Urban road material stock prediction regression method based on gbdt algorithm

Abstract

This invention discloses a regression method for predicting urban road material inventory based on the GBDT algorithm, comprising the following steps: segmenting road network information into regions using ArcMap on GIS road network vector maps at various time points, determining parameters such as road width, pavement thickness, and density and admixture of road construction materials; calculating the material inventory of different road construction materials using Python programming; summarizing and organizing the material inventory data, and obtaining area, population, and economic data within the calculation area to construct a feature variable dataset, while converting the categorical feature variables into binary vectors using One-hot encoding; dividing the sample set into a training set and a validation set; training a material inventory prediction regression model based on the GBDT algorithm; evaluating the model's adaptability and validating it on an independent test set. This invention establishes a prediction regression model for the material inventory of various road construction materials in the road system, achieving high prediction accuracy.

Description

Technical Field

[0001] This invention belongs to the field of road material metabolism technology, and more specifically, relates to a regression method for predicting urban road material stock based on the GBDT algorithm. Background Technology

[0002] With the continuous development of my country's economy and the ongoing improvement of transportation infrastructure, the scale of the comprehensive transportation network is constantly expanding, and the material stock within the transportation system is also continuously increasing. To maintain its current normal operating status, the nationwide road network requires regular maintenance, reconstruction, and expansion, necessitating a continuous influx of building materials and energy. Predicting the future material input required into the road system is fundamental for optimizing resource allocation and improving efficiency when formulating road development plans. Therefore, for regions with gradually improving road systems, this study focuses on constructing a model capable of predicting the material stock of various types of roads of different grades within a specific region over the next year, thus providing a reference for road system planning and development.

[0003] GBDT (Gradient Boosting Decision Tree) is a representative algorithm in the Boosting family of algorithms, and it is an iterative decision tree algorithm. The GBDT model consists of multiple decision trees, and the final result is obtained by ensemble training and accumulation of each weak machine learning decision tree. It belongs to the ensemble learning algorithm and has strong generalization ability. Compared with traditional regression methods, GBDT has advantages such as higher prediction accuracy, stronger data adaptability, and better adaptability to nonlinear correlated data.

[0004] Currently, there is relatively little research on predictive models for the internal material inventory of road systems in the field of road material metabolism. Summary of the Invention

[0005] The main objective of this invention is to provide a regression method for predicting urban road material inventory based on the GBDT algorithm, which can directly predict the material inventory of various materials of different grades of roads in a certain area in the coming year without obtaining the length of the internal road network.

[0006] To address at least one of the aforementioned technical problems, according to one aspect of the present invention, a method for predicting and regressing urban road material inventory based on the GBDT algorithm is provided, comprising the following steps:

[0007] Step 1. Based on ArcMap, segment the GIS road network vector map of each time point into different areas to obtain road pavement structure and cross-section design data within the calculation area directly or indirectly, determine parameters such as road width, pavement thickness, and density and admixture of road construction materials, and use Python programming to calculate the material inventory of different road construction materials at each time point and in each sub-region.

[0008] Step 2. Summarize and organize the material inventory data, and obtain relevant data on area, population, and economy within the calculation area to construct a feature variable dataset. At the same time, use One-hot encoding to convert the categorical feature variables into binary vectors.

[0009] Step 3. Divide the sample set into a training set and a validation set;

[0010] Step 4. Train the material stock prediction regression model based on the GBDT algorithm;

[0011] Step 5. Evaluate the model's adaptability and validate it on an independent test set.

[0012] Furthermore, the specific steps of segmenting the GIS road network vector map for each time point based on ArcMap in step 1 include:

[0013] Step 111. In the ArcMap main menu, select Geoprocessing, then select Arctoolbox, then select Extract & Analyze in the Analysis Tools, and use the clipping function to segment the road network and node information in the road network vector map;

[0014] Step 112. Select the batch processing function in the clipping function of Step 11, select the road network vector file and node file of a certain time node as input elements, select the sub-region base map as clipping elements, and batch process the clipping to obtain the road network and node files of each sub-region;

[0015] Step 113. Repeat steps 111 and 112 for the GIS road network vector map at each time point to obtain the road network and node shp files for each sub-region at each time point.

