Smart city shared bicycle deployment and operation area planning method and internet of things system

Abstract

The embodiment of the present specification provides a smart city shared bicycle delivery and operation area planning method and an Internet of Things system. The method is executed based on a management platform of the smart city shared bicycle delivery and operation area planning Internet of Things system. The method comprises the following steps: obtaining reference information of human flow of at least one target area, determining predicted human flow of the at least one target area at a target time based on the reference information of human flow, and determining a shared bicycle demand based on the predicted human flow; obtaining shared bicycle historical data of a plurality of reference areas, predicting a shared bicycle change amount of the at least one target area at the target time based on the shared bicycle historical data; and determining a shared bicycle delivery amount of the at least one target area at the target time based on the shared bicycle demand and the shared bicycle change amount.

Description

Technical Field

[0001] This specification relates to the field of Internet of Things (IoT) technology, and in particular to a smart city shared bicycle deployment and operation area planning method and IoT system. Background Technology

[0002] With the development of the sharing economy, shared bicycles have become an increasingly popular mode of transportation. However, as the user base of shared bicycles grows, the number of bicycles deployed by related companies is also expanding rapidly. Countless shared bicycle users crisscross the city every day, and their needs vary at different times and among different users. This presents increasing challenges to the operation and management of shared bicycles. How to achieve the rational deployment, scheduling, and operation of shared bicycles, and reduce operational costs, is an urgent problem to be solved.

[0003] Therefore, we hope to propose a smart city shared bicycle deployment and operation area planning method and Internet of Things system to achieve the automation and intelligence of shared bicycle operation and management. Summary of the Invention

[0004] This specification provides one or more embodiments of a method for planning the deployment and operation area of ​​shared bicycles in a smart city. The method is executed on the management platform of an IoT system for planning the deployment and operation area of ​​shared bicycles in a smart city. The method includes: acquiring pedestrian flow reference information for at least one target area; determining the predicted pedestrian flow for the at least one target area at a target time based on the pedestrian flow reference information; determining the demand for shared bicycles based on the predicted pedestrian flow; acquiring historical shared bicycle data for multiple reference areas; predicting the change in the number of shared bicycles in the at least one target area at the target time based on the historical shared bicycle data; and determining the number of shared bicycles deployed in the at least one target area at the target time based on the demand for shared bicycles and the change in the number of shared bicycles.

[0005] This specification provides an IoT system for smart city shared bicycle deployment and operation area planning through one or more embodiments. The system includes a user platform, a service platform, a management platform, a sensor network platform, and an object platform. The service platform includes multiple service sub-platforms, with different target areas corresponding to different service sub-platforms. The management platform includes a central management platform database and multiple management sub-platforms, each corresponding to a different target area. The sensor network platform includes multiple sensor network sub-platforms, each corresponding to a different target area. The object platform is used to obtain pedestrian traffic parameters for the target areas. The system collects reference information and historical data of shared bicycles in the reference area, and transmits this data to the corresponding management sub-platform based on the sensor network sub-platform corresponding to the target area. The management sub-platform is used to determine the predicted pedestrian flow in the target area at a target time based on the pedestrian flow reference information, and to determine the demand for shared bicycles based on the predicted pedestrian flow. It also determines the change in the number of shared bicycles in the target area at the target time based on the historical data of shared bicycles. Based on the demand and change in the number of shared bicycles, it determines the number of shared bicycles to be deployed. Finally, it transmits the number of shared bicycles deployed to the service platform based on the management platform database. The service platform then transmits the number of shared bicycles deployed to the user platform.

[0006] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the smart city shared bicycle deployment and operation area planning method as described in any of the above embodiments. Attached Figure Description

[0007] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0008] Figure 1 This is an exemplary platform structure diagram of an IoT system for smart city shared bicycle deployment and operation area planning, as shown in some embodiments of this specification.

[0009] Figure 2 This is an exemplary flowchart of a smart city shared bicycle deployment and operation area planning method according to some embodiments of this specification;

[0010] Figure 3 This is an exemplary schematic diagram of the first prediction model structure shown in some embodiments of this specification;

[0011] Figure 4This is another exemplary schematic diagram of the first prediction model shown in some embodiments of this specification;

[0012] Figure 5 This is an exemplary schematic diagram illustrating the determination of changes in shared bicycles based on operational maps, according to some embodiments of this specification;

[0013] Figure 6 This is another exemplary schematic diagram illustrating the determination of changes in shared bicycles based on operational maps according to some embodiments of this specification;

[0014] Figure 7 This is an exemplary flowchart illustrating the adjustment of the operating area according to some embodiments of this specification;

[0015] Figure 8 This is another exemplary flowchart illustrating the adjustment of the operating area according to some embodiments of this specification. Detailed Implementation

[0016] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0017] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0018] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0019] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0020] Figure 1This is an exemplary platform structure diagram of an IoT system for smart city shared bicycle deployment and operation area planning, as shown in some embodiments of this specification. In some embodiments, the IoT system 100 for smart city shared bicycle deployment and operation area planning may include a user platform 110, a service platform 120, a management platform 130, a sensor network platform 140, and an object platform 150.

[0021] User platform 110 can be a user-facing service interface. In some embodiments, user platform 110 can receive information from users and / or service platforms. For example, user platform 110 can receive input from users. As another example, user platform 110 can receive information fed back to users from service platforms, such as the number of shared bicycles deployed in a target area, operating area adjustment strategies, etc. In some embodiments, user platform 110 can be configured to feed back the received information to users. In some embodiments, user platform 110 can be configured to send data and / or instructions to the service platform, for example, sending instructions to determine operating area adjustment strategies.

[0022] Service platform 120 can be a platform for preliminary information processing. In some embodiments, the service platform can be configured to interact with the user platform and the management platform for information and / or data exchange. For example, service platform 120 can obtain user-inputted shared bicycle deployment query commands from user platform 110, upload shared bicycle deployment data to user platform 110, etc. As another example, service platform 120 can send shared bicycle deployment query commands to management platform 130, obtain shared bicycle deployment data and operating area adjustment strategies from management platform 130, etc.

[0023] In some embodiments, the service platform 120 may include multiple service sub-platforms, with different target areas corresponding to different service sub-platforms. In some embodiments, at least one of the multiple service sub-platforms may send a query instruction to the management platform 130 to query the number of shared bicycles deployed in the corresponding target area, in order to obtain the number of shared bicycles deployed in the corresponding target area. In some embodiments, at least one of the multiple service sub-platforms may upload the number of shared bicycles deployed in the corresponding target area to the user platform 110.

[0024] The management platform 130 can refer to an IoT platform that coordinates and manages the connections and collaboration between various functional platforms, providing sensing management and control management. In some embodiments, the management platform 130 can be used to determine the predicted pedestrian flow in at least one target area at a target time based on pedestrian flow reference information, and to determine the demand for shared bicycles based on the predicted pedestrian flow. In some embodiments, the management platform 130 can be used to predict the change in the number of shared bicycles in at least one target area at a target time based on historical shared bicycle data. In some embodiments, the management platform 130 can be used to determine the number of shared bicycles deployed in at least one target area at a target time based on the demand for shared bicycles and the change in the number of shared bicycles. In some embodiments, the management platform 130 can be used to adjust the shared bicycle operation area at a target time based on real-time operational reference data.

[0025] More information on determining predicted pedestrian traffic, changes in shared bicycle usage, the number of shared bicycles deployed, and strategies for adjusting operating areas can be found at [link to relevant documentation]. Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 And its related descriptions.

[0026] In some embodiments, the management platform 130 may include a central management platform database and multiple sub-management platforms. In some embodiments, each sub-management platform corresponds to a different target area, and each sub-management platform can determine relevant information about the corresponding target area at a target time based on data and / or information uploaded by the sensor network platform. For example, each sub-management platform can determine the predicted pedestrian flow of the corresponding target area at a target time based on pedestrian flow reference information uploaded by the sensor network platform. As another example, each sub-management platform can determine the change in the number of shared bicycles in the corresponding target area at a target time based on historical shared bicycle data of a reference area uploaded by the sensor network platform.

