A method, device, equipment and storage medium for predicting transportation capacity

Abstract

The present application discloses a method, apparatus, device, and storage medium for predicting transportation capacity. The method comprises: extracting the transportation starting point and end point of a freight vehicle based on its trajectory data; matching points of interest within a test area with the transportation starting point and end point to obtain the historical number of freight vehicles corresponding to each point of interest within the test area; screening multiple popular points of interest based on the historical number of freight vehicles corresponding to each point of interest; inputting the historical number of freight vehicles corresponding to the multiple popular points of interest into a pre-trained transportation capacity prediction model to obtain prediction results for the number of freight vehicles corresponding to the popular points of interest. According to this method, the historical number of freight vehicles at the point of interest is obtained by analyzing the historical stop data of the vehicle. Based on the historical number of freight vehicles and the pre-trained neural network model, the transportation capacity prediction results for the point of interest in a certain period of time in the future are obtained, which is beneficial for logistics companies or individual vehicle owners to avoid transportation risks and reduce the probability of empty vehicles.

Description

Technical Field

[0001] The present invention relates to the field of logistics and transportation technology, and in particular to a method, device, equipment and storage medium for predicting transportation capacity. Background Art

[0002] With the rise of e-commerce, the improvement of highway infrastructure, and the rapid development of international trade, road freight is becoming increasingly important in the overall freight logistics system. However, the mismatch between the supply and demand of goods and trucks between regions makes it difficult for logistics companies in these regions to accurately estimate the transport capacity between them. This results in a large number of trucks returning empty or being unable to transport full loads due to competition, leading to either overcapacity or undercapacity. Summary of the Invention

[0003] The embodiments of the present application provide a method, apparatus, device, and storage medium for predicting transportation capacity. To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is provided below. This summary is not intended to be a comprehensive review, identify key or important components, or delineate the scope of protection for these embodiments. Its sole purpose is to present some concepts in a simplified form, serving as a prelude to the detailed description that follows.

[0004] In a first aspect, an embodiment of the present application provides a method for predicting transportation capacity, comprising:

[0005] Extract the transportation starting point and end point of the vehicle based on the trajectory data of the freight vehicle;

[0006] Match the points of interest in the area to be tested with the transport starting and ending points to obtain the historical number of freight vehicles corresponding to each point of interest in the area to be tested;

[0007] Filter out multiple popular points of interest based on the historical number of freight vehicles corresponding to each point of interest;

[0008] The historical number of freight vehicles corresponding to multiple popular points of interest is input into the pre-trained traffic capacity prediction model to obtain the prediction results of the number of freight vehicles corresponding to the popular points of interest.

[0009] In an optional embodiment, extracting the transport start and end points of the vehicle based on the trajectory data of the freight vehicle includes:

[0010] Obtain the trajectory data of vehicles in the test area;

[0011] Extracting vehicle parking data based on vehicle trajectory data;

[0012] Delete the stops that are shorter than the preset stop time threshold and the stops that are not logistics stops to obtain the logistics stop point set;

[0013] The stops in the logistics stop set are determined as the transportation starting point and end point in chronological order, two by two.

[0014] In an optional embodiment, after selecting multiple hot spots of interest based on the number of historical freight vehicles corresponding to each spot of interest, the method further includes:

[0015] Get the freight vehicles corresponding to each popular point of interest;

[0016] Query the logistics and transportation company to which the freight vehicle belongs;

[0017] Based on the relationship between each popular point of interest and the historical number of freight vehicles and logistics and transportation companies, the first transportation capacity knowledge graph is constructed.

[0018] In an optional embodiment, after selecting multiple hot spots of interest based on the number of historical freight vehicles corresponding to each spot of interest, the method further includes:

[0019] Get the freight vehicles corresponding to each popular point of interest;

[0020] Query the owner identity information of the freight vehicle;

[0021] Based on the relationship between each popular point of interest and the number of historical freight vehicles and the identity of the vehicle owners, a second transportation capacity knowledge graph is constructed.

[0022] In an optional embodiment, before inputting the historical number of freight vehicles corresponding to the plurality of hot spots of interest into the pre-trained traffic capacity prediction model, the method further includes:

[0023] Training traffic capacity prediction models;

[0024] The traffic capacity prediction model is a Conv-LSTM neural network model.

