Construction engineering project cost management method and system, electronic equipment and storage medium

Abstract

The invention discloses a construction engineering project cost management method and system, electronic equipment and a storage medium, and relates to the technical field of engineering cost. The method comprises the following steps: acquiring historical preliminary constructional engineering drawings corresponding to a plurality of historical constructional engineering; obtaining a plurality of historical prediction stage costs based on the plurality of historical preliminary construction engineering drawings; acquiring historical actual building engineering drawings and historical actual stage cost corresponding to the plurality of historical building engineering; constructing a regression model based on the cost of each prediction stage, the cost of each actual stage, each historical preliminary construction engineering drawing and each historical actual construction engineering drawing; obtaining a current preliminary constructional engineering drawing corresponding to the current constructional engineering; and obtaining the cost of the current prediction stage based on each current preliminary building engineering drawing and the regression model. According to the technical scheme, accurate cost prediction can be provided at the initial stage of the project.

Description

Technical Field

[0001] The present application relates to the field of engineering cost technology, and specifically to a construction project cost management method, system, electronic equipment and storage medium. Background Art

[0002] In the implementation of construction projects, accurate cost estimation is one of the key factors for project success. In the current technical scenario, the cost estimation of construction projects is mainly estimated in the early stage of the project based on the project's construction drawings. However, due to the limited information in the preliminary design stage and the possibility of major changes in the project design plan, there is a large deviation between the preliminary construction drawings and the actual construction drawings, which often leads to large errors in the cost forecast in the early stage of the project.

[0003] Therefore, how to provide accurate cost forecasts at the early stage of a project has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present application provides a construction project cost management method, system, electronic equipment and storage medium, which can provide accurate cost forecasts at the early stage of a project.

[0005] In a first aspect, the present application provides a construction project cost management method, the method comprising: obtaining historical preliminary construction project drawings corresponding to a plurality of historical construction projects; Based on the plurality of said historical preliminary construction engineering drawings, a plurality of historical forecast stage construction costs are obtained; Obtaining actual historical construction engineering drawings and actual historical stage costs corresponding to a plurality of the historical construction projects; Based on the predicted stage costs, the actual stage costs, the historical preliminary construction engineering drawings, and the historical actual construction engineering drawings, a regression model is constructed; Obtain the current preliminary building project drawings corresponding to the current building project; Based on each of the current preliminary construction engineering drawings and the regression model, the current forecast stage cost is obtained.

[0006] By adopting the above technical solution, by obtaining the historical preliminary construction drawings corresponding to multiple historical construction projects, historical data can be systematically collected and sorted, providing comprehensive basic information for subsequent analysis and model construction. By obtaining multiple historical forecast stage costs based on multiple historical preliminary construction drawings, the cost of the construction project can be preliminarily estimated by using the design and cost relationship in the historical data. By obtaining the historical actual construction drawings and historical actual stage costs corresponding to multiple historical construction projects, the historical forecast cost can be compared and analyzed with the actual cost. By constructing a regression model based on the cost of each forecast stage, the cost of each actual stage, each historical preliminary construction drawings and each historical actual construction drawings, a mathematical model that can reflect the relationship between design features and cost is established. By obtaining the current preliminary construction drawings corresponding to the current construction project, the latest design information is incorporated into the prediction model. By obtaining the current forecast stage cost based on each current preliminary construction drawings and the regression model, it is ensured that the model can make predictions based on the specific situation of the current project, so that the prediction results are closer to actual needs, thereby providing accurate cost predictions at the beginning of the project.

[0007] Optionally, obtaining multiple historical forecast stage costs based on multiple historical preliminary construction engineering drawings specifically includes: Based on each of the historical preliminary architectural engineering drawings, a plurality of first key features are obtained; The first basic construction costs corresponding to the respective first key features are determined, and all the first basic construction costs are summed to obtain the construction costs of the respective historical prediction stages.

[0008] By adopting the above technical solution, multiple first key features are obtained based on various historical preliminary architectural engineering drawings, thereby systematically extracting important feature information in architectural design. By determining the first basic cost corresponding to each first key feature and summing up all the first basic costs, the cost of each historical prediction stage is obtained, thereby converting complex design features into specific cost estimates.

[0009] Optionally, the regression model includes a first regression model and a second regression model; constructing the regression model based on the predicted stage costs, the actual stage costs, the historical preliminary construction engineering drawings, and the historical actual construction engineering drawings specifically includes: Based on each of the historical preliminary architectural engineering drawings, a plurality of first key features are obtained; the first key features include first numerical type key features and first non-numerical type key features; Based on each of the historical actual architectural engineering drawings, a plurality of second key features are obtained; the second key features include a second numerical type key feature and a second non-numerical type key feature; Encode each of the first non-numeric key features and each of the second non-numeric key features to obtain a plurality of first vectors and a plurality of second vectors; Subtract each of the first vectors from each of the second vectors to obtain a difference vector corresponding to each of the historical building projects; subtract each of the first value type key features from each of the second value type key features to obtain a difference value corresponding to each of the historical building projects; Based on the construction costs of each predicted stage and the construction costs of each actual stage, obtaining the cost difference corresponding to the historical building project; Based on the plurality of difference vectors and the plurality of cost difference values, constructing the first regression model; Based on the plurality of difference values ​​and the plurality of cost difference values, the second regression model is constructed.

