A construction data analysis and decision-making assistance method and system for traffic engineering projects

Abstract

The present invention relates to the technical field of traffic engineering data analysis, and specifically to a construction data analysis auxiliary decision-making method and system for a traffic engineering project. The present invention trains a neural network model through text information of multidimensional construction data to obtain the construction progress index of different decision points in the current project; obtains the complexity of decision-making for the current project by analyzing the frequency of decision-making in the current project and the influence of surrounding traffic facilities on the current project; then analyzes the time domain span parameters of the decision points to evaluate the rationality of the decisions of each decision point; combines the correlation between the construction data of each dimension of the decision point and the flow data of the surrounding traffic facilities to evaluate the decision reference importance of the construction data of each dimension; and performs dimensionality reduction processing on the multidimensional construction data according to the decision reference importance. The method and system provided by the present invention reduce the probability of important information loss after data dimensionality reduction, and improve the rationality of construction decision-making.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic engineering data analysis, and in particular to a construction data analysis decision-making assistance method and system for a traffic engineering project. Background Art

[0002] With the acceleration of urbanization and the continuous increase in traffic demand, the transportation system has become more and more complex and diverse. Traditional experience and intuitive decision-making methods have been unable to cope with the complexity and uncertainty of modern transportation engineering projects. At the same time, the development of emerging technologies such as big data, artificial intelligence and the Internet of Things has provided new means and tools for decision-making in transportation engineering projects.

[0003] In transportation engineering projects, decision-making support methods use advanced technologies such as big data, artificial intelligence, and machine learning to conduct comprehensive analysis of data in the process of transportation construction, planning, and operation. It aims to provide project managers with a scientific basis through quantitative analysis and prediction, helping them make more accurate and efficient decisions.

[0004] In order to ensure the safety and quality of the construction process of traffic engineering projects, construction data is usually collected and recorded frequently. While ensuring the smooth progress of traffic engineering projects, a large amount of data redundancy with low information density will be generated, resulting in low efficiency of the analysis data-assisted decision-making process, and a certain lag in decision-making compared to the actual construction progress. Usually, data redundancy can be reduced by reducing the dimension of construction data and improving the information density of construction data. However, when reducing the dimension of construction data, the existing technology only reduces the dimension through the distribution characteristics of the data, ignoring the impact of the project construction progress and traffic information on the decision-making process, which may lead to problems such as the loss of key decision data and reduce the rationality of decision-making. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide a construction data analysis and decision-making assistance method and system for a transportation engineering project.

[0006] According to a first aspect of an embodiment of the present invention, a construction data analysis and decision-making assistance method for a traffic engineering project is provided, and the technical solution adopted is specifically as follows:

[0007] Collecting multi-dimensional construction data of different traffic engineering projects, wherein the multi-dimensional construction data includes text information;

[0008] Combined with the text information, a neural network model is trained to obtain construction progress indexes of different decision points in the current traffic engineering project;

[0009] Combined with the construction progress index, the decision frequency of the current traffic engineering project and the influence of surrounding traffic facilities on the current traffic engineering project are analyzed to obtain the decision-making complexity of the current traffic engineering project;

[0010] Analyze the time domain span parameters of the decision points in the current traffic engineering project according to the construction progress index, and then combine the multi-dimensional construction data and the decision making complexity to obtain the decision rationality of the decision points;

[0011] Analyze the correlation between the construction data of each dimension of the decision point and the flow data of surrounding traffic facilities, and then combine the multi-dimensional construction data of the decision point, the construction progress index and the rationality of the decision to obtain the decision reference importance of the construction data of each dimension;

[0012] In combination with the decision reference importance, dimensionality reduction processing is performed on the multi-dimensional construction data.

[0013] In some embodiments of the present invention, collecting multi-dimensional construction data of different traffic engineering projects includes:

[0014] Combined with wireless network sensor technology, multi-dimensional construction data of different traffic engineering projects are collected, including historical data and real-time data;

[0015] Expanding the construction data of different dimensions to the same data volume as the construction data with the largest data volume by interpolation method;

[0016] Using the TF-IDF algorithm to extract keywords from the text information of the construction data, and then using the word2vec algorithm to convert the keywords into word vectors;

[0017] The construction data of each dimension are combined into multi-dimensional construction data.

[0018] In some embodiments of the present invention, the text information is combined to train a neural network model to obtain the construction progress index of different decision points in the current traffic engineering project, including:

[0019] Combined with the name of the traffic engineering project, the text information is scored by manual scoring to obtain a historical construction progress index, and the text information is marked;

[0020] Dividing the labeled text information into a training set and a validation set, inputting the training set data into a neural network for training, using a gradient descent method to train until the loss function converges, completing the neural network training, and inputting the validation set into the neural network for verification;

[0021] The name of the current traffic engineering project and the text information of each decision point are obtained, input into the trained neural network, and the construction progress index of different decision points in the current traffic engineering project is output.

[0022] In some embodiments of the present invention, the decision-making complexity of the current traffic engineering project is obtained by analyzing the decision-making frequency of the current traffic engineering project and the influence of the surrounding traffic facilities on the current traffic engineering project in combination with the construction progress index, including:

[0023] Obtain the number of decisions for the current traffic engineering project, and combine it with the construction progress index to obtain the decision frequency of the current traffic engineering project;

[0024] Obtain the mean of the flow data of the surrounding traffic facilities of the current traffic engineering project, and obtain the degree of influence of the surrounding traffic facilities on the current traffic engineering project;

[0025] The decision-making complexity of the current traffic engineering project is obtained by combining the decision frequency and the impact degree.

