Rail transit cost assessment method, system and storage medium based on big data

Abstract

The present invention discloses a rail transit cost assessment method, system and storage medium based on big data, the method comprising: obtaining rail transit design files from a preset database, determining the corresponding influencing parameter factor group for each engineering project in the design file, and searching for multiple historical influencing parameter value groups corresponding to the influencing parameter factor group, generating influencing factor weights corresponding to the influencing parameter factor group based on each historical influencing parameter value group, correcting the cost values ​​in the project list corresponding to the engineering project, and generating the project cost corresponding to the engineering project; based on the project cost generated for each engineering project, obtaining the transportation cost corresponding to the rail transit and conducting cost assessment on the rail transit. The present invention achieves accurate assessment of the cost of rail transit by comprehensively analyzing the factors that cause cost changes in similar historical engineering projects for each split engineering project, generating influencing factor weights to correct the project cost of the engineering project, and achieving accurate assessment of the cost of rail transit.

Description

Technical Field

[0001] The present invention relates to the field of big data processing technology, and in particular to a rail transit cost assessment method, system and storage medium based on big data. Background Art

[0002] With the rapid advancement of urbanization, major cities are facing increasing pressure on transportation. The convenience, speed and low carbon of urban rail transit are effective ways to solve the pressure of urban transportation. Urban rail transit projects are composed of multi-level systems and are complex system projects, covering more than ten systems such as civil engineering systems (stations, sections, etc.), equipment installation systems (communication, signaling, power supply, water supply and drainage, etc.) and vehicles. The overall project involves many professions, complex engineering conditions, long construction period and large investment amount.

[0003] Although rail transit can effectively solve the pressure of transportation, its continuously rising project cost has also brought heavy financial pressure to governments at all levels, and to a certain extent restricted the sustained and rapid development of urban rail transit. At present, the cost of rail transit is usually divided into the pre-assessment stage, the in-process control stage and the post-event summary stage. If the preliminary assessment is inaccurate, it will affect the in-process control and cause the final settlement to far exceed the expected investment amount. However, the preliminary assessment is difficult to accurately assess due to the actual complexity of rail transit projects. Therefore, how to achieve accurate assessment of rail transit cost has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The main purpose of the present invention is to provide a rail transit cost assessment method, system and storage medium based on big data, aiming to solve the technical problem of how to achieve accurate assessment of rail transit cost.

[0005] To achieve the above object, the present invention provides a rail transit cost assessment method based on big data, and the rail transit cost assessment method based on big data includes:

[0006] Acquire a rail transit design document from a preset database, wherein the design document includes a plurality of construction bid sections, a plurality of bid section projects corresponding to each of the construction bid sections, a plurality of engineering projects corresponding to each of the bid section projects, and a project list corresponding to each of the engineering projects;

[0007] For each of the engineering projects, determine the influencing parameter factor group corresponding to the engineering project, and search for multiple historical influencing parameter value groups corresponding to the influencing parameter factor group in a preset database, and generate an influencing factor weight corresponding to the influencing parameter factor group based on each of the historical influencing parameter value groups;

[0008] Obtaining cost values ​​in a project list corresponding to the engineering project, and modifying each cost value according to the weight of the influencing factor to generate a project cost corresponding to the engineering project;

[0009] Based on the project cost generated for each of the engineering projects, the traffic cost corresponding to the rail transit is obtained, and according to the traffic cost, the cost of the rail transit is evaluated.

[0010] Preferably, the step of generating the impact factor weight corresponding to the impact parameter factor group based on each of the historical impact parameter value groups includes:

[0011] Constructing an influence parameter matrix based on each parameter name in the influence parameter factor group, and generating matrix element values ​​in the influence parameter matrix based on preset cost influence coefficients between each pair of parameters in each historical influence parameter value group;

[0012] The values ​​of each matrix element in the influencing parameter matrix are converted and calculated according to a preset conversion formula to generate a consistency matrix corresponding to the influencing parameter matrix. The preset conversion formula is:

[0013]

[0014] Among them, B ij represents the element in the i-th row and j-th column of the consistency matrix, n represents the number of rows in the influencing parameter matrix, Ai represents the sum of the elements in the i-th row of the influencing parameter matrix, Ak represents the sum of the elements in the k-th row of the influencing parameter matrix, Aj represents the sum of the elements in the j-th row of the influencing parameter matrix, Amax represents the maximum value of the sum of the elements in each row of the influencing parameter matrix, and Amin represents the minimum value of the sum of the elements in each row of the influencing parameter matrix;

[0015] The elements in the consistency matrix are normalized to obtain the influencing factor weights corresponding to the influencing parameter factor group.

[0016] Preferably, the step of generating matrix element values ​​in the influencing parameter matrix based on preset cost influencing coefficients between any two parameters in each historical influencing parameter value group comprises:

[0017] For the same pairwise parameters in each of the historical influencing parameter value groups, the number of positive relationships and the number of negative relationships are counted, and a ratio operation is performed between the number of positive relationships and the number of negative relationships to obtain a positive-negative ratio;

[0018] Obtaining the construction period corresponding to the engineering project, and determining the material variation factor array and the labor variation factor array corresponding to the construction period;

[0019] The preset cost impact coefficients corresponding to the same pairwise parameters in each of the historical impact parameter value groups are averaged, and the result of the average operation is corrected based on the positive and negative ratio, the material variation factor array and the manual variation factor array to obtain the matrix element value. The calculation formula for the correction is:

[0020]

[0021] Among them, Pxy represents the matrix element value of the xth row and yth column in the influencing parameter matrix, θ represents the positive-negative ratio, s1 represents the number of material types in the material variation factor array, and M 1c1 Indicates the proportion of the c1th material in the material variation factor array in the engineering project, ΔP 1c1 represents the price change factor of the c1th material in the material change factor array, s2 represents the number of types of work in the labor change factor array, M 2c2 Indicates the proportion of the c2th type of work in the project in the labor change factor array, ΔP 2c2 represents the price variation factor of the c2th type of work in the labor variation factor array, f(·) represents the weight function of materials and labor, and Axy represents the result of the mean calculation of the preset cost impact coefficients corresponding to the xth parameter and the yth parameter in each of the historical impact parameter value groups.

[0022] Preferably, the step of converting and calculating each matrix element value in the influencing parameter matrix according to a preset conversion formula to generate a consistency matrix corresponding to the influencing parameter matrix comprises:

[0023] The consistency index value of the consistency matrix is ​​calculated based on a preset judgment formula, and the preset judgment formula is:

[0024]

[0025] Among them, μ represents the consistency index value, m represents the number of elements in the consistency matrix, and b ij represents the element in the i-th row and j-th column of the consistency matrix, b ji represents the element in the jth row and ith column of the consistency matrix, w i represents the influence weight value of the parameter corresponding to the i-th row in the influence parameter matrix, w j Represents the influence weight value of the parameter corresponding to the jth row in the influence parameter matrix;

[0026] Compare the consistency index value with a preset index threshold to determine whether the consistency index value is greater than the preset index threshold;

[0027] If it is greater than the preset indicator threshold, the preset cost impact coefficient between each parameter in the historical influencing parameter value group is adjusted based on the preset adjustment rule, and after the adjustment, the step of generating the matrix element value in the influencing parameter matrix based on the preset cost impact coefficient between each parameter in each historical influencing parameter value group is executed.

[0028] Preferably, the step of normalizing the elements in the consistency matrix to obtain the influencing factor weights corresponding to the influencing parameter factor group comprises:

[0029] Performing a ratio operation on each matrix element in the consistency matrix and the sum of the matrix columns where the matrix element is located to obtain a normalized matrix;

[0030] Performing a sum operation on the row elements of each matrix row in the normalized matrix to obtain a sum result corresponding to each matrix row;

[0031] A ratio operation is performed between the sum result and the number of matrix rows of the normalized matrix to obtain the weight of the impact factor.

[0032] Preferably, the step of obtaining the traffic cost corresponding to the rail transit based on the project cost generated for each engineering project, and performing cost evaluation on the rail transit according to the traffic cost comprises:

[0033] The project costs of the engineering projects corresponding to each of the bidding sections are added together to generate the engineering cost of the bidding section engineering, and the project costs of the bidding section engineering corresponding to each of the construction bidding sections are added together to generate the bidding section cost of the construction bidding section;

[0034] The traffic cost corresponding to the rail transit is generated based on the cumulative calculation of the cost of each bidding section, and the engineering cost, the bidding section cost and the traffic cost are evaluated respectively based on the preset engineering cost, the preset bidding section cost and the preset traffic cost.