[0010] By adopting the above technical solution, by encoding each first non-numerical type key feature and each second non-numerical type key feature respectively, a plurality of first vectors and a plurality of second vectors are obtained, thereby converting the non-numerical type feature into a numerical form that can be processed by the model. By subtracting each first vector from each second vector, a difference vector corresponding to each historical building project is obtained, thereby quantifying the difference in non-numerical features between the preliminary design stage and the actual construction stage. By subtracting each first numerical type key feature from each second numerical type key feature, a difference value corresponding to each historical building project is obtained, thereby quantifying the change of numerical features at different stages. By obtaining the cost difference corresponding to the historical building project based on the cost of each predicted stage and the cost of each actual stage, the prediction error is quantified. By constructing a first regression model based on a plurality of difference vectors and a plurality of cost difference values, the influence of non-numerical features on cost difference is accurately captured. By constructing a second regression model based on a plurality of difference values ​​and a plurality of cost difference values, the influence of numerical features on cost difference is accurately reflected.

[0011] Optionally, constructing the first regression model based on the plurality of difference vectors and the plurality of cost difference values ​​specifically includes: The first regression model is trained based on the plurality of difference vectors and the plurality of cost difference values ​​to obtain a first regression model intercept and a first regression model coefficient; wherein the relationship between the difference vector and the cost difference value is expressed by the following formula: Where, ΔC iis the cost difference corresponding to the i-th historical building project, d ij is the jth element in the difference vector corresponding to the i-th historical building project, β0 is the intercept of the first regression model, β j is the first regression model coefficient.

[0012] Optionally, constructing the second regression model based on the plurality of difference values ​​and the plurality of cost difference values ​​specifically includes: The second regression model is trained based on the plurality of difference values ​​and the plurality of cost difference values ​​to obtain a second regression model intercept and a second regression model coefficient; wherein the relationship between the difference value and the cost difference value is expressed by the following formula: ΔC i =α0+α1u i ; Where, ΔC i is the cost difference corresponding to the i-th historical building project, u i is the difference value corresponding to the i-th historical building project, α0 is the intercept of the second regression model, and α1 is the coefficient of the second regression model.

[0013] Optionally, the regression model includes a first regression model and a second regression model; obtaining the current forecast stage cost based on each of the current preliminary construction engineering drawings and the regression model specifically includes: Based on each of the current preliminary architectural engineering drawings, a plurality of third key features are obtained; the third key features include third numerical type key features and third non-numerical type key features; Determine the second basic cost corresponding to each of the third key features, and sum all of the second basic costs to obtain the cost in the current forecast stage; Inputting the plurality of third non-numeric key features into the first regression model to obtain a first current forecast stage cost difference; Inputting the plurality of third value type key features into the second regression model to obtain a second current forecast stage cost difference; Performing a weighted summation of the cost difference values ​​of the first current forecast stage and the cost difference values ​​of the second current forecast stage to obtain a total cost difference value of the current forecast stage; The cost of each current forecast stage is added to the difference between the total cost of each current forecast stage to obtain the cost of the current forecast stage.

[0014] By adopting the above technical solution, by inputting multiple third non-numerical type key features into the first regression model, the first current prediction stage cost difference is obtained, thereby quantifying the impact of non-numerical type features on the current cost. By inputting multiple third numerical type key features into the second regression model, the second current prediction stage cost difference is obtained, thereby quantifying the impact of numerical type features on the current cost. By weighted summing up the cost differences of each first current prediction stage and each second current prediction stage cost difference, the total cost difference of the current prediction stage is obtained, thereby comprehensively considering the impact of all key features on the cost. By adding each current prediction stage cost to each current prediction stage total cost difference, the current prediction stage cost is obtained, thereby comprehensively considering the impact of the basic cost and all features to obtain the final prediction result. This process ensures the comprehensiveness and accuracy of the prediction results.

[0015] Optionally, the obtaining of a plurality of first key features based on each of the historical preliminary architectural engineering drawings specifically includes: Digitally processing the plurality of said historical preliminary architectural engineering drawings to obtain a plurality of historical preliminary digital drawings; A first key feature is extracted from each of the historical preliminary digital drawings.

[0016] By adopting the above technical solution, multiple historical preliminary architectural engineering drawings are digitized to obtain multiple historical preliminary digital drawings, thereby converting paper drawings into digital formats that are easy to process and analyze. By extracting the first key feature from each historical preliminary digital drawing, important feature information in architectural design is systematically obtained.

[0017] In a second aspect of the present application, a construction project cost management system is provided, the system comprising: an acquisition module and a processing module; The acquisition module is used to acquire historical preliminary architectural engineering drawings corresponding to a plurality of historical architectural engineering projects; The processing module is used to obtain multiple historical forecast stage costs based on the multiple historical preliminary construction engineering drawings; The acquisition module is further used to acquire historical actual construction engineering drawings and historical actual stage costs corresponding to the plurality of historical construction projects; The processing module is further used to construct a regression model based on the predicted stage costs, the actual stage costs, the historical preliminary construction engineering drawings and the historical actual construction engineering drawings; The acquisition module is further used to acquire the current preliminary construction engineering drawing corresponding to the current construction engineering; The processing module is also used to obtain the current forecast stage cost based on each of the current preliminary construction engineering drawings and the regression model.

[0018] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any method as described in the first aspect of the present application.

[0019] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program that can be loaded by a processor and execute the method as described in any one of the first aspects of the present application.

