A building budget evaluation method and system based on big data

Abstract

This invention relates to the field of building budget assessment technology, specifically to a building budget assessment method and system based on big data. In use, this invention achieves standardized processing of budget data and precise selection of core parameters through standardized cleaning and the 3σ principle combined with a feature ranking algorithm based on information gain. Simultaneously, based on an innovative combination of process topology construction, transitive closure operation, and Boolean operation, it uncovers easily overlooked implicit processes, filling the gap in traditional methods' inability to identify implicit processes. It constructs a comprehensive consumption vector integrating explicit and implicit process consumption data, and, combined with relative error residual calculation, dual threshold judgment, and a 0-100 score system, achieves a quantitative assessment of budget integrity.

Description

Technical Field

[0001] This invention relates to the field of building budget assessment technology, and specifically to a building budget assessment method and system based on big data. Background Technology

[0002] Construction project budgeting is a core step before construction projects are implemented. Its completeness and accuracy directly determine the effectiveness of project investment control, construction schedule arrangement, and project quality assurance. It is a key foundation for the full life cycle management of construction projects.

[0003] Existing budget assessment methods have limited data processing capabilities. Construction budget list data contains multi-dimensional information such as work processes, quantities, consumption standards, and unit prices. The data volume is huge and the format is inconsistent. Manual cleaning and standardization are inefficient and prone to omissions or misjudgments of abnormal noise data, leading to deviations in subsequent assessment results. Moreover, the extraction of core feature parameters lacks scientific basis and relies heavily on the experience of staff for screening, which easily leads to the omission of parameters that have a significant impact on budget assessment or the introduction of redundant parameters, affecting the accuracy of assessment. Traditional methods cannot effectively uncover the implicit dependencies between construction processes. There are a large number of hidden auxiliary processes that are easily overlooked during construction. Such omissions will lead to incomplete budgets, which in turn will cause problems such as cost overruns and schedule delays during construction. Summary of the Invention

[0004] This invention effectively solves the problems of non-standard data processing, difficulty in removing abnormal data, and reliance on manual feature selection in existing building budget assessments.

[0005] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a big data-based method for building budget assessment, comprising the following steps: S1. Data Acquisition and Preprocessing: Obtain the budget item list data of the target building project, clean and standardize the list data, remove abnormal noise data, and extract core feature parameters. S2. Feature vector construction: Map the preprocessed core feature parameters to a multidimensional feature space to construct an initial multidimensional feature vector representing the project features. S3. Construction of process topology: Based on the standard specification library of construction engineering processes, the dependency relationship and association logic between processes are analyzed, an adjacency matrix representing the process topology is constructed, and the adjacency matrix is ​​binarized. S4. Implicit process derivation: Perform transitive closure operation on the processed adjacency matrix to construct the reachability graph of process dependencies, and derive the set of implicit processes under specific working conditions through Boolean operations on matrix elements. S5. Construction of comprehensive consumption vector: Based on the implicit process set and the initial multidimensional feature vector, construct a comprehensive consumption vector containing complete process logic and generate budget integrity assessment residual values. S6. Evaluation and Result Output: Compare the residual value with the preset evaluation threshold range to determine the logical integrity of the building budget cost, and output the evaluation result including suggestions for filling in any missing items.

[0006] Furthermore, in step S1, the project list data includes process name, quantity of work, labor consumption standard, material consumption standard, machinery consumption standard, and corresponding unit price data. The process name must be completely consistent with the process name in the construction engineering process standard specification library to avoid errors in process dependency resolution due to inconsistent names. The quantity of work is the actual construction quantity of each process of the target project, which needs to be determined in conjunction with construction drawings and on-site survey data to ensure that the data is true and accurate. The labor consumption standard, material consumption standard, and machinery consumption standard are all based on the current industry standards and adjusted according to the market conditions of the project location to ensure that the consumption standards are in line with the actual situation of the project. The corresponding unit price data is the current fair market unit price, which is captured and updated in real time by the big data platform to avoid budget evaluation deviation due to unit price lag. In step S1, the removal of abnormal noise data adopts the 3σ principle, calculating and removing outliers for each expense item in the list data. The calculation formula is as follows: , In the formula, This represents the mean of all expense items in the list data, indicating the normal center of value for that expense item. Specifically, it is the arithmetic mean of all valid sample data for that expense item. The standard deviation of each cost item represents the degree of dispersion of its values. The larger the value, the greater the fluctuation in the value of that expense item. The smaller the value, the more concentrated the values. Here, n represents the sample size for each cost item. For degrees of freedom, adopt An unbiased estimate of the sample standard deviation is performed to avoid estimation bias when the sample size is small, thus meeting the sample statistical requirements of construction budget list data. For each specific value of the expense item, that is, for each valid data point of the expense item in the list, exclude values ​​within the range of... Data points other than those listed above are considered abnormal noise data, specifically abnormal data that exceeds the normal range by too much, does not conform to market rules or actual construction conditions, such as the cost of ordinary wall plastering at a unit price of 10,000 yuan / square meter. In step S1, the core feature parameter extraction adopts a feature importance ranking algorithm based on information gain to screen feature parameters with an impact weight ≥5% on the budget assessment. These include the quantity of work, the unit price of main materials, and the time consumed in each process. The feature importance ranking algorithm based on information gain calculates the information gain value of each feature parameter on the budget assessment result. The higher the information gain value, the greater the impact of the feature parameter on the budget assessment, and vice versa. The screening threshold of ≥5% impact weight is set in combination with the actual needs of construction budget assessment. After verification by a large number of similar projects, a weight of ≥5% can effectively screen out core feature parameters and eliminate redundant parameters with minimal impact on the budget assessment. The quantity of work specifically refers to the total amount of construction work for each process, which is the core basis for budget calculation. The unit price of main materials refers to the unit price of the main materials of the project, such as steel bars, cement, and sand and gravel, which directly affects the total budget cost. The time consumed in each process refers to the estimated construction time of each process, which affects the consumption costs of labor and machinery. The three together constitute the core features of the budget assessment.

