Intelligent Management Method for Portable Boxes Used in Subway Rides

Abstract

The portable box intelligent management method for subway rides provided by this application relates to the technical field of data management. It retrieves the portable box management logs of K stations on the target subway line within a preset historical window, extracts indicators from the logs respectively according to the usage volume, storage volume, and in-cabinet retention volume indicators, traverses each indicator set for indicator central tendency analysis to determine the liquidity factors of the K stations, and then obtains the deployment coefficients of the K stations. By performing fluctuation identification, the steady-state coefficients of the K stations are obtained. The monitoring bandwidth sets of the K stations are determined by using a monitoring step length identifier to identify the deployment and steady-state coefficients, and a portable box monitoring transmission branch is constructed for monitoring management. This solves the problems of congestion during the security check process, inconvenience to passengers, and increased management costs caused by the inability to efficiently and conveniently manage passengers' items, achieving the effects of improving the operation efficiency and safety of the entire subway network and ensuring the riding and management experience.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and particularly to an intelligent management method for a portable box for subway rides. Background Art

[0002] With the rapid development of urban public transportation, the subway, as an important part of urban transportation, has an increasing passenger flow. To ensure public safety, subway security inspection has become an essential procedure for passengers entering the station. Existing subway security inspection methods mainly include X-ray scanning, metal detection, etc. for the packages and items carried by passengers. However, these methods usually can only statically identify potential dangerous goods and cannot effectively manage passengers' non-dangerous personal items, such as some small tools or equipment that must be carried but cannot be taken into the carriage. Although these items do not pose a direct safety threat, they often cause cumbersome operations and passenger complaints during the security inspection process, which not only reduces the efficiency of entering the station but also increases the travel burden of passengers. Moreover, the security inspection standards and enforcement efforts may vary at different subway stations, resulting in inconsistent passenger experiences. Although some subway stations have tried to solve the problem of passengers' item storage by providing temporary storage cabinets, the security of these solutions cannot be fully guaranteed, the flexibility of fixed-point storage is poor, and additional manpower and material resources are required, increasing the operating cost.

[0003] In summary, there are often technical problems in existing subway management, such as the inability to efficiently and conveniently manage passengers' non-dangerous personal items, resulting in congestion during the security inspection process, inconvenience to passengers, and increased management costs. Summary of the Invention

[0004] This application provides an intelligent management method for a portable box for subway rides, aiming to solve the technical problems in existing subway management, such as the inability to efficiently and conveniently manage passengers' non-dangerous personal items, resulting in congestion during the security inspection process, inconvenience to passengers, and increased management costs.

[0005] The intelligent management method for a portable box for subway rides provided by this application includes:

[0006] Retrieve the portable box management logs of K stations on the target subway line within a preset historical window to obtain K historical portable box management logs, where K is an integer greater than or equal to 1; respectively extract indicators from the K historical portable box management logs according to the usage volume indicator, the deposit volume indicator, and the in-cabinet retention volume indicator to obtain K usage volume indicator sets, K deposit volume indicator sets, and K in-cabinet retention volume indicator sets; traverse the K usage volume indicator sets and the K deposit volume indicator sets for indicator central tendency analysis to determine K station liquidity factors, and respectively divide the K station liquidity factors by the sum of the K station liquidity factors to obtain K station allocation coefficients; traverse the K in-cabinet retention volume indicator sets for fluctuation identification to obtain K station steady-state coefficients; use a monitoring step identifier to identify the K station allocation coefficients and the K station steady-state coefficients to determine a set of K station monitoring bandwidths, where the set of K station monitoring bandwidths includes K first station monitoring bandwidths and K second station monitoring bandwidths, and each first station monitoring bandwidth is less than the corresponding second station monitoring bandwidth; construct K station portable box monitoring transmission branches for the target subway line based on the K first station monitoring bandwidths and the K second station monitoring bandwidths; use the K station portable box monitoring transmission branches to monitor and extract the portable box management data of the K stations, and send the extracted data to the background management terminal to manage the portable boxes of the K stations.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The portable box intelligent management method for subway rides provided by this application obtains K historical portable box management logs by retrieving the portable box management logs of K stations on the target subway line within a preset historical window, where K is an integer greater than or equal to 1; respectively extract indicators from the K historical portable box management logs according to the usage volume indicator, the deposit volume indicator, and the in-cabinet retention volume indicator to obtain K usage volume indicator sets, K deposit volume indicator sets, and K in-cabinet retention volume indicator sets; traverse the K usage volume indicator sets and the K deposit volume indicator sets for indicator central tendency analysis to determine K station liquidity factors, and divide each of the K station liquidity factors by the sum of the K station liquidity factors to obtain K station allocation coefficients; traverse the K in-cabinet retention volume indicator sets for fluctuation identification to obtain K station steady-state coefficients; use a monitoring step identifier to identify the K station allocation coefficients and the K station steady-state coefficients to determine a set of K station monitoring bandwidths, where the set of K station monitoring bandwidths includes K first station monitoring bandwidths and K second station monitoring bandwidths, and each first station monitoring bandwidth is less than the corresponding second station monitoring bandwidth; based on the K first station monitoring bandwidths and the K second station monitoring bandwidths, construct K station portable box monitoring transmission branches for the target subway line; use the K station portable box monitoring transmission branches to monitor and extract the portable box management data of the K stations, and send the extracted data to the background management terminal to manage the portable boxes of the K stations, solving the technical problem in existing subway management that it is impossible to efficiently and conveniently manage passengers' non-dangerous personal items, resulting in congestion during the security check process, inconvenience to passengers, and increased management costs. By using portable boxes and performing intelligent management, the technical effect of effectively and conveniently managing passengers' items, dynamically adjusting the resource allocation of each station, thereby improving the operation efficiency and safety of the entire subway network, and ensuring the riding and management experience is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 FIG. is a schematic flow chart of the portable box intelligent management method for subway rides provided by this application.

