhance the model's generalization ability through ensemble learning.

[0070] After the model is evaluated and optimized, it is deployed into the airport management system, enabling it to receive new data and make predictions in real time. Based on the latest flight dynamics, traffic flow, and weather information, the model generates traffic forecasts for the next hour and provides dynamic travel route suggestions to passengers based on these forecasts. For example, if a road is predicted to be congested in the next hour, the system will recommend alternative routes to passengers to avoid congestion.

[0071] Finally, continuous monitoring and updating of the model is equally important. After the model is deployed, the system needs to monitor its performance in actual operation regularly. If the model's prediction error begins to increase over time, it may be because new external factors have not been incorporated into the model's training. At this time, the system needs to collect data again, update or retrain the model to ensure that the model can always maintain high-precision prediction capabilities.

[0072] In summary, by following the steps above, an LSTM model capable of predicting traffic congestion around airports can be trained and deployed. This not only improves the efficiency of airport traffic management but also provides passengers with faster and more accurate travel services, significantly enhancing the overall travel experience.

[0073] In this embodiment, before step S3 of receiving feedback information uploaded by passengers in the area surrounding the airport via the airport management terminal, the method further includes:

[0074] S301: Based on the passenger information pre-entered into the airport management terminal, identify the passenger group corresponding to the passenger, and collect the traffic impact factors of the passenger group on the feedback information. The passenger information specifically includes ticketing information, waiting area entry status and boarding record. The passenger group specifically includes business travelers, family travelers and leisure travelers. The traffic impact factors specifically include travel time, distance and weather conditions.

[0075] S302: Determine whether the traffic influencing factors are within the preset average value of the airport management terminal;

[0076] S303: If not, then based on the existing management strategy of the airport management terminal, establish a dedicated channel window for the passenger group, configure the required service content to the passenger group according to the dedicated channel window, update the dedicated channel window and the required service content to the airport management terminal in real time, and display the usage status of the dedicated channel window and the required service content to the passengers in real time through the airport management terminal. The dedicated channel window specifically includes a dedicated security check channel, a dedicated channel for family passengers, and a dedicated boarding gate.

[0077] In this embodiment, the system identifies the passenger group to which the passenger belongs based on passenger information pre-entered into the airport management terminal. Specifically, passenger groups include business travelers, family travelers, and leisure travelers. The system collects traffic-related factors affecting the feedback information from these passenger groups, including travel time, distance, and weather conditions. The system then determines whether these traffic-related factors are within the pre-set average values ​​of the airport management terminal to execute corresponding steps. For example, when the system determines that the traffic-related factors affecting the passenger group's feedback information are within the pre-set average values ​​of the airport management terminal, the system considers the current traffic conditions and passenger travel conditions to be relatively stable. The system does not need to make emergency adjustments or take additional measures and will continue to implement existing traffic management measures. This includes maintaining the current signal light duration, vehicle scheduling, and traffic flow control strategies without triggering additional traffic management measures or scheduling changes. For example, it continues to operate according to the original signal light timing plan, without needing to open additional lanes or activate temporary parking lots, while continuously monitoring passenger feedback information. The system monitors and analyzes real-time data and traffic conditions to promptly identify potential changes or problems. It regularly updates and analyzes this data to ensure all parameters remain within expected ranges and is prepared to handle any unforeseen circumstances. The system records and analyzes current data and feedback to create baseline data for future comparison and optimization. Feedback information is recorded in the system, and the data model is updated to enable faster response to similar situations in the future. For example, if the system determines that the traffic impact factors on passenger groups are not within the pre-set average values ​​on the airport management terminal, it will consider the current traffic conditions to be affecting passenger travel. The system will then establish dedicated channels for these passenger groups based on the existing management strategies on the airport management terminal. These dedicated channels include dedicated security checkpoints, dedicated channels for families, and dedicated boarding gates. Corresponding service requirements will be configured for each dedicated channel, and the dedicated channels and service requirements will be updated in real-time on the airport management terminal. The airport management terminal will then display the usage status of these dedicated channels and service requirements to passengers in real-time.The system establishes dedicated security checkpoints, family travel lanes, and boarding gates for different passenger groups, effectively diverting passenger flow, reducing congestion in main security areas and boarding gates, shortening passenger queuing time, and improving the efficiency of security checks and boarding. This alleviates overall airport traffic pressure. It also provides personalized and exclusive services to meet the special needs of different passengers, improving passenger satisfaction and travel comfort. The system dynamically adjusts and optimizes the configuration of dedicated lane windows based on real-time data and forecasts, making resource allocation more flexible and effectively responding to changing passenger flow, avoiding resource waste or shortages. Furthermore, the system updates the usage of dedicated lane windows and requested services in real time through the airport management terminal, ensuring passengers receive the latest information. Passengers can promptly understand the opening status and usage of each dedicated lane window, helping them better plan their trips and avoid misunderstandings or inconveniences caused by information delays.

[0078] In this embodiment, step S3, which involves applying a pre-trained prediction model to perform rolling predictions on the latest data in the dynamic data pool and obtaining the prediction results, further includes:

[0079] S31: Calculate the difference between the prediction result and the actual value based on the preset error measurement index, identify the trend of prediction error through the difference value, and collect the prediction deviation for a specific time period from the trend of change.

[0080] S32: Determine whether the prediction deviation has a corresponding prediction weight, wherein the prediction weight specifically includes traffic weight, weather weight and flow weight;

[0081] S33: If so, adjust the model parameters of the prediction model according to the prediction deviation, add corresponding weight conditions to the prediction model in real time, obtain the interaction features between the latest data and the weight conditions, generate the joint impact parameters of the prediction result based on the interaction features, input the joint impact parameters into the prediction model, and reveal the joint impact value of the joint impact parameters on the prediction result. Specifically, the weight conditions include sudden weather data, holidays and special events, traffic conditions and traffic accidents and flight delay information.

[0082] In this embodiment, the system calculates the difference between the predicted result and the actual value based on a pre-set error metric. This difference value is used to identify the trend of the prediction error. The system collects prediction deviations for specific time periods from different trends. Then, the system determines whether the prediction deviation has a corresponding prediction weight. These prediction weights specifically include traffic weights, weather weights, and flow weights, and executes corresponding steps accordingly. For example, if the system determines that the prediction deviation does not have a corresponding prediction weight, it considers that no corresponding traffic weight, weather weight, or flow weight has been found. This indicates that the influencing factors set for these weights do not cover all factors causing the prediction error. The system analyzes the deviation data and error metric to confirm whether the prediction deviation is related to traffic, weather, or flow weights, checks for other unconsidered influencing factors, compares actual and predicted data, uses statistical analysis methods to find potential factors affecting the deviation, and checks for new patterns or anomalies in historical data. Simultaneously, the prediction model is re-evaluated and optimized, checking for structural problems, redesigning the model architecture or changing model parameters, increasing the scope and depth of data collection, including collecting new data types or sources to ensure the model can capture more influencing factors, and expanding data sources, including collecting more real-time or historical data, especially those previously unconsidered data that may affect the prediction results; for example, when the system determines that the prediction deviation has a corresponding prediction weight, the system will consider the prediction deviation to be related to the prediction weight factor, and the system will adjust the model parameters of the prediction model according to the prediction deviation, adding corresponding weight conditions to the prediction model in real time. The weight conditions specifically include sudden weather data, holidays and special events, traffic conditions and traffic accidents and flight delay information, obtaining the interaction characteristics of the latest data and weight conditions, generating joint impact parameters of the prediction results based on the interaction characteristics, inputting the joint impact parameters into the prediction model, and revealing the joint impact value of the joint impact parameters on the prediction results;By adjusting model parameters and adding new weighting conditions in real time, the system can more accurately capture and consider various factors affecting prediction results. This allows the model to better adapt to dynamically changing environments and improves the accuracy of predictions. For example, adding sudden weather data can help the model better predict traffic conditions caused by weather changes. Simultaneously, the system can quickly adjust the model based on emergencies (such as traffic accidents and flight delays) to ensure that prediction results reflect the actual situation in a timely manner, improving responsiveness to emergencies and reducing prediction bias caused by them. This optimizes traffic management and passenger services. Furthermore, by acquiring the interaction characteristics of the latest data and weighting conditions, the system can reveal the relationships between various factors, improving the model's robustness in complex environments and its adaptability to complex and changing environments. This enhances the stability and reliability of prediction results. Finally, by generating joint influence parameters for the prediction results, the system can gain a detailed understanding of the impact of each weighting condition on the prediction results, revealing the combined effect of different weighting conditions on the prediction results. This helps in developing more reasonable traffic management strategies and emergency plans.

