A management method for intelligent scheduling of vehicles on a freight platform based on the Internet of Things

Abstract

The present invention relates to the technical field of freight transportation scheduling management, and particularly to a management method for intelligent scheduling of vehicles on a freight platform based on the Internet of Things. The method includes the following steps: collecting on-vehicle terminal information and performing feature extraction to generate a comprehensive vehicle time-series feature set, drawing a freight route based on these feature sets to obtain basic freight route data, and using this data to reconstruct a real-time map to form a freight map network, performing uniform zoning, analyzing the intersection of regional vehicle trajectories to generate intersection trajectory data, predicting regional vehicle demand based on this, simulating on-demand scheduling to generate relevant data, and comparing the comprehensive vehicle time-series feature set with the simulated scheduling data to identify capacity differences, designing a scheduling plan based on these differences, and implementing intelligent scheduling, thereby realizing the optimized management of transportation capacity. The present invention realizes a more efficient and flexible management method for intelligent scheduling of vehicles on a freight platform.

Description

Technical Field

[0001] The present invention relates to the technical field of freight dispatch management, and particularly to a management method for intelligent dispatch of vehicles on a freight platform based on the Internet of Things. Background Art

[0002] Intelligent vehicle dispatch has become the key to improving freight efficiency and reducing operating costs. However, most current freight platforms still rely on traditional dispatch methods and lack the effective use of real-time data. Such methods not only fail to respond promptly to changes in freight demand but also lead to waste of resources and low dispatch efficiency, thus affecting the operation of the overall transportation chain. Especially during peak periods, the problem of unreasonable vehicle dispatch is particularly prominent, often resulting in delivery delays and a decline in customer satisfaction. In addition, existing dispatch systems often lack real-time monitoring and analysis of vehicle operating conditions, making it impossible to accurately predict vehicle demand. This information lag makes it difficult for freight platforms to achieve refined management and allocate resources according to actual needs. With the acceleration of urbanization, the volatility of freight demand has increased, and traditional static dispatch methods have become increasingly unsuitable. Summary of the Invention

[0003] Based on this, it is necessary to provide a management method for intelligent dispatch of vehicles on a freight platform based on the Internet of Things to solve at least one of the above technical problems.

[0004] To achieve the above object, a management method for intelligent dispatch of vehicles on a freight platform based on the Internet of Things includes the following steps:

[0005] Step S1: Collect on-vehicle terminal information; extract features from the on-vehicle terminal information to obtain on-vehicle feature information data; perform time series decomposition on the on-vehicle feature information data to generate a vehicle comprehensive time series feature set;

[0006] Step S2: Draw a basic freight route based on the vehicle comprehensive time series feature set to obtain basic freight route data; perform real-time map reconstruction on the vehicle comprehensive time series feature set according to the basic freight route data to generate a freight map network;

[0007] Step S3: Uniformly partition the freight map network to obtain a partitioned freight map network; perform regional vehicle trajectory intersection analysis on the vehicle comprehensive time series feature set based on the partitioned freight map network to generate intersection trajectory data;

[0008] Step S4: Predict vehicle demand for the partitioned freight map network according to the intersection trajectory data and the vehicle comprehensive time series feature set to obtain regional vehicle demand data; perform on-demand dispatch simulation on the vehicle comprehensive time series feature set based on the regional vehicle demand data to generate simulated on-demand dispatch data;

[0009] Step S5: Compare the vehicle comprehensive time - series feature set with the simulated on - demand scheduling data to obtain the transport capacity difference data; design a scheduling plan for the simulated on - demand scheduling data based on the transport capacity difference data to generate a vehicle scheduling plan;

[0010] Step S6: Intelligently schedule the vehicles on the freight platform according to the vehicle scheduling plan to implement the transport capacity optimization management method.

[0011] By collecting in - vehicle terminal information and extracting features, the present invention can obtain the real - time status and operation characteristics of vehicles. The vehicle comprehensive time - series feature set generated after time - series decomposition provides rich basic data for subsequent data analysis, supports multi - dimensional information mining and analysis, enhances the timeliness and accuracy of data. The combination of freight route drawing and basic freight route data makes the visualization of the transport path more intuitive. The freight map network generated by real - time map reconstruction provides spatial information support for optimized scheduling, can effectively reflect the current traffic conditions and road usage. The evenly partitioned freight map network provides a clear framework for the analysis of the intersection of regional vehicle trajectories. The generated intersection trajectory data can reveal the activity rules of different vehicles in the region, support the dynamic monitoring of vehicle flow, enhance the sensitivity to freight demand within the region. Vehicle demand prediction is based on intersection trajectory data and the vehicle comprehensive time - series feature set, can accurately identify the fluctuations in transport demand in different regions. The formed regional vehicle demand data helps the on - demand scheduling simulation, promotes the intelligence of scheduling strategies, optimizes resource allocation. The comparison of transport capacity resources can clearly identify the regions with insufficient and excessive transport capacity by analyzing the vehicle comprehensive time - series feature set and the simulated on - demand scheduling data. The generated transport capacity difference data provides a basis for scheduling plan design, ensuring the reasonable allocation and effective utilization of resources. Intelligent scheduling is implemented according to the vehicle scheduling plan, realizing the efficient management of vehicles on the freight platform, optimizing the transport capacity configuration, improving the overall scheduling efficiency, and enhancing the response ability and service quality of the freight platform.

[0012] Preferably, step S1 includes the following steps:

[0013] Step S11: Collect the vehicle position information in real - time to obtain the original position data; monitor and collect the vehicle operation status to obtain the original status data; dynamically collect the vehicle load condition to obtain the original load data;

[0014] Step S12: Integrate the original position data, original status data and original load data to obtain the in - vehicle terminal information; perform unified format conversion processing on the in - vehicle terminal information to obtain the standard format information;

[0015] Step S13: Extract features from the standard format information to obtain the in - vehicle feature information data; perform periodic analysis on the in - vehicle feature information data to obtain the periodic data;

[0016] Step S14: Statistically analyze the trend of change of the cycle data to generate a cycle change trend; perform time series recombination on the cycle data and the cycle change trend to obtain a vehicle comprehensive time series feature set.

[0017] Through real-time collection of vehicle position information, operating status, and load conditions, the original data obtained by the present invention provides rich basic information for subsequent analysis, ensuring the accuracy and timeliness of the data. The ability of dynamic monitoring improves the real-time control of the vehicle state, enhances the transparency of the transportation process, and the data integration and unified format conversion enable effective combination of data from different sources to form vehicle terminal information. The generation of standard format information improves the standardization and consistency of data processing, simplifies the complexity of subsequent data analysis, ensures the availability of data in each link, feature extraction and periodic analysis provide an important basis for in-depth understanding of vehicle operation characteristics, and the generated vehicle feature information data and cycle data can reflect the performance of the vehicle under different conditions, supporting the precise evaluation and optimization of vehicle performance. The trend change statistics and time series recombination can reveal the laws and potential problems of vehicle operation, and the generated cycle change trend provides a scientific basis for dynamic scheduling decisions, promoting the comprehensive monitoring of the vehicle operation state, realizing the sensitive capture of the change in transport capacity and the improvement of the prediction ability.

[0018] Preferably, step S2 includes the following steps:

[0019] Step S21: Extract path points from the vehicle comprehensive time series feature set to obtain path point data; statistically analyze the frequency of the path point data to generate path frequency data;

[0020] Step S22: Calculate the path density of the path point data based on the path frequency data to generate path density data; draw a freight route based on the path density data for the path point data to obtain basic freight route data;

[0021] Step S23: Perform grid processing on the basic freight route data to obtain a road network grid; identify road network nodes for the road network grid to generate road network node data;

[0022] Step S24: Analyze the connectivity of the road network node data to obtain connectivity data; calculate the road network weight for the connectivity data based on a preset road network weight assignment to generate road network weight data;

[0023] Step S25: Construct a topological relationship for the basic freight route data based on the road network weight data to obtain topological structure data; perform network integration on the topological structure data to generate a freight map network.

[0024] By extracting path points from the vehicle comprehensive timing feature set, the generated path point data lays a foundation for subsequent path analysis. Frequency statistics provide a quantitative analysis of the usage of each path point. The path frequency data reflects the usage frequency of different paths. Based on the path frequency data, path density calculation is performed, and the generated path density data reveals the transportation intensity of different regions. The basic freight route data obtained from freight route drawing provides an intuitive basis for effective transportation planning. Grid processing of the basic freight route data generates a road network grid, which provides a structured basis for subsequent analysis. Road network node identification can clarify the key nodes in the transportation network, and connectivity analysis can reveal the connection situation between different nodes in the road network. The generated connectivity data provides a basis for the overall performance evaluation of the transportation network. Road network weight calculation is based on preset weight assignment. The road network weight data is combined with the basic freight route data, and the constructed topological relationship lays a foundation for the structural optimization of the transportation network. The generation of topological structure data provides a clear framework for the integration of the freight map network, which helps to achieve efficient transportation management and scheduling decisions.

[0025] Preferably, step S3 includes the following steps:

[0026] Step S31: Identify the boundary of the freight map network to obtain the map network boundary; calculate the area of the freight map network according to the map network boundary to generate the global map network area;

[0027] Step S32: Based on the global map network area, perform a four-equal division area cut on the freight map network to obtain a partitioned freight map network;

[0028] Step S33: Map the vehicle comprehensive timing feature set to each region according to the partitioned freight map network to obtain a partitioned vehicle timing feature set; extract vehicle trajectories from the partitioned vehicle timing feature set to generate partitioned vehicle trajectories belonging to the regions;

[0029] Step S34: Identify the intersection points of the partitioned vehicle trajectories belonging to the regions to obtain vehicle intersection point data; perform an intersection trajectory statistics on the partitioned vehicle trajectories belonging to the regions based on the vehicle intersection point data to generate intersection trajectory data.