[0026] In some embodiments of the present invention, according to the construction progress index, the time domain span parameter of the decision point in the current traffic engineering project is analyzed, and then combined with the multi-dimensional construction data and the decision making complexity, the decision rationality of the decision point is obtained, including:

[0027] Draw the average flow data curve of the surrounding traffic facilities of the current traffic engineering project and the construction data curve of each dimension in the multi-dimensional construction data, and segment the curve using the decision points in the current traffic engineering project to obtain the decision segments;

[0028] Calculating the difference of the construction progress indexes of the two decision points corresponding to the decision segment, and calculating the time difference between the two decision points corresponding to the decision segment, and calculating the ratio of the difference of the construction progress indexes to the time difference to obtain the time domain span parameter of the front-end decision point corresponding to the decision segment;

[0029] Calculate the mean square error of the multidimensional construction data of the two decision points corresponding to the decision segment to obtain the construction data influencing parameter of the front-end decision point corresponding to the decision segment;

[0030] The decision rationality of the front-end decision point corresponding to the decision segment is obtained by combining the time domain span parameter, the construction data influencing parameter and the decision-making complexity.

[0031] In some embodiments of the present invention, analyzing the correlation between the construction data of each dimension of the decision point and the flow data of surrounding traffic facilities includes:

[0032] Obtaining the absolute value of the derivative of each sampling point in the decision segment of the flow data curve and the construction data curve of each dimension, and obtaining a flow data derivative array and a construction data derivative array of each dimension;

[0033] Arrange the flow data derivative array and the construction data derivative array of each dimension according to the time sequence of the decision segment to obtain a flow data sequence group and a construction data sequence group of each dimension;

[0034] The Pearson correlation coefficient between the construction data sequence group of each dimension and the corresponding flow data sequence group in time series is calculated to obtain the correlation between the construction data of each dimension of the front-end decision point corresponding to the decision segment and the flow data of the surrounding traffic facilities.

[0035] In some embodiments of the present invention, the correlation between the construction data of each dimension of the decision point and the flow data of the surrounding transportation facilities is analyzed, and then the multi-dimensional construction data of the decision point, the construction progress index and the rationality of the decision are combined to obtain the decision reference importance of the construction data of each dimension, including:

[0036] Analyze the correlation between the construction data of each dimension of the decision point and the flow data of surrounding transportation facilities;

[0037] Calculate the mean square error between the derivative array of the construction data of each dimension in the decision segment and the derivative array of the construction data of each dimension in the previous decision segment, and obtain the first influence index of the construction data of each dimension of the front-end decision point corresponding to the decision segment on the current decision point;

[0038] Calculate the difference between the construction progress index of the decision point and the construction progress index of the current point to be decided, and obtain a second influence index of the construction progress index of the decision point on the current point to be decided;

[0039] According to the correlation, the first influence index and the second influence index, combined with the decision rationality of the decision point, all the decision points are traversed to obtain the decision reference importance of the construction data of each dimension.

[0040] In some embodiments of the present invention, in combination with the decision reference importance, dimensionality reduction processing is performed on the multi-dimensional construction data, including:

[0041] The multi-dimensional construction data are organized into a matrix A with n rows and m columns, wherein n is the number of construction data and m is the number of data dimensions;

[0042] Zero-mean the construction data of each dimension of matrix A to obtain matrix B;

[0043] By multiplying the decision reference importance of the construction data in each dimension with the data of the corresponding dimension of matrix B, and using the data priority to further adjust the difference between the data, the adjustment matrix is ​​obtained. ;

[0044] Calculate the adjustment matrix The covariance matrix C of , and find the eigenvalues ​​and eigenvectors of the covariance matrix C;

[0045] Arrange the eigenvectors into a matrix by row from top to bottom according to the corresponding eigenvalue size, and take the first k rows to form the matrix P, where k is determined by the cumulative contribution rate of the eigenvalue;

[0046] Multiply matrix P by matrix A to obtain data reduced to k dimensions.

[0047] According to a second aspect of an embodiment of the present invention, a construction data analysis and decision-making support system for a traffic engineering project is provided, comprising: a memory and a processor, wherein:

[0048] The memory is used to store program code;

[0049] The processor is used to read the program code stored in the memory and execute the method described in the first aspect of the embodiment of the present invention.

[0050] In some embodiments of the present invention, the processor comprises:

[0051] A multi-dimensional construction data collection module, used to collect multi-dimensional construction data of different traffic engineering projects, wherein the multi-dimensional construction data includes text information;

[0052] A construction progress analysis module, used to combine the text information, train a neural network model, and obtain construction progress indexes of different decision points in the current traffic engineering project;

[0053] A decision-making complexity analysis module is used to analyze the decision-making frequency of the current traffic engineering project and the influence of surrounding traffic facilities on the current traffic engineering project in combination with the construction progress index, so as to obtain the decision-making complexity of the current traffic engineering project;

[0054] A decision rationality analysis module is used to analyze the time domain span parameters of the decision points in the current traffic engineering project according to the construction progress index, and then combine the multi-dimensional construction data and the decision making complexity to obtain the decision rationality of the decision points;

[0055] A decision reference importance analysis module is used to analyze the correlation between the construction data of each dimension of the decision point and the flow data of surrounding traffic facilities, and then combine the multi-dimensional construction data of the decision point, the construction progress index and the rationality of the decision to obtain the decision reference importance of the construction data of each dimension;

[0056] The dimension reduction processing module is used to perform dimension reduction processing on the multi-dimensional construction data in combination with the decision reference importance.