[0016] Furthermore, in step 1, parameters such as road width, pavement thickness, and density and dosage of road construction materials are determined. Based on the road pavement structure and cross-sectional design data within the calculation area, or according to the "Urban Road Engineering Design Code," on-site surveys are conducted, and relevant literature is reviewed to determine the road width W corresponding to grade j road. j The thickness H of the first layer of the road cross section jl Widening value W for one side of intersecting roads j1s and W j2s The compaction degree c of the i-th type of mixture iand dry density ρ i The content α of the k-th material in the i-th mixture k Parameters such as these.

[0017] Furthermore, the step 1, which involves using Python programming to calculate the material inventory of different road construction materials, includes:

[0018] Step 131. Save the dbf files of road networks and nodes from the ArcGIS vector map file as xlsx files. Use the pandas library in Python to call the file summary calculation to obtain the total length L of roads of level j. j The number of T-shaped intersections is N3, and the number of cross-shaped intersections is N4;

[0019] Step 132. Calculate the total volume RS of the i-th type of mixture for different road grades within the road system at each time point. i :

[0020]

[0021] Where j represents the road grade, n1 represents the number of road grade types, l represents the l-th pavement structure layer, n2 represents the total number of pavement structure layers, and L j W represents the length of road of grade j. j H represents the road width corresponding to road level j. jl This indicates the thickness of the l-th layer in the cross-section of a road of grade j.

[0022] Step 133. Calculate the total volume RI of the i-th mixture that has been repeatedly calculated within the intersection area. i :

[0023]

[0024] Where j1 and j2 represent the road grades of the two intersecting roads (j2 has a higher road grade than j1), W j1 W j2 W represents the width of the intersecting roads. j1s W j2s H represents the widening value on one side of the intersecting roads. j2 P represents the thickness of the higher-grade road in two intersecting roads. j1j2 This represents the number of road intersection transitions for road grades j1 and j2:

[0025]

[0026] In the above formula, N3 represents the number of T-shaped intersections and N4 represents the number of cross-shaped intersections.

[0027] Step 134. Calculate the total stock of the i-th type of mixture within the road system, MS.i :

[0028] MS i =c i ρ i [RS i -RI i (4)

[0029] Among them, c i ρ i These represent the compaction degree and dry density of the i-th mixture, respectively.

[0030] Step 135. Calculate the total stock of material k in the road system, MS. k :

[0031] MS k =∑MS i ·α k (5)

[0032] Where, α k This represents the content of the k-th material in the i-th mixture.

[0033] Furthermore, the material inventory data mentioned in step 2 is summarized and organized, and relevant data on area, population, and economy within the calculation area are obtained to construct a feature variable dataset, specifically including:

[0034] Step 211. Summarize and organize the material inventory data obtained in Step 1 along with its corresponding time node, sub-region, road grade, material type, surface layer thickness, base layer thickness, road width, and other parameters.

[0035] Step 212. Locate the statistical yearbook or relevant statistical data of the calculation area to obtain the area, population, and economic data indicators of each sub-region within the calculation area, and summarize and organize each material stock data according to the above.

[0036] Step 213. Construct a feature variable dataset based on the various data indicators related to material inventory mentioned above.

[0037] Furthermore, in step 2, the classification feature variables are converted into binary vectors using One-hot encoding. Specifically, the two classification feature variables, road grade and material type, are represented using the One-hot encoding principle to ensure that each state of each sample has only one bit that is state 1, and the rest of the bits are 0.

[0038] Furthermore, in step 3, the sample set is divided into a training set and a validation set. The training set and validation set are randomly divided using the train_test_split() function in the sklearn library, and the ratio of the number of samples in the training set to the number of samples in the validation set is set.

[0039] Furthermore, the material stock prediction regression model trained based on the GBDT algorithm in step 4 uses time node, surface layer thickness, base layer thickness, road width, area, population, and GDP as prediction variables, while also considering material type and road grade as classification feature variables. After training, the training set accuracy and validation set accuracy of the model are obtained.