[0025] In an optional embodiment, the historical number of freight vehicles corresponding to multiple popular points of interest is input into a pre-trained traffic capacity prediction model to obtain a prediction result of the number of freight vehicles corresponding to the popular points of interest, including:

[0026] A spatiotemporal matrix is constructed based on the historical number of freight vehicles corresponding to multiple popular points of interest. The constructed spatiotemporal matrix is input into the convolutional layer to obtain the extracted spatial features.

[0027] Input the extracted spatial features into the LSTM layer to obtain the extracted spatiotemporal features;

[0028] The extracted spatiotemporal features are input into the normalization layer to obtain the predicted number of freight vehicles corresponding to the popular points of interest.

[0029] In an optional embodiment, it further includes:

[0030] The prediction results of the number of freight vehicles corresponding to the first traffic capacity knowledge graph, the second traffic capacity knowledge graph, and popular points of interest are displayed on the front end.

[0031] In a second aspect, an embodiment of the present application provides a traffic capacity prediction device, comprising:

[0032] An extraction module is used to extract the transportation starting point and end point of the vehicle based on the trajectory data of the freight vehicle;

[0033] A matching module is used to match the points of interest in the test area with the transport starting point and end point to obtain the number of historical freight vehicles corresponding to each point of interest in the test area;

[0034] A screening module is used to screen out multiple popular points of interest based on the number of historical freight vehicles corresponding to each point of interest;

[0035] The prediction module is used to input the historical number of freight vehicles corresponding to multiple popular points of interest into the pre-trained traffic capacity prediction model to obtain the prediction results of the number of freight vehicles corresponding to the popular points of interest.

[0036] In a third aspect, an embodiment of the present application provides a traffic capacity prediction device, comprising a processor and a memory storing program instructions, wherein the processor is configured to execute the traffic capacity prediction method provided in the above embodiment when executing the program instructions.

[0037] In a fourth aspect, an embodiment of the present application provides a computer-readable medium having computer-readable instructions stored thereon, and the computer-readable instructions are executed by a processor to implement a method for predicting traffic capacity provided in the above embodiment.

[0038] The technical solutions provided by the embodiments of the present application may have the following beneficial effects:

[0039] The traffic capacity prediction method provided in the embodiment of the present application analyzes the vehicle's transportation starting point and end point based on the vehicle's historical trajectory data, and then matches it with the industry's points of interest to obtain the historical number of freight vehicles at the industry's points of interest, and constructs a traffic capacity knowledge graph between logistics companies, vehicle owners, points of interest, and the number of vehicles. Then, the capacity situation in the future period is predicted based on the number of freight vehicles at the points of interest and the preset neural network model. Through the analysis of historical capacity and the prediction of future capacity, it is beneficial for logistics companies or individual car owners to avoid their transportation risks, reduce the probability of empty vehicles, improve the transportation efficiency of the entire industry in the region, and increase the profit margins of transportation companies or individual car owners. It is also convenient for regional regulatory authorities to focus on the freight vehicles of large logistics companies in the regulatory industry and the freight routes of related large POI gathering points (ports, factories, logistics parks, etc.) in the regulatory industry.

[0040] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0042] Figure 1 is a flow chart illustrating a method for predicting transportation capacity according to an exemplary embodiment;

[0043] Figure 2 is a flow chart illustrating a method for predicting transportation capacity according to an exemplary embodiment;

[0044] Figure 3 is a schematic diagram of a first transportation capacity knowledge graph according to an exemplary embodiment;

[0045] Figure 4 is a schematic diagram of a second transportation capacity knowledge graph according to an exemplary embodiment;

[0046] Figure 5 is a schematic structural diagram of a traffic capacity prediction device according to an exemplary embodiment;

[0047] Figure 6 is a schematic structural diagram of a traffic capacity prediction device according to an exemplary embodiment;

[0048] Figure 7 It is a schematic diagram of a computer storage medium according to an exemplary embodiment. DETAILED DESCRIPTION

[0049] The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them.

[0050] It should be understood that the embodiments described are only a portion of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0051] In the following description, unless otherwise indicated, identical numbers in different figures represent identical or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of systems and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0052] This application is designed based on actual vehicle operation data scenarios, aiming to address the existing issues of excess and insufficient transport capacity between urban areas and between logistics companies and popular points of interest. A traffic capacity prediction method based on the conv1D-LSTM neural network structure comprehensively predicts traffic capacity within the test area, providing an accurate and clear regional capacity diagram to address the narrow profit margins caused by empty loads in the transportation industry. This application provides effective data support and statistics for individual vehicle owners, logistics companies, and government freight regulators, thereby expanding the activity of freight logistics across the entire region, ensuring the stability of the commodity market, and improving the overall efficiency and profit margins of the industry.