[0020] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: By obtaining the historical preliminary construction drawings corresponding to multiple historical construction projects, historical data can be systematically collected and sorted, providing comprehensive basic information for subsequent analysis and model construction. By obtaining multiple historical forecast stage costs based on multiple historical preliminary construction drawings, the cost of the construction project can be preliminarily estimated by using the design and cost relationship in the historical data. By obtaining the historical actual construction drawings and historical actual stage costs corresponding to multiple historical construction projects, the historical forecast cost can be compared and analyzed with the actual cost. By constructing a regression model based on the cost of each forecast stage, the cost of each actual stage, each historical preliminary construction drawings, and each historical actual construction drawings, a mathematical model that can reflect the relationship between design features and cost is established. By obtaining the current preliminary construction drawings corresponding to the current construction project, the latest design information is incorporated into the prediction model. By obtaining the current forecast stage cost based on each current preliminary construction drawing and the regression model, it is ensured that the model can make predictions based on the specific situation of the current project, so that the prediction results are closer to actual needs, thereby providing accurate cost predictions at the beginning of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is one of the flow charts of a construction project cost management method provided in an embodiment of the present application; Figure 2 This is the second flow chart of a construction project cost management method provided in an embodiment of the present application; Figure 3 This is a flowchart of a construction project cost management method provided by an embodiment of the present application; Figure 4 It is a structural diagram of a construction project cost management system provided by an embodiment of the present application; Figure 5 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0022] Explanation of the reference numerals: 1. Acquisition module; 2. Processing module; 500. Electronic device; 501. Processor; 502. Communication bus; 503. User interface; 504. Network interface; 505. Memory. DETAILED DESCRIPTION

[0023] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0024] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0025] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" can explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0026] This application provides a construction project cost management method, referring to Figure 1 , which shows one of the flow diagrams of a construction project cost management method provided in an embodiment of the present application. The method includes steps S1-S6, which are as follows: Step S1: Obtaining historical preliminary architectural engineering drawings corresponding to a plurality of historical architectural engineering projects.

[0027] Specifically, in order to establish an accurate project cost prediction model, a large amount of historical construction project data must be collected and processed first. These data include preliminary design drawings of multiple historical construction projects, which record the design intent and construction details at the time and have important reference value. By analyzing these historical drawings, key features can be extracted to provide data support for subsequent cost prediction.

[0028] Step S2: Based on a plurality of historical preliminary construction engineering drawings, a plurality of historical forecast stage costs are obtained.

[0029] In order to accurately predict the cost of a construction project, it is necessary to extract key features based on multiple historical preliminary construction project drawings and use these features to calculate the cost of the historical prediction stage. The purpose of this step is to provide a reliable reference for current and future project cost predictions by analyzing historical data.

[0030] In one possible implementation, refer to Figure 2 , which shows the second flow chart of a construction project cost management method provided by an embodiment of the present application. Step S2 specifically includes steps S21-S22: Step S21: Based on each historical preliminary architectural engineering drawing, a plurality of first key features are obtained.

[0031] In a possible implementation, step S21 specifically includes the following steps: A plurality of historical preliminary architectural engineering drawings are digitized to obtain a plurality of historical preliminary digital drawings.

[0032] Extract the first key features from each historical preliminary digital drawing.

[0033] Specifically, the process of obtaining multiple historical forecast stage costs based on multiple historical preliminary construction engineering drawings is as follows: First, multiple first key features are obtained based on each historical preliminary construction engineering drawing. These key features are an important basis for predicting the cost. In order to extract these features, multiple historical preliminary construction engineering drawings are first digitized to obtain multiple historical preliminary digital drawings. The digitization process includes converting paper drawings into digital images using a high-resolution scanner and saving them in high-quality file formats such as TIFF or PDF. Then, these digital drawings are pre-processed using professional image processing software, including removing noise and adjusting contrast and brightness to improve image clarity and readability.

[0034] Extracting the first key features from each historical preliminary digital drawing is the core work of the next step. These features include the basic structure of the building, material usage, design specifications, etc. These features can be automatically extracted from the drawings through optical character recognition (OCR) technology and image recognition technology. The extracted key features will be saved in a structured database for subsequent statistical analysis and model training.

[0035] Step S22: determining the first basic cost corresponding to each first key feature, and summing up all the first basic costs to obtain the cost of each historical prediction stage.

[0036] Specifically, first, the first key features extracted from the historical preliminary digital drawings are used, which include the basic structure of the building, material usage, design specifications, etc. Each key feature has a corresponding cost record in the historical data. By analyzing these records, a mapping relationship between the key features and the basic cost can be established.

[0037] In order to achieve this mapping relationship, statistical analysis and machine learning methods can be used. Specifically, the historical data is first cleaned and sorted to ensure the accuracy and consistency of the data. Then, the model is trained based on historical data using algorithms such as linear regression models and decision tree models, so that the model can predict the corresponding cost based on the extracted key features.

[0038] During the model training process, historical data needs to be divided into training sets and validation sets, and the performance of the model needs to be evaluated through cross-validation to ensure the generalization ability and prediction accuracy of the model. After the training is completed, the model can input the first key feature and output the corresponding first basic cost.

[0039] Next, all the first basic costs of each construction project are summed up to obtain the historical forecast stage cost of the project. Specifically, for each construction project, the sum of the basic costs corresponding to all key features is calculated, which is the total cost forecast value of the project in the preliminary design stage.

[0040] Step S3: Obtain historical actual construction engineering drawings and historical actual stage costs corresponding to multiple historical construction projects.

[0041] Specifically, when building a project cost prediction model, it is crucial to obtain historical actual construction engineering drawings and historical actual stage costs. The purpose of this step is to compare the historical predicted stage costs with the actual costs incurred, so as to calibrate the model and ensure the accuracy and reliability of the prediction.