[0010] Figure 2 FIG. is a schematic flow chart of determining the station liquidity factor in the portable box intelligent management method for subway rides provided by this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0011] In order to make the objectives, technical solutions, and advantages of this application clearer, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0012] To effectively reduce security risks, small tool-like items are frequently seized during subway security checks. Although these items may not necessarily pose high risks, they are restricted from being carried according to regulations, leading to a large number of passenger complaints and disputes, affecting the smoothness of security checks and the management burden, and at the same time reducing the passenger experience. The existing security check policies have not effectively solved the problem that passengers cannot take the subway due to carrying daily tool-like items, and more user-friendly solutions are needed. Therefore, this application proposes the concept of setting up shared convenience boxes and intelligent management, allowing passengers to temporarily store small tool-like items in special convenience boxes, which are managed and transported through the subway system and retrieved when needed. Moreover, the convenience boxes are equipped with intelligent locks and real-time positioning systems, and are effectively managed through big data and mobile applications, ensuring safety and convenience. This solution improves the operation efficiency of the subway and passenger satisfaction, solves the technical problem in the existing subway management that non-dangerous personal items of passengers cannot be efficiently and conveniently managed, resulting in congestion during the security check process, passenger inconvenience, and increased management costs. By using portable boxes and intelligent management, it achieves the technical effect of effectively and conveniently managing passengers' items, dynamically adjusting the resource allocation of each station, thereby improving the operation efficiency and safety of the entire subway network and ensuring the riding and management experience.

[0013] Embodiment, such as Figure 1 As shown, this application provides a method for intelligent management of portable boxes for subway rides, and the method includes:

[0014] Retrieve the portable box management logs of K stations on the target subway line within a preset historical window to obtain K historical portable box management logs, where K is an integer greater than or equal to 1.

[0015] Specifically, retrieve the management logs of portable boxes at K stations on the target subway line within a preset historical window. Herein, the preset historical window refers to a set specific time range, such as the past week, month, or longer, aiming to analyze and evaluate the usage of portable boxes during a specific period. The K stations refer to the stations on the subway line, which can be K selected representative or key stations, where K is an integer greater than or equal to 1, meaning at least one station is included and can be extended as needed to include more stations. This selection can be based on parameters such as passenger flow, portable box usage frequency, or other relevant parameters at each station. By retrieving the management logs of each station within the preset historical window, detailed data on the usage volume, storage status, and retrieval status of portable boxes at each station can be obtained. These management logs include, but are not limited to, the deposit and retrieval times of portable boxes, the numbers of portable boxes, user information, etc., thus providing an empirical basis for subsequent data analysis and resource allocation. Specifically in operation, this step can be executed through the management system of the subway station. The management system can be configured to automatically extract the above data from the database or allow the operator to manually select specific query conditions to perform data extraction. This flexible data retrieval method enables managers to flexibly adjust the scope of data collection and analysis according to actual operation needs. In summary, by retrieving and obtaining the management logs of portable boxes at K stations, the usage of subway portable boxes can be accurately captured and analyzed, providing data support for further intelligent management and optimization decisions, which not only helps improve the operation efficiency of the subway system but also optimizes the usage experience of passengers.

[0016] Extract indicators from the K historical portable box management logs according to the usage volume indicator, deposit volume indicator, and in-cabinet retention volume indicator respectively, to obtain K usage volume indicator sets, K deposit volume indicator sets, and K in-cabinet retention volume indicator sets.

