Line transport cargo volume prediction method, device, equipment and medium
By combining regression model and monitoring system data, the problem of low prediction accuracy of logistics line transportation cargo volume is solved, and higher prediction accuracy is achieved, providing effective reference information for operations.
Patent Information
- Application Number
- CN202111444196.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-11-30
- Publication Date
- 2024-12-31
- Estimated Expiration
- 2041-11-30
AI Technical Summary
The existing logistics line transportation cargo forecasting methods rely on manual experience or static routing forecasting, and fail to make full use of capacity monitoring data, resulting in low prediction accuracy and difficulty in meeting actual needs.
The method based on the regression model is adopted, combining the historical transportation volume of the line, attribute information and the forecast transportation volume of the monitoring system, and the model is trained through feature engineering and label value, and the transportation volume prediction results are comprehensively determined.
It significantly improves the accuracy of transportation volume forecasts, provides more effective reference information, and provides operators with more accurate capacity forecasts.
Smart Images
Figure CN114154709B_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of data processing, and in particular to a method, device, equipment and medium for predicting line transportation cargo volume. Background Art
[0002] In the field of logistics and freight, empirical prediction methods, monitoring system forecast data, time series prediction methods and machine learning methods are usually used to predict logistics routes and the transportation volume on the corresponding logistics routes. However, the empirical prediction method relies heavily on manual experience, has high labor costs, and has low manual prediction accuracy. The monitoring system forecast data is generated using the static routing method, without considering the actual dynamic factors. In the actual transportation process, the logistics volume is often affected by traffic, weather, human factors and other factors, resulting in the actual arrival time, departure time, and transportation volume of the transportation volume being inconsistent with the results calculated by the static routing, which in turn leads to low accuracy of the capacity forecast data and is difficult to meet the actual forecast needs. The time series prediction method and the machine learning method mainly use the historical transportation volume of the route as input for prediction, but because they do not make full use of the existing capacity monitoring and forecast data, the prediction accuracy is difficult to improve. Summary of the invention
[0003] The embodiments of the present application provide a method, device, equipment and medium for predicting the volume of cargo transported on a route, which can fully utilize the volume of cargo transported on the route predicted by a monitoring system, combined with the volume of cargo transported on the route predicted based on a regression model, to comprehensively determine the prediction result of the volume of cargo transported on the route, which can effectively improve the accuracy of the prediction and provide effective reference information for operators.
[0004] In a first aspect, an embodiment of the present application further provides a method for predicting route transport cargo volume, the method comprising:
[0005] Obtaining the daily transport volume of the route within the first historical period, attribute information of the route, and the forecast transport volume of the route;
[0006] Determine the predicted transport volume of the route according to the transport volume, the attribute information, the predicted transport volume and the trained regression model;
[0007] The transport volume prediction result of the route is determined according to the forecasted transport volume and the predicted transport volume.
[0008] In a second aspect, an embodiment of the present application further provides a route transportation cargo volume prediction device, the device comprising:
[0009] An acquisition module, used to acquire the daily transport volume of the route within a first historical period, attribute information of the route, and the forecast transport volume of the route;
[0010] A determination module, configured to determine the predicted transport volume of the route according to the transport volume, the attribute information, the predicted transport volume and the trained regression model;
[0011] The determination module is used to determine the transport volume prediction result of the route according to the forecasted transport volume and the predicted transport volume.
[0012] In a third aspect, an embodiment of the present application further provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a method for predicting the volume of line transport cargo as provided in any embodiment of the present application is implemented.
[0013] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for predicting the line transport cargo volume provided in any embodiment of the present application is implemented.
