Resource scheduling method and device for parking demand in peak period, and medium

Abstract

The invention discloses a resource scheduling method and device for parking demands in a peak period and a medium, and the method comprises the steps: collecting and preprocessing urban peak multi-source data, and generating a peak data set; feature extraction and prediction are performed on the data set through a prediction agent containing a multi-source feature fusion module and an LSTM-XGBoost hybrid model, and a peak parking demand result is obtained; and inputting the result and real-time parking space data into a resource scheduling agent containing intra-regional scheduling, cross-regional collaboration and user pre-guiding units, and calculating the regional parking space saturation. When the saturation is smaller than a preset threshold value, parking spaces are distributed in the area according to rules, and information is updated; if yes, searching surrounding parking lots in a cross-region manner, and selecting an optimal concurrent reservation instruction. According to the method, the prediction accuracy is greatly improved through multi-source fusion and a hybrid model, and a reliable basis is provided for scheduling.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and in particular to a resource scheduling method, device and medium for parking demand during peak hours. Background Technology

[0002] With the acceleration of urbanization and the surge in motor vehicle ownership, the shortcomings of static traffic (parking management) in the intelligent transportation system are becoming increasingly prominent. In particular, the sudden and concentrated nature of parking demand during peak hours has become a core challenge for urban traffic management.

[0003] Current parking management systems suffer from several problems: First, insufficient accuracy in demand forecasting. Most systems rely solely on historical parking data for modeling, failing to integrate dynamic factors such as real-time traffic congestion indices, urban events (e.g., concerts, hospital peak hours), and weather conditions, resulting in very low accuracy in predicting peak commuting times. Second, passive and rigid resource allocation, employing static rules like first-come, first-served or fixed-area allocation, cannot adapt to fluctuating demand, often leading to an imbalance between saturated parking spaces in core areas and idle spaces in surrounding parking lots. Third, a lack of cross-regional collaboration. Fragmented parking information between regions and the absence of a unified scheduling mechanism not only reduce user travel experience but also cause serious waste of parking resources, hindering the intelligent upgrading of urban static traffic. Summary of the Invention

[0004] This application provides a resource scheduling method, device, and medium for parking demand during peak hours. The method includes: collecting multi-source data during urban peak hours; preprocessing the multi-source data to generate a peak hour dataset; extracting and predicting features from the peak hour dataset according to a preset predictive agent to obtain a peak parking demand prediction result; the predictive agent includes a multi-source feature fusion module and an LSTM-XGBoost hybrid prediction model; inputting the peak parking demand prediction result and real-time parking space data from the multi-source data into the preset resource scheduling agent to obtain the parking space saturation of each area of ​​the city; the resource scheduling agent includes an intra-regional scheduling unit, a cross-regional coordination unit, and a user pre-guidance unit; if the parking space saturation of a certain area is less than a preset threshold, the intra-regional scheduling unit allocates parking spaces according to preset rules and updates the location information of available parking spaces; if the parking space saturation of a certain area is greater than the preset threshold, the cross-regional coordination unit queries parking information within a preset range, selects the optimal target parking lot, and sends a parking space reservation instruction.

[0005] In one example, multi-source data is collected during peak hours in the city. The multi-source data is preprocessed to generate a peak-hour dataset. Specifically, this includes: collecting parking lot operation data during peak hours for a preset historical period as historical parking data; collecting real-time dynamic data and real-time parking space data; where real-time dynamic data includes real-time congestion index of main roads, urban event information, and meteorological data; and real-time parking space data includes the parking space occupancy status of each area; performing outlier removal, feature standardization, temporal alignment, and scene labeling operations on all collected data in sequence to generate the peak-hour dataset; and dividing the peak-hour dataset into a training set, a validation set, and a test set.