[0042] Step S4: construct a regression model based on the cost of each predicted stage, the cost of each actual stage, each historical preliminary construction engineering drawing, and each historical actual construction engineering drawing.

[0043] Specifically, the regression model includes the first regression model and the second regression model. The purpose of building a regression model is to establish a mathematical model that can accurately predict future project costs by analyzing the predicted costs and actual costs in historical data. By comparing the historical predicted stage costs with the actual stage costs, and combining the key features in the preliminary and actual construction engineering drawings, the regression model can identify and correct deviations in cost forecasts and improve the accuracy and reliability of the forecast.

[0044] In one possible implementation, refer to Figure 3 , which shows the third flow chart of a construction project cost management method provided by an embodiment of the present application. Step S4 specifically includes steps S41-S48: Step S41: Based on each historical preliminary architectural engineering drawing, a plurality of first key features are obtained; the first key features include first numerical type key features and first non-numerical type key features.

[0045] Specifically, the specific implementation of step S41 is the same as the specific implementation of step S21, so it will not be described in detail here.

[0046] Step S42: Based on each historical actual construction engineering drawing, a plurality of second key features are obtained; the second key features include a second numerical type key feature and a second non-numerical type key feature.

[0047] Specifically, the specific implementation of step S42 is similar to the specific implementation of step S21, so it will not be described in detail here.

[0048] It should be noted that in step S41 and step S42, the key features of numerical type include specific quantitative indicators such as building area, number of floors, number of rooms, and material consumption. These features can be automatically extracted by image recognition software, for example, by analyzing the annotations and dimension lines in the drawings to obtain data such as building area and number of floors. Non-numerical key features include design style, building purpose, material type, etc. These features usually appear in text form or symbol form and need to be identified from the drawings through OCR technology. For example, by identifying the text descriptions and legend symbols in the drawings, information such as the design style and material type of the building can be extracted.

[0049] Step S43: Encode each first non-numeric key feature and each second non-numeric key feature respectively to obtain a plurality of first vectors and a plurality of second vectors.

[0050] In the process of establishing the project cost prediction model, non-numerical key features (such as design style, building purpose, material type, etc.) cannot be directly input into the regression model. Therefore, it is necessary to encode these non-numerical key features and convert them into a vector format acceptable to the model. This ensures that all features can be effectively used, thereby improving the accuracy and reliability of the prediction model.

[0051] Specifically, the process of encoding each first non-numeric type key feature and each second non-numeric type key feature is as follows: First, the first non-numerical key features extracted from the historical preliminary architectural engineering drawings and the second non-numerical key features extracted from the historical actual architectural engineering drawings are collected and sorted. In order to encode these non-numerical key features, the method is One-Hot Encoding.

[0052] One-hot encoding converts each categorical feature into a binary vector, where the length of each vector is equal to the total number of categories, and only one position is 1, and the rest are 0. For example, if the design style has three types: "modern", "classical", and "simple", they can be encoded as: [1,0,0] (modern), [0,1,0] (classical), [0,0,1] (simple). This encoding method is suitable for features with a limited number of categories and no order.

[0053] When choosing an encoding method, you need to choose it according to the specific situation of the feature. For unordered features with a limited number of categories, one-hot encoding is usually used; for ordered features or features with a large number of categories, label encoding is usually used. After encoding, multiple first vectors and multiple second vectors are obtained, which will be used as inputs to the regression model.

[0054] During the encoding process, an encoder (such as OneHotEncoder or LabelEncoder in Scikit-Learn) is needed for batch processing to ensure the consistency and accuracy of all features. After encoding, all vectors are saved in a structured database for subsequent statistical analysis and model training.

[0055] Step S44: Subtract each first vector from each second vector to obtain a difference vector corresponding to each historical building project.

[0056] First, ensure that all historical preliminary architectural engineering drawings and historical actual architectural engineering drawings have been coded to obtain a plurality of first vectors and second vectors, respectively. The first vector represents non-numerical key features extracted and coded from the preliminary drawings, while the second vector represents non-numerical key features extracted and coded from the actual drawings.

[0057] For each historical building project, find the corresponding first vector and second vector to ensure a one-to-one correspondence between the two. Then, subtract each element in the first vector from the corresponding element in the second vector to obtain a difference vector. For example, assuming that the first vector of a historical building project is f1, f2, f3, and the second vector is g1, g2, g3, then the difference vector is f1-g1, f2-g2, f3-g3. This subtraction operation can quantify the difference between each non-numerical type feature in the prediction stage and the actual stage. After the difference vector is calculated, it is saved in a structured database. These difference vectors will serve as important input data for subsequent regression model training.

[0058] Step S45: Subtract each first value type key feature from each second value type key feature to obtain a difference value corresponding to each historical building project.

[0059] Specifically, first, ensure that the first value type key features extracted from the historical preliminary architectural engineering drawings and the second value type key features extracted from the historical actual architectural engineering drawings have been structured. These value type key features include specific quantitative indicators, such as building area, number of floors, number of rooms, material usage, etc.

[0060] For each historical building project, the corresponding first numerical type key feature and the second numerical type key feature are matched to ensure a one-to-one correspondence between the two. Then, the value of each first numerical type key feature is subtracted from the value of the corresponding second numerical type key feature to obtain a difference value. For example, assuming that the building area of ​​a historical building project in the preliminary design stage is 3,000 square meters (first numerical type key feature), and the building area in the actual construction stage is 3,200 square meters (second numerical type key feature), the difference value is 3,200-3,000=200 square meters.