[0017] Furthermore, the usage volume indicator is the number of portable boxes taken out by users from the station, the deposit volume indicator is the number of portable boxes deposited by users at the station, and the in-cabinet retention volume indicator is the number of portable boxes stored at the station.

[0018] Optionally, immediately following the data collection phase, detailed metric extraction will be performed on the K historical portable case management logs collected. This process mainly involves three key metrics: the usage metric, the deposit metric, and the in-cabinet inventory metric. The usage metric is the number of times a user retrieves a portable case from a specific station. This metric is crucial for measuring the usage frequency of portable cases at each station and providing management with dynamic insights into the demand for portable cases. For example, if the usage metric at a certain station is abnormally high, it indicates that the station may need to increase the supply of more portable cases or conduct more frequent maintenance inspections. The deposit metric refers to the number of portable cases deposited by users at each station. This metric reflects the active utilization of portable cases and provides basic data for predicting the possible storage requirements at each station. Based on this data, the storage capacity of the station can be adjusted to ensure that the needs of passengers are met. The in-cabinet inventory metric represents the total number of portable cases stored at each station at any given point in time. This metric is an important parameter for measuring the utilization rate of the portable case capacity at the station and is particularly crucial for managers to formulate portable case maintenance and scheduling strategies. By monitoring the in-cabinet inventory in real-time, the management system can effectively prevent the over-accumulation or shortage of portable cases and optimize resource allocation. Through the systematic extraction and analysis of the above three metrics, the actual usage needs of passengers at each subway station can be accurately evaluated and responded to, and then the allocation and management strategies of portable cases can be adjusted and optimized. This data-driven approach not only improves the efficiency of resource utilization but also greatly enhances the passenger experience. After implementing these measures, the subway operating unit can manage portable case resources more flexibly and also provides an empirical support basis for possible future expansion.

[0019] Traverse the K usage metric sets and K deposit metric sets for central tendency analysis of the metrics, determine the K station liquidity factors, and divide each of the K station liquidity factors by the sum of the K station liquidity factors to obtain K station deployment coefficients.

[0020] Traverse the K in-cabinet inventory metric sets for fluctuation identification to obtain K station steady-state coefficients.

[0021] Furthermore, perform a central tendency analysis on the usage metrics and deposit metrics. Central tendency analysis mainly refers to determining the aggregation trend or average level of data through statistical methods such as calculating the mean, median, etc. In this step, by traversing the K stations on the subway line, calculate the average usage and deposit of the portable boxes at each station respectively. This analysis can determine the liquidity factor of each station, which measures the liquidity or usage frequency of the portable boxes at each station. Among them, the larger the liquidity factor, the greater the passenger flow at the station, and the relatively more frequent the demand and turnover of the portable boxes. Next, compare the liquidity factor of each station with the sum of the liquidity factors of all stations to calculate the allocation coefficient of each station. This allocation coefficient reflects the proportion of each station on the entire subway line in the allocation of portable box resources and guides how to adjust the allocation of portable boxes according to actual needs. For example, stations with a high liquidity factor may require more portable box supplies or more frequent portable box updates. After that, perform volatility identification on the set of in-cab inventory metrics. The purpose of volatility identification is to determine the stability of the portable box inventory at each station, which is usually achieved by calculating the variance of the inventory and its reciprocal. Specifically, the reciprocal of the variance is defined as the station steady-state coefficient, and the magnitude of this coefficient reflects the degree of fluctuation of the number of portable boxes at the station. The larger the station steady-state coefficient, the smaller the fluctuation of the inventory at the station, that is, the inventory is relatively stable, and such stations do not require frequent monitoring or adjustment of the number of portable boxes; on the contrary, if the steady-state coefficient is small, it indicates that the number of portable boxes at the station fluctuates greatly and requires intensive monitoring and possible portable box allocation. Through the above analysis, the portable box resources at the subway station can be managed and optimized more precisely. To ensure the balance between the supply and demand of portable boxes at each station, the distribution of portable boxes can be dynamically adjusted according to the allocation coefficient and the steady-state coefficient to ensure the effective utilization of portable boxes in the subway system, thereby improving passenger satisfaction and subway operation efficiency.

[0022] Use the monitoring step identifier to identify the K-station allocation coefficients and the K-station steady-state coefficients to determine a set of K-station monitoring bandwidths, where the set of K-station monitoring bandwidths includes K first-station monitoring bandwidths and K second-station monitoring bandwidths, and each first-station monitoring bandwidth is less than the corresponding second-station monitoring bandwidth.