[0006] In one example, based on a pre-defined predictive agent, feature extraction and prediction are performed on the peak-hour dataset to obtain peak parking demand prediction results. Specifically, this includes: extracting weekly periodic features and time-period peak features from historical parking data; extracting congestion delay features and demand increment features from real-time dynamic data; learning all extracted features using an LSTM-XGBoost hybrid prediction model to obtain time-series feature results and nonlinear feature results; fusing the time-series feature results and nonlinear feature results according to pre-defined weights to obtain initial prediction values; and correcting the initial prediction values ​​using an error correction unit based on validation set data to output peak parking demand prediction results.

[0007] In one example, the peak parking demand forecast results and real-time parking space data from multiple sources are input into a preset resource scheduling agent to obtain the parking space saturation of each area of ​​the city. Specifically, this includes: obtaining the predicted number of vehicles in demand for each area from the peak parking demand forecast results; obtaining the actual number of available parking spaces in the corresponding area from the real-time parking space data; and calculating the parking space saturation of each area of ​​the city based on the predicted number of vehicles in demand and the actual number of available parking spaces.

[0008] In one example, if the parking space saturation in a certain area is less than a preset threshold, the scheduling unit within the area allocates parking spaces according to preset rules and updates the location information of available parking spaces. Specifically, this includes: when the parking space saturation in a certain area is less than the preset threshold, reserving a preset proportion of parking spaces for users who have made reservations; after the reservations are allocated, prioritizing the allocation of parking spaces to short-term parking users whose expected parking time is less than or equal to a preset time from the remaining parking spaces; and generating updated location information of available parking spaces based on the allocation results, and publishing it in real time through the guidance screen.

[0009] In one example, if the parking saturation in a certain area exceeds a preset threshold, the cross-regional collaborative unit queries parking information within a preset range, selects the optimal target parking lot, and sends a parking space reservation instruction. Specifically, this includes: when the parking saturation in a certain area is greater than or equal to the preset threshold, querying the real-time status information of all parking lots within a preset distance range around the current area; the real-time status information includes at least the real-time number of available parking spaces and real-time traffic conditions; calculating the comprehensive matching degree for each surrounding parking lot based on the real-time status information; the comprehensive matching degree is calculated by weighting distance factors, traffic condition factors, and parking space availability factors; and selecting the optimal target parking lot with the highest matching degree based on the comprehensive matching degree, and sending an instruction containing the number of reserved parking spaces to the optimal target parking lot.

[0010] In one example, after selecting the optimal target parking lot and sending a parking space reservation instruction, the method further includes: generating user-oriented guidance information within a pre-guidance time window; the guidance information includes the parking space availability status of the destination area, the name, location, and real-time available parking spaces of the recommended target parking lot, as well as a navigation link to the recommended target parking lot; and pushing the guidance information to the user group through an application on the user's terminal.

[0011] In one example, the method also includes: monitoring the parking space occupancy rate and user arrival rate in each area after scheduling; when the actual parking demand in a certain area deviates from the prediction result by more than a preset threshold, re-executing the peak demand prediction and updating the resource scheduling strategy according to the new prediction result.

[0012] On the other hand, embodiments of this application provide a resource scheduling device for parking demand during peak hours, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a resource scheduling method for parking demand during peak hours as described above.

[0013] On the other hand, embodiments of this application provide a non-volatile computer storage medium for resource scheduling of parking demand during peak hours, which stores computer-executable instructions that can execute any of the above-mentioned resource scheduling methods for parking demand during peak hours.

[0014] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: This application significantly improves the accuracy of parking demand prediction during peak hours by integrating multi-source data such as historical parking data, real-time traffic flow, urban events, and meteorological information. It employs an attention mechanism to allocate feature weights and a hybrid LSTM-XGBoost model for collaborative prediction. By introducing a trigger mechanism based on parking space saturation thresholds and dynamic resource allocation through cross-regional collaborative scheduling that combines distance, road conditions, and parking space availability, the utilization rate of parking spaces in core areas during peak hours is increased, while simultaneously boosting the utilization rate of idle parking lots in surrounding areas. This achieves overall optimization and balance of urban parking resources, avoiding resource waste caused by localized saturation and surrounding idleness. Furthermore, by constructing a dual-agent automated collaborative architecture consisting of a predictive agent and a scheduling agent, the entire process from prediction and decision-making to scheduling is intelligently managed. This significantly reduces the manpower required for manual scheduling and lowers user complaints related to parking lot operations during peak hours, thereby improving operational management efficiency. Attached Figure Description