[0014] The embodiment of the present application provides a method, device, equipment and medium for predicting the volume of cargo transported by a route. The method includes: obtaining the volume of cargo transported by the route every day in the first historical period, the attribute information of the route and the predicted volume of cargo transported by the route; determining the predicted volume of cargo transported by the route according to the volume of cargo transported, the attribute information, the predicted volume of cargo transported and the trained regression model; determining the volume forecast result of the route according to the predicted volume of cargo transported and the predicted volume of cargo transported. In this way, on the basis of making full use of the volume of cargo transported predicted by the monitoring system, combined with the volume of cargo transported predicted based on the regression model, the volume forecast result of cargo transported on the route can be comprehensively determined, which can effectively improve the accuracy of the forecast and provide effective reference information for the operators. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 This is a flow chart of a method for predicting the volume of cargo transported by a route in an embodiment of the present application;
[0016] Figure 2 is a flow chart of a method for determining a trained regression model in an embodiment of the present application;
[0017] Figure 3 is a schematic diagram of a sliding time window in an embodiment of the present application;
[0018] Figure 4 It is a structural schematic diagram of a route transportation cargo volume prediction device in an embodiment of the present application;
[0019] Figure 5 It is a structural schematic diagram of another device for predicting the volume of cargo transported by a line in an embodiment of the present application;
[0020] Figure 6It is a schematic diagram of the structure of the computer device in the embodiment of the present application. DETAILED DESCRIPTION
[0021] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only the parts related to the present application, rather than all structures, are shown in the accompanying drawings.
[0022] In addition, in the embodiments of the present application, words such as "optionally" or "exemplarily" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "optionally" or "exemplarily" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "optionally" or "exemplarily" is intended to present related concepts in a specific way.
[0023] Figure 1 This is a flowchart of a method for predicting the volume of cargo transported by a route provided in an embodiment of the present application. This method can be applied to logistics transportation scenarios such as land transportation, water transportation, and air transportation. Based on the forecasted volume of cargo transported and the predicted volume of cargo transported on any route in each scenario, the prediction result of the final volume of cargo transported on the route is comprehensively determined, which can effectively improve the prediction accuracy of the volume of cargo transported on the route. Figure 1 As shown, the method may include but is not limited to the following steps:
[0024] S101. Obtain the daily cargo transportation volume of the route within the first historical period, the attribute information of the route, and the predicted cargo transportation volume of the route.
[0025] The route in the embodiment of the present application is the route for cargo transportation in the field of logistics and transportation, and the route can be multiple routes. Furthermore, in different scenarios of land transportation, water transportation, and air transportation, the route can also be understood as a land transportation route (for example, a highway, a railway, etc.), a water transportation route, an air transportation route, etc. The daily transportation volume on the route in the first historical time can be understood as the actual daily transportation volume on the route in the previous period of time (for example, the previous week or the previous month). For example, taking the LINE01 line and the LINE02 line as examples, Table 1 shows the daily transportation volume on these two lines in the first historical time.
[0026] Table 1
[0027]
[0028]
[0029] The attribute information of the line may include the line identification (Identity, ID), line type (for example, whether the line is a branch line or a trunk line), the starting city, destination city, starting area, etc. of the line. For example, the attribute information of the line may be as shown in Table 2.
[0030] Table 2
[0031] Line ID Line Type Origin City Destination City … Origin area LINE01 Trunk Beijing Guangzhou … North China LINE02 Branch Line Beijing Tianjin … North China … … … … … … LINE100 Trunk Harbin Wuhan … Northeast
[0032] The predicted transport volume of the route can be obtained through real-time forecasting of the monitoring system. For example, in the land logistics scenario, it can be obtained through real-time forecasting of the land transport capacity monitoring system, or in the air logistics scenario, it can be obtained through real-time forecasting of the capacity monitoring system in the air transport field. For example, the forecast time is 24 hours in advance or 48 hours in advance, and the predicted transport volume of the route at different time points can be forecasted in advance per hour. For example, the forecast transport volume on this route 1 hour in advance, the forecast transport volume 2 hours in advance... . For example, taking the LINE01 route in the land logistics scenario as an example, the forecast transport volume on this route based on the forecast of the land transport capacity monitoring system is shown in Table 3.
[0033] Table 3
[0034]
[0035]
[0036] S102. Determine the predicted transport volume of the route according to the transport volume, attribute information, predicted transport volume and the trained regression model.
[0037] Exemplarily, the implementation method of this step may include determining the feature engineering matrix of the route based on the transported cargo volume, attribute information and the first sliding time window, and determining the label value of the route based on the transported cargo volume and the predicted transported cargo volume, and then determining the predicted transported cargo volume of the route based on the feature engineering matrix and the label value and the trained regression model.