[0035] Preferably, the step of obtaining the design file of rail transit from the preset database includes:

[0036] Receiving an uploaded design file corresponding to the rail transit, and judging whether the uploaded design file is a standard file according to a standardized format corresponding to the preset database;

[0037] If the uploaded design file is a standard file, the uploaded design file is parsed based on a preset recognition model to generate the design file;

[0038] Among them, the preset recognition model includes a first hidden layer, a second hidden layer, a third hidden layer and a fourth hidden layer, which are respectively used to parse out construction sections, section projects, engineering projects and project lists from the uploaded design files, and output them through the output layer of the preset recognition model.

[0039] Preferably, the step of obtaining the design file of rail transit from the preset database includes:

[0040] Acquire a file sample formed based on a historical standard design file of historical rail transit, and divide the file sample into a training sample and a test sample;

[0041] Training a preset initial model based on the training sample, and when the training reaches a preset time, calculating a loss function value of the preset initial model based on the test sample;

[0042] Determine whether the loss function value is less than or equal to a preset loss threshold, and if so, generate the preset initial model as a preset recognition model;

[0043] If the loss function value is greater than the preset loss threshold, the weight value between the fourth hidden layer and the output layer in the preset initial model is updated, and the update formula is:

[0044]

[0045] Among them, ω represents the updated weight value, ω0 represents the weight value before the update, τ represents the learning rate of the preset initial model, T represents the current cumulative number of tests of the preset initial model, yt represents the test result corresponding to the tth test of the preset initial model, and y0 represents the reference test result corresponding to the test sample;

[0046] For the preset initial model after the weight value is updated, a step of training the preset initial model based on the training sample is performed.

[0047] Furthermore, to achieve the above-mentioned purpose, the present invention also provides a rail transit cost evaluation system based on big data, the rail transit cost evaluation system based on big data includes a storage device, a processor, a communication bus, and a control program stored on the storage device:

[0048] The communication bus is used to realize the connection and communication between the processor and the storage;

[0049] The processor is used to execute the control program to implement the steps of the rail transit cost assessment method based on big data as described above.

[0050] Furthermore, to achieve the above-mentioned purpose, the present invention also provides a storage medium, on which a control program is stored, and when the control program is executed by a processor, the steps of the rail transit cost assessment method based on big data as described above are implemented.

[0051] The rail transit cost assessment method, system and storage medium based on big data of the present invention are pre-set with a preset database, in which the design files of various rail transits are stored, and the design files are stored in the structural form of each construction section, the section project corresponding to each construction section, the engineering project corresponding to each section project, and the project list corresponding to each engineering project. When it is necessary to evaluate the cost of a certain rail transit, the design document of the rail transit is obtained from the preset database, and based on the engineering projects in the design document, the corresponding influencing parameter factor group is determined for each engineering project; at the same time, the preset database also stores various historical files of the completed rail transit, and the historical influencing parameter value group corresponding to the influencing parameter factor group is found from such historical files, and the influencing factor weight corresponding to the influencing parameter factor group is generated based on the historical influencing parameter value group; then the cost data is obtained from the project list corresponding to the engineering project, and the cost value is corrected by the influencing factor weight to obtain the project cost corresponding to the engineering project; then the engineering cost of the project is generated from the project cost of each engineering project, and the section cost of the construction section is generated from the engineering cost of each engineering project, and the section cost of each construction section forms the engineering cost of the rail transit. Among them, the influence factor weights generated according to the historical influence parameter value groups corresponding to the influence parameter factor groups reflect the degree of influence of similar cost influencing factors on the cost of completed projects with similar cost influencing factors as the current rail transit project. The cost values ​​of the project list are corrected by the weights of such influence factors, which can accurately reflect the impact of possible changes in the construction process of rail transit on the cost. In this way, by subdividing the design documents of rail transit into engineering projects, and comprehensively analyzing the factors that cause cost changes in similar historical engineering projects for each engineering project, the influence factor weights are generated to correct the project cost of the engineering project, so that the project cost assessment of each engineering project is more accurate, and the accurate assessment of the cost of rail transit is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a flow chart of the first embodiment of the rail transit cost assessment method based on big data of the present invention;

[0053] Figure 2 It is a flow chart of the second embodiment of the rail transit cost assessment method based on big data of the present invention;

[0054] Figure 3It is a flow chart of the third embodiment of the rail transit cost assessment method based on big data of the present invention;

[0055] Figure 4 It is a structural schematic diagram of the hardware operating environment involved in an embodiment of a rail transit cost evaluation system based on big data of the present invention.

[0056] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0057] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0058] The present invention provides a rail transit cost assessment method based on big data, please refer to Figure 1 , Figure 1 It is a flow chart of the first embodiment of the rail transit cost assessment method based on big data of the present invention.

[0059] The embodiment of the present invention provides an embodiment of a rail transit cost assessment method based on big data. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that here. Specifically, the rail transit cost assessment method based on big data in this embodiment includes:

[0060] Step S10, obtaining rail transit design documents from a preset database, wherein the design documents include multiple construction sections, multiple section projects corresponding to each of the construction sections, multiple engineering projects corresponding to each of the section projects, and a project list corresponding to each of the engineering projects.

[0061] The rail transit cost assessment method based on big data in this embodiment is applied to a background server, which is connected to a preset database, and various files of various rail transits are stored in the preset database, such as cost-related files of various stages such as decision-making, design, construction, and settlement of completed rail transit, cost-related files of decision-making, design, construction, and other stages of rail transit currently under construction, and cost-related files of rail transit that is in the design stage but not yet constructed. The background server obtains the cost-related files of the rail transit that needs to be cost assessed from the preset database, and conducts a comprehensive analysis in combination with the cost-related files of other rail transits with similar influencing factors as the rail transit, so as to achieve accurate cost assessment of the rail transit that needs to be cost assessed.

[0062] Specifically, each rail transit carries an identifier that represents its own uniqueness in the preset database. When there is a need to conduct a cost assessment on a certain rail transit, the relevant personnel can initiate an assessment request to the backend server through a device that is connected to the backend server. After receiving the assessment request, the backend server reads the identifier carried therein, and then generates the read identifier as an acquisition request and transmits it to the database. The database searches for the design file with the identifier and returns it to the backend server. The backend server obtains the design file of the rail transit from the preset database. This type of design file is stored in a five-level structure in the preset database, the first level is rail transit, the second level is construction section, the third level is section engineering, the fourth level is engineering project, and the fifth level is project list. Except for the fourth and fifth levels, the other upper and lower levels are one-to-many relationships. That is, rail transit corresponds to multiple construction sections, such as a station section, a section between two stations, etc.; each construction section corresponds to multiple section projects, such as the station main civil engineering project and station equipment installation project corresponding to a station section; each section project corresponds to multiple engineering projects, such as the retaining foundation pit excavation project, high-pressure jet grouting pile project, underground continuous wall project, etc. in the station main civil engineering project. Each engineering project corresponds to a project list, which includes the expected project situation, construction technology, construction equipment, number of workers, construction period, and their respective price information and total price information. By splitting the design files of each rail transit into a five-level structure for storage, the engineering projects in the rail transit and the estimated cost of each engineering project can be accurately reflected.

[0063] Step S20, for each of the engineering projects, determine the influencing parameter factor group corresponding to the engineering project, and search a preset database for multiple historical influencing parameter value groups corresponding to the influencing parameter factor group, and generate an influencing factor weight corresponding to the influencing parameter factor group based on each of the historical influencing parameter value groups.

[0064] Furthermore, for each split engineering project, the parameter factors that may affect the cost change of rail transit are determined and formed into an influencing parameter factor group. Among them, different engineering projects have different parameter factors that affect the cost change. For example, for the maintenance pit excavation project, the parameter factor that affects the cost change may be the change in rental price caused by the change of excavation machinery and equipment; for the high-pressure rotary jet pile project, the parameter factor that affects the cost change may be the change in rental price caused by the change of spray pile materials or material premium. The parameter factors related to the cost change of each engineering project can be pre-set according to the cost change of each engineering project of historical rail transit, and the corresponding relationship between the engineering project and the parameter factors is formed and stored in the preset database. For example, for the historical rail transit engineering project A, its cost changes due to the change of construction equipment and materials, so the parameter factors of engineering project A are set to the construction equipment change A1 and the material change A2, and A and A1, A2 are formed into a corresponding relationship and stored in the preset database. For each engineering project in rail transit that requires cost evaluation, the corresponding relationship can be used to find various parameter factors that affect the change of its cost, and the found parameter factors can be formed into an influencing parameter factor group.