[0015] To more clearly illustrate the technical solution of this application, some embodiments of this application will be described in detail below with reference to the accompanying drawings, in which: Figure 1 A flowchart illustrating a resource scheduling method for peak-hour parking demand provided in an embodiment of this application; Figure 2 A diagram illustrating a dynamic planning system architecture for parking space resources, provided in this application embodiment, for a resource scheduling method for parking demand during peak hours. Figure 3 This is a schematic diagram of a resource scheduling device for peak-hour parking demand provided in an embodiment of this application. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] Some embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0018] Figure 1 This is a flowchart illustrating a resource scheduling method for peak-hour parking demand, provided as an embodiment of this application. This method can be applied to different business domains. Certain input parameters or intermediate results in this process allow for manual intervention to help improve accuracy.

[0019] The analysis method involved in the embodiments of this application can be implemented by a terminal device or a server, and this application does not impose any special limitations on it. For ease of understanding and description, the following embodiments are all described in detail using a server as an example.

[0020] Based on this Figure 1 The process may include the following steps: S101: Collect multi-source data during peak hours in the city, preprocess the multi-source data, and generate a peak hour dataset.

[0021] In some embodiments of this application, the application first performs data acquisition and preprocessing. Specifically, it collects multi-source data during peak hours in the city, including but not limited to: historical parking data for the past year (such as parking space occupancy rate and parking duration, categorized by scenarios such as commuting, business districts, hospitals, and large-scale events), real-time traffic flow data obtained through a 5G IoT module (such as the main road congestion index, sampled every 5 minutes), event information captured from the city event platform (such as concerts and exhibitions), meteorological data, and real-time parking space status data collected through geomagnetic sensors and cameras (sampled every 20 seconds).

[0022] Furthermore, these multi-source data are preprocessed, including removing outliers such as geomagnetic sensor malfunctions, normalizing features such as congestion index and parking space occupancy rate to the [0,1] interval, performing time-series alignment based on timestamps, and adding scene labels to the data. Finally, a high-quality peak-hour dataset is generated and divided into training, validation, and test sets in a ratio of 75:15:10.

[0023] By integrating heterogeneous data from multiple sources, including historical parking data, real-time traffic data, urban events data, and meteorological data, the problem of single-dimensional data in existing technologies is solved. A refined preprocessing workflow ensures high data quality and consistency, providing a reliable guarantee for the accurate training of subsequent models and overcoming prediction biases caused by incomplete or inaccurate data at the source.

[0024] S102: Based on the preset predictive agent, perform feature extraction and feature prediction on the peak period dataset to obtain the peak parking demand prediction result; the predictive agent includes a multi-source feature fusion module and an LSTM-XGBoost hybrid prediction model.

[0025] In some embodiments of this application, after obtaining the peak-hour dataset, a pre-defined predictive agent is used to extract features and predict the peak-hour dataset. During this process, the multi-source feature fusion module extracts weekly periodic features and time-period peak features from historical data, extracts congestion delay features and demand increment features brought about by urban events from real-time data, and uses an attention mechanism to assign weights to different features (e.g., event features 0.3, traffic features 0.25, historical features 0.45).

[0026] Furthermore, the LSTM-XGBoost hybrid prediction model learns from these fused features. The LSTM network is responsible for capturing temporal patterns, while the XGBoost model is responsible for handling nonlinear bursts. The outputs of the two are fused with a weight of 6:4 to obtain the initial prediction value.

[0027] Finally, this application will also calibrate the initial results based on the error correction rules constructed on the validation set (e.g., adjusting the predicted value by 10% on rainy days), and output more accurate peak parking demand prediction results divided by region and time period. For example, in the Jinan CBD area, from 17:30 to 18:30, the expected demand is 850 vehicles, with a parking space shortage of 200 vehicles.