[0061] This subtraction operation can quantify the difference between the key features of each value type in the prediction stage and the actual stage, reflecting the impact of design changes, construction adjustments and other factors on the actual cost. After the calculation is completed, all the difference values ​​are saved in a structured database, which will serve as important input data for subsequent regression model training.

[0062] Step S46: Based on the cost of each predicted stage and the cost of each actual stage, the cost difference corresponding to the historical construction project is obtained.

[0063] Specifically, first, ensure that the data of all historical construction projects have been collected and processed, including the historical forecast stage costs and the historical actual stage costs. These data can be extracted from the previous steps. The forecast stage cost is the estimated result based on the preliminary design drawings and feature data, while the actual stage cost is the final cost after the actual construction is completed.

[0064] Next, for each historical building project, the corresponding predicted stage cost is matched with the actual stage cost to ensure a one-to-one correspondence between the two. Then, the predicted stage cost of each historical building project is subtracted from the actual stage cost to obtain the cost difference. For example, assuming that the predicted stage cost of a historical building project is 5 million yuan, and its actual stage cost is 5.5 million yuan, then the cost difference of the project is 5.5 million yuan - 5 million yuan = 500,000 yuan.

[0065] This subtraction operation can quantify the difference between the predicted and actual cost of each construction project, reflecting the prediction error. By calculating the cost difference of all historical construction projects, a series of key data can be obtained, which will serve as an important input for subsequent regression model training. The calculated cost difference is saved in a structured database. This data is not only used to evaluate the prediction performance of the model, but also used as the target variable of the regression model.

[0066] Step S47: constructing a first regression model based on the multiple difference vectors and the multiple cost difference values.

[0067] In a possible implementation, step S47 specifically includes the following steps: The first regression model is trained based on multiple difference vectors and multiple cost difference values ​​to obtain the first regression model intercept and the first regression model coefficient; wherein the relationship between the difference vector and the cost difference value is expressed by the following formula: Where, ΔC i is the cost difference corresponding to the i-th historical building project, d ij is the jth element in the difference vector corresponding to the i-th historical building project, β0 is the intercept of the first regression model, β j is the coefficient of the first regression model.

[0068] In order to accurately predict the cost of construction projects, it is necessary to train a regression model based on historical data, and establish the first regression model by analyzing the difference vectors and cost differences of historical construction projects. The goal of this step is to obtain the intercept and coefficient of the regression model so that the model can reflect the impact of feature differences on cost differences, thereby improving the accuracy of the prediction.

[0069] Specifically, the process of training the first regression model based on multiple difference vectors and multiple cost difference values ​​is as follows: First, ensure that the difference vectors and cost difference data of all historical construction projects have been collected. These data include the difference vectors calculated in the previous steps (calculated by the difference between the first non-numeric key feature and the second non-numeric key feature) and the cost difference (calculated by the difference between the predicted stage cost and the actual stage cost).

[0070] Then, use these data to train the first regression model. Select a suitable regression algorithm, such as linear regression, to establish a mapping relationship between the difference vector and the cost difference. The goal of the linear regression model is to find the optimal first regression model intercept (β0) and first regression model coefficient (βj) to minimize the prediction error of the model. The training process can be performed using the ordinary least squares method (OLS) to determine the optimal parameters by minimizing the residual sum of squares.

[0071] During the training process, the data is divided into a training set and a validation set, and the performance of the model is evaluated through cross-validation to ensure that the model has good generalization ability. Through multiple iterations and parameter adjustments, the optimal regression model intercept and coefficient are finally obtained. After the training is completed, the intercept and coefficient of the first regression model are saved in a structured database for subsequent prediction and model optimization.

[0072] Step S48: constructing a second regression model based on the multiple difference values ​​and the multiple cost difference values.

[0073] In a possible implementation, step S48 specifically includes the following steps: The second regression model is trained based on multiple difference values ​​and multiple cost difference values ​​to obtain the second regression model intercept and the second regression model coefficient; wherein the relationship between the difference value and the cost difference is expressed by the following formula: ΔC i =α0+α1u i ; Where, ΔC i is the cost difference corresponding to the i-th historical building project, u i is the difference value corresponding to the i-th historical building project, α0 is the intercept of the second regression model, and α1 is the coefficient of the second regression model.

[0074] In order to improve the accuracy of the project cost prediction model, it is necessary to train a second regression model based on the key feature differences of numerical types and the cost difference. The goal of this step is to establish an accurate mathematical model by analyzing the impact of changes in numerical features on the cost difference, thereby enhancing the reliability of the prediction.

[0075] First, ensure that all the difference values ​​and cost difference data obtained from the previous steps have been collected. These data include the key feature differences of numerical types (such as building area, number of floors, number of rooms, etc.) extracted and calculated from the historical preliminary construction drawings and the historical actual construction drawings, as well as the cost difference calculated from the predicted stage cost and the actual stage cost.

[0076] Next, use these difference values ​​and cost difference data to train the second regression model. Select a suitable regression algorithm, such as linear regression, to establish a mapping relationship between the difference values ​​and the cost difference. The goal of the linear regression model is to find the optimal second regression model intercept (α0) and second regression model coefficient (α1) to minimize the prediction error of the model. The training process uses the ordinary least squares method (OLS) to determine the optimal parameters by minimizing the residual sum of squares.

[0077] During the training process, the data is divided into a training set and a validation set, and the performance of the model is evaluated through cross-validation to ensure that the model has good generalization ability. Through multiple iterations and parameter adjustments, the optimal regression model intercept and coefficient are finally obtained. After the training is completed, the intercept and coefficient of the second regression model are saved in a structured database for subsequent prediction and model optimization.