[0030] Through the boundary recognition of the freight map network, the generated map network boundary provides a clear boundary for subsequent regional analysis. The calculation of the global map network area lays a foundation for understanding the scale and characteristics of the entire transportation network. Based on the global map network area, the four-equal-area division is carried out to obtain the partitioned freight map network, which provides a clear structure for the optimized management of different transportation regions, ensuring the balance and coordination among regions. Using the partitioned freight map network to map the vehicle comprehensive time series feature set to each region, the generated partitioned vehicle time series feature set can accurately reflect the running states of vehicles in each region. By identifying the intersection points of the vehicle trajectories belonging to the partitioned regions, the obtained vehicle intersection data reveals the interaction situations of different vehicles within the same region. The statistical analysis of the intersecting trajectories based on the vehicle intersection data provides a quantitative basis for analyzing the vehicle flow patterns, promoting the insight and improvement of the transportation efficiency within the region.

[0031] Preferably, step S4 includes the following steps:

[0032] Step S41: Perform the aggregation calculation of the attribution regions for the intersecting trajectory data to obtain the regional density data; divide the time periods of the vehicle comprehensive time series feature set to generate the time-divided vehicle feature data;

[0033] Step S42: Perform the feature correlation analysis on the regional density data and the time-divided vehicle feature data to obtain the trajectory density feature data; predict the vehicle demands for the partitioned freight map network according to the trajectory density feature data to obtain the regional vehicle demand data;

[0034] Step S43: Perform the resource allocation calculation on the regional vehicle demand data to obtain the on-demand resource allocation plan;

[0035] Step S44: Based on the on-demand resource allocation plan, perform the on-demand scheduling simulation on the vehicle comprehensive time series feature set to generate the simulated on-demand scheduling data.

[0036] Through the attribution area aggregation calculation of the intersecting trajectory data, the generated area density data effectively reflects the concentration degree of vehicle activities in each area. The time-sharing vehicle feature data generated by time period division provides a detailed perspective on the vehicle operation conditions within different time periods. Through the feature correlation analysis of the area density data and the time-sharing vehicle feature data, the obtained trajectory density feature data reveals the vehicle flow trend within different time periods and areas, providing a quantitative basis for the dynamic changes of transportation demands. Based on the trajectory density feature data, vehicle demand prediction is carried out for the zoned freight map network, resource allocation calculation is performed on the regional vehicle demand data, and the generated on-demand resource allocation plan provides scientific support for ensuring the efficient utilization of resources, enabling the reasonable allocation of vehicle resources according to actual demands, optimizing transportation efficiency and service quality. Based on the on-demand resource allocation plan, on-demand scheduling simulation is carried out for the vehicle comprehensive time series feature set, and the generated simulated on-demand scheduling data provides a reference for actual scheduling decisions, enabling the dynamic adjustment and optimization of vehicle operations, and enhancing the flexibility and response speed of the overall logistics operation.

[0037] Preferably, step S42 includes the following steps:

[0038] Step S421: Conduct correlation analysis on the area density data and the time-sharing vehicle feature data to obtain correlation data; perform feature combination processing on the correlation data to generate combined feature data;

[0039] Step S422: Based on the correlation data, conduct correlation importance ranking on the combined feature data to obtain ranked feature data; perform screening and integration on the ranked feature data to generate trajectory density feature data;

[0040] Step S423: Deduce the regional freight supply volume for the trajectory density feature data to obtain predicted regional freight supply data; based on the predicted regional freight supply data, conduct supply digestion calculation on the zoned freight map network to generate digested supply and demand data;

[0041] Step S424: According to the digested supply and demand data, conduct vehicle demand statistics on the zoned freight map network to obtain regional vehicle demand data.

[0042] Through the correlation analysis of the regional density data and the time-sharing vehicle characteristic data, the generated correlation data reveals the relationship between vehicle demand and regional characteristics. The combined feature data obtained after the feature combination processing provides a multi-dimensional information perspective for subsequent analysis. Based on the correlation data, the combined feature data is sorted according to the correlation importance, and the obtained sorted feature data provides a basis for identifying key factors. The trajectory density feature data after screening and integration further optimizes the feature set, ensuring the pertinence and effectiveness of subsequent analysis. The regional freight supply volume is deduced from the trajectory density feature data, and the generated predicted regional freight supply data provides an important basis for the supply-demand balance. Based on these data, the supply and digestion calculation of the partitioned freight map network is carried out, and the generated digestion supply-demand data provides a quantitative reference for resource allocation and scheduling decisions. According to the digestion supply-demand data, the vehicle demand in the partitioned freight map network is counted, and the obtained regional vehicle demand data provides a real-time basis for dynamic scheduling, which can timely reflect the changes in freight demand, support the optimization and adjustment of the scheduling plan, and improve the flexibility and response ability of logistics management.

[0043] Preferably, step S423 includes the following steps:

[0044] Perform regional division calculation on the trajectory density feature data to obtain regional distribution data;

[0045] Based on the trajectory density feature data, analyze the changing trend of the trajectory density of the regional distribution data to generate the changing trend of the trajectory density;

[0046] Map the supply volume to the changing trend of the trajectory density to obtain the changing trend of the supply volume;

[0047] Based on the trajectory density feature data, calculate the supply volume from the changing trend of the supply volume to generate the predicted regional freight supply data;

[0048] Based on the predicted regional freight supply data, perform road network mapping and matching on the partitioned freight map network to obtain road network matching data;

[0049] Perform freight supply-demand balance calculation on the road network matching data to obtain freight supply-demand balance data;

[0050] Based on the freight supply-demand balance data, predict the supply digestion demand of the predicted regional freight supply data to generate digestion supply-demand data.

[0051] Through the regional division calculation of the trajectory density feature data, the generated regional distribution data provides a basis for analyzing the transportation demands of different regions, can clearly show the vehicle activities in each region. Based on the trajectory density feature data, the trend analysis of the trajectory density is carried out on the regional distribution data, and the generated trajectory density change trend reveals the dynamic changes of vehicle activities, providing important information for understanding the evolution of transportation modes. The supply quantity mapping is carried out on the trajectory density change trend, and the obtained supply quantity change trend provides an intuitive basis for predicting the future freight supply level. Based on the predicted regional freight supply data, the road network mapping matching is carried out on the zonal freight map network, and the generated road network matching data provides support for the optimization and adjustment of the actual transportation routes. The freight supply-demand balance calculation is carried out on the road network matching data, and the obtained freight supply-demand balance data provides a quantitative basis for subsequent scheduling decisions. Based on the freight supply-demand balance data, the supply digestion demand prediction is carried out on the predicted regional freight supply data, and the generated digestion supply-demand data provides an important reference for dynamic scheduling and real-time decision-making, improving the flexibility and response ability of freight management.

[0052] Preferably, step S5 includes the following steps:

[0053] Step S51: Calculate the transport capacity of the vehicle comprehensive time series feature set to obtain the existing transport capacity data; calculate the scheduling transport capacity of the simulated on-demand scheduling data to obtain the scheduling transport capacity data;

[0054] Step S52: Perform spatio-temporal distribution mapping on the existing transport capacity data to obtain the existing transport capacity distribution data; perform regional scheduling demand analysis on the scheduling transport capacity data to obtain the regional demand data;

[0055] Step S53: Calculate the matching degree between the existing transport capacity distribution data and the regional demand data to obtain the matching degree data; perform a comparison of the transport capacity differences on the vehicle comprehensive time series feature set and the simulated on-demand scheduling data based on the matching degree data to obtain the transport capacity difference data;

[0056] Step S54: Plan the scheduling routes of the simulated on-demand scheduling data according to the transport capacity difference data to obtain the route planning data; perform conflict detection on the route planning data to obtain the scheduling conflict data;

[0057] Step S55: Adjust the route planning data based on the scheduling conflict data to generate an adjusted scheduling route; formulate a scheduling plan for the adjusted scheduling route to generate a vehicle scheduling plan.

[0058] Through the calculation of the transportation capacity of the vehicle integrated timing feature set, the generated existing transportation capacity data provides a basis for evaluating the current transportation capacity. The scheduling transportation capacity calculation of the simulated on-demand scheduling data reveals the actual transportation capacity demand during the scheduling process. The spatio-temporal distribution mapping of the existing transportation capacity data generates the existing transportation capacity distribution data, which can clearly display the transportation capacity distribution in different regions. The regional scheduling demand analysis of the scheduling transportation capacity data provides a basis for identifying key scheduling regions. The matching degree calculation of the existing transportation capacity distribution data and the regional demand data generates the matching degree data, which provides a quantitative basis for evaluating the rationality of resource allocation. Based on the matching degree data, the vehicle integrated timing feature set and the simulated on-demand scheduling data are compared for transportation capacity differences. According to the transportation capacity difference data, the scheduling route of the simulated on-demand scheduling data is planned. The obtained route planning data provides a scientific basis for optimizing the vehicle driving path. Conflict detection can identify potential scheduling problems. Based on the scheduling conflict data, the route planning data is adjusted. The generated adjusted scheduling route provides a guarantee for ensuring the smoothness and efficiency of the transportation process. The formulation of the scheduling plan combines the adjusted scheduling route, and the finally formed vehicle scheduling plan provides a systematic solution for realizing transportation capacity optimization and efficient management.