[0083] It should be noted that the interaction features between the latest data and the weighting conditions are obtained, and the joint influence parameters of the prediction result are generated based on the interaction features. A specific example is as follows:

[0084] Assuming we need to predict traffic flow in the area surrounding the airport, the following factors are temporarily included:

[0085] Real-time data: traffic flow data, weather data, holiday information, traffic accident data;

[0086] Weighting criteria: traffic flow weight, weather weight, and holiday weight;

[0087] Traffic flow data: Vehicle flow during the current time period;

[0088] Weather data: Current weather conditions (such as rainfall, temperature);

[0089] Holiday Information: Is it currently a holiday?

[0090] Traffic accident data: Information on recently occurring traffic accidents;

[0091] Weather and Traffic Flow: Analysis revealed that traffic flow decreased by 20% during rainfall.

[0092] Holidays and Traffic Flow: Traffic flow increases by 15% during holidays;

[0093] Traffic accidents and traffic flow: When a traffic accident occurs, traffic flow decreases by 30%;

[0094] Combined impact parameters: Considering the combined effects of rainfall, holidays, and traffic accidents, the calculated combined impact parameters are as follows: Under the conditions of rainfall and holidays, traffic flow is expected to decrease by 25%;

[0095] Update the model: Input the joint impact parameters (25% reduction) into the traffic flow prediction model;

[0096] Validation results: After the model was updated, the prediction results showed that the traffic flow did indeed decrease as expected, thus improving the accuracy of the prediction;

[0097] In summary, by acquiring the interaction characteristics of the latest data and weight conditions, and generating joint influence parameters, the system can more accurately reflect the actual situation, enabling the prediction model to be dynamically adjusted to consider the comprehensive influence of multiple factors, thereby improving the accuracy of prediction results and optimizing traffic management decisions.

[0098] In this embodiment, step S5, which generates the dynamically recommended route required by the passenger and synchronizes the dynamically recommended route from the airport management terminal to the passenger's preset terminal, further includes:

[0099] S51: Based on the passenger's preset transportation preferences on the airport management terminal, obtain the passenger's feedback on the dynamically recommended route;

[0100] S52: Determine whether the feedback content agrees to the application of the dynamic recommended route;

[0101] S53: If not, then the route content that does not include the traffic preference in the dynamic recommended route is blocked, the specific time difference between the route content and the best recommended route is generated, and the specific time difference is synchronized to the passenger's preset mobile terminal in real time.

[0102] In this embodiment, the system obtains feedback from passengers regarding dynamically recommended routes based on their pre-set travel preferences on the airport management terminal. The system then determines whether this feedback indicates agreement to the application of the dynamically recommended route and executes corresponding steps accordingly. For example, if the system determines that a passenger's feedback indicates agreement to the application of the dynamically recommended route, the system assumes that the passenger agrees to adaptively replace the currently used navigation route with the dynamically recommended route. The system confirms the passenger's feedback to ensure explicit agreement, receives and parses the feedback information provided by the passenger through the voice recognition application on the airport management terminal, verifies the feedback content to ensure agreement to the recommended route, and updates the passenger's navigation system settings on the airport management terminal based on the confirmed dynamically recommended route, setting the recommended route as the current navigation route. If the passenger uses a navigation application on a mobile terminal, the system synchronizes the dynamically recommended route to that application, generates the latest navigation route data, and pushes it to the passenger's terminal device, updating the navigation system with detailed information about the dynamically recommended route (such as path, estimated time, traffic conditions, etc.). When the system determines that a passenger disagrees with the application of a dynamically recommended route, it assumes that the passenger does not agree to use that route because it does not reflect the passenger's preferred transportation options. The system will then filter out routes that do not reflect the passenger's preferences and generate a time difference between the current route and the optimal recommended route. This time difference will be synchronized in real-time to the passenger's pre-set mobile device. The system ensures that the passenger's personal preferences and transportation preferences are respected and considered, improving passenger satisfaction and comfort. By filtering out routes that do not align with the passenger's preferences, the system can optimize its recommendation strategy, improving the accuracy and practicality of the recommendations. Simultaneously, generating and synchronizing the time difference between the current route and the optimal recommended route allows passengers to clearly understand the time difference between choosing the dynamically recommended route and the optimal route, helping them make more informed decisions and improving the passenger experience. Providing time difference information also enhances the system's transparency, allowing passengers to understand the specific differences between different route choices and increasing the system's trustworthiness. Passengers can quickly assess whether they are willing to accept the dynamically recommended route or choose another route based on the time difference information, thereby improving decision-making efficiency.

[0103] In this embodiment, step S2, which determines whether the dynamic data pool can execute the preset data update strategy of the airport management terminal, further includes:

[0104] S21: Based on the simulated load preset by the airport management terminal, obtain the maximum number of concurrent requests that the dynamic data pool can accommodate;

[0105] S22: Determine whether the dynamic data pool can operate stably for a preset time period when it is at the maximum concurrent request volume;

[0106] S23: If so, monitor the performance parameters of the dynamic data pool in real time, apply preset rapid simulated traffic input to the dynamic data pool, and record the abnormal information of the dynamic data pool under different loads. The performance parameters specifically include response time, throughput and resource usage, and the abnormal information specifically includes timeout, failed requests and resource exhaustion.