[0027] In some embodiments, each management sub-platform can upload the number of shared bicycles deployed and the operation area adjustment strategy for the corresponding target area to the management platform database. In some embodiments, the management platform database can aggregate or upload the number of shared bicycles deployed and the operation area adjustment strategy to the service platform by region.

[0028] The sensor network platform 140 can be a platform that enables interaction between the management platform and the object platform. In some embodiments, the sensor network platform 140 can receive instructions from the management platform to obtain pedestrian flow reference information, shared bicycle historical data, and operational reference data, and then send the instructions to the object platform. In some embodiments, the sensor network platform 140 can be used to receive pedestrian flow reference information, shared bicycle historical data, and operational reference data from the object platform, and then upload the received pedestrian flow reference information, shared bicycle historical data, and operational reference data to the management platform.

[0029] In some embodiments, the sensor network platform 140 may include multiple sensor network sub-platforms, each of which corresponds to a different target area. In some embodiments, each sensor network sub-platform corresponds one-to-one with each management sub-platform and one-to-one with each object sub-platform.

[0030] In some embodiments, each sensor network sub-platform can interact with its corresponding management sub-platform and object sub-platform for information and / or data exchange. For example, each sensor network sub-platform can receive instructions from its corresponding management sub-platform to obtain pedestrian flow reference information, shared bicycle historical data, and operational reference data, and then send these instructions to its corresponding object sub-platform. As another example, each sensor network sub-platform can receive pedestrian flow reference information, shared bicycle historical data, and operational reference data uploaded by its corresponding object sub-platform, and then upload these data to its corresponding management sub-platform.

[0031] The object platform 150 can be a functional platform for generating sensing information and ultimately executing control information. In some embodiments, the object platform 150 can be configured as a monitoring device to acquire pedestrian flow reference information, historical data of shared bicycles, and operational reference data. For example, road surveillance cameras based on target areas can acquire shared bicycle transfer data, pedestrian flow change data, etc. In some embodiments, the object platform 150 can include object sub-platforms corresponding to different target areas, and each object sub-platform can be implemented by a monitoring device or a sensing device. The object sub-platforms corresponding to different areas can upload the collected pedestrian flow reference information, historical data of shared bicycles, and operational reference data to the corresponding sensor network sub-platform, which then uploads them to the management sub-platform for processing. Different management sub-platforms can issue instructions to the object sub-platforms based on the corresponding sensor network sub-platforms to collect pedestrian flow reference information, historical data of shared bicycles, and operational reference data for that area, which are then executed by the corresponding object sub-platforms.

[0032] It should be noted that the above description of the IoT system and its modules for the deployment and operation area planning of smart city shared bicycles is for convenience only and should not be construed as limiting this specification to the scope of the embodiments described. It is understood that those skilled in the art, after understanding the principles of this system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles.

[0033] Figure 2 This is an exemplary flowchart illustrating a smart city shared bicycle deployment and operation area planning method according to some embodiments of this specification. In some embodiments, process 200 can be executed by a management platform 130. Figure 2 As shown, process 200 includes the following steps:

[0034] Step 210: Obtain pedestrian flow reference information for at least one target area, determine the predicted pedestrian flow for at least one target area at a target time based on the pedestrian flow reference information, and determine the demand for shared bicycles based on the predicted pedestrian flow.

[0035] The target area can refer to the area where shared bicycles need to be operated and managed. For example, the target area can be a residential area, office area, subway entrance, hospital, school, business district, tourist attraction, etc. where shared bicycles have been deployed.

[0036] Pedestrian flow reference information refers to relevant information that influences changes in pedestrian flow. For example, different weather conditions may affect people's travel, thus affecting outdoor pedestrian flow. Different areas of popularity will also attract different people, resulting in varying pedestrian volumes.

[0037] In some embodiments, the pedestrian flow reference information may include at least one of the following: heat level of the target area, weather information, social activity information, and area type.

[0038] The popularity of a target area can refer to the degree of attention and preference people have for that area. In some embodiments, the popularity of a target area can be obtained through city heat maps. In some embodiments, the popularity of a target area can also be obtained through the attention and collection levels of users on third-party platforms (such as travel information platforms).

[0039] Weather information refers to weather-related information for a target area at a target time. Weather information may include temperature, rainfall, sunshine, etc. In some embodiments, weather information may be obtained through a climate-related third-party platform (such as a weather forecasting platform).

[0040] Social activity information can refer to information about various public activities held in a target area. Social activity information may include the activity time, location, scale, and number of participants. In some embodiments, social activity information can be obtained through relevant government activity management and approval platforms, or through the event organizer's relevant media platforms (such as WeChat official accounts, Weibo, etc.).

[0041] The area type can refer to the category to which the target area belongs, such as office area, residential area, entertainment venue, public service venue, etc. In some embodiments, the type of the target area can be determined based on its corresponding function. In some embodiments, the type of the target area can be obtained through a third-party platform (such as the housing and construction department). In some embodiments, different types of target areas have different pedestrian flows at different times. For example, during peak hours, the pedestrian flow at subway entrances is greater than during off-peak hours, while the opposite is true for entertainment venues.

[0042] The target time refers to the time period during which shared bicycle operations and management are required. The target time is a future time period based on the current time. For example, the target time could be 1 hour, 2 hours, 3 hours, or 1 day from the current time.

[0043] Current time refers to the actual time at present. The historical time, target time, and future time mentioned in the real-time examples in this specification are all determined based on the current time.

[0044] Predicted pedestrian traffic can refer to the predicted pedestrian traffic volume in a target area at a target time based on pedestrian traffic reference information. For example, predicted pedestrian traffic could be the pedestrian traffic volume in Tiananmen Square on October 1, 2026, based on pedestrian traffic reference information on October 1, 2025.

[0045] In some embodiments, the management platform 130 can perform statistical analysis based on pedestrian traffic reference information from a second historical time corresponding to the target time to determine the predicted pedestrian traffic. The second historical time can refer to a time period corresponding to the target time that is a certain period prior to the target time (such as one year, one month, etc.). For example, if the target time is 8:00-9:00 on October 1, 2025, then the second historical time could be one year ago, i.e., 8:00-9:00 on October 1, 2024. For example, the pedestrian traffic in the target area during the second historical time corresponding to the target time can be used as the predicted pedestrian traffic.

[0046] In some embodiments, the management platform 130 can output the predicted pedestrian flow of at least one target area at a target time based on pedestrian flow reference information and a first prediction model. In some embodiments, the management platform 130 can determine a pedestrian flow reference information sequence based on pedestrian flow reference information of multiple target areas within at least one target area; based on the pedestrian flow reference information sequence, it outputs a predicted pedestrian flow sequence using the first prediction model; each element in the predicted pedestrian flow sequence corresponds one-to-one with each target area within the multiple target areas. More information on determining predicted pedestrian flow based on the first prediction model can be found at [link to relevant documentation]. Figure 3 And its related descriptions.

[0047] The demand for shared bicycles can refer to the number of shared bicycles needed to meet the riding needs of people in a target area.

[0048] In some embodiments, the management platform 130 can determine the cycling ratio based on historical pedestrian traffic and historical ridership, and then determine the demand for shared bicycles based on the product of predicted pedestrian traffic and the cycling ratio. For example, if the cycling ratio determined based on historical pedestrian traffic in a certain area is 40%, and the predicted pedestrian traffic is 10,000 people, then the demand for shared bicycles in that area can be determined to be 4,000 people.

[0049] In some embodiments, the management platform 130 can determine the demand for shared bicycles based on a first prediction model. In some embodiments, the first prediction model includes a pedestrian flow prediction layer and a demand prediction layer: the pedestrian flow prediction layer is used to process pedestrian flow reference information to determine the predicted pedestrian flow; the demand prediction layer is used to process the predicted pedestrian flow to determine the demand for shared bicycles in at least one target area.