[0023] Exemplarily, a monitoring step recognizer is then used to analyze the deployment coefficients and steady-state coefficients of the K stations to determine the monitoring bandwidth sets for each station. This step is a core part of optimizing the data monitoring frequency and resource allocation in the subway portable box management system. The monitoring step recognizer is a model built based on machine learning technology for identifying and predicting the optimal data monitoring frequency. The so-called monitoring bandwidth refers to the time interval of data monitoring, that is, how often the data is collected. In the present invention, the specific task of the monitoring step recognizer is to determine the suitable monitoring bandwidth for each station according to the deployment coefficient and steady-state coefficient of the station. The deployment coefficient is calculated based on the station mobility factor, which reflects the relative proportion of portable box flow at the station in the entire subway network. The steady-state coefficient is calculated based on the volatility of the in-cab inventory, indicating the fluctuation stability of the number of portable boxes at the station. These two coefficients jointly affect the monitoring density that a station should have in the portable box management system.

[0024] In a specific implementation, the monitoring step recognizer outputs the corresponding monitoring bandwidth sets by inputting the deployment coefficients and steady-state coefficients of the K stations. This set includes the first monitoring bandwidth and the second monitoring bandwidth for each station. Among them, the first monitoring bandwidth is set to a shorter time interval for collecting data more frequently in the case of high mobility or large fluctuations; while the second monitoring bandwidth is longer and is suitable for a stable environment with low mobility or small fluctuations to reduce unnecessary data processing and storage pressure. For example, if a station has a high deployment coefficient, indicating that the portable boxes at this station are used and transferred frequently, the monitoring step recognizer may recommend a shorter first monitoring bandwidth. On the contrary, if a station has a high steady-state coefficient, indicating that the number of its portable boxes is relatively stable, the monitoring step recognizer may recommend a longer second monitoring bandwidth. In this way, the monitoring step recognizer can provide an optimized and dynamically adjusted monitoring strategy for each station, ensuring that the portable box management system can effectively respond to the actual needs of different stations, while avoiding resource waste and improving the overall efficiency and response ability of the system. This intelligent monitoring strategy is the key to improving the efficiency of subway portable box management and helps to achieve more refined and adaptive operation management.

[0025] Construct the K-station portable box monitoring and transmission branches of the target subway line based on the K first-station monitoring bandwidths and the K second-station monitoring bandwidths.

[0026] Use the K-station portable box monitoring and transmission branches to monitor and extract the portable box management data of the K stations, and send the extracted data to the background management terminal to manage the portable boxes of the K stations.

[0027] Specifically, the monitoring transmission branch refers to a data transmission channel established specifically for the monitoring of portable boxes at subway stations. Each station will have different data transmission frequencies and priorities according to the settings of its first monitoring bandwidth and second monitoring bandwidth. The first monitoring bandwidth, usually shorter, is used for frequent data updates at stations with frequent user activities or high mobility of portable boxes; while the second monitoring bandwidth is longer and is applicable to stations with relatively fewer activities or more stable use of portable boxes. By constructing these monitoring transmission branches, dynamic monitoring of the usage of portable boxes at each station can be achieved. This monitoring includes not only the access times of portable boxes, but also multi-dimensional data such as time, frequency, and user type. These data are regularly sent to the background management terminal through the monitoring transmission branch, and the background management terminal uses these data for comprehensive analysis to optimize the allocation of portable box resources across the entire subway network. In actual operation, for example, a station with high mobility, such as the city center or a major transportation hub, may be set with a shorter first monitoring bandwidth to ensure that the usage of portable boxes can be quickly fed back to the background management system. On the contrary, for a relatively remote or less frequently used station, the data update frequency may be lower, so the second monitoring bandwidth will be longer. Once the data is extracted by the monitoring transmission branch and sent to the background management terminal, the background system will adjust the allocation of portable boxes according to the received real-time information, such as reconfiguring the number of portable boxes and adjusting the maintenance plan. This data-driven management method not only improves the efficiency and accuracy of portable box management, but also enhances the subway system's response ability to changes in passenger demand, ultimately achieving the goals of improving passenger satisfaction and optimizing operating costs. Through the above steps, the efficient operation of the subway portable box management system is ensured, and the intelligent management of subway portable box resources is realized by finely adjusting the monitoring frequency and optimizing the data flow.

[0028] Furthermore, as Figure 2 shown, traverse the K sets of usage metrics and the K sets of deposit metrics for central tendency analysis of the metrics to determine K station mobility factors, including:

[0029] Traverse the K sets of usage metrics for mean calculation to determine the means of the K usage metrics.

[0030] Use the means of the K usage metrics as the K initial centroids, and iterate the K initial centroids according to a preset centroid iteration update formula to obtain K stage iteration centroids.

[0031] Continue to iterate based on the K stage iteration centroids until a preset iteration stop condition is met, obtain K updated iteration centroids, and use the K usage metrics corresponding to the K updated iteration centroids as the K concentrated usage metrics.