[0107] In this embodiment, the system obtains the maximum concurrent request volume that the dynamic data pool can accommodate based on the simulated load pre-set by the airport management terminal. Then, the system determines whether the dynamic data pool can stably operate for a pre-set time period when it is at its maximum concurrent request volume, and executes the corresponding steps accordingly. For example, if the system determines that the dynamic data pool is at its maximum concurrent request volume and cannot stably operate for the pre-set time period, the system will consider that the system's load capacity or performance is insufficient and cannot effectively handle high concurrent requests. The system will then use performance monitoring tools to monitor the resource consumption and performance of various parts of the system, identifying the main bottlenecks causing the system to malfunction, such as excessive CPU usage. High concurrency or slow database queries can be addressed by introducing a load balancer to distribute requests across multiple server nodes, reducing the load on individual nodes. Database performance can be improved by adding indexes, optimizing query statements, and using sharding or partitioning. Furthermore, the system can be divided into multiple independent service modules, using a distributed architecture to enhance overall processing capacity. A caching system can be introduced to reduce the frequency of direct database queries and improve data response speed. For non-real-time requests, asynchronous processing mechanisms can be used to queue these requests, reducing the real-time pressure on the system. For example, when the system determines that the dynamic data pool is at its maximum concurrent request volume and can stably operate for a pre-set time period, the system... The system assumes its load capacity is sufficient to handle high concurrency requests. It monitors the performance parameters of the dynamic data pool in real time, including response time, throughput, and resource usage. Pre-set, rapidly simulated traffic is input into the dynamic data pool, recording anomalies under different loads, such as timeouts, failed requests, and resource exhaustion. By continuously monitoring performance parameters like response time, throughput, and resource usage, the system can continuously assess its performance under high loads, ensuring stability during long-term operation. If response time gradually increases or resource usage reaches a critical point, the system can promptly issue alerts and take appropriate measures. To prevent potential performance degradation, the system employs drastic traffic simulation testing. This allows the system to identify potential problems under extreme conditions, such as timeouts, failed requests, and resource exhaustion, thus preventing failures during actual operation. After simulating a large influx of sudden traffic, the system may find that the request failure rate increases under certain specific conditions. This can prompt developers to optimize relevant processing mechanisms. Furthermore, by recording and analyzing anomaly information, the system can enhance its fault tolerance and optimize adaptive adjustment strategies under extreme loads. The system can automatically trigger resource expansion mechanisms before resource exhaustion or adjust timeout parameters when timeout events occur frequently to ensure service continuity.

[0108] In this embodiment, step S4, which determines whether the feedback information matches the response event corresponding to the area surrounding the airport, further includes:

[0109] S41: Based on the mobile terminal of the passenger pre-connected to the airport management terminal, obtain the passenger's location information in the area surrounding the airport;

[0110] S42: Determine whether the location information is within the preset range of the response event;

[0111] S43: If so, the response event is synchronously published to the online devices pre-connected to the airport management terminal, and a corresponding processing time is adaptively generated according to the response event. The online devices specifically include airport traffic equipment and airport passenger equipment.

[0112] In this embodiment, the system obtains the passenger's location information in the area surrounding the airport based on the passenger's mobile terminal pre-connected to the airport management terminal, enabling area-based person locating. The system then determines whether this location information is within a preset range of a response event to execute corresponding steps. For example, if the system determines that the passenger's location information in the airport's surrounding area is not within the preset range of the response event, the system considers the passenger's location outside the scope requiring attention or processing, meaning the feedback information is inaccurate. Therefore, the current response event has no direct impact on the passenger. The system checks and adjusts the preset range of the response event to ensure coverage of all potentially affected areas, expanding the preset range as needed to ensure it includes potentially affected areas. Simultaneously, if the passenger is indeed outside the event's impact range, the system chooses not to send relevant notifications to avoid unnecessary interference. Cases of mismatch between passenger location information and the event range are recorded, generating possible causes, and a passenger feedback mechanism is established, allowing passengers to manually confirm or adjust their location to improve positioning accuracy. Conversely, if the system determines that the passenger's location information in the airport's surrounding area is within the preset range of the response event, the system considers the passenger's location to be affected by the response event, meaning the feedback information is authentic. The system synchronously publishes response events to pre-connected online devices on the airport management terminal. These online devices include airport traffic equipment and airport passenger equipment. The system adaptively generates corresponding processing times based on the response events. By synchronizing response events to these pre-connected online devices in real time, the system can respond quickly and take appropriate measures, shortening event processing time. When traffic congestion or emergencies occur, airport traffic equipment such as traffic lights and signs can be adjusted immediately to reduce the impact on passenger travel. Simultaneously, timely dissemination of response event information to passengers improves their awareness of the current situation and helps them make safer decisions. If severe weather events affect the area surrounding the airport, the system can issue warnings to passengers through passenger equipment such as information screens or mobile applications, advising them to adjust their travel plans. Based on the adaptively generated processing times for response events, the system can rationally allocate resources, optimize airport traffic management, and avoid resource waste or untimely dispatch. During peak periods, if the system predicts severe congestion in a certain area, it can pre-arrange more transportation or open additional lanes to alleviate pressure. Through real-time updates and responses, the system can reduce passenger waiting time and provide a smoother and more comfortable travel experience.

[0113] In this embodiment, step S1, which integrates and summarizes the real-time data pre-collected by the airport management terminal to form a dynamic data pool, includes:

[0114] S11: Based on the traffic organization preset by the airport management terminal, the traffic organization is uniformly deployed on a pre-built cloud data platform, and real-time data uploaded by the traffic organization to the cloud data platform is collected. The traffic organization specifically includes traffic management departments, airport management departments, and weather forecasting agencies.

[0115] S12: Determine whether the real-time data has a corresponding timestamp format;

[0116] S13: If not, the obsolete data that does not have the timestamp format will be transferred to the information silo preset by the cloud data platform, and the public access of the obsolete data in the information silo will be restricted according to the access permissions preset by the airport management terminal.

[0117] In this embodiment, the system is based on pre-set traffic organization in the airport management terminal. This traffic organization specifically includes traffic management departments, airport management departments, and weather forecasting agencies. These traffic organizations are uniformly deployed on a pre-built cloud data platform. The system collects real-time data uploaded by these traffic organizations to the cloud data platform. Then, the system determines whether this real-time data has the corresponding timestamp format to execute the corresponding steps. For example, when the system determines that the real-time data uploaded by the traffic organizations to the cloud data platform has the corresponding timestamp format, the system considers this data to have been correctly labeled with time information and can accurately reflect the time point of data generation or collection. The system will then check whether all uploaded data not only has timestamp information... The data is formatted with timestamps and includes all necessary fields and content to ensure data integrity. A data validation program is run to confirm that all data correctly includes timestamps and other necessary information. Real-time data with timestamps is synchronized and integrated in chronological order to ensure temporal consistency. All uploaded data is sorted by timestamp to ensure data processing follows chronological logic. Timestamp-verified data is stored on a cloud data platform and backed up regularly to ensure data security. A regular backup mechanism is implemented to prevent data loss and ensure rapid recovery in case of system failure or data loss. For example, if the system determines that the real-time traffic organization data uploaded to the cloud data platform does not have timestamps, the system will perform a backup. If the data lacks the required timestamp format, the system will consider it to be without time information and unable to accurately reflect the corresponding collection time. The system will then transfer this invalid data to a pre-defined information silo on the cloud data platform. Access to this invalid data will be restricted based on pre-set access permissions on the airport management terminal. By transferring data without the correct timestamp format to the information silo, the system avoids contaminating the overall dataset with low-quality data, ensuring sufficient accuracy and timeliness of the data used for analysis and decision-making. If some traffic flow data fails to be correctly timestamped due to format issues, the system will isolate this data to prevent it from affecting the accuracy of the traffic flow prediction model. Meanwhile, by restricting public access to obsolete data, the system can prevent the misuse of this unverified data, thereby avoiding erroneous decisions or judgments caused by incorrect data. When airport management makes traffic scheduling decisions, misusing data without time stamps may lead to unreasonable resource allocation. Isolating such data can effectively avoid this situation. Furthermore, isolating data without timestamps in information silos can provide a safe storage area for subsequent data cleaning and repair, making it possible to reuse the data after repair. In other words, authorized technical personnel can periodically check the data in information silos, timestamp the parts that can be repaired, and reuse them, rather than discarding the data completely.