[0057] Compared with the prior art, the construction data analysis and decision-making assistance method and system for a traffic engineering project provided by the present invention have the following beneficial effects:

[0058] The present invention obtains the construction data of each dimension in the construction process of the current traffic engineering project, and combines the text information of the construction data of each dimension to train a neural network, and evaluates the construction progress of different decision points in the current traffic engineering project; then combines the construction progress and the impact of public travel on the current traffic engineering project to analyze the complexity of decision-making in the current traffic engineering project; and by analyzing the time domain span parameters of the decision points in the current traffic engineering project, combined with the complexity of decision-making and multidimensional construction data, evaluates the rationality of the decision of each decision point; finally, by analyzing the impact of the flow data of the surrounding traffic facilities on the current traffic engineering project, combined with the multidimensional construction data and the construction progress index and the rationality of the decision, calculates the decision reference importance of the construction data of each dimension; and adjusts the dimensionality reduction process of the multidimensional data in combination with the decision reference importance of the construction data of each dimension. The dimensionality reduction process combines the construction progress and construction data change rules of the actual construction process of the traffic engineering project and the impact of the surrounding public travel, reduces the probability of important information loss after the dimensionality reduction of the multidimensional construction data, provides a data reference with high information density for decision-making, and improves the rationality of construction decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 A basic flow chart of a construction data analysis and decision-making assistance method for a traffic engineering project provided by an embodiment of the present invention;

[0061] Figure 2A schematic diagram of an average flow data curve of a surrounding traffic facility and a construction data curve of each dimension in multi-dimensional construction data and a decision segment provided by an embodiment of the present invention;

[0062] Figure 3 A schematic diagram of the basic composition of a construction data analysis and decision support system for a transportation engineering project provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the construction data analysis decision-making assistance method and system for a traffic engineering project proposed by the present invention, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. Terms such as "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the article or device including the element.

[0065] The specific scheme of the construction data analysis and decision-making assistance method for a traffic engineering project provided by the present invention is described in detail below with reference to the accompanying drawings.

[0066] See also Figure 1 , which shows the basic process of a construction data analysis and decision-making assistance method for a transportation engineering project provided by an embodiment of the present invention.

[0067] like Figure 1 As shown, an embodiment of the present invention provides a construction data analysis and decision-making assistance method for a traffic engineering project, which specifically includes:

[0068] S100: Collect multi-dimensional construction data of different traffic engineering projects, where the multi-dimensional construction data includes text information.

[0069] Combined with wireless network sensor technology, multi-dimensional construction data of different transportation engineering projects are collected, including historical data and real-time data. Among them, multi-dimensional construction data includes: construction progress data, cost data, quality inspection data, human resources data, environmental monitoring data and / or construction area geographic information. Specifically:

[0070] Construction progress data: including project completion status, construction log, progress plan, etc.;

[0071] Cost data: including material purchase costs, labor wages, machinery and equipment rental and use fees, etc.;

[0072] Quality inspection data: including raw material inspection reports, construction process quality inspection records, finished product acceptance reports, etc.;

[0073] Human resources data: including the number of workers, skills, working hours, etc.;

[0074] Environmental monitoring data: including real-time monitoring data of environmental parameters such as temperature, humidity, wind speed, and noise;

[0075] Geographic information of the construction area: including location information of roads, buildings, terrain, traffic lights, signboards, surveillance cameras and other facilities.

[0076] Since the sampling frequencies of construction data of different dimensions are different, the data volumes in the construction data of different dimensions are not the same. To facilitate the subsequent processing and use of the data, this embodiment uses an interpolation method to expand the construction data of different dimensions to the same data volume as the construction data with the largest data volume, where the data volume is mainly reflected by the sampling interval.

[0077] Multidimensional construction data includes text information and digital information. In order to facilitate the subsequent data processing and use, the TF-IDF algorithm is used to extract keywords from the text information of the construction data, and then the word2vec algorithm is used to convert the keywords into word vectors.

[0078] Finally, the construction data of each dimension are combined into multi-dimensional construction data.

[0079] S200: Combine text information, train a neural network model, and obtain construction progress indexes of different decision points in the current traffic engineering project.

[0080] For the multi-dimensional construction data of different transportation engineering projects, the importance of construction data of different dimensions that need to be referred to when making engineering decisions is different under different construction progresses. Therefore, in order to analyze the auxiliary decision-making process in combination with construction data of different dimensions, it is necessary to analyze the construction progress of the current transportation engineering project.

[0081] The construction progress is often difficult to reflect directly through construction data, and needs to be analyzed through the latest text information of the traffic engineering project. Therefore, the construction progress can be evaluated by training a neural network model combined with the text information of the current traffic engineering project.

[0082] The training process of the neural network model for evaluating construction progress is as follows:

[0083] Collect multi-dimensional text information data of several traffic engineering projects with different construction progress, that is, the text information obtained in step S100; combine the name of the traffic engineering project, use manual scoring to score the text information, obtain the historical construction progress index, and mark the text information. Among them, the scoring value can be: 0.01, 0.02, 0.03, ... 1, a total of 100 score types, corresponding to 100 types of construction progress, and the scoring result is the construction progress index. The higher the construction progress index, the later the construction progress of the current decision.