[0007] Further, in step S2, the initial multidimensional feature vector is constructed using the filtered core feature parameters as the vector dimension. The initial vector is expressed by the following formula: Where n is the number of core feature parameters, and to eliminate dimensional differences, the Z-Score standardization formula is used to process the vector components to obtain an initial multidimensional feature vector. The calculation formula for the standardization process is as follows: In the formula, These are the standardized feature parameter values. Let be the original value of the i-th core feature parameter. Let be the historical mean of the i-th core feature parameter. The historical standard deviation of the i-th core feature parameter is used. The standardized initial multidimensional feature vector is used for subsequent process topology construction and residual calculation, providing standardized data support for budget evaluation.

[0008] Furthermore, in step S3, the dependencies between processes include two types: mandatory association between preceding and subsequent processes, and collaborative association between parallel processes. Mandatory association between preceding and subsequent processes specifically means that the subsequent process can only begin construction after the preceding process is completely completed, and cannot be carried out in advance. For example, formwork erection is a mandatory preceding process for concrete pouring, and concrete pouring can only be carried out after the formwork erection is completed. Collaborative association between parallel processes specifically means that two or more processes can be carried out simultaneously without affecting each other, and can collaboratively promote the construction progress. For example, steel bar binding and water and electricity pipeline pre-embedding can be carried out in parallel without waiting for each other. The core of the analysis of the two types of dependencies is to combine the building construction process standards to ensure that the analysis results conform to the actual construction process and have no logical errors. In step S3, the adjacency matrix is ​​denoted as A, which is an m×m matrix, where m is the total number of explicit processes in the project. The construction logic of the adjacency matrix is ​​as follows: when the process... For process In the direct preceding process, matrix elements When process i and process j have no direct dependency, the matrix elements 0. Binarization of the adjacency matrix retains only two values, 0 and 1, to clarify the connectivity between processes, eliminate the interference of the workload on the logical judgment of the process, and adapt to the actual needs of building budget assessment. After binarization, the matrix elements only represent whether there is a direct dependency, without involving quantitative indicators such as workload and time consumption, ensuring the accuracy of the logical judgment of the process.

[0009] Further, in step S4, the transitive closure operation employs the Warshall algorithm to operate on the binarized adjacency matrix to obtain the transitive closure matrix. The operation formula is as follows: In the formula, The closure matrix is ​​used to represent all dependencies between processes, including direct and indirect dependencies. The initial adjacency matrix is ​​the same as the total number of explicit operations; that is, m is the number of rows or columns in the initial adjacency matrix. Let be the matrix order, and be the square of the initial adjacency matrix A, representing the indirect dependency between processes via one intermediate process. This represents an indirect dependency relationship involving two intermediate processes, and so on. The indirect dependencies between m-1 intermediate processes can be represented by the above calculations, which can completely cover all dependencies between processes without omission. In step S4, the reachability between processes is determined by Boolean operations on the matrix elements: when the closure matrix is ​​passed... medium elements And the initial adjacency matrix Corresponding element When, explain the process. With process An indirect dependency relationship exists, leading to the derivation of implicit procedures. Combining this with a building construction technology knowledge base, implicit procedures matching the current project are selected, forming a set of implicit procedures. ,in The number of implicit processes is determined by Boolean operations, which simultaneously satisfy two conditions: the element of the transitive closure matrix is ​​1 and the corresponding element of the initial adjacency matrix is ​​0. Implicit processes are auxiliary processes that are necessary to carry out during construction but are easily overlooked, such as concrete curing and cleaning after formwork removal. Although these processes are not explicitly listed in the budget, their absence will affect the construction quality and schedule, and are considered budget omissions. The construction technology knowledge base covers the complete construction processes of various construction projects and can accurately match them according to project type and construction technology to ensure that the selected implicit processes are fully compatible with the target project.