[0053] The following is a detailed description of the traffic capacity prediction method provided by the embodiment of the present application with reference to the accompanying drawings. Figure 1 , the method specifically includes the following steps.

[0054] S101 extracts the transportation starting point and end point of the vehicle based on the trajectory data of the freight vehicle.

[0055] In one possible implementation method, the historical trajectory data of freight vehicles in the area to be tested is first obtained, for example, the GPS trajectory data of freight vehicles in Qingdao in the past 10 days is obtained. Among them, the GPS trajectory point is the vehicle position data reported in real time by the GPS (Global Positioning System) device deployed on the vehicle. Usually, when the GPS device reports in real time, the reporting time interval can be set according to the actual situation. The preferred time interval of this application is 10S. It should be noted that in addition to using the vehicle-mounted GPS device to report the vehicle's trajectory data in real time, this application can also use Beidou equipment to locate the vehicle trajectory and report it.

[0056] The obtained trajectory data includes basic trajectory data such as license plate number, latitude and longitude, driving speed, time, and direction, and then the vehicle stop data is extracted from the vehicle trajectory data.

[0057] Specifically, points with a speed of zero are obtained from the trajectory information and merged according to the time interval threshold and offset threshold. For example, multiple stops within 10 minutes are merged, and multiple stops within 50 meters are merged to obtain the stops represented by the centroid. This step can be used to aggregate small stops within a certain range to obtain vehicle stop data.

[0058] Furthermore, it also includes obtaining road network data, POI location data, etc. in the area to be tested, and screening logistics stops based on the POI location data.

[0059] Specifically, stops with a duration shorter than a preset threshold are removed. For example, stops with a duration of less than 30 minutes are filtered out. Stops that are not logistics stops are then removed. Each stop is iterated over, and the distance from each stop to the highway service area and toll booth is calculated. Stops with a distance less than 500 meters are removed to obtain a set of logistics stops. The stops in the set are then sequentially determined as transport starting and ending points, in chronological order.

[0060] In an exemplary scenario, there are six stops in the logistics stop set, which are stop 1, stop 2, stop 3, stop 4, stop 5, and stop 6 in chronological order. Stop 1 is the transportation starting point, stop 2 is the transportation end point, stop 3 is the transportation starting point, stop 4 is the transportation end point, stop 5 is the transportation starting point, and stop 6 is the transportation end point.

[0061] S102 matches the points of interest in the area to be measured with the transport starting point and end point to obtain the number of historical freight vehicles corresponding to each point of interest in the area to be measured.

[0062] In one possible implementation, after obtaining the vehicle's transportation starting point and end point based on the vehicle's trajectory data analysis, the points of interest related to the logistics industry in the test area are matched with the vehicle's transportation starting point and end point respectively, and then the number of historical freight vehicles corresponding to the points of interest related to the logistics industry in the test area is obtained.

[0063] Specifically, the distance between each point of interest related to the logistics industry and each starting point and each destination is calculated. If the distance between the point of interest and the starting point or destination is less than a preset distance threshold, the point of interest is determined to match the starting point or destination. The preset distance threshold can be 50-200 meters, and the embodiment of this application does not specifically limit the value of the distance threshold. The number of transportation starting points and destinations that match each point of interest is then counted to obtain the historical number of freight vehicles corresponding to each point of interest.

[0064] In an exemplary scenario, the point of interest "Qingdao Iron and Steel Group Co., Ltd." matches 10 transportation starting points and 5 transportation end points, so the number of historical freight vehicles for this point of interest is 15.

[0065] According to this step, the historical number of freight vehicles at the points of interest in the area to be measured within a preset time period can be counted.

[0066] S103 selects multiple hot points of interest based on the historical number of freight vehicles corresponding to each point of interest.

[0067] In one possible implementation, after obtaining the historical number of freight vehicles at each point of interest, the vehicle ratio of each point of interest is calculated. The vehicle ratio can be calculated according to the following formula:

[0068] Vehicle ratio = historical number of freight vehicles at the point of interest / total number of vehicles in the area to be measured.