[0084] The neural network structure is a 5-layer fully connected neural network. In other embodiments, it can be set to a neural network of other structures. This embodiment does not specifically limit it. The loss function is a cross entropy function. The marked text information is divided into a training set and a validation set in a ratio of 7:3. The training set data is input into the neural network for training. The training is performed using the gradient descent method until the loss function converges and the neural network training is completed. The validation set is input into the trained neural network for verification.

[0085] The name of the current traffic engineering project and the text information of each decision point are obtained, input into the trained neural network, and the construction progress index of different decision points in the current traffic engineering project is output.

[0086] So far, the construction progress index of different decision points in the current traffic engineering project has been obtained. It should be noted that the decision points here are historical decision points that have been completed, and all subsequent decision points are historical decision points unless otherwise specified.

[0087] In traffic engineering projects with different construction progress and flow data of traffic facilities around the project, the reference weights of construction data of different dimensions for construction decisions are different, so it is also necessary to analyze the decision-making reference importance of construction data of each dimension under the current construction progress. The specific analysis method includes steps S300 to S500.

[0088] S300: Combined with the construction progress index, the decision frequency of the current traffic engineering project is analyzed, and the influence of the surrounding traffic facilities on the current traffic engineering project is analyzed to obtain the decision-making complexity of the current traffic engineering project.

[0089] The decision-making of different traffic engineering projects is affected by the actual construction complexity of the project and external interference during the construction. To evaluate the reference importance of construction data of different dimensions to current decision-making, it is necessary to evaluate the degree to which the decision-making of each traffic engineering project is affected by the construction complexity and external interference. Specific evaluation methods include:

[0090] Get the number of decisions for the current traffic engineering project ; Since the construction progress is linear, a parameter that can represent the current construction progress is needed here, so the construction progress index of the latest decision point in the current traffic engineering project is combined , calculate the ratio of the number of decisions of the current traffic engineering project to the construction progress index of the latest decision point. This ratio represents the average number of construction decisions made for the current traffic engineering project under the unit construction progress index. The larger the value, the more frequent the construction decision making of the current traffic engineering project. Therefore, the decision frequency of the current traffic engineering project is For different traffic engineering projects, the number of decisions can often reflect the overall construction complexity of the current traffic engineering project: the more decisions are made and the more frequent the construction decisions are, the more decisive issues the current traffic engineering project needs to face during the construction process, which means that the overall complexity of the construction process of this traffic engineering project may be relatively high.

[0091] Obtain the flow data of the surrounding traffic facilities of the current traffic engineering project, and calculate the mean flow data of the surrounding traffic facilities of the current traffic engineering project , as the mean value of the flow data, it indicates the degree of influence of surrounding traffic facilities on the current traffic engineering project, that is, the degree of influence of public transportation on the current traffic engineering project. The larger the value, the stronger the influence of public travel on the current traffic engineering project. For the construction process of traffic engineering projects in different locations, the larger the flow data of the surrounding traffic facilities, the greater the influence of public travel on the decision-making of this traffic engineering project, which indirectly increases the complexity of decision-making in the construction process of this traffic engineering project.

[0092] Therefore, combining the decision frequency and impact, the decision-making complexity of the current transportation engineering project is obtained. Further, the calculation formula for the decision-making complexity is constructed as follows:

[0093]

[0094] In the formula, Indicates the complexity of decision making in current transportation engineering projects; It represents the mean value of the flow data of the surrounding traffic facilities of the current traffic engineering project; Indicates the number of decisions made in current transportation engineering projects; Represents the construction progress index of the latest decision point of the current transportation engineering project.

[0095] The complexity of decision making The larger the value, the higher the complexity of decision-making for the current traffic engineering project, and the larger the amount of data that needs to be retained in the process of reducing the dimensionality of multi-dimensional construction data. By calculating the complexity of decision-making for traffic engineering projects, an analytical basis is provided for the subsequent judgment of the importance of decision-making references for construction data of different dimensions.

[0096] S400: According to the construction progress index, the time domain span parameters of the decision points in the current traffic engineering project are analyzed, and then the rationality of the decision points is obtained by combining the multi-dimensional construction data and the complexity of decision making.

[0097] For the construction process of transportation engineering projects, the construction decision-making is not always correct in the project. After the decision is made, the unadopted construction decision needs to be adjusted in time. Therefore, for unreasonable decisions, the changes in construction data after the decision is made need to be weakened in the process of evaluating the importance of decision reference.

[0098] Based on the above analysis, this embodiment analyzes the time domain span parameters of the decision points in the current traffic engineering project according to the construction progress index, and then combines the multi-dimensional construction data and the complexity of decision making to obtain the rationality of the decision points. Further including:

[0099] Draw the average flow data curve of the surrounding traffic facilities of the current traffic engineering project and the construction data curve of each dimension in the multi-dimensional construction data (multi-dimensional construction data draws multiple construction data curves), where the horizontal axis represents time, the vertical axis represents the average flow of the surrounding traffic facilities or the construction data of this dimension, and the average flow represents the average value of the flow data of multiple traffic facilities; and use the decision points in the current traffic engineering project to segment the curve to obtain decision segments, such as Figure 2 As shown, the horizontal axis of the dotted line represents the time of the decision point, and the curve between the two dotted lines represents the decision segment.