[0059] Preferably, step S53 includes the following steps:

[0060] Perform data alignment processing on the existing transportation capacity distribution data and the regional demand data to obtain aligned basic data;

[0061] Perform regional matching calculation on the aligned basic data to obtain regional matching data;

[0062] Perform time matching calculation on the aligned basic data based on the regional matching data to obtain time matching data;

[0063] Perform capacity matching calculation on the aligned basic data based on the time matching data to generate capacity matching data;

[0064] Perform matching degree calculation on the aligned basic data according to the capacity matching data to generate matching degree data;

[0065] Perform threshold slicing on the matching degree data to obtain threshold classification data;

[0066] Perform transportation capacity difference identification on the vehicle integrated timing feature set and the simulated on-demand scheduling data based on the threshold classification data to obtain difference degree data;

[0067] Perform data integration on the difference degree data to generate transportation capacity difference data.

[0068] Through data alignment processing of existing transport capacity distribution data and regional demand data, the generated aligned basic data provides a unified reference framework for subsequent analysis, ensuring the consistency and comparability between different data sources. By performing regional matching calculations on the aligned basic data, the obtained regional matching data provides a basis for identifying the effectiveness of resource allocation within each region, supporting the dynamic analysis of transport capacity and demand within a specific region. Based on the regional matching data, time matching calculations are carried out to generate time matching data, which provides support for analyzing the time characteristics of transport capacity and demand, ensuring the timely capture of fluctuations in transport demand. Based on the time matching data, capacity matching calculations are performed on the aligned basic data to generate capacity matching data, which provides a basis for evaluating the utilization efficiency of transport capacity in each region and can identify regions with insufficient or excessive transport capacity. According to the capacity matching data, matching degree calculations are performed on the aligned basic data to generate matching degree data, which provides a quantitative basis for evaluating the matching degree between transport capacity and demand, supporting the optimization and adjustment of scheduling strategies. By performing threshold slicing on the matching degree data, the obtained threshold classification data provides a clear standard for subsequent identification of transport capacity differences. Based on the threshold classification data, transport capacity differences are identified for the vehicle comprehensive time series feature set and simulated on-demand scheduling data, and data integration is performed on the difference degree data to generate transport capacity difference data, which provides a basis for comprehensively evaluating the relationship between the current transport capacity and demand, promoting the accuracy and scientific nature of scheduling decisions, and improving the efficiency and flexibility of overall logistics management.

[0069] Preferably, step S55 includes the following steps:

[0070] Identify conflict sections from the scheduling conflict data to obtain conflict section data; retrieve alternative routes for the route planning data based on the conflict section data to generate alternative route data;

[0071] Allocate time windows to the alternative route data to obtain time allocation data; reorganize the route planning data based on the time allocation data to generate an adjusted scheduling route;

[0072] Make a vehicle allocation plan for the adjusted scheduling route to obtain vehicle allocation data;

[0073] Formulate a vehicle scheduling plan based on the time allocation data for the vehicle allocation data to generate a vehicle scheduling plan.

[0074] By identifying conflict sections from scheduling conflict data, the generated conflict section data provides a basis for identifying potential transportation obstacles, enables the specific areas affecting vehicle travel to be clarified, supports subsequent route optimization and adjustment, retrieves alternative routes based on the conflict section data for the route planning data, and the generated alternative route data provides diverse options for resolving conflicts, ensuring the flexibility to respond to emergencies, reducing the risk of transportation delays, allocating time windows for the alternative route data, and the obtained time allocation data provides a scientific basis for the travel time of each alternative route, ensuring the rationality and feasibility of the transportation plan, promoting the improvement of the overall scheduling efficiency, reorganizing the route planning data based on the time allocation data, and the generated adjusted scheduling route provides clear guidance for optimizing the vehicle travel path, can effectively reduce travel time and costs, improve transportation efficiency, making a vehicle allocation plan for the adjusted scheduling route, and the obtained vehicle allocation data provides a basis for ensuring the rational use of resources, can reasonably allocate vehicle resources according to actual needs, improving the flexibility and effectiveness of services, formulating a vehicle scheduling plan based on the time allocation data for the vehicle allocation data, and the generated vehicle scheduling plan provides a systematic solution for achieving transportation capacity optimization and efficient management, enhancing the adaptability and response speed of the logistics platform in a dynamic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a schematic diagram of the step - by - step process of a management method for intelligent vehicle scheduling of a freight platform based on the Internet of Things;

[0076] Figure 2 is Figure 1 a detailed implementation step - by - step schematic diagram of step S2 in

[0077] Figure 3 is Figure 1 a detailed implementation step - by - step schematic diagram of step S3 in

[0078] The realization, functional features, and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0079] The technical method of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0080] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0081] It should be understood that although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0082] To achieve the above object, please refer to Figures 1 to 3 , a management method for intelligent scheduling of freight platform vehicles based on the Internet of Things, comprising the following steps:

[0083] Step S1: Collect on-vehicle terminal information; extract features from the on-vehicle terminal information to obtain on-vehicle feature information data; perform time series decomposition on the on-vehicle feature information data to generate a vehicle comprehensive time series feature set;

[0084] Step S2: Draw a freight route for the vehicle comprehensive time series feature set to obtain basic freight route data; perform real-time map reconstruction on the vehicle comprehensive time series feature set according to the basic freight route data to generate a freight map network;

[0085] Step S3: Uniformly partition the freight map network to obtain a partitioned freight map network; perform regional vehicle trajectory intersection analysis on the vehicle comprehensive time series feature set based on the partitioned freight map network to generate intersection trajectory data;

[0086] Step S4: Perform vehicle demand prediction on the partitioned freight map network according to the intersection trajectory data and the vehicle comprehensive time series feature set to obtain regional vehicle demand data; perform on-demand scheduling simulation on the vehicle comprehensive time series feature set based on the regional vehicle demand data to generate simulated on-demand scheduling data;

[0087] Step S5: Compare the vehicle comprehensive time series feature set and the simulated on-demand scheduling data to obtain capacity difference data; design a scheduling plan for the simulated on-demand scheduling data based on the capacity difference data to generate a vehicle scheduling plan;

[0088] Step S6: Intelligently dispatch the vehicles on the freight platform according to the vehicle dispatching plan to implement the capacity optimization management method.

[0089] By collecting on-vehicle terminal information and performing feature extraction, the present invention can obtain the real-time status and operation characteristics of vehicles. The vehicle comprehensive time series feature set generated after time series decomposition provides rich basic data for subsequent data analysis, supports multi-dimensional information mining and analysis, enhances the timeliness and accuracy of data. The freight route drawing combines with the basic freight route data, making the visualization of the transportation route more intuitive. The freight map network generated by real-time map reconstruction provides spatial information support for optimized dispatching, can effectively reflect the current traffic conditions and road usage. The evenly partitioned freight map network provides a clear framework for the analysis of the intersection of regional vehicle trajectories. The generated intersection trajectory data can reveal the activity rules of different vehicles in the region, support the dynamic monitoring of vehicle flow, enhance the sensitivity to freight demand in the region. The vehicle demand prediction is based on the intersection trajectory data and the vehicle comprehensive time series feature set, can accurately identify the fluctuations in transportation demand in different regions. The formed regional vehicle demand data helps the on-demand dispatching simulation, promotes the intelligence of dispatching strategies, optimizes resource allocation. The comparison of transport capacity resources can clearly identify the regions with insufficient and excessive transport capacity by analyzing the vehicle comprehensive time series feature set and the simulated on-demand dispatching data. The generated transport capacity difference data provides a basis for the design of the dispatching plan, ensuring the reasonable allocation and effective utilization of resources. The intelligent dispatching is implemented according to the vehicle dispatching plan, realizing the efficient management of the vehicles on the freight platform, optimizing the transport capacity configuration, improving the overall dispatching efficiency, and enhancing the response ability and service quality of the freight platform.

[0090] In the embodiment of the present invention, refer to Figure 1 , which is the step flow schematic diagram of a management method for intelligent dispatching of vehicles on a freight platform based on the Internet of Things. In this example, the management method for intelligent dispatching of vehicles on a freight platform based on the Internet of Things includes the following steps:

[0091] Step S1: Collect on-vehicle terminal information; perform feature extraction on the on-vehicle terminal information to obtain on-vehicle feature information data; perform time series decomposition on the on-vehicle feature information data to generate a vehicle comprehensive time series feature set;

[0092] In this embodiment, a high-precision GPS module and multiple sensors are installed in the in-vehicle terminal to collect real-time data such as the vehicle's position information, speed, acceleration, and fuel consumption. The data collection frequency is set to once per second. The data is transmitted to the central processing system through a wireless network (such as 4G / 5G). The central processing system receives and stores the original data, and uses data preprocessing techniques for data cleaning and denoising to ensure the accuracy of the information. Then, feature extraction is performed on the cleaned data, and key feature information is extracted from it using machine learning algorithms (such as decision trees and random forests). Finally, the extracted feature information is decomposed by time series to generate a vehicle comprehensive time series feature set, which contains the dynamic state data of the vehicle at different time periods.

[0093] Step S2: Draw a freight route for the vehicle comprehensive time series feature set to obtain basic freight route data; perform real-time map reconstruction on the vehicle comprehensive time series feature set according to the basic freight route data to generate a freight map network;

[0094] In this embodiment, the location information in the vehicle comprehensive time series feature set is used to draw a freight route using GIS (Geographic Information System) software. First, each location point passed by the vehicle is connected to form a path line. Then, the length and travel time of each line are calculated based on the path line to generate basic freight route data. Subsequently, based on these basic data, the driving state of the vehicle is monitored in real time using a real-time map API (Application Programming Interface), and the real-time map is reconstructed to generate a freight map network, ensuring that the map network reflects the latest traffic conditions and route information for subsequent scheduling decisions.