[0010] Further, in step S5, the method for constructing the comprehensive consumption vector is as follows: the consumption parameters of each process in the implicit process set T, including labor, material and machinery consumption, are mapped into vector components and spliced ​​and fused with the initial multidimensional feature vector obtained in step S2 to form a comprehensive consumption vector containing complete process logic. The method for obtaining the consumption parameters of the hidden process is as follows: combining the knowledge base of building construction technology and historical data of similar projects, the labor consumption standard, material consumption standard and machinery consumption standard of each hidden process are determined. Then, combined with the engineering quantity and market unit price of the target project, the specific values ​​of the consumption parameters of each hidden process are calculated. The vector component mapping adopts the same mapping rule as the initial multidimensional feature vector to ensure that the format and dimension of the hidden process consumption vector components are consistent with the initial multidimensional feature vector components. When splicing and fusion, the splicing is strictly carried out in the order of initial multidimensional feature vector first and hidden process consumption vector components last to ensure that the comprehensive consumption vector can completely represent the consumption of all processes and provide complete data support for the subsequent residual value calculation. The budget integrity assessment residual value The relative error calculation method is used, which quantifies the difference between the comprehensive consumption vector and the actual budget vector. The calculation formula is as follows: In the formula, This is the result of standardizing the actual budget vector. This is the result of standardizing the overall consumption vector. The L2 norm of a vector represents its magnitude. It is calculated by taking the square root of the sum of the squares of all its components. The magnitude of the L2 norm directly reflects the overall range of values ​​the vector can take. The residual value calculated using the above formula... The range of values ​​is , The closer it is to 0, the smaller the difference between the actual budget and the complete process consumption, and the more complete the budget is.

[0011] Furthermore, in step S6, the preset evaluation threshold range is divided into first-level thresholds. and secondary threshold ,satisfy The threshold is set according to the actual needs of building budget assessment and is used to distinguish different levels of budget logical integrity. Among them, the first-level threshold... and secondary threshold The specific value was determined through statistical analysis, combining budget assessment data from a large number of similar construction projects and industry standards. The specific range of values ​​is as follows: , The threshold can be dynamically adjusted according to the project type, such as residential or public buildings, and the project scale, such as small or large. After adjustment, it needs to be re-verified to ensure the reasonableness of the threshold. The core function of the threshold is to serve as a standard for judging the integrity of the budget, clarify the degree of budget defects corresponding to different residual values, and provide a clear basis for the output of subsequent evaluation results.

[0012] Further, in step S6, when the residual value Right now When the construction budget cost calculation logic is deemed complete, an assessment result indicating that the budget is qualified is output. This result shows that the current budget list has no omissions or abnormalities and can be directly used for project construction. Right now When a logical flaw is identified in the construction budget, an assessment report is output containing suggestions for supplementing hidden work processes. These suggestions include the name of the hidden work process, its consumption parameters, and estimated costs, facilitating staff to quickly fill in any missing items in the budget. If a serious logical flaw is found in the construction budget, a budget rejection instruction and a revision plan are output. The revision plan includes detailed information on filling in missing items in the budget, correcting abnormal data, and adjusting consumption parameters. After the staff improves the budget according to the revision plan, it is re-evaluated.

[0013] Furthermore, the assessment results also include a budget integrity score, which is based on residual values. The calculation shows that the score ranges from 0 to 100 points, with a higher score for smaller residual values. This score directly reflects the completeness of the construction budget and outputs the estimated costs for each missing item, providing data support for budget completion. The specific formula for calculating the budget completeness score is as follows: For example, when δ=0.03, the score is 97 points. When =0.10, the score is 90 points. When the value is 0.20, the score is 80 points. The corresponding rating levels are: 90-100 points for budget completeness, 70-89 points for minor omissions, and below 70 points for serious deficiencies. This allows staff to quickly and intuitively judge the budget completeness level. The estimated cost of each missing item is calculated by combining the consumption parameters of the hidden process and the current market unit price. Specifically, it is the sum of the labor, material, and machinery consumption costs for each missing process. The error range of the estimated cost is controlled within ±5%, ensuring accurate data support for budget completion.