[0100] After segmenting the data curve, the data changes in different decision segments also represent the changes in the impact of the decision represented by the front-end decision segment point on the construction process. The smaller the time span of the decision segment, the shorter the time interval between the decision and the next decision, and the more likely the next decision is to be a revised strategy for the current decision; the smaller the difference in project progress, the less the actual progress of the current decision, and the lower the rationality of the construction of the current decision. Therefore, the difference in the construction progress index of the two decision points corresponding to the decision segment is calculated, and the time difference between the two decision points corresponding to the decision segment is calculated. The difference in the construction progress index is calculated by the ratio of the time difference to the difference, and the time domain span parameter of the decision point in the current traffic engineering project is obtained.

[0101] In addition to time and construction progress, the difference in multidimensional construction data between the two decision points corresponding to the decision segment is also an important basis for evaluating the rationality of the decision. For construction decisions with a low degree of rationality, in order to ensure the quality of the project in traffic engineering projects, it is usually necessary to restore the construction status as much as possible when the correction strategy is executed. Therefore, the mean square error of the multidimensional construction data of the two decision points corresponding to the decision segment is calculated to obtain the construction data influencing parameters of the front-end decision point corresponding to the decision segment.

[0102] At the same time, the higher the complexity of decision-making, the higher the probability of irrationality in construction decisions. Therefore, the rationality of the decision at each decision point is evaluated by combining the time domain span parameters and construction data influencing parameters of the decision point and the complexity of decision-making of the current transportation engineering project.

[0103] Furthermore, the calculation formula for the decision rationality of the decision point is constructed as follows:

[0104]

[0105] In the formula, Indicates The rationality of the decision at each decision point; Indicates The time span parameter of the decision points; Indicates The construction data of each decision point affects the parameters; Represents the complexity of decision making in current transportation engineering projects.

[0106] Time span parameter of decision point The smaller the difference, the smaller the difference in project progress per unit time, which means that the actual progress of the current decision is less, and the construction rationality of the current decision should be lower; the construction data of the decision point affects the parameters The smaller it is, the lower the rationality of the current decision, which means that in order to ensure the quality of the project in the traffic engineering project, it is usually necessary to restore the construction status as much as possible when the correction strategy is implemented; the complexity of decision-making in the current traffic engineering project The larger it is, the higher the probability of irrationality in construction decisions. By combining the construction data influencing parameters and time domain span parameters and comparing them with the complexity of decision making, the decision rationality index of a certain decision point is obtained.

[0107] S500: Analyze the correlation between the construction data of each dimension of the decision point and the flow data of the surrounding transportation facilities, and then combine the multi-dimensional construction data and the construction progress index of the decision point and the rationality of the decision to obtain the decision reference importance of the construction data of each dimension.

[0108] During the construction of a transportation project, changes in construction data of different dimensions may be affected by the flow data of surrounding transportation facilities: Generally, when the flow data of surrounding transportation facilities is large, in order to reduce the impact of the construction process on public travel, some construction progress may be slowed down. When evaluating the importance of construction data of different dimensions to construction decisions, the impact of construction measures on public travel represented by construction data of different dimensions is one of its components.

[0109] Therefore, this embodiment analyzes the correlation between the construction data of each dimension of the decision point and the flow data of the surrounding traffic facilities, and then combines the multi-dimensional construction data and construction progress index of the decision point and the rationality of the decision to obtain the decision reference importance of the construction data of each dimension.

[0110] Among them, the correlation between the construction data of each dimension of the decision point and the flow data of the surrounding traffic facilities is analyzed, specifically: draw the average flow data curve of the surrounding traffic facilities of the current traffic engineering project and the construction data curve of each dimension in the multi-dimensional construction data. And use the decision points in the current traffic engineering project to segment the curve to obtain the decision segment. Obtain the derivatives of each sampling point in the decision segment of the flow data curve and the construction data curve of each dimension, and take their absolute values ​​to obtain the flow data derivative array and the construction data derivative array of each dimension; arrange the flow data derivative array and the construction data derivative array of each dimension according to the time sequence of the decision segment to obtain the flow data sequence group and the construction data sequence group of each dimension; calculate the Pearson correlation coefficient between the construction data sequence group of each dimension and the corresponding flow data sequence group in the time sequence, and take its absolute value to obtain the correlation between the construction data of each dimension of the front-end decision point corresponding to the decision segment and the flow data of the surrounding traffic facilities.

[0111] According to the above analysis, when the correlation between the change rate of construction data of different dimensions at each sampling point and the change rate of the flow data of surrounding transportation facilities is higher, it means that the construction process represented by the construction data of the current dimension is relatively greatly affected by public travel under the current decision-making, and the flow data of surrounding transportation facilities needs to be combined to arrange engineering tasks during the construction process.

[0112] In addition, during the construction process of a traffic engineering project, the data dimensions of the construction data that change faster after the decision is made can usually represent the construction tasks that are given priority in the decision, and the construction of traffic engineering projects is usually continuous (such as road repair). Therefore, the changes in construction data in different dimensions of decision segments with similar construction progress can analyze the data dimensions with higher reference value under the current decision-making. Therefore, this embodiment calculates the mean square error between the derivative array of construction data in each dimension in the decision segment and the derivative array of construction data in each dimension in the previous decision segment, and obtains the first influence index of the construction data of each dimension of the front-end decision point corresponding to the decision segment on the current decision point. In addition, before calculating the mean square error between the two decision segments, the amount of data in the two decision segments is compared, and the decision segment with less data is interpolated to make the data amount of the two decision segments the same.