[0038] S103. Determine the transport volume forecast result of the route according to the forecasted transport volume and the predicted transport volume.
[0039] After the predicted transport volume of the route is obtained based on the trained regression model, the predicted transport volume is added to the forecasted transport volume, and the sum of the two is determined as the predicted result of the transport volume of the route.
[0040] The embodiment of the present application provides a method for predicting the volume of cargo transported by a route, the method comprising: obtaining the volume of cargo transported by the route every day in the first historical period, the attribute information of the route and the predicted volume of cargo transported by the route; determining the predicted volume of cargo transported by the route according to the volume of cargo transported, the attribute information, the predicted volume of cargo transported and the trained regression model; determining the predicted volume of cargo transported by the route according to the predicted volume of cargo transported and the predicted volume of cargo transported. In this way, on the basis of making full use of the volume of cargo transported by the route predicted by the monitoring system, combined with the predicted volume of cargo transported by the corresponding route predicted based on the regression model, the volume of cargo forecast on the route is comprehensively determined, which can effectively improve the accuracy of the forecast and provide effective reference information for the operators.
[0041] like Figure 2 As shown, in one example, the implementation method of obtaining the regression model trained in the above step S102 may include but is not limited to the following methods:
[0042] S201. Obtain the daily cargo volume transported by the route in the second historical period, the historical forecast cargo volume transported by the route, and attribute information of the route.
[0043] The second historical time in this step is the time when the training samples are selected in the regression model training phase. For example, assuming that the second historical time is 3 months, the daily transport volume on the route in the first 3 months is obtained. In other words, the difference between the first historical time and the second historical time is that the first historical time is the time when the samples are selected in the regression model use phase, and the second historical time is the time when the samples are selected in the regression model training phase. The time lengths of the first historical time and the second historical time can be the same or different. The form of the daily transport volume of the route in the second historical time can be shown in Table 1. Among them, the above samples can be understood as the daily transport volume on each route.
[0044] The historical forecasted cargo volume of the route is the forecasted volume of cargo transported on the route every day during the second historical time by the monitoring system, that is, the predicted cargo volume transported on each route in the sample every day. Furthermore, the time granularity of the historical forecasted cargo volume can be hourly, for example, 24 hours or 48 hours in advance, and the forecasted cargo volume of the route is updated every hour. The forecast record format of the historical forecasted cargo volume of the route can be as shown in Table 3.
[0045] The attribute information of the route can be applied to the regression model training stage to combine with other data for model training, and can also be applied to the regression model use stage to combine with other data to predict the predicted transportation volume on the route in the next 24 hours or 48 hours at the current moment.
[0046] S202: Determine a label value of the route according to the daily cargo volume transported by the route in the second historical period and the historical forecast cargo volume transported by the route.
[0047] Exemplarily, the method of determining the line label value in this step may include subtracting the line's historical forecasted transport volume from the line's daily transport volume within the second historical time, and determining the difference as the line's current label value. For example, taking the forecasted transport volume on the line predicted by the monitoring system 48 hours in advance as an example, as shown in Table 4, the actual transport volume of the line every day within the second historical time minus the forecasted transport volume on the line 1 to 48 hours in advance is the label value corresponding to the line. Among them, a positive label value indicates that the forecasted transport volume is less than the actual transport volume of the line every day within the second historical time, and a negative label value indicates that the forecasted transport volume is less than the actual transport volume of the line every day within the second historical time. Each line can have multiple labels for model training every day. As shown in Table 4, LINE0 has 48 label values that can be used for regression model training.
[0048] Table 4
[0049]
[0050]
[0051] It should be noted that the above Table 1 only shows the label values of LINE01 that are 1 to 48 hours in advance every day. The table may also include label values of other lines that are 1 to 48 hours in advance every day.
[0052] Optionally, in step S102, the label value of the route is determined according to the daily transport volume in the first historical time and the forecast transport volume of the route, and the above method can also be adopted. That is, the forecast transport volume of the route is subtracted from the daily transport volume in the first historical time, and the difference is determined as the current label value of the route.