[0032] Perform an analysis of the central tendency of the K sets of deposit quantity indicators to determine K central deposit quantity indicators.

[0033] Perform a weighted calculation on the K central usage quantity indicators and the K central deposit quantity indicators to obtain the K station liquidity factors.

[0034] Furthermore, the method further includes:

[0035] ;

[0036] where is the centroid of the stage iteration, is the set of usage quantity indicators, is the i-th usage quantity indicator in the set of usage quantity indicators, is the starting centroid, is the Gaussian function.

[0037] In a specific embodiment, traverse the set of usage quantity indicators for K stations and calculate the average value of the usage quantity indicators for each station. The purpose of this step is to determine the average usage level of each station and provide a baseline for subsequent data analysis. The average value of the usage quantity indicators refers to the average number of times the portable boxes at each station are used within a certain period of time, which is an intuitive indicator to measure the usage frequency of the station. Next, taking these average values of the usage quantity indicators as the starting points, that is, as the starting centroids, use a preset centroid iteration update formula to iteratively update these starting centroids. Here, the centroid iteration update is a commonly used technique in cluster analysis for optimizing and refining the representation of the usage quantity data of each station. By gradually adjusting these centroids, the subtle changes in the usage patterns of each station can be captured more accurately. Specifically, the centroid iteration update formula is , where is the centroid of the stage iteration, is the set of usage quantity indicators, is the i-th usage quantity indicator in the set of usage quantity indicators, is the starting centroid, is the Gaussian function. This iterative process will continue until a preset iterative stop condition is met, such as the change amplitude of the centroid being less than a set threshold, which indicates that the centroid has stabilized and the changes brought about by further iteration are negligible. At this time, the finally obtained centroid is defined as the updated iterative centroid, which represents the concentrated usage metrics of each station and provides a more accurate description of the station usage pattern. At the same time, a similar central tendency analysis is performed on the set of deposit metrics for the K stations to determine the concentrated deposit metrics for each station, which is a measure of the central tendency of the portable box deposit behavior at each station and provides data support for understanding the portable box replenishment and demand at the stations. Finally, these obtained concentrated usage metrics and concentrated deposit metrics are weighted and calculated to obtain the liquidity factor for each station. In this step, the weighted calculation takes into account the importance and liquidity requirements of different stations to ensure that the liquidity factor can truly reflect the activity level of the stations in the subway network and the complexity of portable box management. Through the above detailed data processing and analysis steps, a scientific and systematic data support system can be provided for subway portable box management, making the allocation and scheduling of portable boxes more accurate and efficient, thereby improving the operation efficiency of the entire subway system and passenger satisfaction. This method not only enhances the ability of data-driven decision-making but also optimizes resource utilization and service quality.

[0038] Furthermore, the method further includes:

[0039] Traverse and calculate the distances from the set of K usage metric means to the K-stage iterative centroids to obtain a set of K-stage centroid deviation distances.

[0040] Based on the K-stage iterative centroids, continue the iteration to obtain K updated-stage iterative centroids and obtain a set of K updated-stage centroid deviation distances.

[0041] Respectively calculate the sum of the K updated-stage distances of the set of K updated-stage centroid deviation distances and the sum of the K-stage distances of the set of K-stage centroid deviation distances. When the sum of the K updated-stage distances is greater than or equal to the sum of the K-stage distances, the preset iterative stop condition is met, and the K updated-stage iterative centroids are taken as the K updated iterative centroids.

[0042] Specifically, traverse and calculate the distances from the K sets of mean usage metrics to the K stage iteration centroids. The purpose is to evaluate the deviation degree between the current centroid position and the actual data points (i.e., the means of the usage metrics). The so-called distance refers to the Euclidean distance, which is a method to measure the straight-line distance between two points. Calculating these distances can show whether the change in the centroid position after each iteration is close enough to the actual usage pattern of the user. Then, continue to iterate based on the obtained K stage iteration centroids, which means updating the centroid position of each station again based on the iteration result of the previous stage to try to reduce the deviation between the actual usage data points and the centroid. This iterative process is carried out through a preset algorithm, such as the least squares method or the gradient descent method, aiming to optimize the usage efficiency and responsiveness of the entire system. After that, obtain the updated set of centroid deviation distances for each stage. This set records the distances between the mean usage metrics of each station and the updated centroid after a new round of iteration. Tracking these distances is crucial for evaluating the efficiency of the iterative process because it directly reflects whether the adjustment of the iterative algorithm is moving in the direction of reducing the overall deviation. Further, calculate the sum of the K updated stage distances for the K updated stage centroid deviation distance sets and the sum of the K stage distances for the K stage centroid deviation distance sets respectively. This step involves comparing two sets of data: one is the total distance before a new round of iteration, and the other is the total distance after iteration. By this comparison, it can be judged whether the iteration still needs to continue. Finally, when the sum of the K updated stage distances is greater than or equal to the sum of the K stage distances, it indicates that further iteration has not significantly improved the accuracy of the centroid. At this time, the preset iteration stop condition is met, which means that the current K updated iteration centroids are accurate enough relative to the actual usage data and can be used as the final model parameters for subsequent portable box management decisions. Through the above steps, it is ensured that the usage pattern of portable boxes at subway stations is accurately captured by a data-driven method, thereby optimizing the configuration and management of portable boxes and improving service quality and user satisfaction.