[0050] More information on determining the demand for shared bicycles based on the first prediction model can be found at [link to relevant documentation]. Figure 4 And its related descriptions.

[0051] Step 220: Obtain historical data of shared bicycles in multiple reference areas, and based on the historical data of shared bicycles, predict the change in the number of shared bicycles in at least one target area at the target time.

[0052] A reference area can refer to other areas related to the target area. For example, a reference area can include other areas connected to the target area by roads. Or, for example, a reference area can include other areas similar to the target area.

[0053] Historical data for shared bicycles can refer to operational data related to shared bicycles at a specific historical time. For example, historical data might include changes in the number of shared bicycles and road conditions in a reference area at that specific historical time. The first historical time refers to a period preceding the target time. For instance, if the target time is 9:00-10:00 AM on October 1, 2025, then the first historical time could be a period preceding 9:00 AM on October 1, 2025.

[0054] In some embodiments, shared bicycle historical data may include at least one of the following: the number of shared bicycles, pedestrian traffic, road environment data, and shared bicycle transfer data for a first historical time in multiple reference areas.

[0055] Road environment data may include information such as road accessibility, closures, and regulations. In some embodiments, when a road is impassable, the number of shared bicycles moving on the road is 0.

[0056] Shared bike transfer data includes the number of bikes transferred and the direction of transfer. The direction of transfer refers to the movement from one area to another. For example, the direction of transfer could mean that shared bikes move from area A to area B, or from area B to area A, etc.

[0057] In some embodiments, the shared bicycle historical data may also include at least one of the following: the number of shared bicycles, pedestrian traffic, road environment data, and shared bicycle transfer data in the target area at a first historical time.

[0058] In some embodiments, historical data on shared bicycles can be obtained through road monitoring. For example, monitoring video of the target area at a first historical time can be obtained based on road monitoring, and the video can be processed using an image recognition model to determine the number of shared bicycles, pedestrian flow, and shared bicycle transfer data. In some embodiments, shared bicycles in the target area may involve multiple operating companies. The number of shared bicycles and shared bicycle transfer data of different operating companies can be identified separately using an image recognition model, and then aggregated to obtain the total number of shared bicycles and total shared bicycle transfer data for the target area. In some embodiments, historical data on shared bicycles can be obtained based on a traffic management platform. For example, road closure and control information can be obtained based on a traffic management platform. In some embodiments, historical data on shared bicycles can also be obtained based on the backend data of multiple shared bicycle operating companies. For example, vehicle location data and trajectory data can be obtained based on the backend operational data of multiple companies; then data with the same location and trajectory can be integrated as the overall number of shared bicycles and shared bicycle transfer data.

[0059] The change in the number of shared bicycles can refer to the change in the number of shared bicycles in a target area within a target time period. In some embodiments, the change in the number of shared bicycles can reflect the riding popularity or demand in the target area. For example, a large change in the number of shared bicycles indicates high demand. In some embodiments, the change in the number of shared bicycles can include increases and decreases. An increase refers to the number of shared bicycles flowing from other areas to the target area, and a decrease refers to the number of shared bicycles flowing from the target area to other areas.

[0060] In some embodiments, the management platform 130 can determine the historical changes in shared bicycles based on historical data, and then determine the changes in shared bicycles based on the historical changes. For example, the average of the historical changes in shared bicycles over a certain period of multiple days can be used to determine the changes in shared bicycles for the corresponding future period.

[0061] In some embodiments, the management platform 130 can construct an operation map based on at least one target area, multiple reference areas, roads connecting the at least one target area and the multiple reference areas, and historical data of shared bicycles; process the operation map based on a second prediction model to determine the change in the number of shared bicycles in at least one target area at a target time; the second prediction model is a machine learning model. More information on determining the change in the number of shared bicycles based on the operation map can be found in [link to relevant documentation]. Figure 5 , Figure 6 And its related descriptions.

[0062] Step 230: Based on the demand for shared bicycles and the change in the number of shared bicycles, determine the number of shared bicycles to be deployed in at least one target area at the target time.

[0063] The number of shared bikes deployed can refer to the number of shared bikes that need to be deployed to a target area.

[0064] In some embodiments, the management platform 130 can determine the number of shared bicycles deployed by subtracting the demand for shared bicycles from the change in the number of shared bicycles. In some embodiments, when the change in the number of shared bicycles is an increase, the change in the number of shared bicycles is a positive value. This is equivalent to subtracting the increase from the demand for shared bicycles to obtain the number of shared bicycles deployed. In some embodiments, when the change in the number of shared bicycles is a decrease, the change in the number of shared bicycles is a negative value. This is equivalent to adding the decrease to the demand for shared bicycles to obtain the number of shared bicycles deployed. In some embodiments, the management platform 130 can also determine the number of shared bicycles deployed using other methods, which are not limited in this specification.

[0065] In some embodiments, the smart city shared bicycle deployment and operation area planning method further includes obtaining real-time operation reference data of shared bicycles in at least one target area, and adjusting the shared bicycle operation area for the target time based on the real-time operation reference data.

[0066] Real-time operational reference data refers to real-time data related to the operation of shared bicycles, which can be used to assist in the management of shared bicycle operating areas.

[0067] In some embodiments, real-time operational reference data may include at least one of the following: environmental information, weather information, policy information, time information, social activity information, and user cycling information for at least one target area.

[0068] Environmental information may include road construction information, road damage information (such as landslides), and congestion information within the operating area. Policy information may include control information, lockdown information (such as epidemic lockdowns), and bans on cycling. Time information refers to the time corresponding to the real-time operational reference data; the real-time operational reference data differs at different times. User cycling information may refer to the user's historical cycling trajectory information, frequency of cycling out of the operating area, etc. Weather information and social activity information can be found in the description in step 210.

[0069] In some embodiments, real-time operational reference data can be obtained through various means. For example, it can be obtained through third-party platforms such as traffic management platforms, government management platforms, and weather forecasting platforms.

[0070] The shared bicycle operating area refers to the area where shared bicycles can be ridden and parked.

[0071] In some embodiments, the management platform 130 can adjust the operating area in various ways based on real-time operational reference data. In some embodiments, the management platform 130 can adjust the operating area based on any one of the real-time operational reference data. For example, it can determine whether to adjust the operating area and the adjustment range based on the user's cycling trajectory (i.e., the frequency of cycling out of the operating area) in the user's cycling information.

[0072] For example, the operating area can be adjusted by comparing the frequency of riding out of the operating area with a preset frequency threshold; if the frequency exceeds the preset frequency threshold, the operating area can be adjusted. Further, the adjustment range can be determined based on the area corresponding to the most frequent riding trajectory. In some embodiments, the management platform 130 can comprehensively adjust the operating area based on multiple real-time operational reference data. For example, a weighted sum can be performed based on the degree of influence of multiple real-time operational reference data on the adjustment of the operating area, and the adjustment strategy can be determined based on the summation result.

[0073] In some embodiments, the management platform 130 can determine the priority of operating area adjustments based on real-time operational reference data; determine the operating area adjustment strategy based on the operating area adjustment priority; and adjust the shared bicycle operating areas for a target time based on the operating area adjustment strategy. More information about adjusting operating areas can be found at [link to relevant documentation]. Figure 7 And its related descriptions.

[0074] In some embodiments, real-time operational reference data further includes predicted pedestrian flow and changes in shared bicycle usage. In some embodiments, the management platform 130 can acquire predicted pedestrian flow and changes in shared bicycle usage in real time; perform a weighted summation of the predicted pedestrian flow and changes in shared bicycle usage to determine the weighted summation result; determine an operational zone adjustment strategy based on the weighted summation result; and adjust the shared bicycle operational zone for a target time based on the operational zone adjustment strategy. More information on adjusting operational zones can be found in [link to relevant documentation]. Figure 8 And its related descriptions.