[0113] Furthermore, the difference between the construction progress index of the decision point and the construction progress index of the current point to be decided is calculated to obtain a second influence index of the construction progress index of the decision point on the current point to be decided.

[0114] Then, according to the correlation, the first influence index and the second influence index, combined with the decision rationality of the decision point, all decision points are traversed to obtain the decision reference importance of the construction data in each dimension.

[0115] Furthermore, the calculation formula for the decision reference importance of construction data in each dimension is constructed as follows:

[0116]

[0117] in, represents the decision reference importance of the construction data of the jth dimension; Indicates The rationality of the decision at each decision point; Indicates The correlation between the construction data of the jth dimension of a decision point and the flow data of surrounding transportation facilities; Indicates The first impact index of the construction data of the jth dimension of the decision point on the current decision point; Indicates The difference between the construction progress index of the decision point and the construction progress index of the current decision point indicates the The second impact index of the construction progress index of the decision point on the current decision point; represents the number of decision points; represents the adjustment coefficient. In this embodiment, =1; represents the linear normalization function.

[0118] According to the above analysis, at different decision points, the decision reference importance of construction data of different dimensions is affected by the flow data of surrounding transportation facilities and previous decision points. The first After combining the parameters of the construction data points of the jth dimension of the decision points, As the weight, the features of the decision points that are similar to the current decision point and have a greater degree of rationality are magnified, and the decision reference importance of the construction data of each dimension is calculated by means of weighted mean.

[0119] S600: Combined with the decision reference importance, dimensionality reduction processing is performed on multi-dimensional construction data.

[0120] Combined with the importance of decision reference, the multi-dimensional construction data is processed by dimensionality reduction. The specific dimensionality reduction method is: first, the multi-dimensional construction data is organized into an n-row and m-column matrix A by column, where n is the number of construction data and m is the number of data dimensions; then the construction data of each dimension of matrix A is zero-meaned, that is, each data point in the row of matrix A is subtracted from the mean of each row of data to obtain matrix B. In order to retain the dimensional data features of the construction data that are more important for the progress decision reference of the current transportation engineering project as much as possible during the dimensionality reduction of multi-dimensional construction data, it is necessary to analyze the dimensionality reduction matrix in combination with the optimization priority of construction data of different dimensions. Therefore, through the decision reference importance of construction data of each dimension Multiply the data of the corresponding dimension (row) of matrix B, and use the data priority to further adjust the difference between the data to obtain the adjustment matrix ; Then calculate the adjustment matrix The covariance matrix C of the covariance matrix C is obtained, and the eigenvalues ​​and eigenvectors of the covariance matrix C are obtained; the eigenvectors are arranged into a matrix by row from top to bottom according to the corresponding eigenvalues, and the first k rows are taken to form a matrix P, where k is determined by the cumulative contribution rate of the eigenvalues. In this embodiment, the cumulative contribution rate threshold is set to 85% based on the empirical value; finally, the matrix P is multiplied by the matrix A to obtain the matrix D, which is the data after dimensionality reduction to k dimensions.

[0121] After dimensionality reduction processing of multi-dimensional construction data, a decision-making support model is introduced to assist in the construction decision-making of the current transportation engineering project, reducing the interference of redundant data on the analysis process and improving the efficiency of decision-making support analysis.

[0122] Based on the same inventive concept as the above method, this embodiment also provides a construction data analysis and decision-making support system for a transportation engineering project.

[0123] See also Figure 3 , which shows the basic composition of a construction data analysis and decision-making support system for a traffic engineering project provided by an embodiment of the present invention.

[0124] like Figure 3 As shown, a construction data analysis and decision-making support system for a traffic engineering project includes: a memory 10 and a processor 20, wherein:

[0125] A memory 10, used for storing program codes;

[0126] The processor 20 is used to read the program code stored in the memory 10 and execute the collection of multi-dimensional construction data of different traffic engineering projects, the multi-dimensional construction data including text information; combine the text information to train the neural network model to obtain the construction progress index of different decision points in the current traffic engineering project; combine the construction progress index to analyze the decision frequency of the current traffic engineering project and the influence of the surrounding traffic facilities on the current traffic engineering project to obtain the decision-making complexity of the current traffic engineering project; analyze the time domain span parameters of the decision points in the current traffic engineering project according to the construction progress index, and then combine the multi-dimensional construction data and the decision-making complexity to obtain the decision rationality of the decision points; analyze the correlation between the construction data of each dimension of the decision point and the flow data of the surrounding traffic facilities, and then combine the multi-dimensional construction data of the decision point and the construction progress index and the decision rationality to obtain the decision reference importance of the construction data of each dimension; combine the decision reference importance to perform dimensionality reduction processing on the multi-dimensional construction data.