[0095] Step S3: Uniformly partition the freight map network to obtain a partitioned freight map network; perform regional vehicle trajectory intersection analysis on the vehicle comprehensive time series feature set based on the partitioned freight map network to generate intersection trajectory data;

[0096] In this embodiment, the generated freight map network is uniformly partitioned using a spatial segmentation algorithm (such as Voronoi diagram or K-means clustering) to divide the map area into several uniform partitions to ensure the balance of transportation demand and resource allocation in each partition. Subsequently, based on the partitioned map network, regional vehicle trajectory intersection analysis is performed using the vehicle comprehensive time series feature set, and a time series data matching algorithm is used to identify the intersections of different vehicle trajectories to generate intersection trajectory data, ensuring comprehensive monitoring and recording of the activities of vehicles in each partition.

[0097] Step S4: Predict the vehicle demand for the partitioned freight map network based on the intersection trajectory data and the vehicle comprehensive time series feature set to obtain regional vehicle demand data; perform on-demand scheduling simulation on the vehicle comprehensive time series feature set based on the regional vehicle demand data to generate simulated on-demand scheduling data;

[0098] In this embodiment, according to the intersection trajectory data and the vehicle comprehensive time-series feature set, the time-series analysis method (such as the ARIMA model) is used to predict the vehicle demand for the partitioned freight map network. First, the historical data is analyzed for seasonality and trend to generate regional vehicle demand data. Subsequently, based on the regional vehicle demand data, a simulation tool (such as AnyLogic) is used to conduct on-demand scheduling simulation to generate simulated on-demand scheduling data, ensuring that vehicle scheduling can adapt to the demand changes in different time periods and regions, and improving the flexibility and accuracy of scheduling.

[0099] Step S5: Compare the vehicle comprehensive time-series feature set and the simulated on-demand scheduling data for transport capacity resources to obtain transport capacity difference data; design a scheduling plan for the simulated on-demand scheduling data based on the transport capacity difference data to generate a vehicle scheduling plan;

[0100] In this embodiment, when comparing the vehicle comprehensive time-series feature set and the simulated on-demand scheduling data for transport capacity resources, first, a transport capacity resource database is established, including information such as the number, type, and load of vehicles. Then, a comparison algorithm (such as the difference analysis method) is used to calculate the difference between the existing transport capacity and the demand to generate transport capacity difference data. Based on the transport capacity difference data, an optimization algorithm (such as the genetic algorithm) is used to design a scheduling plan to ensure that the generated vehicle scheduling plan can effectively match the current transport capacity demand and resource allocation, and improve the overall transport efficiency.

[0101] Step S6: Intelligently schedule the vehicles on the freight platform according to the vehicle scheduling plan to implement the transport capacity optimization management method.

[0102] In this embodiment, according to the generated vehicle scheduling plan, the scheduling management system is used to intelligently schedule the vehicles on the freight platform. First, the scheduling plan is input into the scheduling system, and the system automatically assigns vehicles to specific transportation tasks through algorithms (such as the shortest path algorithm) to ensure that the vehicles can complete freight tasks efficiently and in a timely manner. Then, the vehicle status during the transportation process is monitored in real time, and the Internet of Things technology (IoT) is used for data feedback and adjustment to ensure that the scheduling plan can be dynamically optimized according to real-time situations, ultimately achieving transport capacity optimization management and improving the quality and efficiency of transportation services.

[0103] Preferably, step S1 includes the following steps:

[0104] Step S11: Collect the vehicle position information in real time to obtain the original position data; monitor and collect the vehicle running status to obtain the original status data; dynamically collect the vehicle load condition to obtain the original load data;

[0105] Step S12: Integrate the position raw data, status raw data, and load raw data to obtain in-vehicle terminal information; perform unified format conversion processing on the in-vehicle terminal information to obtain standard format information;

[0106] Step S13: Extract features from the standard format information to obtain in-vehicle feature information data; perform periodic analysis on the in-vehicle feature information data to obtain periodic data;

[0107] Step S14: Perform change trend statistics on the periodic data to generate a periodic change trend; perform time series recombination on the periodic data and the periodic change trend to obtain a vehicle comprehensive time series feature set.

[0108] In this embodiment, the real-time position of the vehicle is collected through the in-vehicle GPS module (Global Positioning System), and the geographical coordinate points of each vehicle are uploaded to the Internet of Things platform at a predetermined frequency to generate raw position data. At the same time, the OBD (On-Board Diagnostics) is used to detect the status information of the vehicle engine, fuel level, battery power, etc., and the instant status data of each parameter is uploaded to generate raw status data. In addition, the in-vehicle sensor is combined to measure the current load of the vehicle, the pressure sensor measures the weight of the goods and uploads the data to form raw load data. During this process, the data acquisition module in the in-vehicle acquisition system continuously obtains and aggregates all monitoring information, and uses the cellular network to upload data in real time to ensure the timeliness of the data. The central processing system of the Internet of Things platform integrates the received vehicle position raw data, status raw data, and load raw data, and distributes them to their corresponding vehicle information databases at the platform end according to the vehicle number through the data synchronization system, completing the aggregation of in-vehicle terminal information. This step uses the data integration module to perform multi-source aggregation of all information according to the tag structures of position, status, and load. At the same time, the data cleaning module standardizes the fields and data types of each type of data, and converts the data format into the standard format information of the system. For example, the position data is unified into the form of longitude and latitude, the status data is unified into integer or boolean type, and the load data is converted into kilograms (kg) unit. The Internet of Things platform uses the feature extraction algorithm to analyze the standard format information, extracts the main feature information of the vehicle from the data. The position feature extraction calculates the average driving speed of the vehicle through the relative change rate of the position coordinates. The status feature extraction comprehensively calculates the fuel economy of the vehicle using the engine power and fuel consumption information provided by the OBD. The load feature extraction calculates the stability of the load weight through the real-time load change. The feature extraction system aggregates the extracted information into in-vehicle feature information data, and performs periodic analysis based on this information. The periodic analysis module of the platform statistically analyzes information such as the vehicle driving distance, status change frequency, and load fluctuation amplitude at a fixed period (such as every 24 hours) to obtain periodic data. The periodic data consists of various statistical indicators of the vehicle, so as to be provided for the dispatching system to use. Based on the periodic data for statistical analysis, the change trends of the vehicle in multiple periods are identified, and moving average and regression analysis are respectively performed on the periodic data such as driving distance, fuel consumption, and load to obtain the periodic change trends. The generated periodic change trends are used to identify the long-term stability and reliability of the vehicle operation. Subsequently, the time series reorganization module performs time series arrangement and reorganization on the periodic data and the periodic change trends, and reorganizes the time points of the periodic data into a continuous time series based on the time axis of each vehicle, so as to obtain the comprehensive time series feature set of the vehicle.

[0109] Preferably, step S2 includes the following steps:

[0110] Step S21: Extract path points from the vehicle comprehensive timing feature set to obtain path point data; perform frequency statistics on the path point data to generate path frequency data;

[0111] Step S22: Calculate path density for the path point data based on the path frequency data to generate path density data; draw freight routes for the path point data according to the path density data to obtain basic freight route data;

[0112] Step S23: Perform grid processing on the basic freight route data to obtain a road network grid; identify road network nodes for the road network grid to generate road network node data;

[0113] Step S24: Analyze the connectivity of the road network node data to obtain connectivity data; calculate road network weights for the connectivity data based on a preset road network weight assignment to generate road network weight data;

[0114] Step S25: Construct a topological relationship for the basic freight route data according to the road network weight data to obtain topological structure data; perform network integration on the topological structure data to generate a freight map network.

[0115] As an example of the present invention, refer to Figure 2 , in this example, step S2 includes:

[0116] Step S21: Extract path points from the vehicle comprehensive timing feature set to obtain path point data; perform frequency statistics on the path point data to generate path frequency data;

[0117] In this embodiment, based on the vehicle comprehensive timing feature set, the system uses a path recognition algorithm to extract the main path points formed by the vehicle during driving one by one, marks the geographical coordinates of each path point through the positioning module, and obtains a path point data set containing information such as longitude and latitude; after the path point extraction is completed, the frequency statistics module of the platform analyzes the path point data, calculates the number of times each path point is passed by identifying the appearance frequency of the vehicle at different path points, and generates path frequency data. The frequency data consists of the coordinates of the path point and the corresponding frequency value, providing data support for subsequent path optimization and scheduling planning.

[0118] Step S22: Calculate path density for the path point data based on the path frequency data to generate path density data; draw freight routes for the path point data according to the path density data to obtain basic freight route data;

[0119] In this embodiment, the platform uses a path density calculation algorithm to calculate the density degree within a region for the frequency information in the path point data. By dividing the equally spaced grid regions, the path points with high frequencies are taken as the dense regions to generate path density data. The path density data contains the density values of each grid and is used to determine the high-frequency routes of freight vehicles. Subsequently, based on the path density data, the platform draws the basic freight routes, and marks the most commonly used freight routes by connecting the path points with high density. The basic freight route data is a complete linear structure, including information such as the sorting of path points, path directions, and path densities, providing a stable freight route network structure for subsequent route optimization and path planning.

[0120] Step S23: Perform grid processing on the basic freight route data to obtain a road network grid; identify road network nodes for the road network grid to generate road network node data.

[0121] In this embodiment, the basic freight route data is segmented by a grid algorithm to decompose the coverage area of the freight route into interconnected small grid units, generating road network grid data. This data structures the spatial distribution of the entire freight route, enabling each grid to represent the corresponding geographical area and contain the route point data within the grid. After the grid processing, the intersection points or key sections between grids are further identified by a road network node identification module, and these intersection points are marked as road network nodes. Each road network node data contains the coordinates of the node, the numbers of adjacent grids, and the route information passing through, providing basic data for subsequent road network analysis.