[0065] In addition, it should be noted that the same engineering project in different rail transit is located in different regions and has different geological construction conditions, which leads to different parameter factors affecting the cost. For example, for underground continuous wall projects, different construction equipment may be required in southern and northern cities due to geological differences, resulting in different parameter factors affecting the cost. At the same time, the economic differences in different regions will also lead to different impacts on the cost of labor and materials. Therefore, there should be similarities between historical rail transit and rail transit in geology and economy, so that the determined group of influencing parameter factors can accurately reflect the impact on the cost of rail transit.

[0066] Specifically, the impact of geology and economy can be considered in advance, and the geological data related to the construction in the areas where the various engineering projects split by rail transit are located, such as soil structure, soil moisture content, etc., can be detected as geological parameters of the areas corresponding to the engineering projects. At the same time, the changes in economic data related to rail transit in the areas where rail transit is located over the years are analyzed, such as obtaining the annual changes in the labor prices of each worker, the annual changes in various materials, etc., as economic parameters of the areas corresponding to the engineering projects. Then, based on such geological parameters and economic parameters of the engineering projects, various types of data of historical rail transit engineering projects stored in the preset database are searched, and historical rail transit with similar geological parameters and economic parameters is determined as the historical rail transit that matches the rail transit that currently needs to be cost evaluated, so as to find the various parameter factors that affect the cost changes from the corresponding relationships of its various engineering projects to form an influencing parameter factor group.

[0067] Understandably, for the same project, different parameter factors have different degrees of influence on its cost changes. Some parameter factors have a large degree of influence, resulting in large cost changes, while other parameter factors have a small degree of influence, resulting in small cost changes. The specific impact size can be reflected by the historical cost change values ​​of rail transit. While forming a corresponding relationship between the project and the parameter factors, the specific cost change numerical proportion caused by the parameter factors is also formed into the corresponding relationship. For example, for the above-mentioned project A, if the construction equipment change A1 causes the cost change numerical proportion of project A to be a1, and the material change A2 causes the cost change numerical proportion of project A to be a2, then a corresponding relationship is formed between A, A1, A2, a1, and a2. While determining the influencing parameter factor group corresponding to the project through the corresponding relationship, the corresponding relationship can also be used to determine the corresponding impact value size of each parameter factor. Arrange the impact values ​​according to the order of arrangement of the parameter factors in the influencing parameter factor group to obtain a historical impact parameter value group corresponding to the influencing parameter factor group. Then, by processing the values ​​in such historical influencing parameter value groups of multiple historical rail transits, influencing factor weights corresponding to the influencing parameter factor groups are formed, and the influencing factor weights are used to reflect the impact of each parameter factor in each influencing parameter factor group on the cost of the engineering project.

[0068] Step S30, obtaining cost values ​​in the project list corresponding to the engineering project, and modifying each cost value according to the weight of the influencing factor to generate the project cost corresponding to the engineering project.

[0069] Further, the project list corresponding to the project includes the cost value required for the construction of the project according to the design documents. The cost value does not take into account the design changes that may be involved in the actual construction process of the project, so that the cost value may be inaccurate. Therefore, in order to make the cost value more accurate, the backend server obtains the cost value corresponding to the project list from the preset database, and corrects the corresponding cost value by the influence factor weight to obtain the corrected accurate project cost of the project. Among them, the obtained cost value is preferably the single cost value of the project split, and the influence factor weight can only be corrected for the price of the cost change. For example, the obtained cost value includes B1 and B2 labor price values, C1 and C2 material price values. The influence factor weight includes the weight D1 of the B2 labor price change and the weight D2 of the C1 material price change, then the B2 labor price value is corrected by D1, and the C2 material price value is corrected by D2. Then, the corrected single cost values ​​are added to obtain the corrected project cost of the project.

[0070] Step S40, based on the project cost generated for each of the engineering projects, obtain the traffic cost corresponding to the rail transit, and perform a cost assessment on the rail transit according to the traffic cost.

[0071] Furthermore, after each engineering project of the rail transit has generated its own project cost, the transportation cost reflecting the overall construction price of the rail transit can be generated by adding up the costs of each project, and then the cost of the rail transit can be evaluated based on the produced transportation cost. This cost evaluation is to evaluate whether the overall construction price of the rail transit is too high, as well as whether the construction price of each construction section and each section project is too high. In order to evaluate the construction sections and section projects, it is necessary to add up the costs of each project according to the hierarchical structure divided by the rail transit design documents in the preset database. Specifically, the steps of obtaining the transportation cost corresponding to the rail transit based on the project cost generated for each engineering project, and conducting a cost evaluation of the rail transit based on the transportation cost include:

[0072] Step S41, summing up the project costs of the engineering projects corresponding to each bidding section to generate the engineering cost of the bidding section, and summing up the engineering costs of the bidding section projects corresponding to each construction bidding section to generate the bidding section cost of the construction bidding section;

[0073] Step S42, generating the traffic cost corresponding to the rail transit based on the cumulative calculation of the cost of each bid section, and evaluating each of the engineering costs, each of the bid section costs and the traffic cost based on the preset engineering cost, the preset bid section cost and the preset traffic cost.

[0074] Furthermore, after each engineering project split from the section project generates its own corresponding project cost, the engineering cost of the section project can be obtained by adding up the costs of each project. After each section project split from the construction section generates its own corresponding project cost, the section cost of the construction section can be obtained by adding up the costs of each project. Then, the section costs of each construction section of the rail transit are added up to obtain the entire transportation cost of the rail transit and conduct a cost assessment of the rail transit. Because the project cost of each engineering project is corrected by the corresponding weight of the influencing factor, the accuracy of each project cost is ensured, so that the engineering cost, section cost and transportation cost formed by each project cost are also more accurate, thereby ensuring the accuracy of the assessment.

[0075] Specifically, preset traffic costs, preset bid section costs and preset engineering costs are pre-set to evaluate the traffic costs, bid section costs and engineering costs. The generated traffic cost is compared with the preset traffic cost to determine whether the traffic cost is higher than the preset traffic cost. If it is higher than the preset traffic cost, it means that the construction price of rail transit may be too high. The cost of each bid section is compared with the preset bid section cost to find the target bid section cost that exceeds the preset bid section cost. The existence of the target bid section cost makes the traffic cost higher than the preset traffic cost. Then, the engineering cost for generating the target bid section cost is compared with the preset traffic cost and the preset engineering cost to determine the target engineering cost that is higher than the preset engineering cost. This type of target engineering cost makes the bid section cost higher than the preset bid section cost. The total traffic cost can be made lower than the preset traffic cost by adjusting and controlling the cost of each item in the target engineering cost. Moreover, in order to further rationally plan the use of funds, for situations where the amount is not higher than , comparison and judgment must also be carried out to ensure that the transportation cost is within the preset transportation cost range, while the costs of each project are within the preset project cost range, and the costs of each section are within the preset section cost range, so that the rail transit is reasonable not only in terms of the total cost, but also in terms of each section and project.

[0076] The rail transit cost assessment method based on big data implemented in this embodiment is pre-set with a preset database, in which the design files of various rail transits are stored, and the design files are stored in the structure form of each construction section, the section project corresponding to each construction section, the engineering project corresponding to each section project, and the project list corresponding to each engineering project. When it is necessary to evaluate the cost of a certain rail transit, the design document of the rail transit is obtained from the preset database, and based on the engineering projects in the design document, the corresponding influencing parameter factor group is determined for each engineering project; at the same time, the preset database also stores various historical files of the completed rail transit, and the historical influencing parameter value group corresponding to the influencing parameter factor group is found from such historical files, and the influencing factor weight corresponding to the influencing parameter factor group is generated based on the historical influencing parameter value group; then the cost data is obtained from the project list corresponding to the engineering project, and the cost value is corrected by the influencing factor weight to obtain the project cost corresponding to the engineering project; then the engineering cost of the project is generated from the project cost of each engineering project, and the section cost of the construction section is generated from the engineering cost of each engineering project, and the section cost of each construction section forms the engineering cost of the rail transit. Among them, the influence factor weights generated according to the historical influence parameter value groups corresponding to the influence parameter factor groups reflect the degree of influence of similar cost influencing factors on the cost of completed projects with similar cost influencing factors as the current rail transit project. The cost values ​​of the project list are corrected by the weights of such influence factors, which can accurately reflect the impact of possible changes in the construction process of rail transit on the cost. In this way, by subdividing the design documents of rail transit into engineering projects, and comprehensively analyzing the factors that cause cost changes in similar historical engineering projects for each engineering project, the influence factor weights are generated to correct the project cost of the engineering project, so that the project cost assessment of each engineering project is more accurate, and the accurate assessment of the cost of rail transit is achieved.