[0043] Furthermore, traverse the K sets of in-cabinet inventory metrics for fluctuation identification to determine the K station steady-state factors for the K stations, including:

[0044] Traverse the K sets of in-cabinet inventory metrics for fluctuation variance calculation, and take the reciprocals of the calculation results as the K station steady-state factors respectively.

[0045] Divide each of the K station steady-state factors by the sum of the K station steady-state factors to obtain the K station allocation coefficients.

[0046] Optionally, traverse the K in-cab inventory indicator sets to calculate the fluctuation variance. In this step, the in-cab inventory indicator of each station, i.e., the number of portable boxes recorded within a given time, is used to calculate the variance. Variance is a measure in statistics that indicates the degree of dispersion of a set of values and can show the consistency and stability of the changes in the number of portable boxes at each station. The formula for variance is to sum the squares of the differences between each value and the average and then take the average. The larger the result, the greater the fluctuation of the data. Then, the reciprocal of the calculated variance is defined as the station stability factor. In statistics, the reciprocal of variance can be regarded as a measure of stability. The smaller the variance, the larger its reciprocal, indicating that the data is more stable. Therefore, the magnitude of the station stability factor directly reflects the stability of the number of portable boxes at each station. A station with a higher stability factor indicates that the number of its portable boxes fluctuates less and is relatively more stable; conversely, it fluctuates more and has lower stability. Immediately afterwards, in order to further apply these stability factors to the actual management of portable boxes, it is necessary to compare the stability factor of each station with the sum of the stability factors of all stations to obtain the station deployment coefficient. This coefficient is calculated by dividing the stability factor of a single station by the sum of the stability factors of all stations and is used to indicate the relative stability weight of each station in the entire subway network. This coefficient can be used to dynamically adjust the allocation of portable boxes, giving priority to stations with large fluctuations and low stability for resource allocation to enhance the adaptability and responsiveness of the entire system. Through the above analysis and calculation, not only can the storage stability of portable boxes at each station be accurately evaluated, but also the resource allocation can be optimized based on the actual stability data, improving the overall efficiency and effectiveness of subway portable box management. This data-based method ensures the scientificity and practicality of resource allocation and significantly improves the quality of subway portable box services and passenger satisfaction.

[0047] Furthermore, use a monitoring step identifier to identify the K station deployment coefficients and the K station stability coefficients to determine a set of K station monitoring bandwidths, including:

[0048] Obtain multiple sample station deployment coefficients, multiple sample station stability coefficients, and multiple sample sets of station monitoring bandwidths as training data.

[0049] Use the training data to perform supervised training on a framework constructed based on a feedforward neural network to learn the mapping relationship between the station deployment coefficients, the station stability coefficients, and the set of station monitoring bandwidths until the training converges to obtain the trained monitoring step identifier.