[0122] Step S24: Perform connectivity analysis on the road network node data to obtain connectivity data; calculate the road network weights for the connectivity data based on a preset road network weight assignment to generate road network weight data.

[0123] In this embodiment, a connectivity analysis algorithm is applied to the road network node data to analyze the connection relationships between each road network node to determine the accessibility of paths between different grids. The information such as the starting point, ending point, and path length of the accessible paths is converted into connectivity data. After the connectivity analysis is completed, a weight calculation module assigns weight values to the connectivity data. Combining with a preset road network weight model, considering factors such as road length, traffic flow, and path stability, different weights are assigned to each path to generate road network weight data. Each path weight data contains information such as the connecting nodes of the path and the path weight value, enabling the path weight data to guide vehicles to give priority to high-weight paths in path selection.

[0124] Step S25: Construct a topological relationship for the basic freight route data according to the road network weight data to obtain topological structure data; perform network integration on the topological structure data to generate a freight map network.

[0125] In this embodiment, a topological relationship construction algorithm is used to comprehensively process the basic freight route data and road network weight data, and a topological structure data is constructed based on the connection relationship between nodes and paths. This topological structure data describes the overall network relationship of the freight route through an ordered combination of nodes, paths, and weights, enabling the topological information of each node to form a tight structured connection with its neighboring nodes and paths. After the topological structure data is completed, the network integration module of the system integrates this data and merges the topological structure data existing in all freight route networks into a complete freight map network. The freight map network is a multi-level structure, including the mutual relationship between nodes, paths, and weights, and provides complete freight route structure information.

[0126] Preferably, step S3 includes the following steps:

[0127] Step S31: Identify the boundary of the freight map network to obtain the map network boundary; calculate the regional area of the freight map network based on the map network boundary to generate the global map network area;

[0128] Step S32: Quadrisect the regional cutting of the freight map network based on the global map network area to obtain a partitioned freight map network;

[0129] Step S33: Map the vehicle comprehensive time series feature set to each region according to the partitioned freight map network to obtain a partitioned vehicle time series feature set; extract vehicle trajectories from the partitioned vehicle time series feature set to generate attributed partition vehicle trajectories;

[0130] Step S34: Identify the intersection points of the attributed partition vehicle trajectories to obtain vehicle intersection point data; perform intersection trajectory statistics on the attributed partition vehicle trajectories based on the vehicle intersection point data to generate intersection trajectory data.

[0131] As an example of the present invention, refer to Figure 3 , in this example, step S3 includes:

[0132] Step S31: Identify the boundary of the freight map network to obtain the map network boundary; calculate the regional area of the freight map network based on the map network boundary to generate the global map network area;

[0133] In this embodiment, a boundary recognition algorithm is used to scan and recognize the boundary area of the freight map network. By positioning the outer boundary of the nodes and path ranges of the map network, a complete boundary line is calculated point by point to obtain the boundary data of the map network. After the boundary recognition is completed, the system uses a regional calculation tool based on the boundary data and uses the polygon area formula to calculate the sum of the areas of each segment, and summarizes to obtain the global map network area. In the calculation of the regional area, the system calculates the area product for each boundary path and adjacent node to ensure the data accuracy of the global map network area. The global map network area data includes the area value and boundary coordinate information.

[0134] Step S32: Based on the global map network area, the freight map network is divided into four equal-area regions to obtain a partitioned freight map network.

[0135] In this embodiment, through a map network segmentation tool, the freight map network is divided into four equal parts based on the global map network area. The map network is divided into four equal-area regions. To achieve the four-equal effect, the system uses a regional segmentation algorithm with the center point of the global map network as the starting point, and sequentially divides the global map into quadrants. The boundary paths and nodes within each quadrant are assigned to the corresponding regions to obtain a partitioned freight map network. The partitioned freight map network contains the boundary, nodes, and path information of each partition, and the partition structure ensures a reasonable layout of the freight routes.

[0136] Step S33: Based on the partitioned freight map network, each area mapping is performed on the vehicle comprehensive time series feature set to obtain a partitioned vehicle time series feature set; vehicle trajectory extraction is performed on the partitioned vehicle time series feature set to generate a partition-owned vehicle trajectory.

[0137] In this embodiment, based on the data of the partitioned freight map network, the corresponding geographical location information in the vehicle comprehensive time series feature set is subjected to partition mapping processing. For the running trajectory of each vehicle, the coincidence between its position information and the partition boundary is detected in turn. The time series features of the vehicle are mapped to the corresponding partition through the geographical coordinate matching method to obtain a partitioned vehicle time series feature set. The partitioned vehicle time series feature set contains the driving data, stop points, and time node information of the vehicle in each partition; after the mapping is completed, the system uses a trajectory extraction tool to extract the trajectories of the partitioned vehicle time series feature set, and connects all the trajectories belonging to the same region in the time series feature set to generate a partition-owned vehicle trajectory. The partition-owned vehicle trajectory data contains content such as the path node sequence and driving time sequence within the partition.

[0138] Step S34: Identify the intersection points of the partition-owned vehicle trajectories to obtain vehicle intersection point data; perform statistics on the intersecting trajectories of the partition-owned vehicle trajectories based on the vehicle intersection point data to generate intersecting trajectory data.

[0139] In this embodiment, the intersection point recognition algorithm is used to analyze each point of the vehicle trajectory in the attributed partition to detect the intersection situation between trajectories. The node coordinates of each vehicle trajectory in the attributed partition are compared one by one to identify the trajectory intersection points and generate vehicle intersection point data. The vehicle intersection point data includes the coordinates of the intersection points, the identifiers of the involved trajectories, and the number of intersection times. After the recognition is completed, the system performs statistics on the intersecting trajectories based on the vehicle intersection point data, calculates the intersection intensity and frequency between each trajectory by counting the occurrence frequency of the intersection points and the path connection situation between the intersection points, and generates intersecting trajectory data. The intersecting trajectory data includes the frequency of the intersecting trajectories, the path coverage situation, and the time distribution.

[0140] Preferably, step S4 includes the following steps:

[0141] Step S41: Perform attribution region aggregation calculation on the intersecting trajectory data to obtain region density data; divide the time period of the vehicle comprehensive time series feature set to generate time-sharing vehicle feature data;

[0142] Step S42: Perform feature correlation analysis on the region density data and the time-sharing vehicle feature data to obtain trajectory density feature data; predict the vehicle demand for the partitioned freight map network based on the trajectory density feature data to obtain regional vehicle demand data;

[0143] Step S43: Perform resource allocation calculation on the regional vehicle demand data to obtain an on-demand resource allocation plan;

[0144] Step S44: Perform on-demand scheduling simulation on the vehicle comprehensive time series feature set based on the on-demand resource allocation plan to generate simulated on-demand scheduling data.

[0145] In this embodiment, the intersection trajectory data is aggregated according to the regional location by using the aggregation calculation algorithm, the intersection trajectory frequency of each region is extracted and counted, all the trajectory data within the region is divided by the coordinate boundary of the attribution region. After the region grouping is completed, the accumulation calculation of the trajectory quantity is performed to generate the regional density data. The regional density data includes indicators such as the trajectory frequency and trajectory coverage rate of each region. On this basis, the time period division method is adopted for the vehicle comprehensive time series feature set, the vehicle driving data is divided according to the set time period, and the time-sharing vehicle feature data is generated through the segmented time window. The time-sharing vehicle feature data includes the vehicle driving path, stop points, and driving time within each time period. The time period division is convenient for analyzing the vehicle flow changes in different time periods. The regional density data and the time-sharing vehicle feature data are subjected to correlation analysis through the feature correlation analysis tool, the trajectory density calculation algorithm is used to calculate the vehicle density characteristics of each region in different time periods, the trajectory density mapping is performed on each path of the partitioned freight map network to generate the trajectory density feature data. The trajectory density feature data includes the vehicle flow density, path coverage rate, and change trend within each time period of each region. By combining the trajectory density feature data with the historical vehicle demand situation, the system performs vehicle demand prediction on each region in the partitioned freight map network based on the linear regression and time series prediction model to obtain the regional vehicle demand data. The regional vehicle demand data includes the estimated vehicle demand quantity, demand peak time period, and average demand change rate of each region. The regional vehicle demand data is analyzed by the resource allocation calculation module. According to the vehicle demand quantity, demand time period, and estimated demand peak situation of each region, the resource allocation algorithm is used to perform dynamic allocation calculation on the vehicle quantity required for each region. The on-demand resource allocation plan includes the vehicle quantity allocation, scheduling priority, and scheduling time arrangement of each region. In the specific operation, first, the demand difference between regions is calculated, and the vehicle resources are scheduled through the optimized allocation strategy of the vehicle quantity to make the vehicle demand within each region match the actual allocated quantity. The on-demand resource allocation plan ensures the reasonable allocation of vehicle resources among different regions and avoids the waste of too much or too little vehicle resources. The on-demand scheduling simulation tool is used to perform scheduling simulation operations on the vehicle comprehensive time series feature set based on the on-demand resource allocation plan. During the simulation process, the vehicle path and time are dynamically adjusted according to the vehicle resource allocation situation of each region. By mapping the vehicle allocation data in the on-demand resource allocation plan to the vehicle comprehensive time series feature set, the vehicle path and within-region are re-planned for each time period. In the on-demand scheduling simulation, the driving route, stop points, and scheduling time of the vehicle are adjusted in sequence to ensure that the vehicle resources in each region match their demand intensity, and the simulated on-demand scheduling data is generated. The simulated on-demand scheduling data includes the scheduling path, arrival time, stop region, and scheduling times of each vehicle.