[0075] This specification describes several embodiments that adjust the shared bicycle operating area using various real-time operational reference data. This method allows for adjustments to the operating area based on changes in the external environment, people's social behavior, and user riding needs; making the adjustments more precise, better meeting people's needs, and improving the user experience.

[0076] Some embodiments in this specification predict pedestrian traffic using pedestrian traffic reference information, determine the demand for shared bicycles based on pedestrian traffic information, determine the change in shared bicycle availability using historical shared bicycle data, and determine the deployment quantity based on the demand and change in shared bicycle availability. This method considers multiple factors related to bicycle deployment, making the determined bicycle deployment quantity more accurate; while meeting people's riding needs, it reduces the difficulty and cost of operation and management, making the operation and management of shared bicycles efficient and orderly. By adjusting the operating area through multiple operational reference data, it better balances meeting people's riding needs with controlling operating costs.

[0077] Figure 3 This is an exemplary schematic diagram of a first prediction model structure shown in some embodiments of this specification.

[0078] In some embodiments, determining the predicted pedestrian flow of at least one target area at a target time based on pedestrian flow reference information includes: based on pedestrian flow reference information, outputting the predicted pedestrian flow of at least one target area at a target time through a first prediction model; the first prediction model is a machine learning model.

[0079] In some embodiments, the first prediction model may be at least one of Convolutional Neural Networks (CNN), Deep Neural Networks (DNN), Recurrent Neural Networks (RNN), or other custom models.

[0080] like Figure 3 As shown, the input to the first prediction model 320 can be pedestrian flow reference information 310, and the output is predicted pedestrian flow 330. For details regarding pedestrian flow reference information, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0081] In some embodiments, the first prediction model can be obtained through training with a first training sample and a first label. The first training sample may be historical pedestrian flow reference information for multiple target areas, and the first label is the historical actual pedestrian flow for each target area. In some embodiments, historical surveillance videos and / or images can be obtained through road monitoring, and the first label can be determined based on an image recognition model.

[0082] In some embodiments, the training of the first prediction model can be performed by the management platform 130. In some embodiments, the management platform 130 can input the first training sample into the initial first prediction model, construct a loss function based on the output of the initial first prediction model and the first label; update the parameters of the initial first prediction model based on the loss function using gradient descent or other methods; until a preset condition is met, the trained first prediction model is obtained. The preset condition can be that the loss function converges or the training reaches the maximum number of iterations.

[0083] In some embodiments, determining the predicted pedestrian flow of at least one target area at a target time based on pedestrian flow reference information includes: determining a pedestrian flow reference information sequence based on pedestrian flow reference information of multiple target areas within the at least one target area; outputting a predicted pedestrian flow sequence based on the pedestrian flow reference information sequence using a first prediction model; and each element in the predicted pedestrian flow sequence corresponds one-to-one with each target area within the multiple target areas.

[0084] In some embodiments, the management platform 130 may determine a sequence of pedestrian flow reference information based on pedestrian flow reference information of multiple target areas in at least one target area.

[0085] A pedestrian flow reference information sequence can refer to a queue composed of multiple pedestrian flow information items arranged in a certain way. For example, pedestrian flow reference information can be sorted according to time sequence, geographical location, etc., to obtain a pedestrian flow reference information sequence. In some embodiments, the management platform 130 can number multiple target areas and arrange the pedestrian flow reference information corresponding to the multiple target areas based on the target area numbers to obtain a pedestrian flow reference information sequence.

[0086] A predicted pedestrian flow sequence can refer to a queue formed by sorting predicted pedestrian flow according to certain rules. For example, sorting the predicted pedestrian flow corresponding to multiple target areas based on target area numbers can form a predicted pedestrian flow sequence.

[0087] In some embodiments, the management platform 130 can output a predicted pedestrian flow sequence based on a pedestrian flow reference information sequence and a first prediction model; each element in the predicted pedestrian flow sequence corresponds one-to-one with each of the multiple target areas. (Continue to refer to...) Figure 3 The input to the first prediction model 320 can be a pedestrian flow reference information sequence 310, and the output is a predicted pedestrian flow sequence 330.

[0088] In some embodiments, when the first prediction model is used to process pedestrian flow reference information sequences, the first training sample can be a sequence of sample pedestrian flow reference information composed of historical pedestrian flow reference information from multiple target areas, arranged according to area numbers; the first label is a sequence composed of historical actual pedestrian flow from multiple target areas. Specific training methods can be found in the relevant description of training the first prediction model for processing pedestrian flow reference information, and will not be elaborated upon here.

[0089] Some embodiments in this specification demonstrate that by inputting a sequence of pedestrian flow reference information from multiple target areas into the model, the predicted pedestrian flow of multiple target areas can be obtained simultaneously, thereby improving prediction efficiency.

[0090] In some embodiments, the first prediction model may include multiple first prediction sub-models, each of which corresponds one-to-one with a target area; each first prediction sub-model is used to process the pedestrian flow reference information of the corresponding target area to determine the predicted pedestrian flow of the corresponding target area. Figure 3 As shown, the first prediction model 320 may include the first prediction sub-model 320-1, the first prediction sub-model 320-2, ..., the first prediction sub-model 320-n.

[0091] Continue to refer to Figure 3 The first prediction sub-model 320-1 takes as input the pedestrian flow reference information 310-1 for its corresponding first target area and outputs as the predicted pedestrian flow 330-1 for its corresponding first target area. The first prediction sub-model 320-2 takes as input the pedestrian flow reference information 310-2 for its corresponding second target area and outputs as the predicted pedestrian flow 330-2 for its corresponding second target area. ... The first prediction sub-model 320-n takes as input the pedestrian flow reference information 310-n for its corresponding Nth target area and outputs as the predicted pedestrian flow 330-n for its corresponding Nth target area.

[0092] In some embodiments, a first prediction sub-model can be trained by training a first prediction model. In some embodiments, the first training samples for training the first prediction sub-model are pedestrian flow reference information for multiple second historical times in the target area corresponding to each first prediction sub-model, and the first label is the actual pedestrian flow for each second historical time.

[0093] In some embodiments, the first prediction model may include an embedding layer and a prediction layer. The embedding layer is used to embed pedestrian flow reference information to obtain a pedestrian flow feature vector; the prediction layer is used to process the pedestrian flow feature vector to determine the predicted pedestrian flow.

[0094] In some embodiments, the embedding layer and prediction layer of the first prediction model can be obtained through joint training. In some embodiments, the management platform 130 can train the embedding layer and prediction layer based on the first training samples and the first label. In some embodiments, the management platform 130 can input the first training samples into the initial embedding layer to obtain an initial pedestrian flow feature vector; input the initial pedestrian flow feature vector into the prediction layer to obtain an initial predicted pedestrian flow; construct a loss function based on the initial predicted pedestrian flow and the first label, and update the parameters of the embedding layer and the prediction layer simultaneously. Through parameter updates, the trained embedding layer and prediction layer are obtained. For details regarding the first training samples and the first label, please refer to the relevant description above.

[0095] By training a first prediction sub-model based on historical pedestrian flow reference information for multiple target areas, each target area's first prediction sub-model can learn the differentiated characteristics of that area, improving its performance. Then, based on the trained first prediction sub-model, pedestrian flow can be predicted, thus improving prediction accuracy.

[0096] Some embodiments in this specification predict pedestrian traffic in a target area using a first prediction model. This leverages the self-learning capabilities of machine learning models to identify patterns in large amounts of data, thereby improving prediction efficiency and accuracy. The demand for and changes in shared bicycles are closely related to pedestrian traffic. Predicting pedestrian traffic provides reliable data support for subsequent predictions of shared bicycle demand and changes in shared bicycle traffic.

[0097] In some embodiments, the first prediction model includes a pedestrian flow prediction layer and a demand prediction layer: the pedestrian flow prediction layer is used to process pedestrian flow reference information to determine the predicted pedestrian flow; the demand prediction layer is used to process the predicted pedestrian flow to determine the demand for shared bicycles in at least one target area.