[0028] By employing a multi-source feature fusion module and an attention mechanism, the system can intelligently weigh the impact of different factors on parking demand, avoiding errors caused by averaging. The LSTM-XGBoost hybrid model leverages the combined advantages of LSTM in capturing temporal patterns and XGBoost in handling nonlinear bursts of features. Combined with scenario-based error correction, this significantly improves prediction accuracy, providing a scientific and reliable basis for subsequent scheduling decisions.

[0029] S103: Input the peak parking demand prediction results and real-time parking space data from multi-source data into a preset resource scheduling intelligent agent to obtain the parking space saturation in various areas of the city; the resource scheduling intelligent agent includes an intra-regional scheduling unit, a cross-regional coordination unit, and a user pre-guidance unit.

[0030] In some embodiments of this application, after obtaining the peak parking demand prediction results, this application inputs the above prediction results together with real-time parking space data into a preset resource scheduling agent to calculate the parking space saturation of each area.

[0031] Specifically, the scheduling AI obtains the predicted number of vehicles in each area from the peak parking demand forecast and the actual number of available parking spaces in the corresponding area from the real-time parking space data. It then calculates the specific parking space saturation using a formula and compares this saturation with a preset 85% threshold to determine the subsequent scheduling direction. The formula is:

[0032] By calculating the intuitive indicator of parking space saturation and using a preset threshold of 85% as the decision trigger point, the entire dispatching system gains foresight and proactivity. This ensures that dispatching actions (whether within or across regions) are initiated based on evidence, are timely, and are accurate, avoiding blind and delayed dispatching.

[0033] S104-S105: If the parking space saturation in a certain area is less than a preset threshold, parking spaces are allocated according to preset rules through the regional scheduling unit, and the location information of available parking spaces is updated. If the parking space saturation in a certain area is greater than a preset threshold, parking information within a preset range is queried through the cross-regional coordination unit, the optimal target parking lot is selected, and a parking space reservation instruction is sent.

[0034] In some embodiments of this application, dynamic scheduling is performed based on the saturation judgment result. If the parking space saturation of a certain area is less than 85%, the scheduling unit in the area is activated to allocate parking spaces according to the rules of priority for reserved users (reserving 15% of parking spaces) and priority for short-term parking (parking ≤ 3 hours), and the location information of available parking spaces is immediately updated through the guidance screen at the entrance of the parking lot.

[0035] Meanwhile, if the saturation of a certain area is greater than or equal to 85%, the cross-regional collaborative unit is triggered to automatically query the real-time available parking spaces and road conditions of parking lots within 3-5 kilometers. The optimal target parking lot is selected by weighted calculation of distance, road conditions, and parking space availability (weights can be 0.4, 0.3, 0.3), and an instruction containing the number of reserved parking spaces (e.g., 100) is sent to it.

[0036] It should also be noted that after selecting the optimal target parking lot and sending the parking space reservation instruction, this application also generates user-oriented guidance information within the pre-guidance time window. The guidance information includes the parking space availability status of the destination area, the name, location, and real-time available parking spaces of the recommended target parking lot, as well as the navigation link of the recommended target parking lot. Finally, the guidance information is pushed to the user group through the application on the user's terminal.

[0037] It should also be noted that this application will monitor the parking space occupancy rate and user arrival rate in each area after scheduling. When the actual parking demand in a certain area deviates from the prediction result by more than a preset threshold, the peak demand prediction will be re-executed, and the resource scheduling strategy will be updated according to the new prediction result.

[0038] By calculating the overall matching degree of distance, road conditions, and parking space availability, the optimal target is selected, ensuring that the recommended alternative parking lots for users are not only available but also have the lowest overall cost, greatly improving user compliance with guidance suggestions. Through reservation priority and short-term priority rules, for users, reservations guarantee parking space certainty and improve the travel experience; short-term priority accelerates parking space turnover and meets temporary needs such as shopping and medical visits. For operators, the reservation mechanism can serve as a value-added service, directly increasing the economic benefits of a single parking lot during peak hours.