[0040] Furthermore, in step 5, the model is evaluated for adaptability and validated on an independent test set. The trained material inventory prediction model is used to make predictions on the independent test set and compared with the actual values. The mean relative error (MRE) and the R-squared coefficient of determination (R2) are calculated, and the model adaptability is evaluated through error analysis.

[0041] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the urban road material inventory prediction regression method based on the GBDT algorithm of the present invention.

[0042] According to another aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the urban road material inventory prediction regression method based on the GBDT algorithm of the present invention.

[0043] Compared with the prior art, the present invention has at least the following beneficial effects:

[0044] This invention processes ArcGIS road network information and uses a bottom-up material flow analysis method to calculate the raw data of material stock for different sub-regions, different road construction materials, and different road grades at various time points. Finally, based on the dataset, a material stock prediction model is constructed using the GBDT algorithm. The method provided by this invention can reasonably predict the material stock of various road construction materials in a certain region at a future time point, even with nonlinear datasets and unknown road network route lengths. The prediction error is within a reasonable range, and the prediction accuracy is high. This invention not only considers the impact of road-related parameters on the material stock of the urban road system but also the influence of external factors such as regional area, population, and economy, providing a reference for planning the future construction and development direction of urban road systems. Attached Figure Description

[0045] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments will be briefly described below. Obviously, the drawings described below only relate to some embodiments of the present invention and are not intended to limit the present invention.

[0046] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0047] Figure 2 This is a schematic diagram of the road network vector map of a certain city in Embodiment 1 of the present invention, segmented using ArcMap.

[0048] Figure 3 This is a flowchart of the material inventory calculation process in this invention;

[0049] Figure 4 This is a comparison chart of predicted values ​​and actual values ​​obtained using the GBDT model on the test set in an embodiment of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention.

[0051] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0052] like Figure 1-4 As shown,

[0053] Example 1:

[0054] This embodiment presents a regression method for predicting urban road material inventory based on the GBDT algorithm, such as... Figure 1 As shown, it includes the following steps:

[0055] In step 1, the road network information is segmented by region based on the GIS road network vector map at each time point using ArcMap. The specific steps include:

[0056] Step 111. In the ArcMap main menu, select Geoprocessing, then select Arctoolbox, then select Extract & Analyze in the Analysis Tools, and use the clipping function to segment the road network and node information in the road network vector map;

[0057] Step 112. Select the batch processing function in the clipping function of Step 11, select the road network vector file and node file of a certain time node as input elements, select the sub-region base map as clipping elements, and batch process the clipping to obtain the road network and node files of each sub-region;

[0058] Step 113. Repeat steps 111 and 112 for the GIS road network vector map at each time point to obtain the road network and node shp files for each sub-region at each time point.

[0059] Specifically, in this embodiment, the preset area is Nanjing City, and the time frame is one year. Using the clipping function in ArcMap, the GIS road network vector information of Nanjing City from 2014 to 2021 is segmented according to the administrative divisions as described in step 1. Figure 2 The image shows the segmentation results of road network information in the 2017 road network vector map of Nanjing.

[0060] In step 1, the road pavement structure and cross-sectional design data within the calculation area are obtained directly or indirectly to determine parameters such as road width, pavement thickness, and density and dosage of road construction materials.

[0061] Specifically, in this embodiment, in accordance with the "Code for Design of Urban Road Engineering", on-site surveys were conducted and relevant literature was reviewed. The road width, the thickness of the first layer of the road cross section, and the widening value of one side of intersecting roads are summarized in Tables 1, 2, and 3. The density and admixture parameters of road construction materials are summarized in Table 4.