[0098] In some embodiments, such as Figure 4 As shown, the first prediction model may include a pedestrian flow prediction layer 420 and a demand prediction layer 440.

[0099] In some embodiments, the pedestrian flow prediction layer 420 processes the pedestrian flow reference information 410 to determine the predicted pedestrian flow 430. In some embodiments, the demand prediction layer 440 processes the predicted pedestrian flow 430 to determine the shared bicycle demand 450 for at least one target area. For details regarding the pedestrian flow reference information, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0100] In some embodiments, the pedestrian flow prediction layer 420 and the demand prediction layer 440 can be obtained through separate training. Figure 3 The first prediction model / first prediction sub-model described herein shares parameters, and its training can be performed by... Figure 3 This is achieved through the method described in the document for training the first prediction model / first prediction sub-model. In some embodiments, the third training sample for training the demand prediction layer 440 may be the pedestrian flow in multiple target areas at a second historical time; the third label may be the actual shared bicycle demand in multiple target areas at a second historical time. The third label may be determined based on the operational data of major shared bicycle operating companies.

[0101] In some embodiments, the management platform 130 can input a third training sample into the initial demand prediction layer, construct a loss function based on the output of the initial demand prediction layer and the third label, update the parameters of the initial demand prediction layer based on the loss function using gradient descent or other methods, and obtain the trained demand prediction layer until the loss function converges or the training reaches its maximum number of iterations.

[0102] In some embodiments, the pedestrian flow prediction layer and the demand prediction layer can be obtained through joint training. In some embodiments, the fourth training sample for jointly training the pedestrian flow prediction layer and the demand prediction layer is pedestrian flow reference information for at least one target area at a second historical time; the fourth label is the actual shared bicycle demand for at least one target area at the second historical time. In some embodiments, the fourth training sample can be input into the initial pedestrian flow prediction layer to obtain the initial predicted pedestrian flow for at least one target area at the second historical time; the initial predicted pedestrian flow for at least one target area at the second historical time can be input into the initial demand prediction layer to obtain the initial shared bicycle demand for at least one target area at the second historical time. A loss function is constructed based on the initial shared bicycle demand and the fourth label, and the parameters of the initial pedestrian flow prediction layer and the initial demand prediction layer are updated synchronously. Through parameter updates, the trained pedestrian flow prediction layer and demand prediction layer are obtained.

[0103] Some embodiments in this specification incorporate a pedestrian flow prediction layer and a demand prediction layer within a first prediction model. This first prediction model can determine the demand for shared bicycles based on pedestrian flow reference information, thereby improving data processing efficiency and prediction speed.

[0104] Figure 5 This is an exemplary schematic diagram illustrating the determination of changes in shared bicycles based on operational maps, according to some embodiments of this specification.

[0105] In some embodiments, predicting the change in the number of shared bicycles in at least one target area at a target time based on historical data of shared bicycles includes: constructing an operation map based on at least one target area, multiple reference areas, roads connecting at least one target area and multiple reference areas, roads connecting multiple reference areas, and historical data of shared bicycles; processing the operation map based on a second prediction model to determine the change in the number of shared bicycles in at least one target area at a target time; the second prediction model is a machine learning model.

[0106] In some embodiments, the management platform 130 can determine the changes in shared bicycles by constructing an operational map and processing the operational map based on a second prediction model.

[0107] In some embodiments, the management platform 130 can construct an operational map based on at least one target area, multiple reference areas, roads connecting the at least one target area and the multiple reference areas, roads connecting the multiple reference areas, and historical data of shared bicycles. For details regarding the historical data of shared bicycles, please refer to... Figure 2 And its related descriptions.

[0108] An operational map can refer to a graph used to store and reflect shared bicycle operational data and related data. In some embodiments, the characteristics of shared bicycle operational data and related data, as well as the relationships between the data, can be represented in the form of a graph to form an operational map.

[0109] In some embodiments, the operation map uses at least one target area and multiple reference areas as nodes and roads as edges, with the edges being directed edges pointing to the corresponding shared bicycle flow direction; node features include the number of shared bicycles and pedestrian traffic at a first historical time; edge features include shared bicycle transfer data and road environment data at the first historical time.

[0110] In some embodiments, such as Figure 5 As shown, the nodes in the operation map 510 consist of at least one target area node and multiple reference area nodes. The target area nodes are represented by black nodes 510-1, 510-2, and 510-3, while the reference area nodes are represented by multiple white nodes. The characteristics of the target area nodes and reference area nodes include the number of shared bicycles and pedestrian traffic at the first historical time. For details on how to obtain the number of shared bicycles and pedestrian traffic at the first historical time, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0111] In some embodiments, the operational map uses roads connecting the target area and multiple reference areas, and roads connecting the multiple reference areas to each other, as edges. For example... Figure 5As shown in the operation graph 510, target areas 510-1, 510-2, and 510-3 are connected to their respective reference areas by edges, and the reference areas are interconnected by edges. The edges are directed, and their direction indicates the direction in which shared bicycles flow (transfer) between the target areas and the reference areas. For example, the edge connecting target area 510-1 and its corresponding reference areas can represent the flow direction of shared bicycles between target area 510-1 and its corresponding reference areas.

[0112] In some embodiments, the features of an edge include shared bicycle transfer data and road environment data at a first historical time. For example, an edge connecting a target area 510-1 and the corresponding multiple reference areas is characterized by the number and direction of shared bicycles flowing (transferring) between the target area 510-1 and the corresponding multiple reference areas at the first historical time, as well as road environment data of the roads connecting the target area 510-1 and the corresponding multiple reference areas.

[0113] In some embodiments, the second prediction model 520 may be at least one of a graph neural network (GNN), a graph convolution network (GCN), a graph attention network (GAN), or other custom models.

[0114] In some embodiments, such as Figure 5 As shown, the input to the second prediction model 520 is the operation map 510, and the output is the change in shared bicycles 530 of at least one node in the target area at the target time. For example, assuming the target time is 9:00-10:00 on October 1, 2025, the historical data of shared bicycles from 8:00-9:00 on October 1, 2025 can be used to construct the map; by inputting the constructed operation map into the second prediction model 520, the change in shared bicycles in the target area during the target time of 9:00-10:00 on October 1, 2025 can be obtained.

[0115] In some embodiments, the management platform 130 can extract the operational subgraph corresponding to each target region from the operational graph based on each target region node with a neighbor degree of 1. A neighbor degree of 1 means that the node is directly connected to the target region without passing through any intermediate nodes. For example, ... Figure 5As shown, the management platform 130 can extract the operation subgraph corresponding to the target region 510-1 based on the target region 510-1 and with an adjacency degree of 1. The operation subgraph corresponding to the target region 510-1 is shown in the area in the dashed box 510-4, including the target region 510-1, the reference regions adjacent to the target region 510-1, and the edges connecting the target region 510-1 and the corresponding multiple reference regions.

[0116] In some embodiments, the management platform 130 can process the operation submap corresponding to each target area based on the second prediction model 520 to determine the change in shared bicycles in each target area. For example, the management platform 130 can input the operation submap corresponding to target area 510-1 into the second prediction model 520 to obtain the change in shared bicycles in target area 510-1 at the target time.

[0117] In some embodiments, the second prediction model 520 can be obtained through training. In some embodiments, the second training samples for training the second prediction model 520 can be multiple sample operation maps constructed based on shared bicycle historical data from multiple first historical times. For example, if the target time is 13:00-14:00 on October 1, 2025, then the multiple first historical times can be 6:00-7:00, 7:00-8:00, ..., 11:00-12:00, 12:00-13:00, etc. on October 1, 2025. The second label can be the actual change in shared bicycles corresponding to each sample operation map, that is, the actual change in shared bicycles in the next historical time after each first historical time. The second label can be determined by processing historical road surveillance videos using an image recognition model. For example, a map can be constructed based on the historical data of shared bicycles from 6:00 to 7:00 on October 1, 2025, the first historical time. The second label can be the actual change in the number of shared bicycles from 7:00 to 8:00 on October 1, 2025, the next historical time.