[0014] On the other hand, the present invention also provides a building budget evaluation system based on big data, which includes: a data acquisition and preprocessing module, a feature vector construction module, a process topology construction module, an implicit process derivation module, a comprehensive consumption vector construction module, and an evaluation and result output module; The data acquisition and preprocessing module is used to acquire the budget item list data of the target building project, perform cleaning and standardization preprocessing operations, remove abnormal noise data, and extract core feature parameters; The feature vector construction module is used to map the preprocessed core feature parameters to a multidimensional feature space, construct an initial vector, and perform Z-Score standardization to obtain an initial multidimensional feature vector. The process topology construction module is used to parse the dependencies between processes based on the standard specification library of building engineering processes, construct an adjacency matrix and perform binarization on it; The implicit process derivation module is used to perform transitive closure operation on the binarized adjacency matrix, construct a process dependency reachability graph, and derive the set of implicit processes. The comprehensive consumption vector construction module is used to construct a comprehensive consumption vector and calculate the budget integrity assessment residual value based on the implicit process set and the initial multidimensional feature vector. The comprehensive consumption vector construction module automatically completes vector splicing and fusion. The evaluation and result output module is used to compare the residual value with the preset evaluation threshold range, determine the logical integrity of the building budget cost, and output the evaluation result including suggestions for omissions, budget integrity score and estimated cost of omissions.

[0015] Compared with known public technologies, the technical solution provided by this invention has the following beneficial effects: This invention achieves standardized processing of budget data and accurate screening of core parameters by combining standardized cleaning with the 3σ principle and a feature ranking algorithm based on information gain. At the same time, it discovers easily overlooked implicit processes by using an innovative combination of process topology construction, transitive closure operation and Boolean operation, filling the gap that traditional methods cannot identify implicit processes. It constructs a comprehensive consumption vector that integrates explicit and implicit process consumption data, and combines it with relative error residual calculation, dual threshold judgment and 0-100 score system to achieve quantitative evaluation of budget integrity. Attached Figure Description

[0016] Figure 1 This is a flowchart of a big data-based building budget assessment method according to the present invention; Figure 2 This is a system model diagram of a big data-based building budget evaluation system according to the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] The present invention will now be described in further detail with reference to the accompanying drawings: Example 1: As Figure 1 - Figure 2 As shown, this invention provides a construction budget assessment method based on big data, including the following steps: S1. Data Acquisition and Preprocessing: Obtain the budget item list data of the target building project, clean and standardize the list data, remove abnormal noise data, and extract core feature parameters.

[0020] Furthermore, the project list data includes process names, quantities, labor consumption standards, material consumption standards, machinery consumption standards, and corresponding unit price data. The process names must be completely consistent with the process names in the construction engineering process standard specification library to avoid errors in process dependency resolution due to inconsistencies in names. The quantities are the actual construction quantities of each process in the target project, which need to be determined in conjunction with construction drawings and on-site survey data to ensure the data is true and accurate. The labor consumption standards, material consumption standards, and machinery consumption standards are all based on current industry standards and adjusted according to the market conditions in the project location to ensure that the consumption standards are in line with the actual project. The corresponding unit price data is the current fair market unit price, which is captured and updated in real time by the big data platform to avoid budget assessment deviations due to unit price lag. In step S1, the removal of abnormal noise data adopts the 3σ principle, calculating and removing outliers for each expense item in the list data. The calculation formula is as follows: , In the formula, This represents the mean of all expense items in the list data, indicating the normal center of value for that expense item. Specifically, it is the arithmetic mean of all valid sample data for that expense item. The standard deviation of each cost item represents the degree of dispersion of its values. The larger the value, the greater the fluctuation in the value of that expense item. The smaller the value, the more concentrated the values. Here, n represents the sample size for each cost item. For degrees of freedom, adopt An unbiased estimate of the sample standard deviation is performed to avoid estimation bias when the sample size is small, thus meeting the sample statistical requirements of construction budget list data. For each specific value of the expense item, that is, for each valid data point of the expense item in the list, exclude values ​​within the range of... Data points other than those listed above are considered abnormal noise data. Specifically, these are abnormal data that exceed the normal range by too much, do not conform to market rules or actual construction conditions, such as the cost of ordinary wall plastering at a unit price of 10,000 yuan per square meter.

[0021] Furthermore, the core feature parameter extraction employs a feature importance ranking algorithm based on information gain to screen feature parameters with an impact weight ≥5% on budget assessment. These include project quantity, main material unit price, and process time. The information gain-based feature importance ranking algorithm calculates the information gain value of each feature parameter on the budget assessment result. The higher the information gain value, the greater the impact of the feature parameter on the budget assessment, and vice versa. The screening threshold of ≥5% impact weight is set based on the actual needs of construction budget assessment. After verification by a large number of similar projects, a weight ≥5% can effectively screen core feature parameters while eliminating redundant parameters with minimal impact on budget assessment. Project quantity specifically refers to the total construction volume of each process, which is the core basis for budget calculation. Main material unit price refers to the unit price of the project's main materials such as steel bars, cement, and sand and gravel, which directly affects the total budget cost. Process time refers to the estimated construction time of each process, affecting the consumption costs of labor and machinery. These three together constitute the core features of budget assessment.