[0146] Preferably, step S42 includes the following steps:

[0147] Step S421: Conduct a correlation analysis on the regional density data and the time-sharing vehicle feature data to obtain correlation data; perform feature combination processing on the correlation data to generate combined feature data;

[0148] Step S422: Based on the correlation data, conduct a correlation importance ranking on the combined feature data to obtain ranked feature data; perform screening and integration on the ranked feature data to generate trajectory density feature data;

[0149] Step S423: Deduce the regional freight supply volume from the trajectory density feature data to obtain the predicted regional freight supply data; perform supply digestion calculation on the zonal freight map network based on the predicted regional freight supply data to generate digestion supply and demand data;

[0150] Step S424: Conduct vehicle demand statistics on the zonal freight map network according to the digestion supply and demand data to obtain the regional vehicle demand data.

[0151] In this embodiment, the correlation analysis algorithm is used to quantitatively analyze the relationship between regional density data and time-sharing vehicle characteristic data. By calculating the correlation degree between the two in different time periods and regions, correlation degree data is obtained. The correlation degree data specifically includes the relationship weight value between regional density and vehicle flow and the correlation characteristics of the time period. The correlation analysis is performed using the Dynamic Time Warping (DTW) algorithm. Through the similarity matching analysis of the data curves, the correlation index between regional density and vehicle characteristic data is extracted. Then, the correlation degree data is subjected to feature combination processing. The multi-dimensional feature combination method is used to combine various correlation degree indexes to generate combined feature data. The combined feature data contains multi-dimensional correlation data points, and the weights of each dimension are normalized. Based on the correlation degree data, the correlation importance of each feature dimension of the combined feature data is sorted. The feature importance sorting algorithm is used to divide the priority of features by calculating the contribution value and correlation of each feature in different time periods, and sorted feature data is obtained. The sorted feature data includes the priority sorting and weight value of each feature. Subsequently, the sorted feature data is screened and integrated. By setting a weight threshold, low-weight data is screened out, and high-correlation and high-weight feature data is retained. The selected important features are integrated to generate trajectory density feature data. The trajectory density feature data includes the selected high-correlation feature set and the density information in the vehicle trajectory data, which is used for subsequent supply volume deduction. Through the regional supply deduction model, the regional freight supply volume is deduced from the trajectory density feature data. Based on the historical freight volume and vehicle density in the region, combined with the current trajectory density feature data, the future freight demand is predicted. The deduction model uses a combination of time series analysis and regression analysis. By segmenting the analysis of time series features, the future freight supply volume of each region is predicted, and thus the predicted regional freight supply data is obtained. The predicted regional freight supply data includes the predicted future freight volume demand value of each region and the estimated supply cycle. Subsequently, based on the predicted regional freight supply data, the supply digestion calculation of the zoned freight map network is performed. The supply volume is compared with the estimated demand volume of the region. Through the supply digestion algorithm, the supply and demand balance status in each region is dynamically calculated to generate digestion supply and demand data. The digestion supply and demand data includes the supply and demand matching situation and difference index in each region. According to the supply and demand difference value of each region in the digestion supply and demand data, the vehicle demand in the zoned freight map network is specifically counted through the demand statistics module. For the situation of insufficient regional supply, the quantity and time of vehicle scheduling demand are calculated based on the demand difference. For the regions with supply and demand balance, the vehicle quantity is optimized. The minimum vehicle scheduling algorithm is used to calculate the vehicle demand in each region, and regional vehicle demand data is generated. The regional vehicle demand data includes the total vehicle demand, demand distribution, and specific scheduling demand of each region.

[0152] Preferably, step S423 includes the following steps:

[0153] Perform regional division calculation on the trajectory density feature data to obtain regional distribution data;

[0154] Analyze the trend of trajectory density change for the regional distribution data based on the trajectory density feature data to generate the trend of trajectory density change;

[0155] Perform supply quantity mapping on the trend of trajectory density change to obtain the trend of supply quantity change;

[0156] Estimate the supply quantity based on the trend of supply quantity change according to the trajectory density feature data to generate the predicted regional freight supply data;

[0157] Perform road network mapping and matching on the zonal freight map network based on the predicted regional freight supply data to obtain road network matching data;

[0158] Perform freight supply-demand balance calculation on the road network matching data to obtain freight supply-demand balance data;

[0159] Predict the supply digestion demand for the predicted regional freight supply data based on the freight supply-demand balance data to generate digestion supply-demand data.

[0160] In this embodiment, the trajectory density feature data is used to perform accurate regional division calculation on the freight map. First, based on the spatial coordinates of the vehicle position and trajectory density data on the freight platform, the freight map is gridded and partitioned according to the geographical boundaries, and the regional division is refined in combination with the vehicle trajectory density in the partition. The K-means clustering algorithm is used to perform cluster analysis on the density data points in the region to ensure that the divided regions do not overlap in space and have consistent density characteristics. The regional distribution data contains the spatial boundary information and trajectory density statistics of each region. The time series analysis method is used to perform change trend analysis on the trajectory density feature data of each region in the regional distribution data. First, the trajectory density change value of each region is extracted according to the time period (such as hours, days), and the weighted moving average (WMA) is used to calculate the change trend of the trajectory density feature data of each region. The track density change in different time periods is smoothed by the Average method to generate the track density change trend of each area. The track density change trend includes the density fluctuation of different areas in each time period, which provides basic data for the subsequent supply mapping. According to the track density change trend data, the supply demand of each area in each time period is calculated by the supply mapping model. The supply mapping model maps the increase and decrease of track density fluctuations with the corresponding relationship between vehicle supply demand, and obtains the specific demand reflection of track density change on supply. The quantitative analysis method based on linear regression is used to obtain the supply change trend value through regression calculation. The supply change trend data includes the predicted supply of each area in different time periods. The value provides the dynamic changes in the demand for regional freight vehicles. According to the supply change trend data generated in the previous stage, the forecast analysis model is used to estimate the regional freight supply. The specific steps are to linearly fit the supply change trend with the trajectory density feature data, and generate the supply value in a specific period of time in the future through the trend forecast model that increases hour by hour. The forecast regional freight supply data includes the supply and demand in each time period in the future. The forecast data is weighted to make the supply forecast value more dynamically responsive. The forecast regional freight supply data is mapped to the partitioned freight map network. First, the main freight routes in the map network are screened and matched according to the supply and demand and vehicle trajectory density in the region. The network mapping matching algorithm (Graph The road network matching data obtained contains the freight demand data of each road section in the freight map network and the predicted traffic pressure index, which provides a basis for the subsequent supply and demand balance calculation. The supply and demand balance calculation is performed on the road network matching data. First, the difference between supply and demand is calculated according to the supply and demand imbalance area in the road network matching data. The Lagrangian optimization algorithm is used to perform dynamic balance calculation on the supply and demand balance of each area. The calculation results sort the areas according to the balance value of supply and demand to obtain freight supply and demand balance data.This data includes the supply-demand balance differences and vehicle demand gaps in each region, as well as the freight supply-demand balance status of each road section, and provides data support for the prediction of supply digestion demand. Based on the freight supply-demand balance data, a supply digestion prediction model is used to adjust the supply demand quantity of the region. The model predicts the future supply digestion demand quantity through the balance degree analysis of the supply-demand differences in each region. The specific operations include fitting and adjusting the future supply data according to the current demand gap, and generating digestion supply-demand data. The digestion supply-demand data includes the supply-demand digestion demand quantity and the specific vehicle allocation quantity in each future time period of each region, providing support for the dispatching strategy of the intelligent dispatching system.

[0161] Preferably, step S5 includes the following steps:

[0162] Step S51: Calculate the transport capacity of the vehicle comprehensive time-series feature set to obtain the existing transport capacity data; calculate the dispatching transport capacity of the simulated on-demand dispatching data to obtain the dispatching transport capacity data;

[0163] Step S52: Perform spatio-temporal distribution mapping on the existing transport capacity data to obtain the existing transport capacity distribution data; perform regional dispatching demand analysis on the dispatching transport capacity data to obtain the regional demand data;

[0164] Step S53: Calculate the matching degree between the existing transport capacity distribution data and the regional demand data to obtain the matching degree data; perform a comparison of transport capacity differences on the vehicle comprehensive time-series feature set and the simulated on-demand dispatching data based on the matching degree data to obtain the transport capacity difference data;

[0165] Step S54: Plan the dispatching route of the simulated on-demand dispatching data according to the transport capacity difference data to obtain the route planning data; perform conflict detection on the route planning data to obtain the dispatching conflict data;

[0166] Step S55: Adjust the route of the route planning data based on the dispatching conflict data to generate an adjusted dispatching route; formulate a dispatching plan for the adjusted dispatching route to generate a vehicle dispatching plan.