[0050] Furthermore, to construct an effective monitoring step identifier, sufficient training data needs to be collected, including deployment coefficients, steady-state coefficients, and corresponding monitoring bandwidth sets from multiple sample sites. Here, the deployment coefficient refers to the weight calculated based on the site steady-state factor, reflecting the importance of each site in the subway network and the priority of resource allocation; while the steady-state coefficient describes the stability of the number of portable boxes fluctuating at each site. Together, they determine the monitoring requirements and frequencies of the sites. Next, using these training data, supervised training will be carried out based on the framework of a feedforward neural network. A feedforward neural network is a deep learning model that processes data through forward connections between layers and does not contain feedback connections. This network structure is particularly suitable for handling complex pattern recognition and prediction tasks, such as predicting the optimal monitoring bandwidth for different sites in the present invention. During the training process, the network gradually learns how to predict the appropriate monitoring bandwidth according to the input site deployment coefficients and steady-state coefficients by repeatedly adjusting the network weights. Supervised training means that each input (site deployment coefficients and steady-state coefficients) corresponds to an expected output (monitoring bandwidth set), and the training objective is to minimize the error between the network prediction value and the actual value. This is usually achieved by defining a loss function (such as mean squared error), and then using gradient descent or other optimization algorithms to adjust the network parameters to minimize the loss function value. This training process will continue until the model converges, that is, when the prediction error of the network no longer decreases significantly, or reaches a predetermined training cycle or error threshold. This stage is a crucial link to ensure that the model can reliably predict the monitoring bandwidth and is also a process of model optimization. Through this method, the monitoring step identifier learns how to dynamically adjust the monitoring frequency according to the actual operating conditions and fluctuation characteristics of the sites, making data collection both efficient and meeting actual needs. After obtaining the trained monitoring step identifier, this model will be deployed into the actual subway site portable box monitoring system. In operation, the system will input the current deployment coefficients and steady-state coefficients of each site into the identifier in real time, and the identifier will output the corresponding monitoring bandwidth recommendations. For example, for a given site, if its steady-state coefficient is high (i.e., the inventory fluctuation is small), the monitoring step identifier may recommend a longer monitoring interval, thus reducing unnecessary data processing and communication burdens; conversely, for a site with larger fluctuations, a shorter monitoring interval is recommended to ensure that any important changes can be captured in a timely manner. Such an intelligent monitoring system not only optimizes the use of resources, reduces operation and maintenance costs, but also improves the flexibility and response speed of subway portable box management. It allows subway operators to more accurately understand and respond to the actual needs of passengers, improving service quality and passenger satisfaction. In addition, through continuous data monitoring and analysis, this system can also help operators identify potential operation problems or optimization points, thereby further improving the overall efficiency and safety of the subway system.

[0051] Furthermore, the method further includes:

[0052] Extract the K monitoring extraction data sets of the K stations received by the background management terminal within a preset monitoring window.

[0053] Perform information entropy calculation on the K monitoring extraction data sets to determine K information entropies.

[0054] Based on the magnitudes of the K information entropies, perform resource balancing configuration on the K processing units in the background management terminal for managing the K stations, and use the configured K processing units to manage the portable boxes of the K stations.

[0055] Specifically, in order to effectively manage the subway station portable boxes and optimize resource allocation, a method based on information entropy calculation is introduced to evaluate and adjust the monitoring and management strategies of each station. First, extract the K monitoring extraction data sets of the K stations within a preset monitoring window from the background management terminal. The preset monitoring window refers to a specific time range set by the administrator, such as the past 24 hours or a week. This time window is used to capture all activity data related to the use of portable boxes, including but not limited to the access times, access times, user interaction situations, etc. of the portable boxes. This step is the starting point of the data processing flow and provides the original input for subsequent analysis. Next, perform information entropy calculation on these monitoring extraction data sets to determine the K information entropies of each station. Information entropy is a statistic that measures the randomness or uncertainty of data, and its calculation formula is based on the probabilities of different events in the data set. In the present invention, the information entropy of each station reflects the irregularity and complexity of its portable box usage pattern. A high information entropy means that the portable boxes are used frequently and irregularly during the inspection period, and vice versa indicates a relatively stable usage pattern. Finally, based on the magnitudes of the calculated K information entropies, perform resource balancing configuration on the K processing units in the background management terminal for managing the K stations. This means that the system will dynamically allocate management resources according to the magnitudes of the information entropies of each station, such as increasing the monitoring intensity of stations with high information entropy to ensure that it can effectively respond to those stations with large demand changes and high uncertainties. At the same time, for stations with low information entropy, the resource allocation can be appropriately reduced to optimize the overall operation efficiency. Using the configured K processing units, the system will manage the portable boxes of each station more intelligently and efficiently. In this way, the background system can not only ensure the reasonable allocation of resources, but also improve the service quality and user satisfaction of the entire subway network. This data-driven dynamic management method provides a highly adaptive and responsive management mechanism for the subway portable box system, effectively improving the operation efficiency and the usage experience of passengers.

[0056] Through the technical solutions of the above embodiments, the portable box intelligent management method for subway rides provided by the present application solves the technical problems existing in the existing subway management, such as the inability to efficiently and conveniently manage the non-dangerous personal items of passengers, resulting in congestion during the security check process, inconvenience to passengers, and increased management costs. By using portable boxes and conducting intelligent management, the technical effect of effectively and conveniently managing the items of passengers and dynamically adjusting the resource allocation of each station is achieved, thereby improving the operation efficiency and safety of the entire subway network and ensuring the riding and management experience.

[0057] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0058] The above-described embodiments merely represent several implementation manners of the present application. Their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application.