[0022] S2. Feature Vector Construction: Map the preprocessed core feature parameters to a multidimensional feature space to construct an initial multidimensional feature vector representing the project features.

[0023] Furthermore, the initial multidimensional feature vector uses the selected core feature parameters as the vector dimension to first construct an initial vector, the formula for which is: Where n is the number of core feature parameters, and to eliminate dimensional differences, the Z-Score standardization formula is used to process the vector components to obtain an initial multidimensional feature vector. The calculation formula for the standardization process is as follows: In the formula, These are the standardized feature parameter values. Let be the original value of the i-th core feature parameter. Let be the historical mean of the i-th core feature parameter. The historical standard deviation of the i-th core feature parameter is used. The standardized initial multidimensional feature vector is used for subsequent process topology construction and residual calculation, providing standardized data support for budget evaluation.

[0024] S3. Construction of Process Topology: Based on the standard specification library of construction engineering processes, the dependencies and association logic between processes are analyzed, an adjacency matrix representing the process topology is constructed, and the adjacency matrix is ​​binarized.

[0025] Furthermore, the dependencies between processes include two types: mandatory association between preceding and subsequent processes, and collaborative association between parallel processes. Mandatory association between preceding and subsequent processes specifically means that the subsequent process can only begin construction after the preceding process is completely completed, and cannot be started in advance. For example, formwork erection is a mandatory preceding process for concrete pouring; concrete pouring can only proceed after the formwork erection is completed. Collaborative association between parallel processes specifically means that two or more processes can be carried out simultaneously without affecting each other, and can collaboratively advance the construction progress. For example, steel bar binding and water and electricity pipeline pre-embedding can be carried out in parallel without waiting for each other. The core of resolving these two types of dependencies is to combine building construction process standards to ensure that the analysis results conform to the actual construction process and are free of logical errors.

[0026] Furthermore, the adjacency matrix is ​​denoted as A, and is an m×m matrix, where m is the total number of explicit processes in the project. The construction logic of the adjacency matrix is ​​as follows: when the process... For process In the direct preceding process, matrix elements When process i and process j have no direct dependency, the matrix elements 0. Binarization of the adjacency matrix retains only two values, 0 and 1, to clarify the connectivity between processes, eliminate the interference of the workload on the logical judgment of the process, and adapt to the actual needs of building budget assessment. After binarization, the matrix elements only represent whether there is a direct dependency, without involving quantitative indicators such as workload and time consumption, ensuring the accuracy of the logical judgment of the process.

[0027] S4. Implicit Process Derivation: Perform transitive closure operation on the processed adjacency matrix to construct the reachability graph of process dependencies, and derive the set of implicit processes under specific working conditions through Boolean operations on matrix elements.

[0028] Furthermore, the transitive closure operation employs the Warshall algorithm to operate on the binarized adjacency matrix, obtaining the transitive closure matrix. The operation formula is as follows: In the formula, The closure matrix is ​​used to represent all dependencies between processes, including direct and indirect dependencies. The initial adjacency matrix is ​​the same as the total number of explicit operations; that is, m is the number of rows or columns in the initial adjacency matrix. Let be the matrix order, and be the square of the initial adjacency matrix A, representing the indirect dependency between processes via one intermediate process. This represents an indirect dependency relationship involving two intermediate processes, and so on. The indirect dependencies between m-1 intermediate processes are represented. Through the above calculation, the dependencies between all processes can be completely covered without omission.

[0029] Furthermore, the reachability between processes is determined through Boolean operations on matrix elements: when the closure matrix is ​​passed... medium elements And the initial adjacency matrix Corresponding element When, explain the process. With process An indirect dependency relationship exists, leading to the derivation of implicit procedures. Combining this with a building construction technology knowledge base, implicit procedures matching the current project are selected, forming a set of implicit procedures. ,in The number of implicit processes is determined by Boolean operations, which simultaneously satisfy two conditions: the element of the transitive closure matrix is ​​1 and the corresponding element of the initial adjacency matrix is ​​0. Implicit processes are auxiliary processes that are necessary to carry out during construction but are easily overlooked, such as concrete curing and cleaning after formwork removal. Although these processes are not explicitly listed in the budget, their absence will affect the construction quality and schedule, and are considered budget omissions. The construction technology knowledge base covers the complete construction processes of various construction projects and can accurately match them according to project type and construction technology to ensure that the selected implicit processes are fully compatible with the target project.