[0167] In this embodiment, a detailed calculation of the transportation capacity is performed on the comprehensive time-series feature set of the vehicle. Specifically, by analyzing the feature data such as the working time, load condition, geographical location, and driving distance of the vehicle, and using the weighted average method and the peak analysis method, the transportation capacity provided by the vehicle in different time periods is summarized to obtain the existing transportation capacity data. The existing transportation capacity data includes the specific transportation capacity values of each region in different time periods. At the same time, the transportation capacity of the simulated on-demand scheduling data is calculated. Through the scheduling simulation calculation of the demand and vehicle availability in each time period, the scheduling transportation capacity data is obtained. The scheduling transportation capacity data represents the expected demand transportation volume of each region in different time periods. Based on the existing transportation capacity data, first, through the spatiotemporal mapping model, spatiotemporal mapping is performed. The transportation capacity data of each time period is spatially marked and distributed on the map, and combined with the spatial coordinates, the existing transportation capacity data is converted into the existing transportation capacity distribution data. The existing transportation capacity distribution data includes the transportation capacity distribution of different regions in different time periods. The regional scheduling demand analysis is performed on the scheduling transportation capacity data. The scheduling demand values of each region are classified according to the geographical location, and the demand analysis tool is used to further refine the transportation capacity demand within the region to obtain the regional demand data. This data includes the vehicle demand situation of different regions in each time period. Through the matching degree calculation method (Cosine Similarity), a one-to-one matching is performed on the existing transportation capacity distribution data and the regional demand data. The transportation capacity supply and demand are compared one by one to obtain the matching degree data. The matching degree data includes the supply-demand matching degree of each region in different time periods. Then, based on the matching degree data, a comparison of the transportation capacity differences is performed on the comprehensive time-series feature set of the vehicle and the simulated on-demand scheduling data. By calculating the difference values between the supply and demand in each region, the balance state and deviation between the supply and demand in each region are determined to obtain the transportation capacity difference data. The transportation capacity difference data is used to clarify the shortage or surplus of the transportation capacity in each region in each time period. According to the transportation capacity difference data, the scheduling route planning is performed on the simulated on-demand scheduling data through the optimal path algorithm. First, the optimal driving route is planned for each vehicle according to the vehicle position, demand region, and road traffic conditions, and the route planning data is generated. The route planning data includes the driving route and the estimated arrival time of each vehicle. Then, conflict detection is performed on the route planning data. Conflict detection judges the potential traffic conflict risk by calculating the intersection points and congestion points of each vehicle in the driving route, and generates the scheduling conflict data. The scheduling conflict data includes the conflict regions in each driving route and the time periods when the vehicles are driving, providing a basis for subsequent route adjustment. According to the scheduling conflict data, the original route planning data is adjusted. The specific adjustment operations include arranging the time sequence of the vehicles at the conflict points in the same route to avoid congestion and conflict of the route.Provide a conflict-free driving route for each vehicle through the route re-selection algorithm, generate an adjusted scheduling route, where the adjusted scheduling route includes the final driving route of each vehicle and the corresponding adjusted driving time period, and then, based on the adjusted route, combined with the scheduling requirement data of the vehicle, formulate a vehicle scheduling plan, determine the scheduling order, driving time, and destination area of the vehicle through the scheduling strategy generation method, and the generated vehicle scheduling plan includes the driving route and scheduling plan of each vehicle.

[0168] Preferably, step S53 includes the following steps:

[0169] Perform data alignment processing on the existing transport capacity distribution data and regional demand data to obtain aligned basic data;

[0170] Perform regional matching calculation on the aligned basic data to obtain regional matching data;

[0171] Perform time matching calculation on the aligned basic data based on the regional matching data to obtain time matching data;

[0172] Perform capacity matching calculation on the aligned basic data based on the time matching data to generate capacity matching data;

[0173] Perform matching degree calculation on the aligned basic data according to the capacity matching data to generate matching degree data;

[0174] Perform threshold slicing on the matching degree data to obtain threshold classification data;

[0175] Perform transport capacity difference identification on the vehicle comprehensive time series feature set and simulated on-demand scheduling data based on the threshold classification data to obtain difference degree data;

[0176] Perform data integration on the difference degree data to generate transport capacity difference data.

[0177] In this embodiment, the data alignment algorithm is used to perform data alignment processing on the existing transport capacity distribution data and regional demand data to ensure the consistency of the two sets of data in terms of time and space. The core of the data alignment processing lies in integrating the data information of different time points and spatial regions so that they can be compared and calculated on the same analysis dimension. In the specific implementation, the two data sources are divided by a spatio-temporal mapping tool according to a predetermined time period and geographical coordinates. At the same time, time interpolation operations are performed on the discontinuous regional demand data so that it can be matched to the same time period and spatial range. After the data alignment processing is completed, the generated aligned basic data contains the unified alignment information of the existing transport capacity distribution and regional demand. Regional matching calculation is carried out. Through a matching algorithm (Cosine Similarity), the transport capacity distribution information and regional demand information in the aligned basic data are compared within the same geographical region, and the transport capacity resources and demands are calculated for each region to generate regional matching data. The regional matching calculation is realized through a spatial matching model. By comparing the supply and demand region by region, the supply-demand ratio of each region is calculated. The regional matching data records the matching ratio data of each region in the form of a table for subsequent time matching operations. Time matching calculation is carried out on the aligned basic data within the region to evaluate the matching degree of transport capacity supply and demand at different time periods. Through the time matching algorithm, the transport capacity distribution and demand distribution within each time period are compared one by one, and the supply-demand ratio within each time period is calculated to generate time matching data. The time matching data contains the supply-demand balance information of each region at different time periods. Through the time series analysis tool, the demand peak value and the maximum load point of transport capacity supply within each time period can be obtained. Based on the time matching data, further capacity matching calculation is carried out. Through the capacity analysis tool, the transport capacity capacity data and demand data of each region in the aligned basic data are compared, and the matching calculation is carried out according to the actual transport capacity capacity. The capacity matching data generated by the capacity matching calculation records the ratio of the actual transport capacity to the demand capacity within each time period and region, and the matching differences are marked with numerical values. After the capacity matching data is formed, according to the capacity matching data, the overall matching degree calculation is carried out on the aligned basic data. Through the matching degree calculation model, the supply-demand data of each time period and region are comprehensively compared to generate matching degree data. The matching degree data records the supply-demand situation within each time period and region in the form of a supply-demand percentage. The closer the value is to 1, the more balanced the supply and demand are. After the matching degree data is generated, the data is threshold-segmented through a threshold classification tool. The threshold segmentation classifies and grades the matching degree data through a preset threshold range, marks the regions with high supply-demand matching degree as "balanced regions", and marks the regions with low matching degree as "regions in need of supplementation", and specific supply-demand level marks are made according to the high and low thresholds. The generated threshold classification data is used for the transport capacity difference identification step.The matching classification situations of different regions are presented in tabular form. After the threshold classification data is generated, through the transport capacity difference identification algorithm, the threshold classification data is comprehensively analyzed with the vehicle comprehensive time-series feature set and the simulated on-demand scheduling data. By comparing the scheduling supply-demand gaps in different regions, the transport capacity differences in different time periods and regions are identified. Specifically, the transport capacity difference identification performs a difference operation on the scheduling value and the actual value through a feature extraction tool to generate data on the degree of difference. The data on the degree of difference shows the supply-demand gaps and excess situations in each time period in the form of difference values. The data on the degree of difference is integrated, and the data on the degree of difference in different time periods and regions is summarized through a transport capacity analysis model to generate transport capacity difference data. The transport capacity difference data is used to describe the overall situation of supply-demand balance in each region and mark the key transport capacity gap regions and times. The generated transport capacity difference data provides a basis for subsequent scheduling optimization.

[0178] Preferably, step S55 includes the following steps:

[0179] Identify the conflict road sections from the scheduling conflict data to obtain conflict road section data; retrieve alternative routes for the route planning data based on the conflict road section data to generate alternative route data;

[0180] Allocate time windows to the alternative route data to obtain time allocation data; reorganize the route planning data based on the time allocation data to generate an adjusted scheduling route;

[0181] Make a vehicle allocation plan for the adjusted scheduling route to obtain vehicle allocation data;

[0182] Formulate a scheduling plan for the vehicle allocation data based on the time allocation data to generate a vehicle scheduling plan.

[0183] In this embodiment, the scheduling conflict data is analyzed, and the road section positions with scheduling conflicts are extracted therefrom using a road section conflict recognition algorithm. The conflict road section recognition is based on road network topology data. First, the conflict points in the scheduling conflict data are matched with the road network, and by analyzing the overlapping situation with other scheduling routes, the position and range of the conflict road section are further determined. The generated conflict road section data includes the specific position of each conflict road section, the start and end point coordinates, and the estimated congestion time information of the road section. The conflict road section data serves as the basis for subsequent alternative route retrieval. According to the conflict road section data, alternative route retrieval is performed on the conflict road sections in the route planning data. Through a path planning tool, the shortest path algorithm (such as the Dijkstra algorithm) or other intelligent planning algorithms are used to select alternative routes. When performing alternative route retrieval, non-congested road sections and fast-passage road sections are preferably selected on the premise of ensuring driving safety. The alternative route is selected through iterative calculation of the algorithm. The generated alternative route data records the road sections passed by each path, the estimated driving time, and relevant road condition information. The alternative route data is used in the time window allocation step. After the alternative route data is generated, time allocation processing is performed on the alternative route data based on the time window allocation algorithm. Each alternative route is matched with the time window expected to pass through its road sections to reduce vehicle scheduling during congested times. In specific operations, the peak traffic period of each road section is set as an unavailable window, the passing period of the vehicle during the non-peak period is calculated, and the corresponding time allocation data is generated. The time allocation data includes the passing time nodes and the estimated passing time periods of each alternative route. The time allocation data is used in the subsequent route reorganization step. The time allocation data is integrated with the route planning data through a route reorganization algorithm. First, according to the feasible passing time periods of each road section, the existing routes are adjusted to avoid peak congestion time periods, and the conflict road sections are replaced with alternative routes. By splicing and reorganizing each alternative route, an adjusted scheduling route is formed. The adjusted scheduling route includes a driving path that is reasonable both in time and space, ensuring the timeliness and smoothness of the entire scheduling. The generated adjusted scheduling route records the specific driving road sections, the estimated passing time, and its scheduling priority information. The existing fleet resources are matched with the adjusted scheduling route through a vehicle allocation algorithm to optimize the vehicle configuration. By analyzing the driving duration, fleet capacity, and scheduled scheduling tasks of each scheduling route, the vehicle that best meets the task requirements is assigned to the specified scheduling route. In the specific assignment process, vehicle allocation data is generated by considering factors such as the current state of the fleet, driving mileage, and fuel consumption. The vehicle allocation data includes the route number, departure time, and estimated arrival time corresponding to each vehicle. Through a scheduling plan generation module, a final vehicle scheduling plan is formulated. The formulation of the scheduling plan includes the specific driving route of the vehicle, the passing time nodes of each road section, and the interval time between vehicles. The scheduling plan details the scheduling order, driving route, and departure and arrival time nodes of each vehicle.Ensure the scheduling optimization of the transport capacity is completed within a reasonable time window.