[0077] For further information, please refer to Figure 2 Based on the first embodiment of the rail transit cost evaluation method based on big data of the present invention, a second embodiment of the rail transit cost evaluation method based on big data of the present invention is proposed.

[0078] The second embodiment of the rail transit cost evaluation method based on big data is different from the first embodiment of the rail transit cost evaluation method based on big data in that the step of generating the influence factor weight corresponding to the influence parameter factor group based on each of the historical influence parameter value groups includes:

[0079] Step S21, constructing an influence parameter matrix based on the parameter names in the influence parameter factor group, and generating matrix element values ​​in the influence parameter matrix based on preset cost influence coefficients between two parameters in each historical influence parameter value group;

[0080] Furthermore, the parameter factor names that form the influencing parameter factor group are used as the parameter names of the influencing parameter factor group, and the parameter names are used as matrix rows and matrix columns according to the arrangement order of each parameter name in the influencing parameter factor group to construct an influencing parameter matrix. For example, if the parameter names in the influencing parameter factor group are m1, m2, m3, and m4, the matrix rows are formed in the order of m1, m2, m3, and m4, and the matrix columns are formed in the order of m1, m2, m3, and m4, so as to obtain a 4*4 influencing parameter matrix. For the matrix element values ​​in the influencing parameter matrix, they are generated according to the preset cost influence coefficients between the two parameters in the historical influencing parameter value group. The cost change value proportion in the historical influencing parameter value group is the parameter in the historical influencing parameter value group. The preset cost influence coefficient between the two parameters reflects the degree of influence of one parameter on the cost relative to the other parameter in the two parameters, that is, the degree of influence of the high and low proportions. Different coefficients can be pre-set as preset cost influence coefficients to indicate the degree of influence on the cost. For example, setting a coefficient of 1 means that one parameter has the same influence as another parameter, setting a coefficient of 3 means that one parameter has a slightly greater influence than another parameter, setting a coefficient of 5 means that one parameter has a greater influence than another parameter, setting a coefficient of 7 means that one parameter has a much greater influence than another parameter, and setting a coefficient of 9 means that one parameter has a much greater influence than another parameter. For the opposite size relationship, the reciprocal of the coefficient is taken. For example, if the other parameter has a much greater influence than one parameter, the coefficient is 1 / 9. In addition, the size relationship is determined by the difference in the proportion of the two parameters. Multiple range intervals are preset. The difference in the proportion is considered slightly larger in one interval and large in another interval. After comparing the difference in the proportion, the size of the impact of the quantity parameter can be determined, and then the corresponding preset cost impact coefficient is determined as the matrix element value in the impact parameter matrix.

[0081] In a specific embodiment, for the above 4*4 influencing parameter matrix, the parameters in the corresponding historical influencing parameter value group are the proportions of the cost change values ​​corresponding to m1, m2, m3, and m4. The values ​​of the elements a11, a22, a33, and a44 in the influencing parameter matrix are respectively related to the size relationship between the parameter corresponding to the matrix row m1 and the parameter corresponding to the matrix column m1, the size relationship between the parameter corresponding to the matrix row m2 and the parameter corresponding to the matrix column m2, the size relationship between the parameter corresponding to the matrix row m3 and the parameter corresponding to the matrix column m3, and the size relationship between the parameter corresponding to the matrix row m4 and the parameter corresponding to the matrix column m4. The size relationships of the four are the same, so the values ​​of the matrix elements a11, a22, a33, and a44 are all 1. For the matrix element a12, it involves the size relationship between the parameter corresponding to the matrix row m1 and the parameter corresponding to the matrix column m2. The interval corresponding to the difference between the two proportions is used to determine the relevant preset cost impact coefficient as its element value. The values ​​of matrix elements a13, a14, a23, a24, and a34 are determined in the same manner as a12 and are not described in detail here. The values ​​of matrix elements a21, a31, a32, a41, a42, and a43 are the reciprocals of the values ​​of matrix elements a12, a13, a14, a23, a24, and a34. For example, if the value of matrix element a12 is 3, the value of a21 is 1 / 3.

[0082] Understandably, the historical rail transit corresponding to the rail transit that currently needs to undergo cost assessment has similarities in geological conditions, labor, materials and other factors, but the labor and materials corresponding to the area where the rail transit is located still have their own special variation characteristics. Moreover, there may be multiple historical rail transits corresponding to the rail transit, so that the preset cost impact coefficients corresponding to the same pairwise parameters in different historical influencing parameter value groups are different, so that the values ​​of the same matrix element generated based on the preset cost impact coefficients are also multiple. Therefore, in order to more accurately reflect the impact of the labor and materials of the rail transit itself on the cost, and to better combine the cost impact factors of each historical rail transit to reflect the cost changes of the rail transit, in this embodiment, the step of generating the matrix element values ​​in the influencing parameter matrix based on the preset cost impact coefficients between the pairwise parameters in each historical influencing parameter value group includes:

[0083] Step S211, for the same pairwise parameters in each of the historical influencing parameter value groups, counting the number of positive relationships and the number of negative relationships, and performing a ratio operation between the number of positive relationships and the number of negative relationships to obtain a positive-negative ratio;

[0084] Step S212, obtaining the construction period corresponding to the engineering project, and determining a material variation factor array and a labor variation factor array corresponding to the construction period;

[0085] Step S213, performing mean calculation on the preset cost impact coefficients corresponding to the same pairwise parameters in each of the historical impact parameter value groups, and correcting the result of the mean calculation based on the positive-negative ratio, the material variation factor array and the manual variation factor array to obtain the matrix element value;

[0086] Furthermore, for the same pairwise parameters in each historical influencing parameter value group, the proportion relationship between the two is counted according to the order of arrangement of the parameters. The former is larger than the latter and is considered a positive relationship, while the latter is larger than the former and is considered a negative relationship. For example, for the above m1, m2, m3, and m4, compare the proportion between m1 and m2. If m1 is larger than m2, it means that the impact of m1 on the cost is greater than that of m2, and it is considered a positive relationship. If m1 is smaller than m2, it means that the impact of m1 on the cost is less than that of m2, and it is considered a negative relationship. Count the number of positive relationships and the number of negative relationships to obtain the number of positive relationships and the number of negative relationships, and then perform a ratio operation between the number of positive relationships and the number of negative relationships to obtain the positive-negative ratio. A positive-negative ratio greater than 1 indicates that the number of historical rail transits in which m1 has a greater impact on the cost than m2 is larger, and vice versa.

[0087] Furthermore, the construction period of the engineering projects currently split by rail transit is obtained. The construction period is the estimated construction period of the engineering project, and then the material change factor array and the labor change factor array corresponding to the construction period are determined. Among them, the material change factor array is a set of price factors of various materials used in the engineering project that change during the construction period, and the labor change factor array is a set of price factors of the types of work required by the engineering project that change during the construction period. The reason for the change in material prices may be that the materials are in short supply during the construction period, and other materials are used to replace them, or the price changes caused by environmental factors; the reason for the change in labor prices may be the change in labor prices caused by the shortage of labor during the construction period. Both can be determined based on the changes in the area where the project is located over the years.

[0088] Furthermore, for the same pairwise parameters in each historical influencing parameter value group, the mean operation is performed based on the preset cost impact coefficients corresponding to each pairwise parameter, that is, the preset cost impact coefficients corresponding to each pairwise parameter are added, and the operation result is obtained and the ratio operation is performed with the number of historical influencing parameter value groups to obtain the final mean operation result. Then, the result of the mean operation is corrected by the positive and negative ratio, the material variation factor array and the artificial variation factor array to obtain the final evidence element value. The specific correction calculation formula is as follows (1).

[0089]

[0090] Among them, Pxy represents the matrix element value of the xth row and yth column in the influencing parameter matrix, θ represents the positive-negative ratio, s1 represents the number of material types in the material variation factor array, and M 1c1 Indicates the proportion of the c1th material in the material variation factor array in the engineering project, ΔP 1c1 represents the price change factor of the c1th material in the material change factor array, s2 represents the number of types of work in the labor change factor array, M 2c2 Indicates the proportion of the c2th type of work in the project in the labor change factor array, ΔP 2c2 represents the price variation factor of the c2th type of work in the labor variation factor array, f(·) represents the weight function of materials and labor, and Axy represents the result of the mean calculation of the preset cost impact coefficients corresponding to the xth parameter and the yth parameter in each of the historical impact parameter value groups.