[0118] In some embodiments, the management platform 130 can input the second training samples into the initial second prediction model to obtain the initial shared bicycle change amount. A loss function is constructed based on the initial shared bicycle change amount and the second label, and the parameters of the initial second prediction model are updated using the loss function. Through parameter updates, the trained second prediction model is obtained.

[0119] Some embodiments in this specification demonstrate that by constructing an operational graph, operational data, related data, and the relationships and characteristics between these data can be represented more intuitively. Furthermore, by processing the operational graph using a second prediction model, the model's ability to handle graph structures can be leveraged to improve the accuracy and efficiency of predicting changes in shared bicycle usage.

[0120] Figure 6This is another exemplary schematic diagram illustrating the determination of changes in shared bicycles based on operational maps, according to some embodiments of this specification.

[0121] In some embodiments, the change in shared bicycles is related to the change in pedestrian flow, which is determined based on historical and predicted pedestrian flow.

[0122] The change in pedestrian flow refers to the quantitative change in pedestrian traffic. For example, the change in pedestrian flow can be the difference in pedestrian traffic at two different points in time. For instance, suppose the pedestrian traffic in a target area is 10,000 people at 9:00 AM on October 1, 2025, and 12,000 people at 10:00 AM. Then the change in pedestrian flow in the target area from 9:00 AM to 10:00 AM on October 1, 2025, is 2,000 people. The change in pedestrian flow can be divided into increases and decreases in pedestrian traffic.

[0123] In some embodiments, the change in pedestrian flow can be determined based on historical pedestrian flow and predicted pedestrian flow. Historical pedestrian flow refers to the pedestrian flow at time points prior to the target time. For example, if the target time is 10:00 AM on October 1, 2025, historical pedestrian flow could be the pedestrian flow at 9:00 AM, 9:30 AM, 9:50 AM, etc., on October 1, 2025. In some embodiments, the management platform 130 can subtract the historical pedestrian flow from the predicted pedestrian flow, using the difference as the change in pedestrian flow. For example, if the historical pedestrian flow for the target time is 1000 people, and the predicted pedestrian flow for the target time is 1200 people, then the management platform 130 can subtract 1000 people from 1200 people to obtain a change in pedestrian flow of 200 people, i.e., an increase in pedestrian flow of 200 people.

[0124] Increases and decreases in pedestrian traffic can affect the riding rate of shared bicycles, thus influencing the change in the number of shared bicycles. For example, an increase in pedestrian traffic may lead to an increase in the riding rate of shared bicycles, resulting in a corresponding increase in the number of shared bicycles. Therefore, predicting changes in pedestrian traffic can provide stronger data support for subsequent predictions of changes in the number of shared bicycles.

[0125] In some embodiments, the second prediction model may include a first prediction layer and a second prediction layer. The first prediction layer is used to process the operational map to determine a first change in the number of shared bicycles in at least one target area at a target time; the second prediction layer is used to process the first change in the number of shared bicycles and the change in pedestrian flow to determine a second change in the number of shared bicycles in at least one target area at a target time.

[0126] The first change in shared bike volume refers to the change in shared bike volume in at least one target area predicted based on operational mapping. The second change in shared bike volume is the predicted change in shared bike volume based on the first change in shared bike volume, combined with changes in pedestrian flow; this is the final change in shared bike volume. The second change in shared bike volume is more accurate than the first change in shared bike volume.

[0127] like Figure 6 As shown, the second prediction model may include a first prediction layer 620 and a second prediction layer 650. In some embodiments, the first prediction layer may be a graph neural network (GNN), and the second prediction model may be a deep neural network (DNN). The input of the first prediction layer 620 is the operation map 610, and the output is a first change 630 of shared bicycles in at least one target area at a target time. The input of the second prediction layer 650 is the first change 630 of shared bicycles and the change 640 of pedestrian flow, and the output is a second change 660 of shared bicycles in at least one target area at a target time.

[0128] In some embodiments, the second change in shared bicycles is the final change in shared bicycles.

[0129] In some embodiments, the first prediction layer and the second prediction layer of the second prediction model can be obtained through separate training. In some embodiments, the first prediction layer and... Figure 5 The second prediction model described herein shares parameters, and its training can be performed by... Figure 3 The method of training the second prediction model as described above will not be elaborated here.

[0130] In some embodiments, the fifth training sample for training the second prediction layer of the second prediction model can be the first change in sample shared bicycles and the change in sample pedestrian flow. The first change in sample shared bicycles can be obtained based on the trained first prediction layer. The change in sample pedestrian flow can be determined based on pedestrian flow at multiple first historical time points in the target area and the corresponding pedestrian flow at multiple next historical time points. The first change in sample shared bicycles and the change in sample pedestrian flow are in a one-to-one correspondence in time, i.e., both are obtained based on the same time. For example, pedestrian flow at multiple first historical time points in the target area can be obtained. If the target time is 3 PM on October 1, 2025, then pedestrian flow at 8 AM, 9 AM, 10 AM, 11 AM, 12 PM, 1 PM, 2 PM, and 3 PM on October 1, 2025 can be obtained. Based on the pedestrian flow at 8 AM and 9 AM, 9 AM and 10 AM, 10 AM and 11 AM, ..., 2 PM and 3 PM, multiple sample pedestrian flow changes can be determined respectively. For details on obtaining pedestrian flow, please refer to [link to relevant documentation]. Figure 2And its related descriptions.

[0131] In some embodiments, the fifth label used to train the second prediction layer can be the actual change in the number of shared bicycles corresponding to the training samples. The fifth label can be determined by processing the road surveillance video of the first historical time corresponding to the fifth training sample using an image recognition model.

[0132] In some embodiments, the management platform 130 can input the fifth training sample (sample shared bicycle first change and sample pedestrian flow change) into the initial second prediction layer to obtain the initial shared bicycle second change. A loss function is constructed based on the initial shared bicycle second change and the fifth label to update the parameters of the initial second prediction layer. Based on the parameter update, the trained second prediction layer is obtained.

[0133] In some embodiments, the first and second prediction layers of the second prediction model can be obtained through joint training.

[0134] In some embodiments, the sixth training data for joint training includes multiple sample operation maps and sample pedestrian flow changes constructed based on historical data of shared bicycles at multiple first historical time points. The sixth label is the actual change in the number of shared bicycles in the target area at each first historical time point.

[0135] In some embodiments, the management platform 130 can input the sample operation map into the initial first prediction layer to obtain the first change in shared bicycles output by the first prediction layer; using the first change in shared bicycles as a sample, and the sample change in pedestrian flow as a sample, it can input into the initial second prediction layer to obtain the second change in shared bicycles output by the second prediction layer. A loss function is constructed based on the sixth label and the second change in shared bicycles output by the second prediction layer, and the first and second prediction layers are updated synchronously. Through parameter updates, the trained first and second prediction layers are obtained.

[0136] By jointly training the first and second prediction layers, the model can learn deeper data features, resulting in a better-performing model and improved prediction accuracy.

[0137] Some embodiments in this specification provide strong data support for subsequent predictions of shared bicycle volume changes by determining changes in pedestrian flow. Since changes in shared bicycle volume are influenced by changes in pedestrian flow, adding pedestrian flow changes as input when predicting shared bicycle volume increases the accuracy of the prediction, making the prediction results more consistent with reality.

[0138] In some embodiments, the management platform 130 can determine the number of shared bicycles to be deployed based on the demand for shared bicycles and the change in the number of shared bicycles in at least one target area. For details on determining the number of shared bicycles to be deployed, please refer to [link / reference needed]. Figure 2 And its related descriptions.

[0139] In some embodiments, the shared bicycle deployment volume determined by the management platform 130 is the total number of shared bicycles deployed by multiple operating companies. In some embodiments, the management platform 130 can determine the number of shared bicycles deployed by each operating company in the corresponding target area based on the shared bicycle deployment ratio of each operating company, and send this information to each operating company through the user platform. For example, if the shared bicycle deployment ratios of operating companies A, B, and C in target area 1 are 30%, 30%, and 40%, respectively, and the determined number of shared bicycles deployed in target area 1 at the target time is 1000, then based on the deployment ratio, the number of shared bicycles deployed by operating companies A, B, and C in target area 1 can be determined to be 300, 300, and 400, respectively.