[0069] According to the proportion of the number of vehicles, each point of interest is sorted from high to low, and multiple points of interest with a higher proportion of the number of vehicles are regarded as hot points of interest. For example, the points of interest with the top 50% of the number of vehicles are regarded as hot points of interest.

[0070] Furthermore, the freight vehicles corresponding to each popular point of interest are obtained, and the logistics and transportation companies to which the freight vehicles belong are queried. Based on the relationship between each popular point of interest and the historical number of freight vehicles and logistics and transportation companies, the first transportation capacity knowledge graph is constructed.

[0071] Figure 3 This is a schematic diagram of the first transportation capacity knowledge graph, such as Figure 3 As shown, Qingdao Dinglong Logistics Co., Ltd. has 34 vehicles associated with POI 1 (Qingdao International Bonded Logistics Park), of which 20 vehicles travel from the park to Qingdao Dinglong Logistics Co., Ltd., and 14 vehicles travel from the park to Qingdao International Bonded Logistics Park. Qingdao Dinglong Logistics Co., Ltd. has 8 vehicles associated with POI 2 (Qingdao Large Freight Market), of which 3 vehicles travel from the market to Qingdao Dinglong Logistics Co., Ltd., and 5 vehicles travel from the market to Qingdao Dinglong Logistics Co., Ltd. It can also be seen that POI 3 (Qingdao Iron and Steel Holding Group Co., Ltd.) has only 16 freight vehicles associated with it, while POI 4 (Shandong Port Qingdao Port) has 390 freight vehicles associated with it.

[0072] By constructing a knowledge graph of transportation capacity between popular points of interest, the number of freight vehicles, and logistics and transportation companies, we can intuitively display the transportation situation between logistics and transportation companies and various popular points of interest, and provide data support for the transportation capacity evaluation of logistics and transportation companies.

[0073] Furthermore, the freight vehicles corresponding to each popular point of interest are obtained, the driver information corresponding to the freight vehicles is queried, and the identity of the vehicle owner is obtained based on the driver information. The vehicle owner identity includes logistics enterprise owners, enterprise-affiliated owners, and individual owners.

[0074] Based on the relationship between each popular point of interest and the number of historical freight vehicles and the identity of the vehicle owners, a second transportation capacity knowledge graph is constructed.

[0075] Figure 4 is the second transportation capacity knowledge graph, such as Figure 4As shown, logistics company A has 45 vehicles owned by the company and 22 vehicles owned by affiliated vehicles. Company A's vehicle owners have business dealings with points of interest 1 and 2, while affiliated vehicles have business dealings with points of interest 1, 2, and 3. Logistics company B has 60 vehicles owned by affiliated vehicles and 35 vehicles owned by the company. Among them, affiliated vehicles owned by B have business dealings with points of interest 3 and 4, while company B's vehicle owners have business dealings with point of interest 4. Points of interest 3 and 4 also have business dealings with some individual vehicle owners.

[0076] By building a knowledge graph of popular points of interest, the number of freight vehicles, and the transportation capacity among vehicle owners, logistics and transportation companies can rationally plan transportation tasks for various types of vehicle owners, thereby increasing the profits of logistics companies and vehicle owners.

[0077] S104 inputs the historical number of freight vehicles corresponding to the multiple hot spots of interest into a pre-trained traffic capacity prediction model to obtain prediction results of the number of freight vehicles corresponding to the hot spots of interest.

[0078] In an optional embodiment, before inputting the historical number of freight vehicles corresponding to multiple popular points of interest into the pre-trained traffic capacity prediction model, it also includes: training the traffic capacity prediction model, wherein the traffic capacity prediction model can be a Conv-LSTM neural network model.

[0079] Specifically, the historical number of freight vehicles corresponding to the point of interest is obtained, and the training set and test set are constructed based on the historical number of freight vehicles corresponding to the point of interest. For example, the number of freight vehicles 5 days before the point of interest is used as the training set, and the number of freight vehicles 5 days after the point of interest is used as the test set.

[0080] The specific training process includes constructing a spatiotemporal matrix based on the historical number of freight vehicles corresponding to multiple popular points of interest. The row vectors of the matrix represent the number of freight vehicles corresponding to different points of interest in the same time period, and the column vectors of the matrix represent the number of freight vehicles corresponding to the same point of interest in different time periods. The specific example is as follows:

[0081]

[0082] Among them, N represents the number of freight vehicles corresponding to the point of interest, and M represents the number of popular points of interest.