[0062] Table 1. Road width parameters for each grade

[0063]

[0064]

[0065] Table 2 Thickness parameters of road structural layers at different levels

[0066]

[0067] Table 3 Widening values ​​for one side of intersecting roads

[0068]

[0069] Table 4 Road Construction Material Parameters

[0070]

[0071] In step 1, Python programming is used to calculate the material inventory of different road construction materials. Figure 3 The flowchart for calculating material inventory in this invention is shown, and the calculation steps are as follows:

[0072] Step 131. Save the dbf files of road networks and nodes from the ArcGIS vector map file as xlsx files. Use the pandas library in Python to call the file summary calculation to obtain the total length L of roads of level j. jThe number of T-shaped intersections is N3, and the number of cross-shaped intersections is N4;

[0073] Step 132. Calculate the total volume RS of the i-th type of mixture for different road grades within the road system at each time point. i :

[0074]

[0075] Where j represents the road grade, n1 represents the number of road grade types, l represents the l-th pavement structure layer, n2 represents the total number of pavement structure layers, and L j W represents the length of road of grade j. j H represents the road width corresponding to road level j. jl This indicates the thickness of the l-th layer in the cross-section of a road of grade j.

[0076] Step 133. Calculate the total volume RI of the i-th mixture that has been repeatedly calculated within the intersection area. i :

[0077]

[0078] Where j1 and j2 represent the road grades of the two intersecting roads (j2 has a higher road grade than j1), W j1 W j2 W represents the width of the intersecting roads. j1s W j2s H represents the widening value on one side of the intersecting roads. j2 P represents the thickness of the higher-grade road in two intersecting roads. j1j2 This represents the number of road intersection transitions for road grades j1 and j2:

[0079]

[0080] In the above formula, N3 represents the number of T-shaped intersections and N4 represents the number of cross-shaped intersections.

[0081] Step 134. Calculate the total stock of the i-th type of mixture within the road system, MS. i :

[0082] MS i =c i ρ i [RS i -RI i ]

[0083] Among them, c i ρ i These represent the compaction degree and dry density of the i-th mixture, respectively.

[0084] Step 135. Calculate the total stock of material k in the road system, MS. k :

[0085] MS k =∑MS i ·α k

[0086] Where, α k This represents the content of the k-th material in the i-th mixture.

[0087] Specifically, this example calculates the material stock over the eight years from 2014 to 2021. Roads are classified into four categories: expressways, arterial roads, secondary arterial roads, and local roads. The calculation results are saved as an xlsx file.

[0088] In step 2, the material inventory data is summarized and organized, and data related to area, population, and economy within the calculation area are obtained to construct a feature variable dataset, specifically including:

[0089] Step 211. Summarize and organize the material inventory data obtained in Step 1 along with its corresponding time node, sub-region, road grade, material type, surface layer thickness, base layer thickness, road width, and other parameters.

[0090] Step 212. Locate the statistical yearbook or relevant statistical data of the calculation area to obtain the area, population, and economic data indicators of each sub-region within the calculation area, and summarize and organize each material stock data according to the above.

[0091] Step 213. Construct a feature variable dataset based on the various data indicators related to material inventory mentioned above.

[0092] Specifically, in this embodiment, the selected feature variables are year, road grade, material type, surface layer thickness, base layer thickness, road width, area, population, and GDP. Each material stock data obtained is matched with the above feature variables to form a sample. The total number of samples that can be used in this embodiment is 1584.

[0093] In step 2, one-hot encoding is used to convert the categorical feature variables into binary vectors. Specifically, the two categorical feature variables, road grade and material type, are represented using the one-hot encoding principle to ensure that each state of each sample has only one bit that is state 1, and the rest of the bits are 0.

[0094] In step 3, the sample set is divided into a training set and a validation set. The training set and validation set are randomly divided using the train_test_split() function in the sklearn library, and the ratio of the number of samples in the training set to the number of samples in the validation set is set.

[0095] Specifically, in this embodiment, when training the GBDT material stock prediction model, samples from 2014 to 2020 are used, with a total sample set of 1386. The ratio of the number of samples in the training set to the number of samples in the validation set is 0.2, that is, 1108 samples are selected as training set samples and the other 278 are validation set samples.

[0096] In step 4, a material stock prediction regression model is trained based on the GBDT algorithm. The prediction variables are time node, surface layer thickness, base layer thickness, road width, area, population, and GDP. At the same time, material type and road grade are considered as classification feature variables. After training, the training set accuracy and validation set accuracy of the model are obtained.