[0140] In some embodiments, the management platform 130 can determine the deployment ratio based on the proportion of each company's historical changes in shared bicycles to the total historical changes in shared bicycles in the corresponding area. The historical changes in shared bicycles include the sum of input and output, and can be obtained through aggregated analysis of the operational data of each company. For example, if the total historical changes in shared bicycles in target area 1 are 300, and the historical changes in shared bicycles for operating companies A, B, and C are 60, 90, and 150 respectively, then the deployment ratios for operating companies A, B, and C can be determined to be 20%, 30%, and 50%, respectively.

[0141] Figure 7 This is an exemplary flowchart illustrating the adjustment of the operating area according to some embodiments of this specification. In some embodiments, process 700 may be executed by user platform 110 and management platform 130. Figure 7 As shown, process 700 includes the following steps.

[0142] Step 710: Determine the priority of operational area adjustments based on real-time operational reference data.

[0143] Operational zone adjustment priority refers to the importance and priority order of real-time operational reference data when adjusting the operational zone. For example, the operational zone adjustment priority can be determined based on the importance of each type of real-time operational reference data, deciding which type of real-time operational reference data to use as the primary reference for adjustment.

[0144] In some embodiments, the priority of adjusting operating areas can be determined based on the importance of real-time operational reference data. For example, for each type of real-time operational reference data, they can be sorted according to their importance, and the reference order for adjusting operating areas can be determined based on the sorting. Exemplarily, real-time operational reference data includes environmental information, weather information, policy information, time information, social activity information, and user cycling information for the target area; they can be sorted according to the importance of each type of information to determine the priority. For example, the priority based on importance sorting could be: policy information > environmental information > weather information > social activity information > user cycling information. In some embodiments, the importance of each type of real-time operational reference data can be set based on actual operational needs, or it can be set by default by the management platform 130. When setting the importance level, the impact of different real-time reference data on cycling safety, cycling accessibility, and shared bicycle operating costs can be considered.

[0145] Step 720: Determine the operation area adjustment strategy based on the operation area adjustment priority.

[0146] The operational area adjustment strategy refers to the specific adjustment plan for the operational area. This includes, for example, whether to adjust, the method of adjustment, and the scope of adjustment. The adjustment method can include shrinking the operational area, expanding the operational area, or closing the operational area.

[0147] In some embodiments, the adjustment strategy for operating areas can be determined sequentially based on the priority of the operating area adjustment. For example, firstly, it is determined whether to adjust the operating area, the adjustment method, and the adjustment scope based on the policy information of the first priority. If the result is adjustment, the operating area is directly adjusted according to the policy information. For example, based on the lockdown information in the policy information, the locked-down area is adjusted to a non-operating area. If the result is no adjustment, it is then determined whether to adjust, the adjustment method, and the scope based on the environmental information of the second priority. By proceeding sequentially according to the aforementioned method, the final operating area adjustment strategy can be determined. In some embodiments, the operating area adjustment strategy is related to the real-time reference data used to determine the operating area to be adjusted. For example, if the adjustment of the operating area is determined based on the policy information of the first priority, the corresponding adjustment strategy can be determined based on the specific policy information. If the adjustment of the operating area is determined based on the environmental information of the second priority, the corresponding adjustment strategy can be determined based on the specific environmental information.

[0148] Step 730: Based on the operation area adjustment strategy, adjust the shared bicycle operation area for the target time.

[0149] In some embodiments, the management platform 130 can upload the operation area adjustment strategy to the user platform 110 through the service platform 120. In some embodiments, the user platform 110 can adjust the shared bicycle operation area for a target time based on the specific adjustment method and adjustment range in the operation area adjustment strategy. For example, if the operation area adjustment strategy is adjustment, the adjustment method is expansion, and the adjustment range is 1 kilometer, then the user platform 110 can make specific adjustments to the operation area based on the aforementioned information.

[0150] Some embodiments in this specification determine the adjustment priority of the operating area based on real-time operational reference data, and determine the operating area adjustment strategy based on the adjustment priority, so as to adjust the operating area. In determining the operating area adjustment priority and the operating area adjustment strategy, the impact and degree of various real-time reference data of the operating area on riding and operation management are considered, so as to make the operating area adjustment more accurate and in line with the actual riding needs and the operating management needs of shared bicycles.

[0151] It should be noted that the above description of process 700 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 700 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification. For example, process 700 may include determining the importance of real-time operational reference data.

[0152] Figure 8 This is another exemplary flowchart illustrating the adjustment of the operating area according to some embodiments of this specification. In some embodiments, process 800 may be executed by user platform 110 and management platform 130. Figure 8 As shown, process 800 includes the following steps:

[0153] Step 810: Obtain the predicted pedestrian flow and changes in shared bicycles in real time.

[0154] In some embodiments, the management platform 130 can obtain the predicted pedestrian flow and shared bicycle changes stored in the management platform database in real time. In some embodiments, the determination of the predicted pedestrian flow and shared bicycle changes is also performed by the management platform 130, and the management platform 130 can directly perform the next weighted summation based on the determined predicted pedestrian flow and shared bicycle changes.

[0155] Step 820: Perform a weighted summation of the predicted pedestrian flow and changes in shared bicycles, and determine the weighted summation result.

[0156] In some embodiments, the management platform 130 can perform a weighted summation of the predicted pedestrian flow and changes in shared bicycles based on preset weights to determine the weighted summation result. In some embodiments, the preset weights can be determined based on actual operational needs or set by system default, with the sum of weights being 1.

[0157] Step 830: Based on the weighted summation result, determine the operation area adjustment strategy.

[0158] In some embodiments, the management platform 130 can determine the operation area adjustment strategy by comparing the weighted summation result with a preset adjustment threshold. The preset adjustment threshold can be set by system default or based on actual operational needs. For example, if the weighted summation result is less than the preset adjustment threshold, the operation area is not adjusted; if the weighted summation result is greater than the preset adjustment threshold, the operation area is increased. In some embodiments, the operation area adjustment strategy also includes an adjustment range. In some embodiments, the operation area adjustment range can be determined based on the weighted summation result. For example, a numerical interval can be pre-defined, with each interval corresponding to an adjustment range. The numerical interval to which the weighted summation result belongs is determined, and the adjustment range corresponding to that interval is the required shared bicycle adjustment range. For example, before weighting, the predicted pedestrian flow and the change in shared bicycles are both integers, and after weighting, the weighted summation result is also a single value; the possible numerical intervals can be divided to form multiple numerical intervals.

[0159] Step 840: Based on the operation area adjustment strategy, adjust the shared bicycle operation area for the target time.

[0160] In some embodiments, the management platform 130 can adjust the shared bicycle operating area for a target time based on specific information in the operating area adjustment strategy. For example, the operating area can be adjusted based on the specific adjustment method and adjustment scope in the operating area adjustment strategy.

[0161] Some embodiments in this specification determine the operation area adjustment strategy based on predicted pedestrian flow and changes in shared bicycles, taking into account the impact of pedestrian flow and changes in bicycles on riding demand and riding range, which is more in line with the actual operation area adjustment needs.

[0162] It should be noted that the above description of process 800 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 800 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification. For example, process 800 may include determining the weights for a weighted summation.

[0163] One embodiment of this specification also provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes the smart city shared bicycle deployment and operation area planning method described in this specification embodiment.