[0030] S5. Construction of Comprehensive Consumption Vector: Based on the implicit process set and the initial multidimensional feature vector, a comprehensive consumption vector containing complete process logic is constructed to generate budget integrity assessment residual values.

[0031] Furthermore, the comprehensive consumption vector construction maps the consumption parameters of each process in the implicit process set T, including labor, material, and machinery consumption, into vector components, and splices and fuses them with the initial multidimensional feature vector obtained in step S2 to form a comprehensive consumption vector containing complete process logic. The method for obtaining consumption parameters for implicit processes is as follows: combining the knowledge base of building construction technology and historical data of similar projects, the labor consumption standard, material consumption standard, and machinery consumption standard for each implicit process are determined. Then, combined with the project quantity and market unit price of the target project, the specific values ​​of the consumption parameters for each implicit process are calculated. The vector component mapping adopts the same mapping rule as the initial multidimensional feature vector to ensure that the format and dimension of the implicit process consumption vector components are consistent with those of the initial multidimensional feature vector components. When splicing and merging, the initial multidimensional feature vector is strictly followed in order, followed by the implicit process consumption vector components, to ensure that the comprehensive consumption vector can completely represent the consumption of all processes and provide complete data support for the subsequent residual value calculation.

[0032] Furthermore, the budget integrity assessment residual value The relative error calculation method is used, which quantifies the difference between the comprehensive consumption vector and the actual budget vector. The calculation formula is as follows: In the formula, This is the result of standardizing the actual budget vector. This is the result of standardizing the overall consumption vector. The L2 norm of a vector represents its magnitude. It is calculated by taking the square root of the sum of the squares of all its components. The magnitude of the L2 norm directly reflects the overall range of values ​​the vector can take. The residual value calculated using the above formula... The range of values ​​is , The closer it is to 0, the smaller the difference between the actual budget and the complete process consumption, and the more complete the budget is.

[0033] S6. Evaluation and Result Output: Compare the residual value with the preset evaluation threshold range to determine the logical integrity of the building budget cost, and output the evaluation result including suggestions for filling in any missing items.

[0034] Furthermore, the preset evaluation threshold range is divided into first-level thresholds. and secondary threshold ,satisfy The threshold is set according to the actual needs of building budget assessment and is used to distinguish different levels of budget logical integrity. Among them, the first-level threshold... and secondary threshold The specific value was determined through statistical analysis, combining budget assessment data from a large number of similar construction projects and industry standards. The specific range of values ​​is as follows: , The threshold can be dynamically adjusted according to the project type, such as residential or public buildings, and the project scale, such as small or large. After adjustment, it needs to be re-verified to ensure the reasonableness of the threshold. The core function of the threshold is to serve as a standard for judging the integrity of the budget, clarify the degree of budget defects corresponding to different residual values, and provide a clear basis for the output of subsequent evaluation results.

[0035] Furthermore, when the residual value Right now When the construction budget cost calculation logic is deemed complete, an assessment result indicating that the budget is qualified is output. This result shows that the current budget list has no omissions or abnormalities and can be directly used for project construction. Right now When a logical flaw is identified in the construction budget, an assessment report is output containing suggestions for supplementing hidden work processes. These suggestions include the name of the hidden work process, its consumption parameters, and estimated costs, facilitating staff to quickly fill in any missing items in the budget. If a serious logical flaw is found in the construction budget, a budget rejection instruction and a revision plan are output. The revision plan includes detailed information on filling in missing items in the budget, correcting abnormal data, and adjusting consumption parameters. After the staff improves the budget according to the revision plan, it is re-evaluated.

[0036] Furthermore, the evaluation results also include a budget integrity score, which is based on residual values. The calculation shows that the score ranges from 0 to 100 points, with a higher score for smaller residual values. This score directly reflects the completeness of the construction budget and outputs the estimated costs for each missing item, providing data support for budget completion. The specific formula for calculating the budget completeness score is as follows: For example, when δ=0.03, the score is 97 points. When =0.10, the score is 90 points. When the value is 0.20, the score is 80 points. The corresponding rating levels are: 90-100 points for budget completeness, 70-89 points for minor omissions, and below 70 points for serious deficiencies. This allows staff to quickly and intuitively judge the budget completeness level. The estimated cost of each missing item is calculated by combining the consumption parameters of the hidden process and the current market unit price. Specifically, it is the sum of the labor, material, and machinery consumption costs for each missing process. The error range of the estimated cost is controlled within ±5%, ensuring accurate data support for budget completion.