[0097] Specifically, in this embodiment, the parameters obtained after parameter tuning are shown in Table 5. After training, the final model training set accuracy is 99.99%, and the validation set accuracy is 98.61%.

[0098] Table 5 GBDT Model Parameters

[0099]

[0100]

[0101] In step 5, the model is evaluated for adaptability and validated on an independent test set. The trained material stock prediction model is used to make predictions on the independent test set and compared with the actual values. The mean relative error (MRE) and the R-squared coefficient of determination (R2) of the model are calculated. The adaptability of the model is evaluated through error analysis.

[0102] Specifically, in this embodiment, the sample size of material stock data for 2021 is 198, and the trained model is used to predict the 198 sets of data. Figure 4 A comparison chart of the actual and predicted values ​​is shown. Combining the comparison chart of the actual and predicted material stock, the predicted material stock values ​​are generally close to the actual values, indicating a high degree of model fit and high prediction accuracy. In this embodiment, the calculated model determination coefficient R² = 0.999999 and the overall mean relative error MRE = 0.238288. The model has high accuracy and good fit. Although there is some prediction error, it provides an effective method for predicting the amount of various road construction materials required for a road system in a certain area in the future, and has certain reference value for road metabolism research and further road construction planning.

[0103] Example 2:

[0104] The computer-readable storage medium of this embodiment stores a computer program that, when executed by a processor, implements the steps in the urban road material inventory prediction regression method based on the GBDT algorithm of Embodiment 1.

[0105] The computer-readable storage medium in this embodiment can be an internal storage unit of the terminal, such as the terminal's hard disk or memory; the computer-readable storage medium in this embodiment can also be an external storage device of the terminal, such as a plug-in hard disk, smart memory card, secure digital card, flash memory card, etc. equipped on the terminal; furthermore, the computer-readable storage medium can include both the terminal's internal storage unit and external storage devices.

[0106] The computer-readable storage medium of this embodiment is used to store computer programs and other programs and data required by the terminal. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0107] Example 3:

[0108] The computer device of this embodiment includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the urban road material inventory prediction regression method based on the GBDT algorithm of Embodiment 1.

[0109] In this embodiment, the processor can be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The memory can include read-only memory and random access memory, and provides instructions and data to the processor. A portion of the memory can also include non-volatile random access memory. For example, the memory can also store device type information.

[0110] Those skilled in the art will understand that the content disclosed in the embodiments can be provided as a method, system, or computer program product. Therefore, this solution can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this solution can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage) containing computer-usable program code.

[0111] This solution is described with reference to flowchart illustrations and / or block diagrams of methods and computer program products according to embodiments of this solution. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0112] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0113] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0114] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0115] The examples described herein are merely preferred embodiments of the invention and are not intended to limit the concept and scope of the invention. Any modifications and improvements made by those skilled in the art to the technical solutions of the invention without departing from the design concept of the invention should fall within the protection scope of the invention.

Claims

1. A regression method for predicting urban road material inventory based on the GBDT algorithm, characterized in that, Includes the following steps: Step 1. Based on ArcMap, segment the GIS road network vector map of each time point into different areas to obtain road pavement structure and cross-section design data within the calculation area directly or indirectly, determine road parameters, and use Python programming to calculate the material inventory of different road construction materials at each time point and in each sub-region. Step 2. Summarize and organize the material inventory data, and obtain relevant data on area, population, and economy within the calculation area to construct a feature variable dataset. At the same time, use One-hot encoding to convert the categorical feature variables into binary vectors. Step 3. Divide the sample set into a training set and a validation set; Step 4. Train the material stock prediction regression model based on the GBDT algorithm; Step 5. Evaluate the model's adaptability and validate it on an independent test set.