[0091] Among them, the proportion of various materials in the project and the price change factors of various materials in the material change factor array can exist in the material change factor array in the form of key-value pairs, and the proportion of materials and the price change factors of materials can reflect the size of price changes, thereby reflecting the overall price changes of various materials in the project. At the same time, the proportion of various types of work in the project and the price change factors of various types of work in the labor change factor array can also exist in the labor change factor array in the form of key-value pairs, and the proportion of the type of work refers to the proportion of the number of people in a certain type of work in the total number of employees. The price change factor of the type of work includes the price change caused by the change of the type of work, and the price change caused by the change of the number of people required for the type of work while the type of work remains unchanged. The size of price changes is reflected by the proportion of the type of work and the price change factor of the type of work, thereby reflecting the overall price changes of various types of work in the project. Then, the weight function of materials and labor is used to measure the impact of the changes of the two on the overall cost of the project, and the change weight is obtained. The mean operation result is corrected by the change weight to obtain the matrix element value. In this way, while reflecting the changes in rail transit costs through the cost influencing factors of various historical rail transit costs, it also reflects the impact of labor and materials on the cost of rail transit during the construction period.

[0092] Step S22, performing conversion calculation on each matrix element value in the influencing parameter matrix according to a preset conversion formula to generate a consistency matrix corresponding to the influencing parameter matrix;

[0093] Understandably, the matrix element values ​​in the influencing parameter matrix are generated based on the mean operation results of the preset cost impact coefficients, and there may be inaccuracies in the process of setting the preset cost impact coefficients. For example, even if one parameter is larger than another parameter, it means that its impact on the cost is large, but whether the impact is huge or very large may be inaccurate. Therefore, in order to improve the accuracy, after the values ​​of each matrix element in the influencing parameter matrix are corrected and generated, the influencing parameter matrix needs to be converted into a consistency matrix and a consistency check is performed. Specifically, a preset conversion formula is pre-set, and each matrix element value in the influencing parameter matrix is ​​converted and calculated by the preset conversion formula. After each matrix element value is converted, a consistency matrix corresponding to the influencing parameter matrix is ​​obtained. Among them, the preset conversion formula can be found in the following formula (2).

[0094]

[0095] Among them, B ij represents the element in the i-th row and j-th column of the consistency matrix, n represents the number of rows in the influence parameter matrix, Ai represents the sum of the elements in the i-th row of the influence parameter matrix, Ak represents the sum of the elements in the k-th row of the influence parameter matrix, Aj represents the sum of the elements in the j-th row of the influence parameter matrix, Amax represents the maximum value of the sum of the elements in each row of the influence parameter matrix, and Amin represents the minimum value of the sum of the elements in each row of the influence parameter matrix. For example, for a 3*3 matrix, the elements in the first row are 1, 3, 2, the elements in the second row are 4, 1, 2, and the elements in the third row are 2, 2, 1. If you need to convert to obtain B 12 At this time, i=1, j=2, n=9, A1=6, the value of k is from 1 to 3, A2=7, Amax=7, Amin=5.

[0096] Furthermore, after the influencing parameter matrix is ​​converted into a consistency matrix, the consistency index value of the consistency matrix is ​​calculated to perform a consistency check through the consistency index value. Specifically, the step of converting and calculating each matrix element value in the influencing parameter matrix according to a preset conversion formula to generate a consistency matrix corresponding to the influencing parameter matrix includes:

[0097] Step a1, calculating the consistency index value of the consistency matrix based on a preset judgment formula;

[0098] Step a2, comparing the consistency index value with a preset index threshold to determine whether the consistency index value is greater than the preset index threshold;

[0099] Step a3: If it is greater than the preset indicator threshold, the preset cost impact coefficient between each parameter in the historical impact parameter value group is adjusted based on the preset adjustment rule, and after the adjustment, the step of generating the matrix element value in the impact parameter matrix based on the preset cost impact coefficient between each parameter in the historical impact parameter value group is executed.

[0100] Furthermore, a preset judgment formula for calculating the consistency index value is pre-set, and each element value of the consistency matrix is ​​calculated by the preset judgment formula to obtain the consistency index value of the consistency matrix. The preset judgment formula can be specifically referred to in the following formula (3).

[0101]

[0102] Among them, μ represents the consistency index value, m represents the number of elements in the consistency matrix, and b ij represents the element in the i-th row and j-th column of the consistency matrix, b ji represents the element in the jth row and ith column of the consistency matrix, w i represents the influence weight value of the parameter corresponding to the i-th row in the influence parameter matrix, w j Indicates the influence weight value of the parameter corresponding to the jth row in the influencing parameter matrix. The row and column titles of the influencing parameter matrix are composed of the parameter factors in the influencing parameter factor group. The influence weight value of the parameter corresponding to the row indicates the influence of the parameter factor corresponding to the row title on the cost, which can be set according to the historical cost data of rail transit.

[0103] Furthermore, in order to indicate the size of the consistency index value, a preset index threshold is pre-set, and the calculated consistency index value is compared with the preset index threshold to determine whether the consistency index value is greater than the preset index threshold. If the consistency index value is determined to be greater than the preset index threshold through comparison, it means that the consistency of the consistency matrix is ​​low, the consistency test fails, and the preset cost impact coefficient set between the two parameters in the historical influencing parameter value group is inaccurate. At this time, the preset cost impact coefficient between the two parameters in the historical influencing parameter value group is adjusted according to the preset adjustment rules, and after the adjustment is completed, the matrix element values ​​in the influencing parameter matrix are regenerated according to the adjusted preset cost impact coefficients, and the influencing parameter matrix is ​​converted into a consistency matrix to recalculate the consistency index value until the calculated consistency index value is less than the preset index threshold, so as to ensure that the converted consistency matrix has a high consistency.

[0104] Among them, the preset adjustment rule can be to adjust the preset cost impact coefficients of all pairwise parameters in all historical impact parameter value groups, or to adjust the preset cost impact coefficients of all pairwise parameters in some historical impact parameter value groups, or to adjust the preset cost impact coefficients of some pairwise parameters in all historical impact parameter value groups. In addition, the specific adjustment method can also be set to be differentiated, for example, the preset cost impact coefficients of pairwise parameters in a certain historical impact parameter value group are set to be adjusted in an arithmetic progression, and the preset cost impact coefficients of pairwise parameters in another historical impact parameter value group are set to be adjusted in a geometric ratio, etc.

[0105] Step S23: normalize the elements in the consistency matrix to obtain the influencing factor weights corresponding to the influencing parameter factor group.

[0106] Furthermore, after determining that the consistency index value is less than the preset index threshold and the consistency check passes, each element in the consistency matrix is ​​normalized, and the influencing factor weight corresponding to the influencing parameter factor group is obtained through normalization. Specifically, the step of normalizing the elements in the consistency matrix to obtain the influencing factor weight corresponding to the influencing parameter factor group includes:

[0107] Step S231, performing a ratio operation on each matrix element in the consistency matrix and the sum of the matrix columns where the matrix element is located to obtain a normalized matrix;

[0108] Step S232, performing a sum operation on the row elements of each matrix row in the normalized matrix to obtain a sum result corresponding to each matrix row;

[0109] Step S233: performing a ratio operation on the sum result and the number of matrix rows of the normalized matrix to obtain the weight of the impact factor.

[0110] Furthermore, the sum of each matrix column of the consistency matrix is ​​calculated, and then each matrix element in the consistency matrix is ​​ratioed with the sum of the matrix column in which it is located to obtain a normalized matrix. For example, for a 3*3 matrix that has passed consistency verification, if the matrix elements contained in its first matrix column are 1, 2, and 3, the matrix elements contained in the second matrix column are 4, 5, and 6, and the matrix elements contained in the third matrix column are 7, 8, and 9, then the sum of its first matrix column is 6, the sum of the second matrix column is 15, and the sum of the third matrix column is 24. The results of the ratio operation of each matrix element of the first matrix column and the sum of the first matrix column are 1 / 6, 1 / 3, and 1 / 2, respectively, the results of the ratio operation of each matrix element of the second matrix column and the sum of the second matrix column are 4 / 15, 1 / 3, and 2 / 5, respectively, and the results of the ratio operation of each matrix element of the third matrix column and the sum of the third matrix column are 7 / 24, 1 / 3, and 3 / 8, respectively. In this way, a normalized matrix is ​​obtained.