[0037] Example 2: Figure 1 - Figure 2 As shown in Example 2, a big data-based building budget assessment system is provided. The system includes: a data acquisition and preprocessing module, a feature vector construction module, a process topology construction module, an implicit process derivation module, a comprehensive consumption vector construction module, and an assessment and result output module. The data acquisition and preprocessing module is used to acquire the budget item list data of the target building project, perform cleaning and standardization preprocessing operations, remove abnormal noise data, and extract core feature parameters; The feature vector construction module is used to map the preprocessed core feature parameters to a multidimensional feature space, construct an initial vector and perform Z-Score standardization to obtain an initial multidimensional feature vector. The feature vector construction module has built-in vector mapping and standardization algorithms, which can automatically complete vector construction and standardization. It also has a vector validity verification function. If the verification fails, it will issue a reminder and return to the data acquisition and preprocessing module for reprocessing. The process topology construction module is used to parse the dependencies between processes based on the building engineering process standard specification library, construct an adjacency matrix and perform binarization on it. The process topology construction module is linked with the building engineering process standard specification library in real time and can synchronously obtain the updated data of the specification library to ensure the accuracy of process dependency parsing and adjacency matrix construction. The binarization process can be completed automatically without manual intervention. The implicit process derivation module is used to perform transitive closure operation on the binarized adjacency matrix, construct a process dependency reachability graph, and derive the implicit process set. The implicit process derivation module has a built-in Warshall algorithm, which can quickly complete the transitive closure operation. At the same time, combined with the building construction technology knowledge base, it automatically filters the implicit processes that match the target project and generates a complete set of implicit processes, which can intuitively display the process dependency reachability graph. The comprehensive consumption vector construction module is used to construct a comprehensive consumption vector and calculate the budget integrity assessment residual value based on the implicit process set and the initial multidimensional feature vector. The comprehensive consumption vector construction module automatically completes vector splicing and fusion, and uses a preset residual value calculation formula to quickly calculate the residual value. At the same time, the calculation results are verified to ensure that the residual value is accurate. The assessment and result output module is used to compare the residual value with the preset assessment threshold range to determine the logical integrity of the building budget cost. It outputs assessment results including suggestions for omissions, budget integrity score and estimated cost of omissions. The assessment and result output module can generate a visual assessment report. The report format conforms to industry standards and can be directly printed and exported. It also supports real-time feedback of assessment results, which makes it easy for staff to understand budget defects and make corrections in a timely manner.

[0038] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A construction budget assessment method based on big data, characterized in that, Includes the following steps: S1. Data Acquisition and Preprocessing Obtain the budget item list data of the target building project, clean and standardize the list data, remove abnormal noise data, and extract core feature parameters; S2, Feature Vector Construction The preprocessed core feature parameters are mapped to a multidimensional feature space to construct an initial multidimensional feature vector representing the project features; S3, Process Topology Construction Based on the standard specification library of construction engineering procedures, the dependencies and association logic between procedures are analyzed, an adjacency matrix representing the topological structure of the procedures is constructed, and the adjacency matrix is ​​binarized. S4, Derivation of Hidden Processes Perform transitive closure operation on the processed adjacency matrix to construct a reachability graph of process dependencies, and derive the set of implicit processes under a specific working condition through Boolean operations on matrix elements; S5. Construction of Comprehensive Consumption Vector Based on the set of implicit processes and the initial multidimensional feature vector, a comprehensive consumption vector containing complete process logic is constructed to generate budget integrity assessment residual values; S6. Evaluation and Results Output The residual value is compared with the preset evaluation threshold range to determine the logical integrity of the building budget cost, and the evaluation result including suggestions for filling in the missing items is output.

2. The construction budget evaluation method based on big data according to claim 1, in step S1, the project list data includes process name, engineering quantity, labor consumption standard, material consumption standard, machinery consumption standard and corresponding unit price data; In step S1, the removal of abnormal noise data adopts the 3σ principle, calculating and removing outliers for each expense item in the list data. The calculation formula is as follows: ， In the formula, The mean of all expense items in the list data represents the normal value center for that expense item. Let be the standard deviation of each cost item, representing the degree of dispersion of values, and n be the sample size for each cost item. For degrees of freedom, adopt An unbiased estimate of the sample standard deviation is performed to avoid estimation bias when the sample size is small, thus meeting the sample statistical requirements of construction budget list data. For the specific values ​​of each expense item, exclude values ​​within the range of Data points other than those specified are considered abnormal noise data. In step S1, the core feature parameter extraction adopts a feature importance ranking algorithm based on information gain to screen out feature parameters with an impact weight of ≥5% on budget assessment, including project quantity, unit price of main materials and process time.

3. The construction budget assessment method based on big data according to claim 2, characterized in that, In step S2, the initial multidimensional feature vector is constructed using the filtered core feature parameters as the vector dimension. The formula for the initial vector is as follows: Where n is the number of core feature parameters, and to eliminate dimensional differences, the Z-Score standardization formula is used to process the vector components to obtain the initial multidimensional feature vector. The calculation formula for the standardization process is as follows: In the formula, These are the standardized feature parameter values. Let be the original value of the i-th core feature parameter. Let be the historical mean of the i-th core feature parameter. The historical standard deviation of the i-th core feature parameter is used. The standardized initial multidimensional feature vector is used for subsequent process topology construction and residual calculation, providing standardized data support for budget evaluation.