[0118] Reference Appendix Figure 2An airport ground traffic passenger flow prediction system, as described in one embodiment of the present invention, includes:

[0119] The integration module 10 is used to integrate and summarize the real-time data pre-collected by the airport management terminal to form a dynamic data pool. The real-time data specifically includes flight dynamic data, traffic flow data, weather forecast data, and passenger movement data.

[0120] The judgment module 20 is used to determine whether the dynamic data pool can execute the data update strategy preset by the airport management terminal;

[0121] The execution module 30 is used to, if possible, apply a pre-trained prediction model to perform rolling prediction on the latest data in the dynamic data pool, obtain prediction results, adaptively trigger traffic management measures preset by the airport management terminal based on the prediction results, dynamically schedule traffic resources in the area surrounding the airport, and receive feedback information uploaded by passengers in the area surrounding the airport through the airport management terminal. The traffic management measures specifically include adjusting traffic light duration, optimizing vehicle scheduling, and issuing congestion warnings. The traffic resources specifically include increasing traffic capacity, opening additional lanes, and activating temporary parking lots.

[0122] The second judgment module 40 is used to determine whether the feedback information matches the response event corresponding to the airport surrounding area;

[0123] The second execution module 50 is used to, if yes, collect the flight information pre-entered by the passenger to the airport management terminal, input the traffic data corresponding to the response event and the flight information into the prediction model, generate the dynamic recommended route required by the passenger, and synchronize the dynamic recommended route from the airport management terminal to the passenger's preset terminal, wherein the dynamic recommended route is specifically the fastest transportation route obtained by combining available transportation tools within the airport.

[0124] In this embodiment, the integration module 10 integrates and summarizes real-time data pre-collected by the airport management terminal, specifically including flight dynamic data, traffic flow data, weather forecast data, and passenger movement data, to form a dynamic data pool. Then, the judgment module 20 determines whether the dynamic data pool can execute the data update strategy pre-set by the airport management terminal, and performs the corresponding steps accordingly. For example, if the system determines that the dynamic data pool cannot execute the data update strategy pre-set by the airport management terminal, the system will consider that the processing capacity or storage capacity of the dynamic data pool may have reached its limit, preventing the system from performing new data update operations. Based on the actual situation, the system will suggest that the airport management team increase the system's processing capacity. This could involve increasing storage capacity, such as expanding server resources or optimizing the database structure, while simultaneously compressing data or cleaning up outdated or low-value data to free up storage space, improve system processing efficiency, and perform data update operations in batches to avoid overloading the system by processing large amounts of data at once. For example, when the system determines that the dynamic data pool can execute the data update strategy pre-set by the airport management terminal, the execution module 30 will consider that the dynamic data pool can perform new data update operations. The system will apply a pre-trained prediction model to perform rolling predictions on the latest data in the dynamic data pool. After obtaining the prediction results, it will adaptively trigger the traffic management measures pre-set by the airport management terminal based on these prediction results. The traffic management measures specifically include... The system dynamically manages traffic resources around the airport by adjusting traffic light durations, optimizing vehicle scheduling, and issuing congestion warnings. These resources include increasing transport capacity, opening additional lanes, and activating temporary parking lots. The system receives feedback from passengers in the airport's surrounding area via airport management terminals. By applying a pre-trained predictive model, the system performs rolling predictions based on the latest data in a dynamic data pool. It can capture real-time information such as traffic flow, flight status, and weather changes, enabling it to more accurately predict future traffic conditions and respond quickly, reducing prediction errors and improving the overall system's responsiveness. Furthermore, it adaptively triggers traffic management measures based on prediction results, such as adjusting traffic light durations, optimizing vehicle scheduling, and issuing congestion warnings. The system can dynamically adjust traffic conditions around the airport, reduce traffic congestion, improve traffic flow, and enhance the travel experience for passengers. Guided by the prediction results, the system can dynamically allocate traffic resources, such as increasing traffic capacity, opening additional lanes, and activating temporary parking lots. This can effectively cope with peak passenger flow or emergencies, maximize the use of available resources, optimize traffic operation efficiency, and ensure the orderly operation of traffic in and around the airport. Furthermore, by receiving and processing feedback information uploaded by passengers through the airport management terminal, the actual experience of passengers can directly affect traffic management decisions. This feedback mechanism not only improves passenger participation and satisfaction but also helps managers to more accurately understand passenger needs, thereby optimizing subsequent traffic management strategies.The second judgment module 40 then determines whether these feedback messages match the corresponding response events in the airport's surrounding area, and executes the corresponding steps accordingly. For example, if the system determines that the feedback messages uploaded by passengers in the airport's surrounding area cannot match the corresponding response events, the system will assume that the feedback messages uploaded by passengers may contain errors or misunderstandings, causing the system to be unable to match them with the actual events. The system will verify the accuracy of the feedback messages by cross-validating other data sources (such as traffic monitoring videos and other passenger feedback). If a discrepancy is found, the system will prompt the passenger to resubmit the feedback or automatically adjust the feedback messages. In the case of inaccurate feedback messages, the system will send a request to the passenger to further confirm or provide more detailed information, thereby improving the accuracy of the feedback. At the same time, based on the new feedback messages, the system's response event library is expanded to include these previously uncovered events in the system's processing scope, so that similar feedback can be identified and matched in the future. The mismatched feedback messages are classified and recorded for subsequent system improvements and expansions. Based on this, the system ensures its gradual improvement, covering more possible events. Upon receiving feedback information, it performs data integrity checks to ensure correct data format and complete content. If errors are found, the system requests the passenger to resubmit the feedback. By checking system logs, it analyzes potential problems in the transmission process, such as network latency and data packet loss, and repairs and optimizes them to ensure that feedback information can be accurately transmitted and interpreted. For example, when the system determines that the feedback information uploaded by the passenger in the airport's surrounding area matches the corresponding response event in the airport's surrounding area, the second execution module 50 will consider that the response event corresponding to the feedback information has indeed occurred in the airport's surrounding area. The system will collect the flight information pre-entered by the passenger to the airport management terminal, input the traffic data and flight information corresponding to the response event into the prediction model, and generate the dynamic recommended route currently needed by the passenger. The dynamic recommended route is specifically the fastest transportation route obtained by combining available transportation tools within the airport, and the dynamic recommended route is synchronized from the airport management terminal to the passenger's pre-set mobile terminal.By collecting passenger flight information and combining it with actual response events, the system can generate dynamically recommended routes that best meet the needs of each passenger. This personalized service provides the optimal travel plan based on real-time conditions, reducing passenger waiting time and inconvenience at and around the airport, and improving the overall passenger experience. Simultaneously, based on current traffic conditions and passenger flight times, it generates the fastest transportation routes, ensuring passengers reach their destinations in the shortest possible time. Adjustments to dynamically recommended routes effectively avoid congested sections or delays, maximizing passenger travel efficiency. Furthermore, by collecting and analyzing feedback information in real time, the system can quickly adjust passenger routes after a response event is confirmed. This dynamic response mechanism can quickly provide alternative solutions for passengers in the event of emergencies (such as traffic accidents, temporary road closures, etc.), avoiding unnecessary delays or inconvenience. By providing personalized dynamically recommended routes, the system can more rationally allocate available transportation within the airport. This precise allocation not only reduces unnecessary resource consumption but also optimizes the utilization efficiency of airport transportation resources, avoiding traffic congestion or service deficiencies caused by improper resource allocation.