[0083] The constructed spatiotemporal matrix is then fed into the convolution layer to extract spatial features. The entire convolution process is implemented using the Conv1D function, which requires the input matrix size, kernel length, number of kernels, padding, and activation function. These settings can be customized based on your needs.

[0084] Furthermore, the extracted spatial features are input into the LSTM layer to obtain the extracted spatiotemporal features, and the extracted spatiotemporal features are input into the normalization layer to obtain the predicted number of freight vehicles corresponding to the popular points of interest.

[0085] Then, the test set is used to test, and the accuracy of the prediction results is evaluated by the MAE (mean absolute error) and RMSE (root mean square error) of the capacity prediction results of each point of interest, and a traffic capacity prediction model with better performance is obtained and saved.

[0086] After obtaining the trained traffic capacity prediction model, the historical number of freight vehicles corresponding to multiple popular points of interest are input into the pre-trained traffic capacity prediction model to obtain the prediction results of the number of freight vehicles corresponding to the popular points of interest.

[0087] Specifically, a spatiotemporal matrix is constructed based on the historical number of freight vehicles corresponding to multiple popular points of interest. The constructed spatiotemporal matrix is input into the convolutional layer to obtain the extracted spatial features; the extracted spatial features are input into the LSTM layer to obtain the extracted spatiotemporal features; the extracted spatiotemporal features are input into the normalization layer to obtain the predicted number of freight vehicles corresponding to the popular points of interest.

[0088] Based on this step, a deep neural network model can be used to predict the number of freight vehicles at various popular points of interest within a specific time period in the future. In deep learning, convolutional neural networks can extract spatial features from data; LSTM neural networks can extract temporal features from data; and ConvLSTM neural networks can extract both spatiotemporal and temporal features from data. Therefore, ConvLSTM neural networks can be used to extract the spatiotemporal characteristics of historical traffic capacity, thereby improving the accuracy of capacity forecasts.

[0089] In an optional embodiment, it also includes obtaining a first transportation capacity knowledge graph and a second transportation capacity knowledge graph, and sending the constructed transportation capacity knowledge graph to the front end for display. It can be sent to the front end of a logistics company for display, or it can be sent to the front end of a government department for display. When displaying the data, it can be combined with the road network data in the map, that is, the connection relationship between popular points of interest and logistics companies is displayed on the map. According to the map, popular transportation routes can be intuitively analyzed, and the logistics transportation conditions between regions can be analyzed.

[0090] Optionally, the predicted number of freight vehicles can be marked at multiple popular points of interest to provide data support for logistics companies and vehicle owners to rationally plan transportation routes, arrange freight tasks, analyze carrier risks, and evaluate transportation profits.

[0091] In order to facilitate understanding of the traffic capacity prediction method provided in the embodiment of the present application, the following Figure 2 For explanation. Figure 2As shown, the method includes the following steps.

[0092] First, obtain truck trajectory data, stop data, map POI data, road data, etc. in the test area. Then preprocess the acquired basic data, for example, using the Douglas-Peuker thinning algorithm to thin the truck trajectory data and using the HMM (Hidden Markov Model) to match the trajectory points to the road network.

[0093] Furthermore, the vehicle stop data is filtered, removing stops with short stop times and those at service areas and toll booths to obtain a logistics stop set. These stops are then paired and chronologically determined as transport start and end points. The transport start and end points are then matched with POIs on the map. The number of start and end points corresponding to each POI is counted to determine the number of freight vehicles associated with that POI. This number of freight vehicles associated with each POI is then used to analyze popular POIs.

[0094] Furthermore, a transportation capacity knowledge graph is constructed based on the correspondence between popular POIs and the number of freight vehicles, logistics companies, and vehicle owners.

[0095] Based on the number of freight vehicles corresponding to popular POIs within a certain period of history, a training set and a test set are constructed. The ConvLSTM neural network model is trained and the transport capacity of each POI is predicted based on the trained neural network model.