[0127] Furthermore, the processor 20 includes a multi-dimensional construction data acquisition module 21, a construction progress analysis module 22, a decision-making complexity analysis module 23, a decision rationality analysis module 24, a decision reference importance analysis module 25 and a dimensionality reduction processing module 26. Among them:

[0128] The multi-dimensional construction data collection module 21 is configured to collect multi-dimensional construction data of different traffic engineering projects, and the multi-dimensional construction data includes text information;

[0129] The construction progress analysis module 22 is configured to combine the text information, train the neural network model, and obtain the construction progress index of different decision points in the current traffic engineering project;

[0130] The decision-making complexity analysis module 23 is configured to analyze the decision-making frequency of the current traffic engineering project and the influence of the surrounding traffic facilities on the current traffic engineering project in combination with the construction progress index, so as to obtain the decision-making complexity of the current traffic engineering project;

[0131] The decision rationality analysis module 24 is configured to analyze the time domain span parameters of the decision points in the current traffic engineering project according to the construction progress index, and then combine the multi-dimensional construction data and the complexity of decision making to obtain the decision rationality of the decision points;

[0132] The decision reference importance analysis module 25 is configured to analyze the correlation between the construction data of each dimension of the decision point and the flow data of the surrounding traffic facilities, and then combine the multi-dimensional construction data of the decision point and the construction progress index and the rationality of the decision to obtain the decision reference importance of the construction data of each dimension;

[0133] The dimension reduction processing module 26 is configured to perform dimension reduction processing on the multi-dimensional construction data in combination with the decision reference importance.

[0134] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0135] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A construction data analysis and decision-making support method for a traffic engineering project, characterized in that: The method comprises: Collecting multi-dimensional construction data of different traffic engineering projects, wherein the multi-dimensional construction data includes text information; Combined with the text information, a neural network model is trained to obtain construction progress indexes of different decision points in the current traffic engineering project; Combined with the construction progress index, the decision frequency of the current traffic engineering project and the influence of surrounding traffic facilities on the current traffic engineering project are analyzed to obtain the decision-making complexity of the current traffic engineering project; Analyze the time domain span parameters of the decision points in the current traffic engineering project according to the construction progress index, and then combine the multi-dimensional construction data and the decision making complexity to obtain the decision rationality of the decision points; Analyze the correlation between the construction data of each dimension of the decision point and the flow data of surrounding traffic facilities, and then combine the multi-dimensional construction data of the decision point, the construction progress index and the rationality of the decision to obtain the decision reference importance of the construction data of each dimension; Combining the decision reference importance, performing dimensionality reduction processing on the multi-dimensional construction data; According to the construction progress index, the time domain span parameters of the decision point in the current traffic engineering project are analyzed, and then combined with the multi-dimensional construction data and the decision-making complexity, the decision rationality of the decision point is obtained, including: Draw the average flow data curve of the surrounding traffic facilities of the current traffic engineering project and the construction data curve of each dimension in the multi-dimensional construction data, and segment the curve using the decision points in the current traffic engineering project to obtain the decision segments; Calculating the difference of the construction progress indexes of the two decision points corresponding to the decision segment, and calculating the time difference between the two decision points corresponding to the decision segment, and calculating the ratio of the difference of the construction progress indexes to the time difference to obtain the time domain span parameter of the front-end decision point corresponding to the decision segment; Calculate the mean square error of the multidimensional construction data of the two decision points corresponding to the decision segment to obtain the construction data influencing parameter of the front-end decision point corresponding to the decision segment; Combining the time domain span parameter, the construction data influencing parameter and the decision-making complexity, obtaining the decision rationality of the front-end decision point corresponding to the decision segment; Analyze the correlation between the construction data of each dimension of the decision point and the flow data of surrounding transportation facilities, including: Obtaining the absolute value of the derivative of each sampling point in the decision segment of the flow data curve and the construction data curve of each dimension, and obtaining a flow data derivative array and a construction data derivative array of each dimension; Arranging the flow data derivative array and the construction data derivative array of each dimension according to the time sequence of the decision segment to obtain a flow data sequence group and a construction data sequence group of each dimension; The Pearson correlation coefficient between the construction data sequence group of each dimension and the corresponding flow data sequence group in time series is calculated to obtain the correlation between the construction data of each dimension of the front-end decision point corresponding to the decision segment and the flow data of the surrounding traffic facilities.

2. The construction data analysis and decision-making support method for a traffic engineering project according to claim 1 is characterized in that: Collect multi-dimensional construction data of different transportation engineering projects, including: Combined with wireless network sensor technology, multi-dimensional construction data of different traffic engineering projects are collected, including historical data and real-time data; Expanding the construction data of different dimensions to the same data volume as the construction data with the largest data volume by interpolation method; Using the TF-IDF algorithm to extract keywords from the text information of the construction data, and then using the word2vec algorithm to convert the keywords into word vectors; The construction data of each dimension are combined into multi-dimensional construction data.

3. The construction data analysis and decision-making support method for a traffic engineering project according to claim 1 is characterized in that: Combined with the text information, the neural network model is trained to obtain the construction progress index of different decision points in the current traffic engineering project, including: Combined with the name of the traffic engineering project, the text information is scored by manual scoring to obtain a historical construction progress index, and the text information is marked; Dividing the labeled text information into a training set and a validation set, inputting the training set data into a neural network for training, using a gradient descent method to train until the loss function converges, completing the neural network training, and inputting the validation set into the neural network for verification; The name of the current traffic engineering project and the text information of each decision point are obtained, input into the trained neural network, and the construction progress index of different decision points in the current traffic engineering project is output.

4. The construction data analysis and decision-making support method for a traffic engineering project according to claim 1 is characterized in that: Combined with the construction progress index, the decision frequency of the current traffic engineering project and the influence of surrounding traffic facilities on the current traffic engineering project are analyzed to obtain the decision-making complexity of the current traffic engineering project, including: Obtain the number of decisions for the current traffic engineering project, and combine it with the construction progress index to obtain the decision frequency of the current traffic engineering project; Obtain the mean of the flow data of the surrounding traffic facilities of the current traffic engineering project, and obtain the degree of influence of the surrounding traffic facilities on the current traffic engineering project; The decision-making complexity of the current traffic engineering project is obtained by combining the decision frequency and the impact degree.