2. The method according to claim 1, characterized in that, Step 1, which involves segmenting the GIS road network vector map for each time point into different regions using ArcMap, specifically includes: Step 111. In the ArcMap main menu, select Geoprocessing, then select Arctoolbox, then select Extract & Analyze in the Analysis Tools, and use the clipping function to segment the road network and node information in the road network vector map; Step 112. Select the batch processing function in the clipping function of Step 111, select the road network vector file and node file of a certain time node as input elements, select the sub-region base map as clipping elements, and batch process the clipping to obtain the road network and node files of each sub-region; Step 113. Repeat steps 111 and 112 for the GIS road network vector map at each time point to obtain the road network and node shp files for each sub-region at each time point.

3. The method according to claim 1, characterized in that, Step 1, determining the road parameters, specifically involves: determining the road width W corresponding to road level j. j The thickness H of the first layer of the road cross section jl Widening value W for one side of intersecting roads j1s and W j2s The compaction degree c of the i-th type of mixture i and dry density ρ i The content α of the k-th material in the i-th mixture k .

4. The method according to claim 3, characterized in that, Step 1, which involves using Python programming to calculate the material inventory of different road construction materials, includes: Step 131. Save the dbf files of road networks and nodes from the ArcGIS vector map file as xlsx files. Use the pandas library in Python to call the file summary calculation to obtain the total length L of roads of level j. j The number of T-shaped intersections is N3, and the number of cross-shaped intersections is N4; Step 132. Calculate the total volume RS of the i-th type of mixture for different road grades within the road system at each time point. i : Where j represents the road grade, n1 represents the number of road grade types, 1 represents the l-th pavement structure layer, n2 represents the total number of pavement structure layers, and L j W represents the length of road of grade j. j H represents the road width corresponding to road level j. jl This indicates the thickness of the l-th layer in the cross-section of a road of grade j; Step 133. Calculate the total volume RI of the i-th mixture that has been repeatedly calculated within the intersection area. i : Where j1 and j2 represent the road grades of the two intersecting roads, with j2 having a higher road grade than j1. j1 W j2 W represents the width of the intersecting roads. j1s W j2s H represents the widening value on one side of the intersecting roads. j2 P represents the thickness of the higher-grade road in two intersecting roads. j1j2 This represents the number of road intersection transitions for road grades j1 and j2: In the above formula, N3 represents the number of T-shaped intersections and N4 represents the number of cross-shaped intersections; Step 134. Calculate the total stock of the i-th type of mixture within the road system, MS. i : MS i =c i ρ i [RS i -RI i ] (4) Among them, c i ρ i These represent the compaction degree and dry density of the i-th mixture, respectively; Step 135. Calculate the total stock of material k in the road system, MS. k : MS k =∑MS i ·α k (5) Where, α k This represents the content of the k-th material in the i-th mixture.

5. The method according to claim 1, characterized in that, The material inventory data mentioned in step 2 is summarized and organized, and relevant data on area, population, and economy within the calculation area are obtained. Constructing a feature variable dataset specifically includes: Step 211. Summarize and organize each material inventory data obtained in Step 1 along with its corresponding time node, sub-region, road grade, material type, surface layer thickness, base layer thickness, and road width; Step 212. Locate the statistical yearbook or relevant statistical data of the calculation area to obtain the area, population, and economic data indicators of each sub-region within the calculation area, and summarize and organize each material stock data according to the above. Step 213. Construct a feature variable dataset based on the various data indicators related to material inventory mentioned above.

6. The method according to claim 1, characterized in that, In step 2, one-hot encoding is used to convert the categorical feature variables into binary vectors. The two categorical feature variables, road grade and material type, are represented using the one-hot encoding principle to ensure that each state of each sample has only one bit that is state 1, and the rest of the bits are 0.

7. The method according to claim 1, characterized in that, In step 3, the sample set is divided into a training set and a validation set. The training set and validation set are randomly divided using the train_test_split() function in the sklearn library, and the ratio of the number of samples in the training set to the number of samples in the validation set is set.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by the processor, it implements the steps in the urban road material stock prediction regression method based on the GBDT algorithm as described in any one of claims 1 to 7.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the urban road material inventory prediction regression method based on the GBDT algorithm as described in any one of claims 1 to 7.

Images