[0096] According to the traffic capacity prediction method provided by the embodiment of the present application, the vehicle's transportation starting point and end point are analyzed based on the vehicle's historical trajectory data, and then matched with the industry's points of interest to obtain the historical number of freight vehicles at the industry's points of interest, and a traffic capacity knowledge graph is constructed between logistics companies, vehicle owners, points of interest, and the number of vehicles. Then, the capacity situation in the future period is predicted based on the number of freight vehicles at the points of interest and the preset neural network model. Through the analysis of historical capacity and the prediction of future capacity, it is beneficial for logistics companies or individual car owners to avoid their transportation risks, reduce the probability of empty vehicles, improve the transportation efficiency of the entire industry in the region, and increase the profit margins of transportation companies or individual car owners. It is also convenient for regional regulatory authorities to focus on the freight vehicles of large logistics companies in the regulatory industry and the freight routes of related large POI gathering points (ports, factories, logistics parks, etc.) in the regulatory industry.

[0097] The present application also provides a traffic capacity prediction device, which is used to execute the traffic capacity prediction method of the above embodiment. Figure 5 As shown, the device includes:

[0098] Extraction module 501, for extracting the transport starting point and end point of the vehicle based on the trajectory data of the freight vehicle;

[0099] Matching module 502, for matching points of interest in the area to be measured with transport starting points and end points, and obtaining the number of historical freight vehicles corresponding to each point of interest in the area to be measured;

[0100] A screening module 503 is used to screen out a plurality of hot points of interest based on the number of historical freight vehicles corresponding to each point of interest;

[0101] The prediction module 504 is used to input the historical number of freight vehicles corresponding to multiple hot spots of interest into the pre-trained traffic capacity prediction model to obtain the prediction results of the number of freight vehicles corresponding to the hot spots of interest.

[0102] It should be noted that the traffic capacity prediction device provided in the above embodiment, when executing the traffic capacity prediction method, only uses the division of the above-mentioned functional modules as an example. In actual application, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the traffic capacity prediction device provided in the above embodiment and the traffic capacity prediction method embodiment are based on the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0103] The embodiment of the present application further provides an electronic device corresponding to the traffic capacity prediction method provided in the above embodiment, so as to execute the above traffic capacity prediction method.

[0104] Please refer to Figure 6 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 6 As shown, the electronic device includes: a processor 600, a memory 601, a bus 602 and a communication interface 603, and the processor 600, the communication interface 603 and the memory 601 are connected via the bus 602; the memory 601 stores a computer program that can be run on the processor 600, and when the processor 600 runs the computer program, it executes the traffic capacity prediction method provided by any of the aforementioned embodiments of the present application.

[0105] The memory 601 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage. The system network element communicates with at least one other network element via at least one communication interface 603 (which may be wired or wireless), and may utilize the Internet, a wide area network, a local area network, a metropolitan area network, or the like.

[0106] Bus 602 may be an ISA bus, a PCI bus, or an EISA bus. Buses may be classified as address buses, data buses, and control buses. Memory 601 is used to store programs, and processor 600 executes the programs upon receiving execution instructions. The traffic capacity prediction method disclosed in any of the aforementioned embodiments of the present application may be applied to or implemented by processor 600.

[0107] The processor 600 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor 600 or by software instructions. The above processor 600 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of this application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory 601 , and the processor 600 reads the information in the memory 601 and completes the steps of the above method in combination with its hardware.

[0108] The electronic device provided in the embodiment of the present application and the method for predicting traffic capacity provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0109] The present application also provides a computer-readable storage medium corresponding to the traffic capacity prediction method provided in the above embodiment. Figure 7 The computer-readable storage medium shown is a CD 700, on which a computer program (ie, a program product) is stored. When the computer program is run by a processor, it executes the traffic capacity prediction method provided by any of the aforementioned embodiments.