[0078] Step S5: Obtain the current preliminary construction project drawings corresponding to the current construction project.

[0079] Specifically, in the construction cost forecast, obtaining the preliminary construction drawings of the current construction project is a crucial step. By obtaining and analyzing these drawings, basic data is provided for subsequent cost forecasts.

[0080] Step S6: Based on each current preliminary construction engineering drawing and the regression model, the current forecast stage cost is obtained.

[0081] Specifically, in order to accurately predict the cost of the current construction project, it is necessary to extract key features based on the current preliminary construction project drawings and use the previously constructed regression model for prediction. The purpose of this step is to ensure that the prediction model can accurately estimate the cost based on the latest design information, thereby improving the accuracy of the project's cost prediction.

[0082] In a possible implementation, step S6 specifically includes the following steps: Based on each current preliminary architectural engineering drawing, a plurality of third key features are obtained; the third key features include third numerical type key features and third non-numerical type key features.

[0083] Specifically, the specific implementation of this step is similar to the specific implementation of step S21, so it will not be described in detail here.

[0084] It should be noted that in this step, the key features of numerical type also include specific quantitative indicators such as building area, number of floors, number of rooms, and material consumption. These features can be automatically extracted by image recognition software, for example, by analyzing the annotations and dimension lines in the drawings to obtain data such as building area and number of floors. Non-numerical key features also include design style, building purpose, material type, etc. These features usually appear in text or symbol form and need to be identified from the drawings through OCR technology. For example, by identifying the text descriptions and legend symbols in the drawings, information such as the design style and material type of the building can be extracted.

[0085] The second basic cost corresponding to each third key feature is determined, and all the second basic costs are summed up to obtain the cost in the current prediction stage.

[0086] Specifically, the specific implementation of this step is similar to the specific implementation of step S22, so it will not be described in detail here.

[0087] The plurality of third non-numeric key features are input into the first regression model to obtain a first current prediction stage cost difference.

[0088] Specifically, first, extract multiple third non-numerical type key features from the current preliminary architectural engineering drawings. These features include design style, building purpose, material type, etc. These features need to be extracted through optical character recognition (OCR) technology and image recognition technology, and encoded into a format acceptable to the model.

[0089] Next, the extracted and encoded third non-numeric key features are input into the previously trained first regression model. The first regression model is trained based on the relationship between the non-numeric key features extracted from historical data and the cost difference. By inputting the current non-numeric key features into the model, the corresponding cost difference can be calculated.

[0090] The first regression model will use the first regression model intercept (β0) and the first regression model coefficient (βj) obtained in the training process of the previous steps to calculate the cost difference. That is, given the first regression model intercept (β0), the first regression model coefficient (βj) and the feature vector corresponding to the third non-numeric type key feature, the cost difference corresponding to the current construction project is solved.

[0091] The plurality of third value type key features are input into the second regression model to obtain the cost difference value in the second current prediction stage.

[0092] Specifically, first, multiple third value type key features are extracted from the current preliminary architectural engineering drawings. These features include specific quantitative indicators such as building area, number of floors, number of rooms, material usage, etc. These value type key features can usually be read directly from the drawings and extracted and processed through image recognition technology.

[0093] Next, the extracted third value type key feature is input into the previously trained second regression model. The second regression model is trained based on the relationship between the value type key feature extracted from the historical data and the cost difference. By inputting the current value type key feature into the model, the corresponding cost difference can be calculated.

[0094] The second regression model will use the second regression model intercept (α0) and the second regression model coefficient (α1) obtained in the training process of the previous step to calculate the cost difference. That is, given the second regression model intercept (α0), the second regression model coefficient (α1) and the value corresponding to the third value type key feature, the cost difference corresponding to the current construction project is solved.

[0095] The cost difference values ​​of each first current forecast stage and the cost difference values ​​of each second current forecast stage are weightedly summed to obtain the total cost difference value of the current forecast stage.

[0096] Specifically, first, ensure that the first current prediction stage cost difference and the second current prediction stage cost difference have been calculated. The first current prediction stage cost difference is obtained by inputting the third non-numeric key feature into the first regression model, and the second current prediction stage cost difference is obtained by inputting the third numeric key feature into the second regression model.

[0097] Next, you need to determine the weighting coefficients. These weighting coefficients can be set based on the historical performance of the model or the importance of the features. Common methods include using the coefficient weights obtained during model training, or determining a reasonable weighting ratio based on experience and data analysis.

[0098] Through this weighted summation method, the impact of non-numerical and numerical key features on the cost can be comprehensively considered, making the final total cost difference more representative and accurate.

[0099] The cost of each current forecast stage is added to the difference between the total cost of each current forecast stage to obtain the cost of the current forecast stage.

[0100] Specifically, first, ensure that the basic cost of the current forecast stage and the difference between the total cost of the current forecast stage have been calculated. The basic cost is a preliminary estimate calculated based on the preliminary design drawings and basic feature data through empirical formulas or simple models. The total cost difference is obtained by weighted summing the first current forecast stage cost difference (non-numerical key feature influence) and the second current forecast stage cost difference (numerical key feature influence). Next, add the basic cost and the total cost difference. This addition operation comprehensively considers the impact of the basic cost and all key features on the cost, making the final forecast cost more comprehensive and accurate. Save the calculated cost of the current forecast stage to a structured database for project budget management and decision support.