[0039] It should be noted that, although the embodiments in this application are based on... Figure 1 Steps S101 to S105 will be described sequentially, but this does not mean that steps S101 and S105 must be performed in a strict order. The reason this embodiment follows this order is... Figure 1 The order in which steps S101 to S105 are described is provided to facilitate understanding of the technical solutions of the embodiments of this application by those skilled in the art. In other words, in the embodiments of this application, the order of steps S101 to S105 can be appropriately adjusted according to actual needs.

[0040] pass Figure 1 This application utilizes a multi-source data approach, integrating historical parking data, real-time traffic flow, urban events, and meteorological information. It employs an attention mechanism to allocate feature weights and a hybrid LSTM-XGBoost model for collaborative prediction, significantly improving the accuracy of parking demand prediction during peak hours. By introducing a trigger mechanism based on parking space saturation thresholds and dynamic resource allocation through cross-regional collaborative scheduling that combines distance, road conditions, and parking space availability, the utilization rate of parking spaces in core areas during peak hours is increased, while simultaneously boosting the utilization rate of idle parking lots in surrounding areas. This achieves overall optimization and balance of urban parking resources, avoiding resource waste caused by localized saturation and surrounding idleness. Furthermore, by constructing a dual-agent automated collaborative architecture consisting of a predictive agent and a scheduling agent, the entire process from prediction and decision-making to scheduling is made intelligent. This significantly reduces the manpower required for manual scheduling and lowers user complaints related to parking lot operations during peak hours, thereby improving operational management efficiency.

[0041] Figure 2 This is a diagram illustrating the dynamic planning system architecture for parking space resources, which is a resource scheduling method for peak-hour parking demand provided in an embodiment of this application.

[0042] exist Figure 2 The document demonstrates all the steps involved in the resource scheduling process of this application, including the selection of time periods and the determination of available parking spaces.

[0043] Figure 3A schematic diagram of a resource scheduling device for peak-hour parking demand provided in this application embodiment includes: At least one processor; and, A memory that is communicatively connected to at least one processor; wherein, The memory stores instructions that can be executed by at least one processor, which enable at least one processor to perform a resource scheduling method for peak-hour parking demand, as described above.

[0044] Some embodiments of this application provide a non-volatile computer storage medium for resource scheduling of parking demand during peak hours, storing computer-executable instructions that can execute any of the above-described resource scheduling methods for parking demand during peak hours.

[0045] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0046] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0047] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0048] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0049] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0050] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0051] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0052] Memory may include non-persistent storage in computer-readable media, random access memory (RAM), and non-volatile memory such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0053] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0054] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0055] The above are merely embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the technical principles of this application should fall within the protection scope of this application.

Claims

1. A resource scheduling method for parking demand during peak hours, characterized in that, The method includes: Collect multi-source data during peak hours in the city, preprocess the multi-source data, and generate a peak hour dataset; Based on the preset predictive agent, feature extraction and feature prediction are performed on the peak period dataset to obtain the peak parking demand prediction result; the predictive agent includes a multi-source feature fusion module and an LSTM-XGBoost hybrid prediction model. The peak parking demand prediction results and real-time parking space data from multi-source data are input into a preset resource scheduling intelligent agent to obtain the parking space saturation in various urban areas; the resource scheduling intelligent agent includes an intra-regional scheduling unit, a cross-regional coordination unit, and a user pre-guidance unit; If the parking space saturation in a certain area is less than a preset threshold, the scheduling unit in the area will allocate parking spaces according to preset rules and update the location information of available parking spaces. If the parking space saturation in a certain area exceeds a preset threshold, the cross-regional collaborative unit queries the parking information within the preset range, selects the optimal target parking lot, and sends a parking space reservation instruction.