[0111] Furthermore, the row elements of each matrix row in the normalized matrix are summed to obtain the summed result corresponding to each matrix row in the normalized matrix. For the above 3*3 normalized matrix, the summed result of the first matrix row can be expressed as (1 / 6+4 / 15+7 / 24=0.725), the summed result of the second matrix row can be expressed as (1 / 3, 1 / 3, 1 / 3=1), and the summed result of the third matrix column can be expressed as (1 / 2+2 / 5+3 / 8=1.275). Then, the summed result is used to perform a ratio operation with the number of matrix rows of the normalized matrix, that is, the influence factor weight corresponding to the influencing parameter factor group is obtained. For the above 3*3 matrix, the number of matrix rows is 3, and the influence factor weights obtained by the ratio operation are 0.24, 0.33, and 0.43 respectively. The influence factor weight reflects the degree of influence of each parameter factor in the influencing parameter factor group on the cost.

[0112] This embodiment constructs an influence parameter matrix by combining the degree of influence on the cost reflected by two parameters in multiple historical influence parameter value groups, and the material change factors and labor change factors in the area where the current rail transit is located. While reflecting the changes in the cost of rail transit through the cost influencing factors of each historical rail transit, it also reflects the impact of labor and materials on the cost of rail transit during the construction period. At the same time, the influence parameter matrix is ​​converted into a consistency matrix for consistency testing, and normalized after the consistency test passes to generate influence factor weights to reflect the impact of each factor on the cost. The accuracy of the constructed influence parameter matrix is ​​ensured by consistency testing, so that the generated influence factor weights can accurately reflect the impact of each factor on the cost.

[0113] For further information, please refer to Figure 3Based on the first and second embodiments of the rail transit cost evaluation method based on big data of the present invention, a third embodiment of the rail transit cost evaluation method based on big data of the present invention is proposed.

[0114] The third embodiment of the rail transit cost assessment method based on big data is different from the first and second embodiments of the rail transit cost assessment method based on big data in that the step of obtaining the design file of the rail transit from the preset database includes:

[0115] Step S50, receiving an uploaded design file corresponding to the rail transit, and judging whether the uploaded design file is a standard file according to a standardized format corresponding to the preset database;

[0116] Step S60, if the uploaded design file is a standard file, the uploaded design file is parsed based on a preset recognition model to generate the design file;

[0117] Among them, the preset recognition model includes a first hidden layer, a second hidden layer, a third hidden layer and a fourth hidden layer, which are respectively used to parse out construction sections, section projects, engineering projects and project lists from the uploaded design files, and output them through the output layer of the preset recognition model.

[0118] Furthermore, the preset database stores the design files in a five-level structure, which relies on the recognition and conversion of the design files. The design files of rail transit are designed by different designers from different institutions, and usually have different formats and styles. In order to facilitate the conversion, the preset database is pre-set with convertible standardized formats and standard templates with such standardized formats. Designers can select standard templates in the early stage of design to make the formed design files have standardized formats.

[0119] Furthermore, when the backend server receives the uploaded design file for rail transit, it first identifies the format of the uploaded design file, and compares the identified format with the standardized format pre-set in the preset database. By comparing the two, it is determined whether the uploaded design file is a standard file with a standardized format. If the comparison determines that the formats of the two match, the uploaded design file is determined to be a standard file, which can be identified and converted to a five-level structure for storage. On the contrary, if the comparison determines that the formats of the two do not match, the uploaded design file is determined to be a non-standard file, and the format conversion must be performed before the five-level structure conversion.

[0120] Furthermore, for the recognition conversion of the five-level structure of the standard file, a preset recognition model is pre-trained in the background server. The preset recognition model includes an input layer, a hidden layer and an output layer. The input layer and the output layer are both one layer, and the hidden layer is four layers, namely the first hidden layer, the second hidden layer, the third hidden layer and the fourth hidden layer. The uploaded design file is input through the input layer, and the input layer identifies the rail transit, and the first hidden layer, the second hidden layer, the third hidden layer and the second hidden layer are respectively parsed to decompose the construction section, section project, engineering project and project list in the design file, and output through the output layer.

[0121] The preset recognition model is generated in advance by training a large number of historical rail transit design files in a standardized format. Specifically, the step of obtaining the rail transit design files from the preset database includes:

[0122] Step b1, obtaining a file sample formed based on historical standard design files of historical rail transit, and dividing the file sample into a training sample and a test sample;

[0123] Step b2, training a preset initial model based on the training sample, and when the training reaches a preset time, calculating the loss function value of the preset initial model based on the test sample;

[0124] Step b3, determining whether the loss function value is less than or equal to a preset loss threshold, if it is less than or equal to the preset loss threshold, generating the preset initial model as a preset recognition model;

[0125] Step b4: if the loss function value is greater than a preset loss threshold, updating the weight value between the fourth hidden layer and the output layer in the preset initial model;

[0126] Step b5, for the preset initial model after the weight value is updated, execute the step of training the preset initial model based on the training sample.

[0127] Furthermore, the design documents of historical rail transit in a standardized format are used as historical standard design documents of historical rail transit, and the historical standard design documents are annotated to form file samples. After obtaining such file samples, the background server divides them into training samples and test samples. In addition, a preset initial model and a preset duration representing the duration of a single training are pre-built. The preset initial model is trained with the training samples, and when the training duration reaches the preset duration, the loss function value of the preset initial model is calculated with the test samples. The loss function value can be specifically calculated by means of cross entropy, which represents the difference between the result obtained by parsing the test sample with the preset initial model and the expected result, and is used to reflect the performance of the preset initial model after training.

[0128] Furthermore, a preset loss threshold indicating the performance is preset, and the calculated loss function value is compared with the preset loss threshold to determine whether the loss function value is less than or equal to the preset loss threshold. If it is less than or equal to the preset loss threshold, it means that the difference between the result obtained by the preset initial model for parsing the test sample and the expected result is small, and the performance of the preset initial model after training is good, which can be used for parsing actual rail transit design documents, so the training of the preset initial model is stopped and it is generated as a preset recognition model.

[0129] Furthermore, if the loss function value is greater than the preset loss threshold value after comparison, it means that the difference between the result obtained by the preset initial model for parsing the test sample and the expected result is large, and the performance of the preset initial model after training is poor, and iterative training needs to be continued. At this time, the weight value between the fourth hidden layer and the output layer in the preset initial model is updated, and the preset initial model is iteratively trained through the training sample after the update, and the loss function value is re-calculated through the test sample until the calculated loss function value is less than the preset loss threshold. The training of the preset initial model is stopped and it is generated as the preset recognition model. Among them, in order to ensure the accuracy of the weight value update between the fourth hidden layer and the output layer, an update formula is set in advance through experiments, and the specific details can be seen in the following formula (4).

[0130]

[0131] Among them, ω represents the updated weight value, ω0 represents the weight value before updating, τ represents the learning rate of the preset initial model, T represents the current cumulative number of tests of the preset initial model, yt represents the test result corresponding to the tth test of the preset initial model, and y0 represents the reference test result corresponding to the test sample. By combining the difference between the test results of previous tests and the reference test results identified by the test sample, the weight value between the fourth hidden layer and the output layer in the preset initial model is updated, so that the difference between the test result and the reference test result after each update is reduced, so that the loss function value is gradually reduced and is less than the preset loss threshold. It should be noted that in addition to updating the weight value between the fourth hidden layer and the output layer, the weight values ​​between the third and fourth hidden layers, the second and third hidden layers, the first and second hidden layers, and the input layer and the first hidden layer can also be updated. The updating method can be the same as the weight value between the fourth hidden layer and the output layer, which will not be repeated here.

[0132] This embodiment sets a standardized format for design files, and parses the design files in the standardized format through a preset recognition model formed by pre-training a large number of samples, forming a five-layer structure stored in a preset database, so that the subsequent cost assessment of rail transit based on the split files is more accurate.

[0133] In addition, the embodiment of the present invention also provides a rail transit cost evaluation system based on big data. Figure 4 , Figure 4 It is a structural schematic diagram of the equipment hardware operating environment involved in the implementation scheme of the rail transit cost assessment system based on big data of the present invention.

[0134] like Figure 4 As shown, the rail transit cost assessment system based on big data may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a storage 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The storage 1005 may be a high-speed RAM storage, or a stable storage (non-volatile memory), such as a disk storage. The storage 1005 may also be a storage device independent of the aforementioned processor 1001.

[0135] Those skilled in the art will understand that Figure 4 The hardware structure of the rail transit cost evaluation system based on big data shown in the figure does not constitute a limitation of the rail transit cost evaluation system based on big data, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0136] like Figure 4 As shown, the storage 1005 as a storage medium may include an operating system, a network communication module, a user interface module and a control program. Among them, the operating system is a program for managing and controlling the rail transit cost assessment system based on big data and software resources, supporting the operation of the network communication module, the user interface module, the control program and other programs or software; the network communication module is used to manage and control the network interface 1004; the user interface module is used to manage and control the user interface 1003.