[0101] Reference Figure 4 , which shows a schematic diagram of the structure of a construction project cost management system provided by an embodiment of the present application, the system includes: an acquisition module 1 and a processing module 2; An acquisition module 1 is used to acquire historical preliminary architectural engineering drawings corresponding to a plurality of historical architectural engineering projects; Processing module 2, for obtaining multiple historical forecast stage construction costs based on multiple historical preliminary construction engineering drawings; The acquisition module 1 is also used to obtain historical actual construction engineering drawings and historical actual stage costs corresponding to multiple historical construction projects; The processing module 2 is further used to construct a regression model based on the cost of each predicted stage, the cost of each actual stage, each historical preliminary construction engineering drawing, and each historical actual construction engineering drawing; The acquisition module 1 is also used to obtain the current preliminary construction engineering drawings corresponding to the current construction engineering; The processing module 2 is also used to obtain the current forecast stage cost based on each current preliminary construction engineering drawing and the regression model.

[0102] In a possible implementation, the processing module 2 is further used to obtain a plurality of first key features based on each of the historical preliminary architectural engineering drawings; The processing module 2 is further used to determine the first basic cost corresponding to each first key feature, and sum up all the first basic costs to obtain the cost of each historical prediction stage.

[0103] In a possible implementation, the processing module 2 is further used to obtain a plurality of first key features based on each historical preliminary architectural engineering drawing; the first key features include a first numerical type key feature and a first non-numerical type key feature; the processing module 2 is further used to obtain a plurality of second key features based on each historical actual architectural engineering drawing; the second key features include a second numerical type key feature and a second non-numerical type key feature; The processing module 2 is further used to encode each first non-numeric key feature and each second non-numeric key feature respectively to obtain a plurality of first vectors and a plurality of second vectors; The processing module 2 is further used to subtract each first vector from each second vector to obtain a difference vector corresponding to each historical building project; The processing module 2 is further used to subtract each first value type key feature from each second value type key feature to obtain a difference value corresponding to each historical building project; The processing module 2 is also used to obtain the cost difference corresponding to the historical building project based on the cost of each predicted stage and the cost of each actual stage; The processing module 2 is further used to construct a first regression model based on the multiple difference vectors and the multiple cost difference values; The processing module 2 is further used to construct a second regression model based on multiple difference values ​​and multiple cost difference values.

[0104] In a possible implementation, the processing module 2 is further used to train a first regression model based on multiple difference vectors and multiple cost difference values ​​to obtain a first regression model intercept and a first regression model coefficient; wherein the relationship between the difference vector and the cost difference value is expressed by the following formula: Where, ΔC i is the cost difference corresponding to the i-th historical building project, d ij is the jth element in the difference vector corresponding to the i-th historical building project, β0 is the intercept of the first regression model, β j is the coefficient of the first regression model.

[0105] In a possible implementation, the processing module 2 is further used to train a second regression model based on multiple difference values ​​and multiple cost difference values ​​to obtain a second regression model intercept and a second regression model coefficient; wherein the relationship between the difference value and the cost difference is expressed by the following formula: ΔC i =α0+α1u i ; Where, ΔC i is the cost difference corresponding to the i-th historical building project, u i is the difference value corresponding to the i-th historical building project, α0 is the intercept of the second regression model, and α1 is the coefficient of the second regression model.

[0106] In a possible implementation, the processing module 2 is further used to obtain a plurality of third key features based on each current preliminary construction engineering drawing; the third key features include a third numerical type key feature and a third non-numerical type key feature; the processing module 2 is further used to determine the second basic cost corresponding to each third key feature, and sum all the second basic costs to obtain the current forecast stage cost; The processing module 2 is further used to input the plurality of third non-numeric key features into the first regression model to obtain the cost difference value of the first current prediction stage; The processing module 2 is further used to input the plurality of third value type key features into the second regression model to obtain the cost difference value of the second current prediction stage; Processing module 2 is further used to perform weighted summation of each first current forecast stage cost difference and each second current forecast stage cost difference to obtain a total cost difference of the current forecast stage; The processing module 2 is further used to add the difference between the cost of each current forecast stage and the total cost of each current forecast stage to obtain the cost of the current forecast stage.

[0107] In a possible implementation, the processing module 2 is further used to digitally process a plurality of historical preliminary architectural engineering drawings to obtain a plurality of historical preliminary digital drawings; The processing module 2 is also used to extract the first key feature from each historical preliminary digital drawing.

[0108] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0109] The present application also discloses an electronic device. Figure 5 , Figure 5 The electronic device 500 may include: at least one processor 501 , at least one network interface 504 , a user interface 503 , a memory 505 , and at least one communication bus 502 .

[0110] The communication bus 502 is used to realize the connection and communication between these components.

[0111] The user interface 503 may include a display screen (Display) and a camera (Camera), and the optional user interface 503 may also include a standard wired interface and a wireless interface.

[0112] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0113] Among them, the processor 501 may include one or more processing cores. The processor 501 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 505, and calling data stored in the memory 505. Optionally, the processor 501 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 501 can integrate one or more combinations of a central processing unit (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 501, and it can be implemented by a single chip.

[0114] Among them, the memory 505 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 505 may optionally be at least one storage device located away from the aforementioned processor 501. Refer to Figure 5 , the memory 505 as a computer-readable storage medium may include an operating system, a network communication module, a user interface module and an application program.