[0053] S203. Determine a feature engineering matrix of the route according to the daily transport volume of the route in the second historical time and the attribute information of the route.
[0054] Exemplarily, the implementation of this step may include extracting the characteristic value of the daily transport volume of the route within the second historical time based on the second sliding time window, such as Figure 3 As shown. The second sliding time window slides backward from the current time, and the extracted characteristic values may include the mean, maximum, minimum, median, etc. of the transported cargo volume in the second sliding time window. Optionally, the length of the second sliding time window can be adjusted. For example, a fixed time length is first used as the second sliding time window to determine each characteristic value in the time window, and then the time length of the second sliding time window is adjusted to continue to determine each characteristic value in the time window. Exemplarily, taking LINE01 and LINE02 as examples, Table 4 shows the characteristic values of the transported cargo volume in different second sliding time windows on each line.
[0055] Table 4
[0056]
[0057] By combining the feature values of the route obtained based on the second sliding time window with the attribute information of the route, a feature engineering matrix of the corresponding route can be obtained.
[0058] Correspondingly, the method for implementing the feature engineering matrix of the route in step S102 according to the transported cargo volume, attribute information and the first sliding time window can also adopt the same method as described above, that is, extracting the characteristic value of the transported cargo volume of the route every day within the first historical time based on the first sliding time window; merging the characteristic value with the attribute information to obtain the feature engineering matrix of the route.
[0059] It can be understood that the first sliding time window is used to extract the characteristic value of the daily transportation volume of the route within the first historical time in the regression model use phase, and the second sliding time window is used to extract the characteristic value of the daily transportation volume of the route within the second historical time in the regression model training phase. Optionally, the time lengths of the first sliding time window and the second sliding time window may be the same or different.
[0060] S204: Train a regression model according to the label value and the feature engineering matrix to obtain a trained regression model.
[0061] After obtaining the label value and feature engineering training matrix of the training sample based on the above steps, they are input into the regression model for training, and the model parameters are continuously optimized. Exemplarily, the regression model used in the embodiment of the present application may include an XGBOOST model, a LIGHTGBM model, a linear regression model, etc. The optimized model parameters are obtained based on the training sample training, and the trained regression model is stored in the server for the later prediction of the transport volume of the route.
[0062] In the embodiment of the present application, the difference between the actual daily transport volume on the route and the transport volume predicted by the monitoring system on the route is used as the label value, and the model prediction is performed in combination with the attribute information on the corresponding route. Further, the result of the model prediction is added to the result of the monitoring system forecast, which can make full use of the real-time forecast data of the monitoring system on the route and the model forecast data on the route, thereby effectively improving the prediction accuracy of the transport volume on the route. Offline experiments show that the method provided by the embodiment of the present application is more than 10% more accurate than directly predicting the transport volume of the route, and can achieve real-time prediction with hourly granularity as the object.
[0063] Figure 4 A schematic diagram of a route transport cargo volume prediction device provided in an embodiment of the present application is shown in FIG. Figure 4As shown, the device may include: an acquisition module 401, a determination module 402;
[0064] The acquisition module is used to obtain the daily transport volume of the route within the first historical period, the route's attribute information, and the route's forecast transport volume;
[0065] A determination module, used to determine the predicted transport volume of the route according to the transport volume, attribute information, the predicted transport volume and the trained regression model;
[0066] The determination module is used to determine the transport volume forecast result of the route according to the forecasted transport volume and the predicted transport volume.
[0067] In one example, a determination module is used to determine a feature engineering matrix of a route based on the transported cargo volume, attribute information and a first sliding time window; determine a label value of the route based on the transported cargo volume and the predicted transported cargo volume; and determine the predicted transported cargo volume of the route based on the feature engineering matrix and the label value and a trained regression model.