[0125] In this embodiment, it also includes:

[0126] The data collection module is used to identify the passenger group corresponding to the passenger based on the passenger information pre-entered into the airport management terminal, and to collect the traffic influencing factors of the passenger group on the feedback information. The passenger information specifically includes ticketing information, waiting area entry status and boarding record. The passenger group specifically includes business travelers, family travelers and leisure travelers. The traffic influencing factors specifically include travel time, distance and weather conditions.

[0127] The third judgment module is used to determine whether the traffic influencing factors are within the preset average value of the airport management terminal;

[0128] The third execution module is used to, if not, establish a dedicated channel window for the passenger group based on the original management strategy of the airport management terminal, configure the required service content to the passenger group according to the dedicated channel window, update the dedicated channel window and the required service content to the airport management terminal in real time, and display the usage status of the dedicated channel window and the required service content to the passengers in real time through the airport management terminal. The dedicated channel window specifically includes a dedicated security check channel, a dedicated channel for family passengers, and a dedicated boarding gate.

[0129] In this embodiment, the system identifies the passenger group to which the passenger belongs based on passenger information pre-entered into the airport management terminal. Specifically, passenger groups include business travelers, family travelers, and leisure travelers. The system collects traffic-related factors affecting the feedback information from these passenger groups, including travel time, distance, and weather conditions. The system then determines whether these traffic-related factors are within the pre-set average values ​​of the airport management terminal to execute corresponding steps. For example, when the system determines that the traffic-related factors affecting the passenger group's feedback information are within the pre-set average values ​​of the airport management terminal, the system considers the current traffic conditions and passenger travel conditions to be relatively stable. The system does not need to make emergency adjustments or take additional measures and will continue to implement existing traffic management measures. This includes maintaining the current signal light duration, vehicle scheduling, and traffic flow control strategies without triggering additional traffic management measures or scheduling changes. For example, it continues to operate according to the original signal light timing plan, without needing to open additional lanes or activate temporary parking lots, while continuously monitoring passenger feedback information. The system monitors and analyzes real-time data and traffic conditions to promptly identify potential changes or problems. It regularly updates and analyzes this data to ensure all parameters remain within expected ranges and is prepared to handle any unforeseen circumstances. The system records and analyzes current data and feedback to create baseline data for future comparison and optimization. Feedback information is recorded in the system, and the data model is updated to enable faster response to similar situations in the future. For example, if the system determines that the traffic impact factors on passenger groups are not within the pre-set average values ​​on the airport management terminal, it will consider the current traffic conditions to be affecting passenger travel. The system will then establish dedicated channels for these passenger groups based on the existing management strategies on the airport management terminal. These dedicated channels include dedicated security checkpoints, dedicated channels for families, and dedicated boarding gates. Corresponding service requirements will be configured for each dedicated channel, and the dedicated channels and service requirements will be updated in real-time on the airport management terminal. The airport management terminal will then display the usage status of these dedicated channels and service requirements to passengers in real-time.The system establishes dedicated security checkpoints, family travel lanes, and boarding gates for different passenger groups, effectively diverting passenger flow, reducing congestion in main security areas and boarding gates, shortening passenger queuing time, and improving the efficiency of security checks and boarding. This alleviates overall airport traffic pressure. It also provides personalized and exclusive services to meet the special needs of different passengers, improving passenger satisfaction and travel comfort. The system dynamically adjusts and optimizes the configuration of dedicated lane windows based on real-time data and forecasts, making resource allocation more flexible and effectively responding to changing passenger flow, avoiding resource waste or shortages. Furthermore, the system updates the usage of dedicated lane windows and requested services in real time through the airport management terminal, ensuring passengers receive the latest information. Passengers can promptly understand the opening status and usage of each dedicated lane window, helping them better plan their trips and avoid misunderstandings or inconveniences caused by information delays.

[0130] In this embodiment, the execution module further includes:

[0131] The acquisition unit is used to calculate the difference between the prediction result and the actual value based on a preset error metric, identify the trend of prediction error through the difference value, and collect the prediction deviation for a specific time period from the trend.

[0132] The judgment unit is used to determine whether the prediction deviation has a corresponding prediction weight, wherein the prediction weight specifically includes traffic weight, weather weight and flow weight;

[0133] An execution unit is configured to, if so, adjust the model parameters of the prediction model according to the prediction deviation, add corresponding weight conditions to the prediction model in real time, obtain the interaction features between the latest data and the weight conditions, generate joint impact parameters of the prediction result based on the interaction features, input the joint impact parameters into the prediction model, and reveal the joint impact value of the joint impact parameters on the prediction result. Specifically, the weight conditions include sudden weather data, holidays and special events, traffic conditions and traffic accidents and flight delay information.

[0134] In this embodiment, the system calculates the difference between the predicted result and the actual value based on a pre-set error metric. This difference value is used to identify the trend of the prediction error. The system collects prediction deviations for specific time periods from different trends. Then, the system determines whether the prediction deviation has a corresponding prediction weight. These prediction weights specifically include traffic weights, weather weights, and flow weights, and executes corresponding steps accordingly. For example, if the system determines that the prediction deviation does not have a corresponding prediction weight, it considers that no corresponding traffic weight, weather weight, or flow weight has been found. This indicates that the influencing factors set for these weights do not cover all factors causing the prediction error. The system analyzes the deviation data and error metric to confirm whether the prediction deviation is related to traffic, weather, or flow weights, checks for other unconsidered influencing factors, compares actual and predicted data, uses statistical analysis methods to find potential factors affecting the deviation, and checks for new patterns or anomalies in historical data. Simultaneously, the prediction model is re-evaluated and optimized, checking for structural problems, redesigning the model architecture or changing model parameters, increasing the scope and depth of data collection, including collecting new data types or sources to ensure the model can capture more influencing factors, and expanding data sources, including collecting more real-time or historical data, especially those previously unconsidered data that may affect the prediction results; for example, when the system determines that the prediction deviation has a corresponding prediction weight, the system will consider the prediction deviation to be related to the prediction weight factor, and the system will adjust the model parameters of the prediction model according to the prediction deviation, adding corresponding weight conditions to the prediction model in real time. The weight conditions specifically include sudden weather data, holidays and special events, traffic conditions and traffic accidents and flight delay information, obtaining the interaction characteristics of the latest data and weight conditions, generating joint impact parameters of the prediction results based on the interaction characteristics, inputting the joint impact parameters into the prediction model, and revealing the joint impact value of the joint impact parameters on the prediction results;By adjusting model parameters and adding new weighting conditions in real time, the system can more accurately capture and consider various factors affecting prediction results. This allows the model to better adapt to dynamically changing environments and improves the accuracy of predictions. For example, adding sudden weather data can help the model better predict traffic conditions caused by weather changes. Simultaneously, the system can quickly adjust the model based on emergencies (such as traffic accidents and flight delays) to ensure that prediction results reflect the actual situation in a timely manner, improving responsiveness to emergencies and reducing prediction bias caused by them. This optimizes traffic management and passenger services. Furthermore, by acquiring the interaction characteristics of the latest data and weighting conditions, the system can reveal the relationships between various factors, improving the model's robustness in complex environments and its adaptability to complex and changing environments. This enhances the stability and reliability of prediction results. Finally, by generating joint influence parameters for the prediction results, the system can gain a detailed understanding of the impact of each weighting condition on the prediction results, revealing the combined effect of different weighting conditions on the prediction results. This helps in developing more reasonable traffic management strategies and emergency plans.