[0115] exist Figure 5 In the electronic device 500 shown, the user interface 503 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 501 can be used to call an application stored in the memory 505, and when executed by one or more processors 501, the electronic device 500 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0116] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0117] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0118] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0119] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0120] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0121] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0122] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A construction project cost management method, It is characterized in that The method comprises: Obtaining historical preliminary architectural engineering drawings corresponding to a plurality of historical architectural engineering projects; Based on the plurality of said historical preliminary construction engineering drawings, a plurality of historical forecast stage construction costs are obtained; Obtaining actual historical construction engineering drawings and actual historical stage costs corresponding to a plurality of the historical construction projects; Based on the predicted stage costs, the actual stage costs, the historical preliminary construction engineering drawings, and the historical actual construction engineering drawings, a regression model is constructed; Obtain the current preliminary building project drawings corresponding to the current building project; Based on each of the current preliminary construction engineering drawings and the regression model, the current forecast stage cost is obtained.

2. The method according to claim 1, It is characterized in that Based on the plurality of the historical preliminary construction engineering drawings, a plurality of historical forecast stage costs are obtained, specifically including: Based on each of the historical preliminary architectural engineering drawings, a plurality of first key features are obtained; The first basic construction costs corresponding to the respective first key features are determined, and all the first basic construction costs are summed to obtain the construction costs of the respective historical prediction stages.

3. The method according to claim 1, It is characterized in that The regression model includes a first regression model and a second regression model; the regression model is constructed based on the predicted stage cost, the actual stage cost, the historical preliminary construction engineering drawings and the historical actual construction engineering drawings, specifically including: Based on each of the historical preliminary architectural engineering drawings, a plurality of first key features are obtained; the first key features include first numerical type key features and first non-numerical type key features; Based on each of the historical actual architectural engineering drawings, a plurality of second key features are obtained; the second key features include a second numerical type key feature and a second non-numerical type key feature; Encode each of the first non-numeric key features and each of the second non-numeric key features to obtain a plurality of first vectors and a plurality of second vectors; Subtract each of the first vectors from each of the second vectors to obtain a difference vector corresponding to each of the historical building projects; subtract each of the first value type key features from each of the second value type key features to obtain a difference value corresponding to each of the historical building projects; Based on the construction costs of each predicted stage and the construction costs of each actual stage, obtaining the cost difference corresponding to the historical building project; Based on the plurality of difference vectors and the plurality of cost difference values, constructing the first regression model; Based on the plurality of difference values ​​and the plurality of cost difference values, the second regression model is constructed.

4. The method according to claim 3, It is characterized in that The constructing the first regression model based on the plurality of difference vectors and the plurality of cost difference values ​​specifically includes: The first regression model is trained based on the plurality of difference vectors and the plurality of cost difference values ​​to obtain a first regression model intercept and a first regression model coefficient; wherein the relationship between the difference vector and the cost difference value is expressed by the following formula: Where, ΔC i is the cost difference corresponding to the i-th historical building project, d ij is the jth element in the difference vector corresponding to the i-th historical building project, β 0 is the intercept of the first regression model, β j is the coefficient of the first regression model.

5. The method according to claim 3, It is characterized in that The constructing the second regression model based on the plurality of difference values ​​and the plurality of cost difference values ​​specifically includes: The second regression model is trained based on the plurality of difference values ​​and the plurality of cost difference values ​​to obtain a second regression model intercept and a second regression model coefficient; wherein the relationship between the difference value and the cost difference value is expressed by the following formula: ΔC i =α 0 +α 1 u i ; Where, ΔC i is the cost difference corresponding to the i-th historical building project, u i is the difference value corresponding to the i-th historical building project, α 0 is the intercept of the second regression model, α 1 is the coefficient of the second regression model.

6. The method according to claim 1, It is characterized in that The regression model includes a first regression model and a second regression model; the current forecast stage cost is obtained based on each of the current preliminary construction engineering drawings and the regression model, specifically including: Based on each of the current preliminary architectural engineering drawings, a plurality of third key features are obtained; the third key features include third numerical type key features and third non-numerical type key features; Determine the second basic cost corresponding to each of the third key features, and sum all of the second basic costs to obtain the cost in the current forecast stage; Inputting the plurality of third non-numeric key features into the first regression model to obtain a first current forecast stage cost difference; Inputting the plurality of third value type key features into the second regression model to obtain a second current forecast stage cost difference; Performing a weighted summation of the cost difference values ​​of the first current forecast stage and the cost difference values ​​of the second current forecast stage to obtain a total cost difference value of the current forecast stage; The cost of each current forecast stage is added to the difference between the total cost of each current forecast stage to obtain the cost of the current forecast stage.

7. The method according to claim 2, It is characterized in that Based on each of the historical preliminary architectural engineering drawings, a plurality of first key features are obtained, specifically including: Digitally processing the plurality of said historical preliminary architectural engineering drawings to obtain a plurality of historical preliminary digital drawings; A first key feature is extracted from each of the historical preliminary digital drawings.

8. A construction project cost management system, It is characterized in that The system comprises: an acquisition module and a processing module; The acquisition module is used to acquire historical preliminary architectural engineering drawings corresponding to a plurality of historical architectural engineering projects; The processing module is used to obtain multiple historical forecast stage costs based on the multiple historical preliminary construction engineering drawings; The acquisition module is further used to acquire historical actual construction engineering drawings and historical actual stage costs corresponding to the plurality of historical construction projects; The processing module is further used to construct a regression model based on the predicted stage costs, the actual stage costs, the historical preliminary construction engineering drawings and the historical actual construction engineering drawings; The acquisition module is further used to acquire the current preliminary construction engineering drawing corresponding to the current construction engineering; The processing module is also used to obtain the current forecast stage cost based on each of the current preliminary construction engineering drawings and the regression model.

9. An electronic device, It is characterized in that It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images