[0137] exist Figure 4In the hardware structure of the rail transit cost evaluation system based on big data shown in the figure, the network interface 1004 is mainly used to connect to the system server and communicate data with the system server; the user interface 1003 is mainly used to connect to the client (user end) and communicate data with the client; the processor 1001 can call the control program stored in the storage 1005 and perform the following operations:

[0138] Acquire a rail transit design document from a preset database, wherein the design document includes a plurality of construction bid sections, a plurality of bid section projects corresponding to each of the construction bid sections, a plurality of engineering projects corresponding to each of the bid section projects, and a project list corresponding to each of the engineering projects;

[0139] For each of the engineering projects, determine the influencing parameter factor group corresponding to the engineering project, and search for multiple historical influencing parameter value groups corresponding to the influencing parameter factor group in a preset database, and generate an influencing factor weight corresponding to the influencing parameter factor group based on each of the historical influencing parameter value groups;

[0140] Obtaining cost values ​​in a project list corresponding to the engineering project, and modifying each cost value according to the weight of the influencing factor to generate a project cost corresponding to the engineering project;

[0141] Based on the project cost generated for each of the engineering projects, the traffic cost corresponding to the rail transit is obtained, and according to the traffic cost, the cost of the rail transit is evaluated.

[0142] Furthermore, the step of generating the influence factor weight corresponding to the influence parameter factor group based on each of the historical influence parameter value groups includes:

[0143] Constructing an influence parameter matrix based on each parameter name in the influence parameter factor group, and generating matrix element values ​​in the influence parameter matrix based on preset cost influence coefficients between each pair of parameters in each historical influence parameter value group;

[0144] The values ​​of each matrix element in the influencing parameter matrix are converted and calculated according to a preset conversion formula to generate a consistency matrix corresponding to the influencing parameter matrix. The preset conversion formula is:

[0145]

[0146] Among them, B ijrepresents the element in the i-th row and j-th column of the consistency matrix, n represents the number of rows in the influencing parameter matrix, Ai represents the sum of the elements in the i-th row of the influencing parameter matrix, Ak represents the sum of the elements in the k-th row of the influencing parameter matrix, Aj represents the sum of the elements in the j-th row of the influencing parameter matrix, Amax represents the maximum value of the sum of the elements in each row of the influencing parameter matrix, and Amin represents the minimum value of the sum of the elements in each row of the influencing parameter matrix;

[0147] The elements in the consistency matrix are normalized to obtain the influencing factor weights corresponding to the influencing parameter factor group.

[0148] Furthermore, the step of generating the matrix element values ​​in the influencing parameter matrix based on the preset cost influencing coefficients between the parameters in each historical influencing parameter value group includes:

[0149] For the same pairwise parameters in each of the historical influencing parameter value groups, the number of positive relationships and the number of negative relationships are counted, and a ratio operation is performed between the number of positive relationships and the number of negative relationships to obtain a positive-negative ratio;

[0150] Obtaining the construction period corresponding to the engineering project, and determining the material variation factor array and the labor variation factor array corresponding to the construction period;

[0151] The preset cost impact coefficients corresponding to the same pairwise parameters in each of the historical impact parameter value groups are averaged, and the result of the average operation is corrected based on the positive and negative ratio, the material variation factor array and the manual variation factor array to obtain the matrix element value. The calculation formula for the correction is:

[0152]

[0153] Among them, Pxy represents the matrix element value of the xth row and yth column in the influencing parameter matrix, θ represents the positive-negative ratio, s1 represents the number of material types in the material variation factor array, and M 1c1 Indicates the proportion of the c1th material in the material variation factor array in the engineering project, ΔP 1c1 represents the price change factor of the c1th material in the material change factor array, s2 represents the number of types of work in the labor change factor array, M 2c2 Indicates the proportion of the c2th type of work in the project in the labor change factor array, ΔP 2c2 represents the price variation factor of the c2th type of work in the labor variation factor array, f(·) represents the weight function of materials and labor, and Axy represents the result of the mean calculation of the preset cost impact coefficients corresponding to the xth parameter and the yth parameter in each of the historical impact parameter value groups.

[0154] Furthermore, the step of converting and calculating each matrix element value in the influencing parameter matrix according to a preset conversion formula to generate a consistency matrix corresponding to the influencing parameter matrix includes:

[0155] The consistency index value of the consistency matrix is ​​calculated based on a preset judgment formula, and the preset judgment formula is:

[0156]

[0157] Among them, μ represents the consistency index value, m represents the number of elements in the consistency matrix, and b ij represents the element in the i-th row and j-th column of the consistency matrix, b ji represents the element in the jth row and ith column of the consistency matrix, w i represents the influence weight value of the parameter corresponding to the i-th row in the influence parameter matrix, w j Represents the influence weight value of the parameter corresponding to the jth row in the influence parameter matrix;

[0158] Compare the consistency index value with a preset index threshold to determine whether the consistency index value is greater than the preset index threshold;

[0159] If it is greater than the preset indicator threshold, the preset cost impact coefficient between each parameter in the historical influencing parameter value group is adjusted based on the preset adjustment rule, and after the adjustment, the step of generating the matrix element value in the influencing parameter matrix based on the preset cost impact coefficient between each parameter in each historical influencing parameter value group is executed.

[0160] Furthermore, the step of normalizing the elements in the consistency matrix to obtain the influencing factor weights corresponding to the influencing parameter factor group includes:

[0161] Performing a ratio operation on each matrix element in the consistency matrix and the sum of the matrix columns where the matrix element is located to obtain a normalized matrix;

[0162] Performing a sum operation on the row elements of each matrix row in the normalized matrix to obtain a sum result corresponding to each matrix row;

[0163] A ratio operation is performed between the sum result and the number of matrix rows of the normalized matrix to obtain the impact factor weight.

[0164] Furthermore, the step of obtaining the traffic cost corresponding to the rail transit based on the project cost generated for each engineering project, and conducting a cost assessment on the rail transit according to the traffic cost includes:

[0165] The project costs of the engineering projects corresponding to each of the bidding sections are added together to generate the engineering cost of the bidding section engineering, and the project costs of the bidding section engineering corresponding to each of the construction bidding sections are added together to generate the bidding section cost of the construction bidding section;

[0166] The traffic cost corresponding to the rail transit is generated based on the cumulative calculation of the cost of each bidding section, and the engineering cost, the bidding section cost and the traffic cost are evaluated respectively based on the preset engineering cost, the preset bidding section cost and the preset traffic cost.

[0167] Furthermore, before the step of acquiring the rail transit design file from the preset database, the processor 1001 may call the control program stored in the storage 1005 and perform the following operations:

[0168] Receiving an uploaded design file corresponding to the rail transit, and judging whether the uploaded design file is a standard file according to a standardized format corresponding to the preset database;

[0169] If the uploaded design file is a standard file, the uploaded design file is parsed based on a preset recognition model to generate the design file;

[0170] Among them, the preset recognition model includes a first hidden layer, a second hidden layer, a third hidden layer and a fourth hidden layer, which are respectively used to parse out construction sections, section projects, engineering projects and project lists from the uploaded design files, and output them through the output layer of the preset recognition model.

[0171] Furthermore, before the step of acquiring the rail transit design file from the preset database, the processor 1001 may call the control program stored in the storage 1005 and perform the following operations:

[0172] Acquire a file sample formed based on a historical standard design file of historical rail transit, and divide the file sample into a training sample and a test sample;

[0173] Training a preset initial model based on the training sample, and when the training reaches a preset time, calculating a loss function value of the preset initial model based on the test sample;

[0174] Determine whether the loss function value is less than or equal to a preset loss threshold, and if so, generate the preset initial model as a preset recognition model;

[0175] If the loss function value is greater than the preset loss threshold, the weight value between the fourth hidden layer and the output layer in the preset initial model is updated, and the update formula is:

[0176]

[0177] Among them, ω represents the updated weight value, ω0 represents the weight value before the update, τ represents the learning rate of the preset initial model, T represents the current cumulative number of tests of the preset initial model, yt represents the test result corresponding to the tth test of the preset initial model, and y0 represents the reference test result corresponding to the test sample;

[0178] For the preset initial model after the weight value is updated, a step of training the preset initial model based on the training sample is performed.

[0179] The specific implementation of the rail transit cost evaluation system based on big data of the present invention is basically the same as the various embodiments of the rail transit cost evaluation method based on big data described above, and will not be repeated here.

[0180] The embodiment of the present invention further provides a storage medium having a control program stored thereon, and when the control program is executed by a processor, the steps of the rail transit cost evaluation method based on big data are implemented.