[0135] In this embodiment, the second execution module further includes:

[0136] The acquisition unit is used to acquire the passenger's feedback on the dynamically recommended route based on the passenger's preset transportation preferences on the airport management terminal.

[0137] The second judgment unit is used to determine whether the feedback content agrees to the application of the dynamic recommended route;

[0138] The second execution unit is used to, if not, filter out route content in the dynamically recommended routes that does not include the travel preference, generate a specific time difference between the route content and the best recommended route, and synchronize the specific time difference to the passenger's preset mobile terminal in real time.

[0139] In this embodiment, the system obtains feedback from passengers regarding dynamically recommended routes based on their pre-set travel preferences on the airport management terminal. The system then determines whether this feedback indicates agreement to the application of the dynamically recommended route and executes corresponding steps accordingly. For example, if the system determines that a passenger's feedback indicates agreement to the application of the dynamically recommended route, the system assumes that the passenger agrees to adaptively replace the currently used navigation route with the dynamically recommended route. The system confirms the passenger's feedback to ensure explicit agreement to the application of the dynamically recommended route, receives and parses the feedback information provided by the passenger through the airport management terminal or mobile terminal, verifies the feedback content, and ensures that the feedback indicates agreement to the application of the recommended route. Simultaneously, based on the confirmed dynamically recommended route, the system updates the passenger's navigation system settings on the airport management terminal, setting the recommended route as the current navigation route. If the passenger uses a navigation application on their mobile terminal, the system synchronizes the dynamically recommended route to the application, generates the latest navigation route data, and pushes it to the passenger's terminal device, updating the navigation system with detailed information about the dynamically recommended route (such as path, estimated time, traffic conditions, etc.). For example, when the system determines that a passenger disagrees with the application of a dynamically recommended route, it assumes that the passenger does not agree to use that route because it does not reflect the passenger's preferred transportation preferences. The system will then filter out routes that do not reflect these preferences, generate a time difference between the current route and the optimal recommended route, and synchronize this time difference to the passenger's pre-set mobile device in real time. This ensures that the passenger's personal preferences and transportation preferences are respected and considered, improving passenger satisfaction and comfort. By filtering out routes that do not align with passenger preferences, the system can optimize its recommendation strategy, improving the accuracy and practicality of the recommendations. Simultaneously, generating and synchronizing the time difference between the current route and the optimal recommended route allows passengers to clearly understand the time difference between choosing a dynamically recommended route and the optimal route, helping them make more informed decisions. Providing this time difference information also enhances the system's transparency, allowing passengers to understand the specific differences between different route choices and increasing the system's trustworthiness. Passengers can quickly assess whether they are willing to accept the dynamically recommended route or choose another route based on the time difference information, thereby improving decision-making efficiency.

[0140] In this embodiment, the determination module further includes:

[0141] The second acquisition unit is used to acquire the maximum number of concurrent requests that the dynamic data pool can accommodate based on the simulated load preset by the airport management terminal.

[0142] The third judgment unit is used to determine whether the dynamic data pool can operate stably for a preset time period when it is at the maximum concurrent request volume;

[0143] The third execution unit is used to monitor the performance parameters of the dynamic data pool in real time if the condition is met, apply a preset rapid simulated traffic input to the dynamic data pool, and record the abnormal information of the dynamic data pool under different loads. Specifically, the performance parameters include response time, throughput, and resource usage, and the abnormal information includes timeouts, failed requests, and resource exhaustion.

[0144] In this embodiment, the system obtains the maximum concurrent request volume that the dynamic data pool can accommodate based on the simulated load pre-set by the airport management terminal. Then, the system determines whether the dynamic data pool can stably operate for a pre-set time period when it is at its maximum concurrent request volume, and executes the corresponding steps accordingly. For example, if the system determines that the dynamic data pool is at its maximum concurrent request volume and cannot stably operate for the pre-set time period, the system will consider that the system's load capacity or performance is insufficient and cannot effectively handle high concurrent requests. The system will then use performance monitoring tools to monitor the resource consumption and performance of various parts of the system, identifying the main bottlenecks causing the system to malfunction, such as excessive CPU usage. High concurrency or slow database queries can be addressed by introducing a load balancer to distribute requests across multiple server nodes, reducing the load on individual nodes. Database performance can be improved by adding indexes, optimizing query statements, and using sharding or partitioning. Furthermore, the system can be divided into multiple independent service modules, using a distributed architecture to enhance overall processing capacity. A caching system can be introduced to reduce the frequency of direct database queries and improve data response speed. For non-real-time requests, asynchronous processing mechanisms can be used to queue these requests, reducing the real-time pressure on the system. For example, when the system determines that the dynamic data pool is at its maximum concurrent request volume and can stably operate for a pre-set time period, the system... The system assumes its load capacity is sufficient to handle high concurrency requests. It monitors the performance parameters of the dynamic data pool in real time, including response time, throughput, and resource usage. Pre-set, rapidly simulated traffic is input into the dynamic data pool, recording anomalies under different loads, such as timeouts, failed requests, and resource exhaustion. By continuously monitoring performance parameters like response time, throughput, and resource usage, the system can continuously assess its performance under high loads, ensuring stability during long-term operation. If response time gradually increases or resource usage reaches a critical point, the system can promptly issue alerts and take appropriate measures. To prevent potential performance degradation, the system employs drastic traffic simulation testing. This allows the system to identify potential problems under extreme conditions, such as timeouts, failed requests, and resource exhaustion, thus preventing failures during actual operation. After simulating a large influx of sudden traffic, the system may find that the request failure rate increases under certain specific conditions. This can prompt developers to optimize relevant processing mechanisms. Furthermore, by recording and analyzing anomaly information, the system can enhance its fault tolerance and optimize adaptive adjustment strategies under extreme loads. The system can automatically trigger resource expansion mechanisms before resource exhaustion or adjust timeout parameters when timeout events occur frequently to ensure service continuity.

[0145] In this embodiment, the second determination module further includes:

[0146] The third acquisition unit is used to acquire the location information of the passenger in the area surrounding the airport based on the mobile terminal pre-connected to the airport management terminal of the passenger.

[0147] The fourth judgment unit is used to determine whether the location information is within a preset range of the response event;

[0148] The fourth execution unit is configured to, if so, synchronously publish the response event to the online devices pre-connected to the airport management terminal, and adaptively generate a corresponding processing time based on the response event, wherein the online devices specifically include airport traffic equipment and airport passenger equipment.