[0184] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Thus, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be embraced within the present invention.

[0185] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A management method for intelligent dispatching of freight platform vehicles based on the Internet of Things, characterized in that: The following steps are involved: Step S1: Collect vehicle terminal information; Extracting features from vehicle terminal information to obtain vehicle feature information data; Perform time series decomposition on vehicle feature information data to generate a comprehensive vehicle time series feature set; Step S2: draw a freight route for the comprehensive time series feature set of the vehicle to obtain basic freight route data; According to the basic freight route data, the comprehensive time series feature set of the vehicle is reconstructed in real time to generate a freight map network; wherein step S2 includes the following steps: Step S21: extracting path points from the comprehensive time series feature set of the vehicle to obtain path point data; performing frequency statistics on the path point data to generate path frequency data; Step S22: Calculate the path density of the path point data based on the path frequency data to generate path density data; draw a freight route for the path point data according to the path density data to obtain basic freight route data; Step S23: gridding the basic freight route data to obtain a road network grid; identifying road network nodes on the road network grid to generate road network node data; Step S24: performing connectivity analysis on the road network node data to obtain connectivity data; performing road network weight calculation on the connectivity data based on the preset road network weight assignment to generate road network weight data; Step S25: constructing a topological relationship for the basic freight route data according to the road network weight data to obtain topological structure data; performing network integration on the topological structure data to generate a freight map network; Step S3: evenly partitioning the freight map network to obtain a partitioned freight map network; performing regional vehicle trajectory intersection analysis on the vehicle comprehensive time series feature set based on the partitioned freight map network to generate intersection trajectory data; Step S4: predicting vehicle demand for the partitioned freight map network based on the intersection trajectory data and the vehicle comprehensive time series feature set to obtain regional vehicle demand data; performing on-demand scheduling simulation on the vehicle comprehensive time series feature set based on the regional vehicle demand data to generate simulated on-demand scheduling data; Step S5: compare the vehicle comprehensive time series feature set and the simulated on-demand dispatching data with transport resources to obtain transport capacity difference data; design a dispatching plan for the simulated on-demand dispatching data based on the transport capacity difference data to generate a vehicle dispatching plan; wherein step S5 includes the following steps: Step S51: Calculate the transport capacity of the comprehensive time series feature set of the vehicle to obtain the existing transport capacity data; calculate the dispatch capacity of the simulated on-demand dispatch data to obtain the dispatch capacity data; Step S52: Performing spatiotemporal distribution mapping on the existing transport capacity data to obtain existing transport capacity distribution data; performing regional dispatch demand analysis on the dispatch transport capacity data to obtain regional demand data; Step S53: Calculate the matching degree of the existing capacity distribution data and the regional demand data to obtain matching degree data; compare the capacity difference of the vehicle comprehensive time series feature set and the simulated on-demand scheduling data based on the matching degree data to obtain capacity difference data; Step S54: scheduling route planning for the simulated on-demand scheduling data according to the capacity difference data to obtain route planning data; performing conflict detection on the route planning data to obtain scheduling conflict data; Step S55: adjusting the route planning data based on the scheduling conflict data to generate an adjusted scheduling route; formulating a scheduling plan for the adjusted scheduling route to generate a vehicle scheduling plan; Step S6: Intelligently dispatch the freight platform vehicles according to the vehicle dispatch plan to achieve a capacity optimization management method.

2. The management method for intelligent dispatching of freight platform vehicles based on the Internet of Things according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: real-time acquisition of vehicle location information to obtain original location data; monitoring and acquisition of vehicle operation status to obtain original status data; dynamic acquisition of vehicle load status to obtain original load data; Step S12: integrating the original position data, the original state data and the original load data to obtain vehicle terminal information; performing unified format conversion processing on the vehicle terminal information to obtain standard format information; Step S13: extracting features from the standard format information to obtain vehicle-borne feature information data; performing periodic analysis on the vehicle-borne feature information data to obtain periodic data; Step S14: performing change trend statistics on the periodic data to generate a periodic change trend; performing time series reorganization on the periodic data and the periodic change trend to obtain a comprehensive vehicle time series feature set.

3. The management method for intelligent dispatching of freight platform vehicles based on the Internet of Things according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: performing boundary identification on the freight map network to obtain the map network boundary; calculating the regional area of ​​the freight map network according to the map network boundary to generate the global map network area; Step S32: dividing the freight map network into four equal regions based on the area of ​​the global map network to obtain a partitioned freight map network; Step S33: mapping the comprehensive time series feature set of vehicles to each region according to the partitioned freight map network to obtain the partitioned vehicle time series feature set; extracting vehicle trajectories from the partitioned vehicle time series feature set to generate the vehicle trajectories belonging to the partition; Step S34: performing intersection point identification on the vehicle trajectories of the belonging partition to obtain vehicle intersection point data; performing intersection trajectory statistics on the vehicle trajectories of the belonging partition based on the vehicle intersection point data to generate intersection trajectory data.

4. The management method for intelligent dispatching of freight platform vehicles based on the Internet of Things according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: performing attribution area aggregation calculation on the intersection trajectory data to obtain area density data; dividing the vehicle comprehensive time series feature set into time periods to generate time-divided vehicle feature data; Step S42: performing feature correlation analysis on the regional density data and the time-sharing vehicle feature data to obtain track density feature data; performing vehicle demand forecasting on the partitioned freight map network according to the track density feature data to obtain regional vehicle demand data; Step S43: performing resource allocation calculation on the regional vehicle demand data to obtain an on-demand resource allocation plan; Step S44: Perform on-demand scheduling simulation on the comprehensive time series feature set of the vehicle based on the on-demand resource allocation scheme to generate simulated on-demand scheduling data.

5. The management method for intelligent dispatching of freight platform vehicles based on the Internet of Things according to claim 4 is characterized in that: Step S42 includes the following steps: Step S421: performing correlation analysis on the regional density data and the time-sharing vehicle characteristic data to obtain correlation data; performing feature combination processing on the correlation data to generate combined feature data; Step S422: sorting the combined feature data by relevance importance based on the relevance data to obtain sorted feature data; screening and integrating the sorted feature data to generate trajectory density feature data; Step S423: performing regional freight supply deduction on the trajectory density feature data to obtain predicted regional freight supply data; performing supply digestion calculation on the regional freight map network based on the predicted regional freight supply data to generate digestion supply and demand data; Step S424: Perform vehicle demand statistics on the regional freight map network based on the digested supply and demand data to obtain regional vehicle demand data.

6. The management method for intelligent dispatching of freight platform vehicles based on the Internet of Things according to claim 5 is characterized in that: Step S423 includes the following steps: Perform regional division calculation on the trajectory density feature data to obtain regional distribution data; Based on the trajectory density feature data, the trajectory density change trend of the regional distribution data is analyzed to generate the trajectory density change trend; Map the supply quantity to the trajectory density change trend to obtain the supply quantity change trend; The supply quantity is estimated based on the supply quantity change trend according to the trajectory density feature data, and the freight supply data of the forecast area is generated; Based on the forecasted regional freight supply data, the regional freight map network is matched with the road network mapping to obtain the road network matching data; Carry out freight supply and demand balance calculation on the road network matching data to obtain freight supply and demand balance data; Based on the freight supply and demand balance data, the supply digestion demand forecast is carried out on the freight supply data in the forecast area to generate digestion supply and demand data.

7. The management method for intelligent dispatching of freight platform vehicles based on the Internet of Things according to claim 1 is characterized in that: Step S53 includes the following steps: Perform data alignment on existing capacity distribution data and regional demand data to obtain alignment basic data; Performing regional matching calculation on the aligned basic data to obtain regional matching data; Performing time matching calculation on the alignment basic data based on the regional matching data to obtain time matching data; Perform capacity matching calculation on the aligned basic data based on the time matching data to generate capacity matching data; Calculate the matching degree of the alignment basic data according to the capacity matching data to generate matching degree data; Perform threshold segmentation on the matching degree data to obtain threshold classification data; Based on the threshold classification data, the capacity difference is identified for the comprehensive time series feature set of vehicles and the simulated on-demand dispatch data to obtain the difference degree data; The difference degree data is integrated to generate capacity difference data.

8. The management method for intelligent dispatching of freight platform vehicles based on the Internet of Things according to claim 1 is characterized in that: Step S55 includes the following steps: Identify the conflicting sections of the dispatch conflict data to obtain the conflicting section data; retrieve the alternative routes of the route planning data based on the conflicting section data to generate the alternative route data; Allocate time windows for the alternative route data to obtain time allocation data; reorganize the route planning data based on the time allocation data to generate an adjusted scheduling route; Carry out vehicle allocation planning for the adjusted dispatch route and obtain vehicle allocation data; A scheduling plan is formulated for the vehicle allocation data based on the time allocation data to generate a vehicle scheduling plan.

Images