5. The construction data analysis and decision-making support method for a traffic engineering project according to claim 1 is characterized in that: Analyze the correlation between the construction data of each dimension of the decision point and the flow data of the surrounding traffic facilities, and then combine the multi-dimensional construction data of the decision point, the construction progress index and the rationality of the decision to obtain the decision reference importance of the construction data of each dimension, including: Analyze the correlation between the construction data of each dimension of the decision point and the flow data of surrounding transportation facilities; Calculate the mean square error between the derivative array of the construction data of each dimension in the decision segment and the derivative array of the construction data of each dimension in the previous decision segment, and obtain the first influence index of the construction data of each dimension of the front-end decision point corresponding to the decision segment on the current decision point; Calculate the difference between the construction progress index of the decision point and the construction progress index of the current point to be decided, and obtain a second influence index of the construction progress index of the decision point on the current point to be decided; According to the correlation, the first influence index and the second influence index, combined with the decision rationality of the decision point, all the decision points are traversed to obtain the decision reference importance of the construction data of each dimension.

6. The construction data analysis and decision-making assistance method for a traffic engineering project according to claim 1 is characterized in that: Combined with the decision reference importance, the multi-dimensional construction data is subjected to dimensionality reduction processing, including: The multi-dimensional construction data are organized into a matrix A with n rows and m columns, wherein n is the number of construction data and m is the number of data dimensions; Zero-mean the construction data of each dimension of matrix A to obtain matrix B; By multiplying the decision reference importance of the construction data in each dimension with the data of the corresponding dimension of matrix B, and using the data priority to further adjust the difference between the data, the adjustment matrix is ​​obtained. ; Calculate the adjustment matrix The covariance matrix C of , and find the eigenvalues ​​and eigenvectors of the covariance matrix C; Arrange the eigenvectors into a matrix by row from top to bottom according to the corresponding eigenvalue size, and take the first k rows to form the matrix P, where k is determined by the cumulative contribution rate of the eigenvalue; Multiply matrix P by matrix A to obtain data reduced to k dimensions.

7. A construction data analysis and decision-making support system for a traffic engineering project, characterized in that: The system comprises: a memory and a processor, wherein: The memory is used to store program codes; The processor is configured to read the program code stored in the memory and execute the method according to any one of claims 1 to 6.

8. The construction data analysis and decision-making support system for traffic engineering projects according to claim 7 is characterized in that: The processor comprises: A multi-dimensional construction data collection module, used to collect multi-dimensional construction data of different traffic engineering projects, wherein the multi-dimensional construction data includes text information; A construction progress analysis module, which is used to train a neural network model based on the text information to obtain construction progress indexes of different decision points in the current traffic engineering project; A decision-making complexity analysis module is used to analyze the decision-making frequency of the current traffic engineering project and the influence of surrounding traffic facilities on the current traffic engineering project in combination with the construction progress index, so as to obtain the decision-making complexity of the current traffic engineering project; A decision rationality analysis module is used to analyze the time domain span parameters of the decision points in the current traffic engineering project according to the construction progress index, and then combine the multi-dimensional construction data and the decision making complexity to obtain the decision rationality of the decision points; A decision reference importance analysis module is used to analyze the correlation between the construction data of each dimension of the decision point and the flow data of surrounding traffic facilities, and then combine the multi-dimensional construction data of the decision point, the construction progress index and the rationality of the decision to obtain the decision reference importance of the construction data of each dimension; A dimensionality reduction processing module, used to perform dimensionality reduction processing on multi-dimensional construction data in combination with the decision reference importance; According to the construction progress index, the time domain span parameters of the decision point in the current traffic engineering project are analyzed, and then combined with the multi-dimensional construction data and the decision-making complexity, the decision rationality of the decision point is obtained, including: Draw the average flow data curve of the surrounding traffic facilities of the current traffic engineering project and the construction data curve of each dimension in the multi-dimensional construction data, and segment the curve using the decision points in the current traffic engineering project to obtain the decision segments; Calculating the difference of the construction progress indexes of the two decision points corresponding to the decision segment, and calculating the time difference between the two decision points corresponding to the decision segment, and calculating the ratio of the difference of the construction progress indexes to the time difference to obtain the time domain span parameter of the front-end decision point corresponding to the decision segment; Calculate the mean square error of the multidimensional construction data of the two decision points corresponding to the decision segment to obtain the construction data influencing parameter of the front-end decision point corresponding to the decision segment; Combining the time domain span parameter, the construction data influencing parameter and the decision-making complexity, obtaining the decision rationality of the front-end decision point corresponding to the decision segment; Analyze the correlation between the construction data of each dimension of the decision point and the flow data of surrounding transportation facilities, including: Obtaining the absolute value of the derivative of each sampling point in the decision segment of the flow data curve and the construction data curve of each dimension, and obtaining a flow data derivative array and a construction data derivative array of each dimension; Arranging the flow data derivative array and the construction data derivative array of each dimension according to the time sequence of the decision segment to obtain a flow data sequence group and a construction data sequence group of each dimension; The Pearson correlation coefficient between the construction data sequence group of each dimension and the corresponding flow data sequence group in time series is calculated to obtain the correlation between the construction data of each dimension of the front-end decision point corresponding to the decision segment and the flow data of the surrounding traffic facilities.

Images