[0110] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0111] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the traffic capacity prediction method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0112] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0113] The above embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for predicting transportation capacity, characterized in that: include: Extract the transportation starting point and end point of the vehicle based on the trajectory data of the freight vehicle; Matching points of interest in the area to be measured with the transport starting point and end point to obtain the number of historical freight vehicles corresponding to each point of interest in the area to be measured; Filter out multiple popular points of interest based on the historical number of freight vehicles corresponding to each point of interest; obtain the freight vehicle corresponding to each popular point of interest; query the logistics and transportation company to which the freight vehicle belongs; construct a first transportation capacity knowledge graph based on the relationship between each popular point of interest, the historical number of freight vehicles, and the logistics and transportation company; obtain the freight vehicle corresponding to each popular point of interest; query the identity information of the owner corresponding to the freight vehicle; construct a second transportation capacity knowledge graph based on the relationship between each popular point of interest, the historical number of freight vehicles, and the identity of the owner; Inputting the historical number of freight vehicles corresponding to multiple popular points of interest into a pre-trained traffic capacity prediction model to obtain the predicted results of the number of freight vehicles corresponding to the popular points of interest; including: constructing a spatiotemporal matrix based on the historical number of freight vehicles corresponding to the multiple popular points of interest, wherein the row vectors of the matrix represent the number of freight vehicles corresponding to different points of interest in the same time period, and the column vectors of the matrix represent the number of freight vehicles corresponding to the same point of interest in different time periods; inputting the constructed spatiotemporal matrix into a convolutional layer to obtain extracted spatial features; inputting the extracted spatial features into an LSTM layer to obtain extracted spatiotemporal features; inputting the extracted spatiotemporal features into a normalization layer to obtain the predicted number of freight vehicles corresponding to the popular points of interest; The first traffic capacity knowledge graph, the second traffic capacity knowledge graph, and the freight vehicle number prediction results corresponding to popular points of interest are displayed on the front end; and combined with the road network data in the map.

2. The method according to claim 1, characterized in that Extract the vehicle's transportation starting point and end point based on the freight vehicle's trajectory data, including: Obtain the trajectory data of vehicles in the test area; Extracting vehicle parking data based on vehicle trajectory data; Delete the stops that are shorter than the preset stop time threshold and the stops that are not logistics stops to obtain the logistics stop point set; The stop points in the logistics stop point set are determined as the transportation starting point and end point in pairs according to time sequence.

3. The method according to claim 1, characterized in that Before inputting the historical freight vehicle counts corresponding to multiple popular points of interest into the pre-trained traffic capacity prediction model, the following steps are also included: Training the traffic capacity prediction model; The traffic capacity prediction model is a Conv-LSTM neural network model.

4. A traffic capacity prediction device, characterized in that: include: An extraction module is used to extract the transportation starting point and end point of the vehicle based on the trajectory data of the freight vehicle; A matching module is used to match the points of interest in the test area with the transportation starting point and end point to obtain the number of historical freight vehicles corresponding to each point of interest in the test area; A screening module is used to screen multiple popular points of interest based on the historical number of freight vehicles corresponding to each point of interest; obtain the freight vehicle corresponding to each popular point of interest; query the logistics and transportation company to which the freight vehicle belongs; construct a first transportation capacity knowledge graph based on the relationship between each popular point of interest and the historical number of freight vehicles and the logistics and transportation company; obtain the freight vehicle corresponding to each popular point of interest; query the identity information of the owner corresponding to the freight vehicle; and construct a second transportation capacity knowledge graph based on the relationship between each popular point of interest and the historical number of freight vehicles and the identity of the owner. The prediction module is used to input the historical number of freight vehicles corresponding to multiple popular points of interest into a pre-trained traffic capacity prediction model to obtain prediction results of the number of freight vehicles corresponding to the popular points of interest, including: constructing a spatiotemporal matrix based on the historical number of freight vehicles corresponding to the multiple popular points of interest, where the row vectors of the matrix represent the number of freight vehicles corresponding to different points of interest in the same time period, and the column vectors of the matrix represent the number of freight vehicles corresponding to the same point of interest in different time periods; inputting the constructed spatiotemporal matrix into the convolution layer to obtain extracted spatial features; inputting the extracted spatial features into the LSTM layer to obtain extracted spatiotemporal features; and inputting the extracted spatiotemporal features into the normalization layer to obtain the predicted number of freight vehicles corresponding to the popular points of interest; The first traffic capacity knowledge graph, the second traffic capacity knowledge graph, and the freight vehicle number prediction results corresponding to popular points of interest are displayed on the front end; and combined with the road network data in the map.

5. A traffic capacity prediction device, characterized in that: The system comprises a processor and a memory storing program instructions, wherein the processor is configured to execute the traffic capacity prediction method according to any one of claims 1 to 3 when executing the program instructions.

6. A computer-readable medium, characterized in that Computer-readable instructions are stored thereon, and the computer-readable instructions are executed by a processor to implement a method for predicting traffic capacity as claimed in any one of claims 1 to 3.

Images