[0068] like Figure 5 As shown, in one example, the above device may further include a training module 403;
[0069] The training module may include an acquisition unit and a determination unit;
[0070] The acquisition unit is used to acquire the daily transport volume of the route within the second historical time, the historical forecast transport volume of the route and the attribute information of the route;
[0071] A determination unit, used to determine a label value of the route according to the daily transport volume of the route within the second historical time and the historical forecast transport volume of the route;
[0072] The determination unit is also used to determine the feature engineering matrix of the route according to the daily transport volume of the route within the second historical time and the attribute information of the route; and to train the regression model according to the label value and the feature engineering matrix to obtain the trained regression model.
[0073] Furthermore, the determination unit is also used to subtract the historical forecasted cargo volume of the route from the daily cargo volume transported within the second historical time of the route, and determine the difference as the current label value of the route; and extract the characteristic value of the daily cargo volume transported within the second historical time of the route based on the second sliding time window, merge the characteristic value with the attribute information of the route, and obtain the feature engineering matrix of the route.
[0074] In one example, the determination module is used to subtract the predicted transport volume of the route from the daily transport volume of the route in the first historical time, and use the obtained difference to determine the current label value of the route;
[0075] In one example, the above-mentioned determination module is used to extract the characteristic value of the daily transportation volume of the route within the first historical time based on the first sliding time window; and merge the characteristic value with the attribute information to obtain the feature engineering matrix of the route.
[0076] In one example, a determination module is used to add the forecasted transport volume and the predicted transport volume to obtain the sum of the two; and determine the sum as the transport volume prediction result of the route.
[0077] The above-mentioned route transport cargo volume prediction device can execute Figure 1-Figure 2 The provided method for predicting the volume of cargo transported by a line has corresponding devices and beneficial effects in the method.
[0078] Figure 6 A schematic diagram of the structure of a computer device provided in Example 6 of the present invention, such as Figure 6 As shown, the computer device includes a controller 601, a memory 602, an input device 603, and an output device 604; the number of the controller 601 in the computer device can be one or more. Figure 6 A controller 601 is taken as an example; the controller 601, the memory 602, the input device 603 and the output device 604 in the computer device can be connected by a bus or other means. Figure 6 The example of connecting through bus is taken in the following.
[0079] The memory 602 is a computer-readable storage medium that can be used to store software programs, computer executable programs, and modules, such as Figure 1-Figure 2 Program instructions / modules corresponding to the line transport volume prediction method in the embodiment (e.g., the acquisition module 401 and the determination module 402 in the line transport volume prediction device). The controller 601 executes the software programs, instructions and modules stored in the memory 602 to perform various functions and data processing of the computer device, that is, to implement the above-mentioned line transport volume prediction method.
[0080] The memory 602 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the computer, etc. In addition, the memory 602 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 602 may further include a memory remotely arranged relative to the controller 601, and these remote memories may be connected to the terminal / server via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0081] The input device 603 may be used to receive input digital or character information and generate key signal input related to user settings and function control of the computer device. The output device 604 may include a display device such as a display screen.
[0082] The present application also provides a storage medium containing computer executable instructions, which are used to execute a method for predicting the volume of freight transported by a line when the computer controller executes the instructions. The method includes: Figure 1 , Figure 2 Steps shown.
[0083] Through the above description of the implementation method, the technicians in the relevant field can clearly understand that the present application can be implemented with the help of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application can be essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.
[0084] It is worth noting that the modules included in the above-mentioned line transport cargo volume prediction device are only divided according to functional logic, but are not limited to the above-mentioned division method. As long as the corresponding functions can be achieved, it is not used to limit the protection scope of this application.
[0085] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.
Claims
1. A method for predicting the volume of freight transported by a route, characterized in that: include: Obtaining the daily transport volume of the route within the first historical period, attribute information of the route, and the forecast transport volume of the route; Determine the predicted transport volume of the route according to the transport volume, the attribute information, the predicted transport volume and the trained regression model; Determine the forecast result of the transport volume of the route according to the forecast transport volume and the forecast transport volume; The predicted transport volume of the route is determined according to the transport volume, the attribute information, the predicted transport volume and the trained regression model, including: Determine a feature engineering matrix for the route according to the transported cargo volume, the attribute information and the first sliding time window; Subtract the predicted transportation volume of the route from the daily transportation volume of the route in the first historical period, and determine the difference as the current label value of the route; The predicted transport volume of the route is determined according to the feature engineering matrix, the label values and the trained regression model.