[0164] The beneficial effects that the embodiments of this specification may bring include, but are not limited to: (1) Based on the reference information of pedestrian traffic in each target area and the historical data of shared bicycles, the demand for and changes in the number of shared bicycles are predicted; and then the number of shared bicycles to be deployed is determined based on the demand for and changes in the number of shared bicycles. This method fully considers the impact of various factors on the demand for shared bicycles and improves the accuracy of determining the number of shared bicycles to be deployed. At the same time, the operation area adjustment strategy is determined based on the real-time operation reference data of shared bicycles, real-time pedestrian traffic and changes in the number of shared bicycles, and the operation area is dynamically adjusted to meet the actual operation needs. (2) By training the machine learning model, the large amount of data involved is processed to improve the efficiency and accuracy of data processing. Moreover, data acquisition and data processing are both executed on the physical network platform, realizing the automation and intelligence of shared bicycle operation management and improving the efficiency and effectiveness of operation management. (3) By constructing an operation map to process the historical data of shared bicycles in multiple areas, the correlation between data and the data flow direction are made more intuitive and clear. Then, the graph model is used to process the graph to improve the processing efficiency and processing effect.

[0165] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0166] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0167] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0168] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0169] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0170] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0171] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for planning the deployment and operation area of ​​shared bicycles in smart cities, characterized in that, The management platform based on the IoT system for smart city shared bicycle deployment and operation area planning is implemented, including: Obtain pedestrian flow reference information for at least one target area; based on the pedestrian flow reference information, output the predicted pedestrian flow for the at least one target area at a target time using a first prediction model, wherein the first prediction model is a machine learning model; determine the demand for shared bicycles based on the predicted pedestrian flow. Obtain historical data of shared bicycles from multiple reference areas, and construct an operation map based on the at least one target area, the multiple reference areas, the roads connecting the at least one target area and the multiple reference areas, the roads connecting the multiple reference areas, and the historical data of shared bicycles; The operation map uses the at least one target area or the multiple reference areas as nodes, and the roads connecting the target area and the multiple reference areas or the roads connecting the multiple reference areas as edges. The edges are directed edges. The node features of the nodes include the number of shared bicycles and pedestrian traffic at a first historical time. The edge features include shared bicycle transfer data and road environment data at the first historical time. The operational map is processed based on a second prediction model to determine the change in the number of shared bicycles in the at least one target area during the target time. The second prediction model is a machine learning model. The change in the number of shared bicycles is related to the change in pedestrian flow, which is determined based on historical pedestrian flow and the predicted pedestrian flow. The change in the number of shared bicycles refers to the change in the number of shared bicycles in the target area during the target time, which reflects the riding popularity or riding demand in the target area. The change in the number of shared bicycles includes both increases and decreases. The second prediction model includes a first prediction layer and a second prediction layer; the first prediction layer is used to process the operation map to determine the first change in the number of shared bicycles in the at least one target area at the target time; the second prediction layer is used to process the first change in the number of shared bicycles and the change in the number of pedestrians to determine the second change in the number of shared bicycles in the at least one target area at the target time. Based on the demand for shared bicycles and the change in the number of shared bicycles, the number of shared bicycles deployed in the at least one target area at the target time is determined.

2. The method according to claim 1, characterized in that, The method further includes obtaining real-time operation reference data of shared bicycles in the at least one target area, and adjusting the shared bicycle operation area for the target time based on the real-time operation reference data.

3. The method according to claim 1, characterized in that, The first prediction model includes a pedestrian flow prediction layer and a demand prediction layer: The pedestrian flow prediction layer is used to process the pedestrian flow reference information to determine the predicted pedestrian flow. The demand prediction layer is used to process the predicted pedestrian flow and determine the demand for shared bicycles in the at least one target area.

4. The method according to claim 2, characterized in that, The shared bicycle operating area that adjusts the target time based on the real-time operation reference data includes: Based on the aforementioned real-time operational reference data, the priority for adjusting the operational areas is determined; Based on the aforementioned operational area adjustment priority, an operational area adjustment strategy is determined; Based on the aforementioned operating area adjustment strategy, the operating areas for shared bicycles at the target time are adjusted.

5. The method according to claim 4, characterized in that, The real-time operational reference data also includes the predicted pedestrian flow and the change in the number of shared bicycles; the adjustment of the shared bicycle operation area for the target time based on the real-time operational reference data includes: Real-time acquisition of the predicted pedestrian flow and the change in the number of shared bicycles; The predicted pedestrian flow and the change in the number of shared bicycles are weighted and summed to determine the weighted summation result; Based on the weighted summation result, the operational area adjustment strategy is determined; Based on the aforementioned operating area adjustment strategy, the operating areas for shared bicycles at the target time are adjusted.

6. The method according to claim 1, characterized in that, The smart city shared bicycle deployment and operation area planning IoT system also includes a user platform, a service platform, a sensor network platform, and an object platform; The service platform includes multiple service sub-platforms; different target regions correspond to different service sub-platforms. The management platform includes a central management platform database and multiple sub-management platforms; The sensor network platform includes multiple sensor network sub-platforms; different target areas correspond to different sensor network sub-platforms; different sensor network sub-platforms correspond to different management sub-platforms; The pedestrian flow reference information of the target area and the historical data of shared bicycles in the reference area are obtained based on the object platform and uploaded to the corresponding management sub-platform based on the sensor network sub-platform corresponding to the target area. The method for planning the deployment and operation areas of shared bicycles in smart cities includes: The number of shared bicycles deployed in the target area at the target time is transmitted to the service platform through the management platform database, and then uploaded to the user platform based on the service platform.

7. A smart city shared bicycle deployment and operation area planning Internet of Things system, characterized in that, This includes user platforms, service platforms, management platforms, sensor network platforms, and object platforms; The service platform includes multiple service sub-platforms, with different target areas corresponding to different service sub-platforms; The management platform includes a central management platform database and multiple management sub-platforms, wherein each of the multiple management sub-platforms corresponds to a different target region; The sensor network platform includes multiple sensor network sub-platforms, and each of the multiple sensor network sub-platforms corresponds to a different target area; The object platform is used to obtain pedestrian flow reference information and shared bicycle historical data of multiple reference areas in the target area, and transmit them to the corresponding management sub-platform based on the sensor network sub-platform corresponding to the target area; The management sub-platform is used to output the predicted pedestrian flow of the target area at a target time based on the pedestrian flow reference information and through a first prediction model, where the first prediction model is a machine learning model; determine the demand for shared bicycles based on the predicted pedestrian flow; and construct an operation map based on the target area, the multiple reference areas, roads connecting the target area and the multiple reference areas, roads connecting the multiple reference areas, and the historical data of shared bicycles; wherein the operation map uses the target area or the multiple reference areas as nodes, and the roads connecting the target area and the multiple reference areas or the roads connecting the multiple reference areas as edges, where the edges are directed edges, the node features of the nodes include the number of shared bicycles and pedestrian flow at a first historical time, and the edge features include shared bicycle transfer data and road environment data at the first historical time; and processes the operation map based on a second prediction model to determine the demand for shared bicycles in the target area at the target time. The shared bicycle change rate; the second prediction model is a machine learning model; wherein, the shared bicycle change rate is related to the change in pedestrian flow, and the change in pedestrian flow is determined based on historical pedestrian flow and the predicted pedestrian flow; the shared bicycle change rate refers to the change in the number of shared bicycles in the target area within the target time, used to reflect the riding popularity or riding demand in the target area, and the shared bicycle change rate includes increases and decreases; the second prediction model includes a first prediction layer and a second prediction layer; the first prediction layer is used to process the operation map to determine the first change rate of shared bicycles in the target area within the target time; the second prediction layer is used to process the first change rate of shared bicycles and the change in pedestrian flow to determine the second change rate of shared bicycles in the target area within the target time; the shared bicycle deployment quantity is determined based on the shared bicycle demand and the shared bicycle change rate; and the shared bicycle deployment quantity is transmitted to the service platform based on the management platform database; The service platform is used to transmit the number of shared bicycles deployed to the user platform.

8. A computer-readable storage medium storing computer instructions, wherein when a computer reads the computer instructions in the storage medium, the computer executes a smart city shared bicycle deployment and operation area planning method as described in claim 1.

Images