[0149] In this embodiment, the system obtains the passenger's location information in the area surrounding the airport based on the passenger's mobile terminal pre-connected to the airport management terminal. The system then determines whether this location information is within a preset range of the response event to execute corresponding steps. For example, if the system determines that the passenger's location information in the area surrounding the airport is not within the preset range of the response event, the system considers the passenger's location to be outside the scope requiring attention or processing, meaning the feedback information is unreliable. Therefore, the current response event has no direct impact on the passenger. The system checks and adjusts the preset range of the response event to ensure coverage of all potentially affected areas, expanding the preset range of the response event according to the actual situation to ensure it includes potentially affected areas. Simultaneously, if the passenger is indeed outside the event's impact range, the system chooses not to send relevant notifications to avoid unnecessary interference. It records cases of mismatch between passenger location information and the event range, generates possible causes, and establishes a passenger feedback mechanism, allowing passengers to manually confirm or adjust their location to improve positioning accuracy. Conversely, if the system determines that the passenger's location information in the area surrounding the airport is within the preset range of the response event, the system considers the passenger's location to be affected by the response event, meaning the feedback information is authentic. The system will then... The system synchronously publishes response events to pre-connected online devices on the airport management terminal. These devices include airport traffic equipment and passenger equipment. The system adaptively generates corresponding processing times based on the response events. By synchronizing response events to these pre-connected devices in real time, the system can respond quickly and take appropriate measures, shortening event processing time. When traffic congestion or emergencies occur, airport traffic equipment such as traffic lights and signs can be adjusted immediately to reduce the impact on passenger travel. Simultaneously, timely dissemination of response event information to passengers improves their awareness of the current situation and helps them make safer decisions. If severe weather events affect the area surrounding the airport, the system can issue warnings to passengers through passenger equipment such as information screens or mobile applications, advising them to adjust their travel plans. Based on the adaptively generated processing times for response events, the system can rationally allocate resources, optimize airport traffic management, and avoid resource waste or untimely dispatch. During peak periods, if the system predicts severe congestion in a certain area, it can pre-arrange more transportation or open additional lanes to alleviate pressure. Through real-time updates and responses, the system can reduce passenger waiting time and provide a smoother and more comfortable travel experience.

[0150] In this embodiment, the integration module further includes:

[0151] The second data acquisition unit is used to collect real-time data uploaded to the cloud data platform based on the traffic organization preset by the airport management terminal, and to uniformly deploy the traffic organization on the pre-built cloud data platform. The traffic organization specifically includes traffic management departments, airport management departments and weather forecasting agencies.

[0152] The fifth judgment unit is used to determine whether the real-time data has the corresponding timestamp format;

[0153] The fifth execution unit is used to transfer obsolete data that does not have the timestamp format to the information silo preset by the cloud data platform if no, and to restrict public access of the obsolete data in the information silo according to the access permissions preset by the airport management terminal.

[0154] In this embodiment, the system is based on pre-set traffic organization in the airport management terminal. This traffic organization specifically includes traffic management departments, airport management departments, and weather forecasting agencies. These traffic organizations are uniformly deployed on a pre-built cloud data platform. The system collects real-time data uploaded by these traffic organizations to the cloud data platform. Then, the system determines whether this real-time data has the corresponding timestamp format to execute the corresponding steps. For example, when the system determines that the real-time data uploaded by the traffic organizations to the cloud data platform has the corresponding timestamp format, the system considers this data to have been correctly labeled with time information and can accurately reflect the time point of data generation or collection. The system will then check whether all uploaded data not only has timestamp information... The data is formatted with timestamps and includes all necessary fields and content to ensure data integrity. A data validation program is run to confirm that all data correctly includes timestamps and other necessary information. Real-time data with timestamps is synchronized and integrated in chronological order to ensure temporal consistency. All uploaded data is sorted by timestamp to ensure data processing follows chronological logic. Timestamp-verified data is stored on a cloud data platform and backed up regularly to ensure data security. A regular backup mechanism is implemented to prevent data loss and ensure rapid recovery in case of system failure or data loss. For example, if the system determines that the real-time traffic organization data uploaded to the cloud data platform does not have timestamps, the system will perform a backup. If the data lacks the required timestamp format, the system will consider it to be without time information and unable to accurately reflect the corresponding collection time. The system will then transfer this invalid data to a pre-defined information silo on the cloud data platform. Access to this invalid data will be restricted based on pre-set access permissions on the airport management terminal. By transferring data without the correct timestamp format to the information silo, the system avoids contaminating the overall dataset with low-quality data, ensuring sufficient accuracy and timeliness of the data used for analysis and decision-making. If some traffic flow data fails to be correctly timestamped due to format issues, the system will isolate this data to prevent it from affecting the accuracy of the traffic flow prediction model. Meanwhile, by restricting public access to obsolete data, the system can prevent the misuse of this unverified data, thereby avoiding erroneous decisions or judgments caused by incorrect data. When airport management makes traffic scheduling decisions, misusing data without time stamps may lead to unreasonable resource allocation. Isolating such data can effectively avoid this situation. Furthermore, isolating data without timestamps in information silos can provide a safe storage area for subsequent data cleaning and repair, making it possible to reuse the data after repair. In other words, authorized technical personnel can periodically check the data in information silos, timestamp the parts that can be repaired, and reuse them, rather than discarding the data completely.

[0155] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for predicting passenger flow in airport ground transportation, characterized in that, Includes the following steps: Based on the real-time data pre-collected by the airport management terminal, the real-time data is integrated and summarized to form a dynamic data pool. The real-time data specifically includes flight dynamic data, traffic flow data, weather forecast data, and passenger movement data. Determine whether the dynamic data pool can execute the data update strategy preset by the airport management terminal; If possible, the pre-trained prediction model is applied to perform rolling predictions on the latest data in the dynamic data pool to obtain prediction results. Based on the prediction results, the traffic management measures preset by the airport management terminal are adaptively triggered to dynamically schedule traffic resources in the area surrounding the airport. The airport management terminal receives feedback information uploaded by passengers in the area surrounding the airport. Specifically, the traffic management measures include adjusting traffic light durations, optimizing vehicle scheduling, and issuing congestion warnings. The traffic resources specifically include increasing traffic capacity, opening additional lanes, and activating temporary parking lots. Determine whether the feedback information matches the response event corresponding to the area surrounding the airport; If so, the flight information pre-entered by the passenger to the airport management terminal is collected, the traffic data corresponding to the response event and the flight information are input into the prediction model, the dynamic recommended route required by the passenger is generated, and the dynamic recommended route is synchronized from the airport management terminal to the passenger's preset terminal. Specifically, the dynamic recommended route is the fastest transportation route obtained by combining available transportation tools within the airport. The step of determining whether the feedback information matches the response event corresponding to the area surrounding the airport further includes: Based on the mobile terminal pre-connected to the airport management terminal, the location information of the passenger in the area surrounding the airport is obtained; Determine whether the location information is within a preset range of the response event; If so, the response event will be synchronously published to the online devices pre-connected to the airport management terminal, and a corresponding processing time will be adaptively generated based on the response event. The online devices specifically include airport traffic equipment and airport passenger equipment.

2. The airport ground traffic passenger flow prediction method according to claim 1, characterized in that, After the step of receiving feedback information uploaded by passengers in the area surrounding the airport through the airport management terminal, the method further includes: Based on the passenger information pre-entered into the airport management terminal, the passenger group corresponding to the passenger is identified, and the traffic influencing factors of the passenger group on the feedback information are collected. The passenger information specifically includes ticketing information, waiting area entry status and boarding record. The passenger group specifically includes business travelers, family travelers and leisure travelers. The traffic influencing factors specifically include travel time, distance and weather conditions. Determine whether the traffic-influencing factors are within the preset average value of the airport management terminal; If not, then based on the existing management strategy of the airport management terminal, a dedicated channel window for the passenger group is established. Based on the dedicated channel window, the passenger group is configured with the required service content. The dedicated channel window and the required service content are updated to the airport management terminal in real time. The airport management terminal displays the usage status of the dedicated channel window and the required service content to the passengers in real time. Specifically, the dedicated channel window includes a dedicated security check channel, a dedicated channel for family passengers, and a dedicated boarding gate.