2. The method according to claim 1, characterized in that The determining of the feature engineering matrix of the line comprises: Extracting characteristic values of the daily transport volume of the route within the first historical time based on the first sliding time window; The eigenvalues are combined with the attribute information to obtain a feature engineering matrix for the line.
3. The method according to claim 1 or 2, characterized in that: The determining of the transport volume prediction result of the route according to the forecast transport volume and the predicted transport volume includes: Adding the forecasted transport volume to the predicted transport volume to obtain a sum of the two; The sum is determined as the forecast result of the transport volume of the route.
4. A method for predicting the volume of freight transported by a route, characterized in that: include: Obtaining the daily transport volume of the route within the first historical period, attribute information of the route, and the forecast transport volume of the route; Determine the predicted transport volume of the route according to the daily transport volume of the route in the first historical period, the attribute information, the predicted transport volume, and the trained regression model; Determine the forecast result of the transport volume of the route according to the forecast transport volume and the forecast transport volume; The trained regression model is determined in the following way: Obtaining the daily transport volume of the route within the second historical period, the historical forecast transport volume of the route, and attribute information of the route; Subtract the historical forecasted cargo volume of the route from the daily cargo volume of the route during the second historical period, and determine the difference as the current label value of the route; Determine a feature engineering matrix of the route according to the daily transport volume of the route within the second historical period and the attribute information of the route; A regression model is trained according to the label value and the feature engineering matrix to obtain a trained regression model.
5. The method according to claim 4, characterized in that The determining of the feature engineering matrix of the line comprises: Extracting characteristic values of the daily transport volume of the route within the second historical time based on the second sliding time window; The eigenvalues are combined with the attribute information of the line to obtain a feature engineering matrix of the line.
6. The method according to claim 4 or 5, characterized in that: The determining of the transport volume prediction result of the route according to the forecast transport volume and the predicted transport volume includes: Adding the forecasted transport volume to the predicted transport volume to obtain a sum of the two; The sum is determined as the forecast result of the transport volume of the route.
7. A route transport cargo volume prediction device, characterized in that: include: An acquisition module, used to acquire the daily transport volume of the route within a first historical period, attribute information of the route, and the forecast transport volume of the route; A determination module, configured to determine the predicted transport volume of the route according to the transport volume, the attribute information, the predicted transport volume and the trained regression model; The determination module is used to determine the transport volume prediction result of the route according to the forecast transport volume and the predicted transport volume; The determination module is specifically used for: Determine a feature engineering matrix for the route according to the transported cargo volume, the attribute information and the first sliding time window; Subtract the predicted transportation volume of the route from the daily transportation volume of the route in the first historical period, and determine the difference as the current label value of the route; The predicted transport volume of the route is determined according to the feature engineering matrix, the label values and the trained regression model.
8. A route transport cargo volume prediction device, characterized in that: include: An acquisition module, used to acquire the daily transport volume of the route within a first historical period, attribute information of the route, and the forecast transport volume of the route; A determination module, configured to determine the predicted transport volume of the route according to the daily transport volume of the route within the first historical period, the attribute information, the predicted transport volume, and the trained regression model; The determination module is used to determine the transport volume prediction result of the route according to the forecast transport volume and the predicted transport volume; The device also includes a training module, which includes an acquisition unit and a determination unit; The acquisition unit is used to acquire the daily transport volume of the route within the second historical period, the historical forecast transport volume of the route and the attribute information of the route; The determining unit is used to subtract the historical forecast transportation volume of the route from the daily transportation volume of the route during the second historical period, and determine the obtained difference as the current label value of the route; The determining unit is further used to determine the feature engineering matrix of the route according to the daily transport volume of the route within the second historical period and the attribute information of the route; And, a regression model is trained according to the label value and the feature engineering matrix to obtain a trained regression model.
9. A computer device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for predicting the volume of line transport cargo as described in any one of claims 1 to 6 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting the line transportation cargo volume as described in any one of claims 1-6 is implemented.