2. The method according to claim 1, characterized in that, The process of collecting multi-source data during peak hours in the city, preprocessing the multi-source data, and generating a peak-hour dataset specifically includes: Collect parking lot operation data during peak hours for a preset historical period in the city as historical parking data; Collect real-time dynamic data and real-time parking space data; the real-time dynamic data includes the real-time congestion index of main roads, urban event information, and meteorological data; the real-time parking space data includes the parking space occupancy status of each area. All collected data are sequentially processed for outlier removal, feature standardization, temporal alignment, and scene labeling to generate a peak period dataset. The peak period dataset is divided into a training set, a validation set, and a test set.

3. The method according to claim 1, characterized in that, The step of extracting and predicting features from the peak-hour dataset based on a preset predictive agent to obtain peak parking demand prediction results specifically includes: Extract the weekly periodicity and time-period peak characteristics of historical parking data; Extract congestion delay characteristics and demand increment characteristics from real-time dynamic data; By using the LSTM-XGBoost hybrid prediction model, all extracted features are learned to obtain time-series feature results and nonlinear feature results; The time-series feature results and nonlinear feature results are fused according to preset weights to obtain an initial prediction value; Based on the validation set data, the initial predicted values ​​are corrected using the error correction unit, and the peak parking demand prediction results are output.

4. The method according to claim 1, characterized in that, The step of inputting the peak parking demand prediction results and real-time parking space data from multi-source data into a preset resource scheduling agent to obtain the parking space saturation in various urban areas specifically includes: Obtain the predicted number of vehicles in each region from the peak parking demand prediction results; Obtain the actual number of available parking spaces in the corresponding area from the real-time parking space data; Based on the predicted number of vehicles in demand and the actual number of available parking spaces, the parking saturation of each area of ​​the city is calculated.

5. The method according to claim 1, characterized in that, If the parking space saturation in a certain area is less than a preset threshold, the scheduling unit within that area allocates parking spaces according to preset rules and updates the location information of available parking spaces, specifically including: When the parking space saturation in a certain area is less than a preset threshold, a preset proportion of parking spaces will be reserved for users who have made reservations. Once the reservation users have been allocated, the remaining parking spaces will be prioritized for short-term parking users whose expected parking duration is less than or equal to the preset duration. Based on the allocation results, updated information on available parking spaces is generated and published in real time through the guidance screen.

6. The method according to claim 1, characterized in that, If the parking space saturation in a certain area exceeds a preset threshold, the system queries parking information within a preset range through a cross-regional collaborative unit, selects the optimal target parking lot, and sends a parking space reservation instruction. Specifically, this includes: When the parking space saturation of a certain area is greater than or equal to a preset threshold, query the real-time status information of all parking lots within a preset distance range around the current area; the real-time status information includes at least the real-time number of available parking spaces and real-time traffic conditions. Based on the real-time status information, a comprehensive matching degree is calculated for each surrounding parking lot; the comprehensive matching degree is obtained by weighted calculation based on distance factor, road condition factor and parking space availability factor. Based on the overall matching degree, the optimal target parking lot with the highest matching degree is selected, and an instruction containing the number of reserved parking spaces is sent to the optimal target parking lot.

7. The method according to claim 1, characterized in that, After selecting the optimal target parking lot and sending a parking space reservation instruction, the method further includes: Within the pre-guided time window, user-oriented guidance information is generated; the guidance information includes the parking availability status of the destination area, the name, location, and real-time available parking spaces of the recommended target parking lot, as well as a navigation link to the recommended target parking lot. The guidance information is pushed to the user group through an application on the user's terminal.

8. The method according to claim 1, characterized in that, The method further includes: After monitoring and scheduling, the parking space occupancy rate and user arrival rate in each area are monitored. When the actual parking demand in a certain area deviates from the prediction result by more than a preset threshold, the peak demand prediction is re-executed, and the resource scheduling strategy is updated according to the new prediction result.

9. A resource scheduling device for parking demand during peak hours, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform a resource scheduling method for peak-hour parking demand as described in any one of claims 1-8.

10. A resource scheduling storage medium for parking demand during peak hours, storing computer-executable instructions, characterized in that, The computer-executable instructions are capable of executing a resource scheduling method for peak-hour parking demand as described in any one of claims 1-8.

Images