3. The airport ground transportation passenger flow prediction method according to claim 1, characterized in that, The step of applying a pre-trained prediction model to perform rolling predictions on the latest data in the dynamic data pool and obtaining the prediction results further includes: The difference between the predicted result and the actual value is calculated based on the preset error metric. The trend of the prediction error is identified by the difference value, and the prediction deviation for a specific time period is collected from the trend. Determine whether the prediction deviation has a corresponding prediction weight, wherein the prediction weight specifically includes traffic weight, weather weight, and flow weight; If so, the model parameters of the prediction model are adjusted according to the prediction deviation, corresponding weight conditions are added to the prediction model in real time, the interaction features between the latest data and the weight conditions are obtained, the joint impact parameters of the prediction results are generated based on the interaction features, the joint impact parameters are input into the prediction model, and the joint impact value of the joint impact parameters on the prediction results is revealed. Specifically, the weight conditions include sudden weather data, holidays and special events, traffic conditions and traffic accidents and flight delay information.

4. The airport ground transportation passenger flow prediction method according to claim 1, characterized in that, The step of generating the dynamically recommended route required by the passenger and synchronizing the dynamically recommended route from the airport management terminal to the passenger's preset terminal further includes: Based on the passenger's preset transportation preferences on the airport management terminal, obtain the passenger's feedback on the dynamically recommended route; Determine whether the feedback indicates agreement to apply the dynamically recommended route; If not, then routes that do not include the travel preference in the dynamically recommended routes are blocked, the specific time difference between the route content and the best recommended route is generated, and the specific time difference is synchronized to the passenger's preset mobile device in real time.

5. The airport ground transportation passenger flow prediction method according to claim 1, characterized in that, The step of determining whether the dynamic data pool can execute the data update strategy preset by the airport management terminal further includes: Based on the simulated load preset by the airport management terminal, the maximum number of concurrent requests that the dynamic data pool can accommodate is obtained; Determine whether the dynamic data pool can operate stably for a preset period of time when it is at the maximum concurrent request volume; If so, the performance parameters of the dynamic data pool are monitored in real time, and a preset rapid simulated traffic is input into the dynamic data pool. Abnormal information of the dynamic data pool under different loads is recorded. The performance parameters specifically include response time, throughput, and resource usage, and the abnormal information specifically includes timeouts, failed requests, and resource exhaustion.

6. The airport ground traffic passenger flow prediction method according to claim 1, characterized in that, The step of integrating and summarizing the real-time data pre-collected based on the airport management terminal to form a dynamic data pool includes: Based on the traffic organization preset by the airport management terminal, the traffic organization is uniformly deployed on a pre-built cloud data platform, and real-time data uploaded by the traffic organization to the cloud data platform is collected. The traffic organization specifically includes traffic management departments, airport management departments, and weather forecasting agencies. Determine whether the real-time data has the corresponding timestamp format; If not, the obsolete data that does not have the timestamp format will be transferred to the information silo preset by the cloud data platform, and public access to the obsolete data in the information silo will be restricted according to the access permissions preset by the airport management terminal.

7. An airport ground traffic passenger flow prediction system, characterized in that, include: The integration module is used to integrate and summarize the real-time data pre-collected by the airport management terminal to form a dynamic data pool. The real-time data specifically includes flight dynamic data, traffic flow data, weather forecast data, and passenger movement data. The judgment module is used to determine whether the dynamic data pool can execute the data update strategy preset by the airport management terminal; The execution module is used to, if possible, apply a pre-trained prediction model to perform rolling predictions on the latest data in the dynamic data pool, obtain prediction results, and adaptively trigger traffic management measures preset by the airport management terminal based on the prediction results, dynamically dispatch traffic resources in the area surrounding the airport, and receive feedback information uploaded by passengers in the area surrounding the airport through the airport management terminal. The traffic management measures specifically include adjusting traffic light durations, optimizing vehicle scheduling, and issuing congestion warnings. The traffic resources specifically include increasing traffic capacity, opening additional lanes, and activating temporary parking lots. The second judgment module is used to determine whether the feedback information matches the response event corresponding to the area surrounding the airport. The second execution module is used to collect the flight information pre-entered by the passenger to the airport management terminal if the condition is met, input the traffic data corresponding to the response event and the flight information into the prediction model, generate the dynamic recommended route required by the passenger, and synchronize the dynamic recommended route from the airport management terminal to the passenger's preset terminal. Specifically, the dynamic recommended route is the fastest transportation route obtained by combining available transportation tools within the airport. The second judgment module also includes: The third acquisition unit is used to acquire the location information of the passenger in the area surrounding the airport based on the mobile terminal pre-connected to the airport management terminal of the passenger. The fourth judgment unit is used to determine whether the location information is within a preset range of the response event; The fourth execution unit is configured to, if so, synchronously publish the response event to the online devices pre-connected to the airport management terminal, and adaptively generate a corresponding processing time based on the response event, wherein the online devices specifically include airport traffic equipment and airport passenger equipment.

8. The airport ground traffic passenger flow prediction system according to claim 7, characterized in that, Also includes: The data collection module is used to identify the passenger group corresponding to the passenger based on the passenger information pre-entered into the airport management terminal, and to collect the traffic influencing factors of the passenger group on the feedback information. The passenger information specifically includes ticketing information, waiting area entry status and boarding record. The passenger group specifically includes business travelers, family travelers and leisure travelers. The traffic influencing factors specifically include travel time, distance and weather conditions. The third judgment module is used to determine whether the traffic influencing factors are within the preset average value of the airport management terminal; The third execution module is used to, if not, establish a dedicated channel window for the passenger group based on the original management strategy of the airport management terminal, configure the required service content to the passenger group according to the dedicated channel window, update the dedicated channel window and the required service content to the airport management terminal in real time, and display the usage status of the dedicated channel window and the required service content to the passengers in real time through the airport management terminal. The dedicated channel window specifically includes a dedicated security check channel, a dedicated channel for family passengers, and a dedicated boarding gate.

9. The airport ground traffic passenger flow prediction system according to claim 7, characterized in that, The execution module further includes: The acquisition unit is used to calculate the difference between the prediction result and the actual value based on a preset error metric, identify the trend of prediction error through the difference value, and collect the prediction deviation for a specific time period from the trend. The judgment unit is used to determine whether the prediction deviation has a corresponding prediction weight, wherein the prediction weight specifically includes traffic weight, weather weight and flow weight; An execution unit is configured to, if so, adjust the model parameters of the prediction model according to the prediction deviation, add corresponding weight conditions to the prediction model in real time, obtain the interaction features between the latest data and the weight conditions, generate joint impact parameters of the prediction result based on the interaction features, input the joint impact parameters into the prediction model, and reveal the joint impact value of the joint impact parameters on the prediction result. Specifically, the weight conditions include sudden weather data, holidays and special events, traffic conditions and traffic accidents and flight delay information.

Images