[0181] The storage medium of the present invention may be a computer-readable storage medium, and its implementation method is substantially the same as the above-mentioned embodiments of the rail transit cost assessment method based on big data, and will not be described in detail here.

[0182] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims. All equivalent structures or equivalent process changes made using the contents of the specification and drawings of the present invention, or directly or indirectly used in other related technical fields, are protected by the present invention.

Claims

1. A rail transit cost assessment method based on big data, characterized in that: The rail transit cost assessment method comprises: Acquire a rail transit design document from a preset database, wherein the design document includes a plurality of construction bid sections, a plurality of bid section projects corresponding to each of the construction bid sections, a plurality of engineering projects corresponding to each of the bid section projects, and a project list corresponding to each of the engineering projects; For each of the engineering projects, determine the influencing parameter factor group corresponding to the engineering project, and search for multiple historical influencing parameter value groups corresponding to the influencing parameter factor group in a preset database, and generate an influencing factor weight corresponding to the influencing parameter factor group based on each of the historical influencing parameter value groups; Obtaining cost values ​​in a project list corresponding to the engineering project, and modifying each cost value according to the weight of the influencing factor to generate a project cost corresponding to the engineering project; Based on the project cost generated for each engineering project, obtain the traffic cost corresponding to the rail transit, and perform a cost assessment on the rail transit according to the traffic cost; The step of generating the influence factor weight corresponding to the influence parameter factor group based on each of the historical influence parameter value groups includes: Constructing an influence parameter matrix based on each parameter name in the influence parameter factor group, and generating matrix element values ​​in the influence parameter matrix based on preset cost influence coefficients between each pair of parameters in each historical influence parameter value group; The values ​​of each matrix element in the influencing parameter matrix are converted and calculated according to a preset conversion formula to generate a consistency matrix corresponding to the influencing parameter matrix. The preset conversion formula is: Among them, B ij represents the element in the i-th row and j-th column of the consistency matrix, n represents the number of rows in the influencing parameter matrix, Ai represents the sum of the elements in the i-th row of the influencing parameter matrix, Ak represents the sum of the elements in the k-th row of the influencing parameter matrix, Aj represents the sum of the elements in the j-th row of the influencing parameter matrix, Amax represents the maximum value of the sum of the elements in each row of the influencing parameter matrix, and Amin represents the minimum value of the sum of the elements in each row of the influencing parameter matrix; Normalizing the elements in the consistency matrix to obtain the influencing factor weights corresponding to the influencing parameter factor group; Wherein, the step of generating the matrix element values ​​in the influencing parameter matrix based on the preset cost influencing coefficients between the parameters in each historical influencing parameter value group comprises: For the same pairwise parameters in each of the historical influencing parameter value groups, the number of positive relationships and the number of negative relationships are counted, and a ratio operation is performed between the number of positive relationships and the number of negative relationships to obtain a positive-negative ratio; Obtaining the construction period corresponding to the engineering project, and determining the material variation factor array and the labor variation factor array corresponding to the construction period; The preset cost impact coefficients corresponding to the same pairwise parameters in each of the historical impact parameter value groups are averaged, and the result of the average operation is corrected based on the positive and negative ratio, the material variation factor array and the manual variation factor array to obtain the matrix element value. The calculation formula for the correction is: Among them, Pxy represents the matrix element value of the xth row and yth column in the influence parameter matrix. represents the positive-negative ratio, s1 represents the number of material types in the material variation factor array, M 1c1 Indicates the proportion of the c1th material in the material variation factor array in the engineering project, ΔP 1c1 represents the price change factor of the c1th material in the material change factor array, s2 represents the number of types of work in the labor change factor array, M 2c2 Indicates the proportion of the c2th type of work in the project in the labor change factor array, ΔP 2c2 represents the price variation factor of the c2th type of work in the labor variation factor array, f(·) represents the weight function of materials and labor, and Axy represents the result of the mean calculation of the preset cost impact coefficients corresponding to the xth parameter and the yth parameter in each of the historical impact parameter value groups.

2. The rail transit cost assessment method according to claim 1, characterized in that: After the step of converting and calculating the values ​​of each matrix element in the influencing parameter matrix according to a preset conversion formula to generate a consistency matrix corresponding to the influencing parameter matrix, the following steps are included: The consistency index value of the consistency matrix is ​​calculated based on a preset judgment formula, and the preset judgment formula is: Among them, μ represents the consistency index value, m represents the number of elements in the consistency matrix, and b ij represents the element in the i-th row and j-th column of the consistency matrix, b ji represents the element in the jth row and ith column of the consistency matrix, w i represents the influence weight value of the parameter corresponding to the i-th row in the influence parameter matrix, w j Represents the influence weight value of the parameter corresponding to the jth row in the influence parameter matrix; Compare the consistency index value with a preset index threshold to determine whether the consistency index value is greater than the preset index threshold; If it is greater than the preset indicator threshold, the preset cost impact coefficient between each parameter in the historical influencing parameter value group is adjusted based on the preset adjustment rule, and after the adjustment, the step of generating the matrix element value in the influencing parameter matrix based on the preset cost impact coefficient between each parameter in each historical influencing parameter value group is executed.

3. The rail transit cost assessment method according to claim 1, characterized in that: The step of normalizing the elements in the consistency matrix to obtain the influencing factor weights corresponding to the influencing parameter factor group comprises: Performing a ratio operation on each matrix element in the consistency matrix and the sum of the matrix columns where the matrix element is located to obtain a normalized matrix; Performing a sum operation on the row elements of each matrix row in the normalized matrix to obtain a sum result corresponding to each matrix row; A ratio operation is performed between the sum result and the number of matrix rows of the normalized matrix to obtain the impact factor weight.

4. The rail transit cost assessment method according to any one of claims 1 to 3, characterized in that: The step of obtaining the traffic cost corresponding to the rail transit based on the project cost generated for each engineering project, and conducting a cost assessment on the rail transit according to the traffic cost comprises: The project costs of the engineering projects corresponding to each of the bidding sections are added together to generate the engineering cost of the bidding section engineering, and the project costs of the bidding section engineering corresponding to each of the construction bidding sections are added together to generate the bidding section cost of the construction bidding section; The traffic cost corresponding to the rail transit is generated based on the cumulative calculation of the cost of each bidding section, and the engineering cost, the bidding section cost and the traffic cost are evaluated respectively based on the preset engineering cost, the preset bidding section cost and the preset traffic cost.

5. The rail transit cost assessment method according to any one of claims 1 to 3, characterized in that: The step of obtaining the design file of rail transit from the preset database includes: Receiving an uploaded design file corresponding to the rail transit, and determining whether the uploaded design file is a standard file according to a standardized format corresponding to the preset database; If the uploaded design file is a standard file, the uploaded design file is parsed based on a preset recognition model to generate the design file; Among them, the preset recognition model includes a first hidden layer, a second hidden layer, a third hidden layer and a fourth hidden layer, which are respectively used to parse out construction sections, section projects, engineering projects and project lists from the uploaded design files, and output them through the output layer of the preset recognition model.

6. The rail transit cost assessment method according to claim 5, characterized in that: The step of obtaining the design file of rail transit from the preset database includes: Acquire a file sample formed based on a historical standard design file of historical rail transit, and divide the file sample into a training sample and a test sample; Training a preset initial model based on the training sample, and when the training reaches a preset time, calculating a loss function value of the preset initial model based on the test sample; Determine whether the loss function value is less than or equal to a preset loss threshold, and if so, generate the preset initial model as a preset recognition model; If the loss function value is greater than the preset loss threshold, the weight value between the fourth hidden layer and the output layer in the preset initial model is updated, and the update formula is: Wherein, ω represents the updated weight value, ω0 represents the weight value before the update, τ represents the learning rate of the preset initial model, T represents the current cumulative number of tests of the preset initial model, yt represents the test result corresponding to the tth test of the preset initial model, and y0 represents the reference test result corresponding to the test sample; For the preset initial model after the weight value is updated, a step of training the preset initial model based on the training sample is performed.

7. A rail transit cost assessment system based on big data, characterized in that: The rail transit cost assessment system includes a storage device, a processor, a communication bus, and a control program stored in the storage device: The communication bus is used to realize the connection and communication between the processor and the storage; The processor is used to execute the control program to implement the steps of the rail transit cost assessment method based on big data as described in any one of claims 1-6.

8. A storage medium, characterized in that: The storage medium stores a control program, and when the control program is executed by the processor, the steps of the rail transit cost assessment method based on big data as described in any one of claims 1 to 6 are implemented.

Images