Claims

1. An intelligent management method for a portable case for subway riding, characterized in that: The method comprises: Retrieve the portable box management logs of K stations of the target subway line within a preset historical window to obtain K historical portable box management logs, where K is an integer greater than or equal to 1, and the portable box is used to store items that must be carried but cannot be brought into the carriage; Extracting indicators from the K historical portable box management logs according to the usage indicator, the deposit indicator and the cabinet inventory indicator respectively, to obtain K usage indicator sets, K deposit indicator sets and K cabinet inventory indicator sets, wherein the usage indicator is the number of portable boxes taken out from the site by the user, the deposit indicator is the number of portable boxes deposited by the user at the site, and the cabinet inventory indicator is the number of portable boxes stored at the site; Traversing the K usage indicator sets and the K deposit indicator sets to perform indicator concentration trend analysis, determine K site liquidity factors, and respectively divide the K site liquidity factors by the sum of the K site liquidity factors to obtain K site allocation coefficients; Traversing the K cabinet inventory index sets to identify fluctuations, and obtaining K station steady-state coefficients; Using a monitoring step identifier to identify the K site deployment coefficients and the K site steady-state coefficients, and determine a set of K site monitoring bandwidths, wherein the set of K site monitoring bandwidths includes K first site monitoring bandwidths and K second site monitoring bandwidths, and each first site monitoring bandwidth is smaller than the corresponding second site monitoring bandwidth; Constructing K station portable box monitoring transmission branches of the target subway line based on the K first station monitoring bandwidths and the K second station monitoring bandwidths; Using the K-site portable box monitoring transmission branches to monitor and extract the portable box management data of the K sites, and sending the extracted data to the background management end to manage the portable boxes of the K sites; Traversing the K usage indicator sets and the K deposit indicator sets to perform indicator concentration trend analysis, and determining K site liquidity factors, including: Traversing the K usage indicator sets to perform mean calculations and determine the mean values ​​of the K usage indicators; Taking the K usage index means as K starting mass centers, iterating the K starting mass centers according to a preset mass center iterative update formula to obtain K stage iterative mass centers; Iteration is continued based on the K stage iteration centroids until a preset iteration stop condition is met, K update iteration centroids are obtained, and K usage indicators corresponding to the K update iteration centroids are used as K concentrated usage indicators; Performing an indicator concentration trend analysis on the K deposit amount indicator sets to determine K concentrated deposit amount indicators; Performing weighted calculation on the K centralized usage indicators and the K centralized deposit indicators to obtain the K site liquidity factors; include: ； in, is the stage iteration centroid, is a set of usage indicators, is the i-th usage indicator in the usage indicator set, is the starting centroid, is a Gaussian function.

2. The intelligent management method for a portable case for subway riding according to claim 1, characterized in that: include: Traversing and calculating the distances from the K usage indicator mean value sets to the K stage iteration centroids, and obtaining the K stage centroid deviation distance sets; Continue to iterate based on the K stage iterative centroids to obtain K update stage iterative centroids, and obtain a set of deviation distances of the K update stage centroids; The sum of the K update stage distances of the K update stage centroid deviation distance set and the sum of the K stage distances of the K stage centroid deviation distance set are calculated respectively. When the sum of the K update stage distances is greater than or equal to the sum of the K stage distances, the preset iteration stop condition is met, and the K update stage iteration centroids are used as the K update iteration centroids.

3. The intelligent management method for a portable case for subway riding according to claim 1, characterized in that: Traversing the K cabinet inventory index sets to identify fluctuations, and determining K station steady-state factors of the K stations, including: Traversing the K cabinet inventory index sets to perform fluctuation variance calculations, and taking the inverse of the calculation results as the steady-state factors of the K sites respectively; The K site deployment coefficients are obtained by respectively dividing the K site steady-state factors by the sum of the K site steady-state factors.

4. The intelligent management method for a portable case for subway riding according to claim 1, characterized in that: The K site allocation coefficients and the K site steady-state coefficients are identified by using a monitoring step length identifier to determine a set of K site monitoring bandwidths, including: Obtaining multiple sample site deployment coefficients, multiple sample site steady-state coefficients, and multiple sample site monitoring bandwidth sets as training data; The framework constructed based on the feedforward neural network is supervised and trained using the training data to learn the mapping relationship between the site allocation coefficient and the site steady-state coefficient and the site monitoring bandwidth set until the training converges to obtain the trained monitoring step identifier.

5. The intelligent management method for a portable case for subway riding according to claim 1, characterized in that: include: Extracting K monitoring extraction data sets of the K sites received by the background management terminal within a preset monitoring window; Performing information entropy calculation on the K monitoring and extraction data sets to determine K information entropies; Based on the sizes of the K information entropies, resource balancing is performed on the K processing units in the background management terminal that manage the K sites, and the configured K processing units are used to manage the portable boxes of the K sites.

Images