4. The construction budget assessment method based on big data according to claim 3, characterized in that, In step S3, the inter-process dependencies include two types: forced association between preceding and subsequent processes, and collaborative association between parallel processes. In step S3, the adjacency matrix is ​​denoted as A, which is an m×m matrix, where m is the total number of explicit processes in the project. The construction logic of the adjacency matrix is ​​as follows: when the process... For process In the direct preceding process, matrix elements When process i and process j have no direct dependency, the matrix elements 0. Binarization of the adjacency matrix retains only 0 and 1 values, which is used to clarify the connectivity between processes, eliminate the interference of the scale of the project on the logical judgment of the process, and adapt to the actual needs of building budget assessment.

5. The construction budget evaluation method based on big data according to claim 4, characterized in that, In step S4, the transitive closure operation uses the Warshall algorithm to operate on the binarized adjacency matrix to obtain the transitive closure matrix. The operation formula is as follows: In the formula, To pass the closure matrix, The initial adjacency matrix, The order of the matrix; In step S4, the reachability between processes is determined by Boolean operations on the matrix elements: when the closure matrix is ​​passed... medium elements And the initial adjacency matrix Corresponding element When, explain the process. With process An indirect dependency relationship exists, leading to the derivation of implicit procedures. Combining this with a building construction technology knowledge base, implicit procedures matching the current project are selected, forming a set of implicit procedures. ,in This refers to the number of hidden processes.

6. The construction budget evaluation method based on big data according to claim 5, characterized in that, In step S5, the method for constructing the comprehensive consumption vector is as follows: the consumption parameters of each process in the implicit process set, including labor, material and machinery consumption, are mapped into vector components and spliced ​​and fused with the initial multidimensional feature vector obtained in step S2 to form a comprehensive consumption vector containing complete process logic. The budget integrity assessment residual value The relative error calculation method is used, which quantifies the difference between the comprehensive consumption vector and the actual budget vector. The calculation formula is as follows: In the formula, This is the result of standardizing the actual budget vector. This is the result of standardizing the overall consumption vector. The L2 norm of a vector is used to characterize the magnitude of the vector.

7. The construction budget evaluation method based on big data according to claim 6, characterized in that, In step S6, the preset evaluation threshold range is divided into first-level thresholds. and secondary threshold ,satisfy The threshold is set according to the actual needs of building budget assessment and is used to distinguish different levels of budget logic integrity.

8. The construction budget evaluation method based on big data according to claim 7, characterized in that, In step S6, when the residual value When the construction budget cost logic is deemed complete, an assessment result indicating that the budget is qualified is output. When a minor logical flaw is identified in the construction budget cost, an assessment report is output containing suggestions for completing hidden construction processes. If a serious logical flaw is found in the construction budget, a budget rejection instruction and a correction plan will be output.

9. The construction budget assessment method based on big data according to claim 8, wherein in step S6, the assessment result further includes a budget integrity score, the score being based on residual values. The calculation shows that the score ranges from 0 to 100 points. The smaller the residual value, the higher the score, which is used to intuitively reflect the completeness of the construction budget cost. At the same time, the estimated cost of each missing item is output to provide data support for budget completion.

10. A big data-based building budget assessment system, based on any one of claims 1-9, characterized in that, The system includes: a data acquisition and preprocessing module, a feature vector construction module, a process topology construction module, an implicit process derivation module, a comprehensive consumption vector construction module, and an evaluation and result output module; The data acquisition and preprocessing module is used to acquire the budget item list data of the target building project, perform cleaning and standardization preprocessing operations, remove abnormal noise data, and extract core feature parameters. The feature vector construction module is used to map the preprocessed core feature parameters to a multidimensional feature space, construct an initial vector, and perform Z-Score standardization to obtain an initial multidimensional feature vector. The process topology construction module is used to parse the dependencies between processes based on the standard specification library of building engineering processes, construct an adjacency matrix and perform binarization on it; The implicit process derivation module is used to perform transitive closure operation on the binarized adjacency matrix, construct a process dependency reachability graph, and derive the set of implicit processes. The comprehensive consumption vector construction module is used to construct a comprehensive consumption vector and calculate the budget integrity assessment residual value based on the set of implicit processes and the initial multidimensional feature vector. The evaluation and result output module is used to compare the residual value with the preset evaluation threshold range, determine the logical integrity of the building budget cost, and output the evaluation result including suggestions for omissions, budget integrity score and estimated cost of omissions.

Images