An engineering cost scheme intelligent comparison and selection method and system for fabricated buildings

Abstract

The application discloses an engineering cost scheme intelligent comparison and selection method and system for fabricated buildings, and belongs to the technical field of fabricated building engineering; the application comprises collecting multi-dimensional basic data of a whole life cycle of a fabricated building project, and performing pretreatment operation on the multi-dimensional basic data to generate model input data; the application realizes scientific quantification and empowerment of multi-dimensional cost influence factors such as a component standardization rate, a transportation radius and assembly efficiency in combination with an analytic hierarchy process, completely discards a traditional mode relying on experience judgment of cost personnel, greatly reduces subjective errors of comparison and selection results, can adapt to differentiated needs of fabricated buildings in different climate zones and different building types, provides scientific and accurate digital support for cost scheme decision of fabricated building projects, and effectively meets the scheme decision needs of rapid iteration of fabricated buildings.

Description

Technical Field

[0001] This invention relates to the field of prefabricated building engineering technology, specifically to an intelligent comparison and selection method and system for engineering cost schemes of prefabricated buildings. Background Technology

[0002] Cost comparison for prefabricated buildings is a crucial step in the field of prefabricated building engineering technology. Its cost structure has significant unique characteristics, encompassing specific stages such as prefabricated component production (approximately 30%-40%), transportation (10%-15%), and on-site assembly (25%-35%). Compared to cast-in-place buildings, it adds the costs of factory production of components and supply chain management. Currently, the comparison work in this field mainly relies on the experience and judgment of cost estimators, and the supporting cost estimation software is mostly developed for cast-in-place buildings. At the same time, existing cost comparison tools mostly focus on single aspects such as material cost calculation, and a full-process cost analysis system adapted to prefabricated buildings has not yet been formed.

[0003] Existing technologies and practices related to cost comparison in prefabricated construction have several shortcomings. First, the comparison results are highly subjective and lack precision. Traditional methods struggle to systematically quantify cost-influencing factors across multiple dimensions, such as component standardization rates, transportation radius, and assembly process complexity. Second, the tools and software have poor compatibility. Traditional cost estimation software cannot meet the needs of dynamic cost analysis across the entire prefabricated construction chain, and existing comparison tools lack integrated analysis capabilities for costs throughout the design, production, transportation, and construction processes. Third, the comparison process is inefficient. Due to the lack of an intelligent optimization mechanism based on project characteristics, the cost comparison cycle for a single prefabricated construction project is typically 3-5 days, making it difficult to adapt to the rapid iteration of decision-making requirements in prefabricated construction. Furthermore, traditional manual comparison methods have a high cost error rate, failing to provide accurate decision support for projects.

[0004] To address the aforementioned technical shortcomings, a solution is proposed. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent comparison and selection method and system for engineering cost schemes of prefabricated buildings to solve the problems mentioned above.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent comparison system for engineering cost schemes in prefabricated buildings, comprising the following steps:

[0007] S1. Collect multi-dimensional basic data of the entire life cycle of prefabricated building projects, and perform preprocessing operations to generate model input data;

[0008] S2. Based on the cost composition characteristics of prefabricated buildings, a three-level cost influencing factor index system is constructed. The weight of each influencing factor in the index system is calculated by the analytic hierarchy process to obtain the weight value of the cost influencing factor.

[0009] S3. Using the model input data and the weight values ​​of cost influencing factors as input, construct a twin-architecture intelligent comparison model. The intelligent comparison model includes a cost prediction sub-model and a scheme optimization sub-model, which respectively complete the cost prediction and cost scheme optimization of prefabricated buildings.

[0010] S4. Input the cost data of the prefabricated building schemes to be compared into the intelligent comparison model, calculate the multi-dimensional cost data of each scheme, and comprehensively score each scheme to be compared by the approximation ideal solution ranking method. Output the optimal cost scheme based on the scoring results, and generate a visual decision report that includes cost composition analysis and sensitivity analysis.

[0011] Furthermore, the multi-dimensional basic data in S1 includes design parameter data, material cost data, production process data, transportation and logistics data, and construction management data. The design parameter data includes component type, component size, and component quantity. The material cost data includes main material price, auxiliary material price, and material loss rate. The production process data includes production process type, prefabrication plant capacity, and production energy consumption. The transportation and logistics data includes transportation distance, transportation mode, and loading and unloading costs. The construction management data includes construction machinery type, labor cost, and construction period. The preprocessing operation specifically involves using the three Sigma criterion to remove outliers, using the Z-score method to standardize the data, and using the minimum-maximum method to normalize the data. After completing these operations in sequence, the standardized and normalized new data is obtained and marked as model input data.

[0012] Furthermore, the three-level cost influencing factor index system in S2 includes a component characteristic layer, a supply chain layer, and a construction layer. The component characteristic layer includes component standardization rate and component integration degree indicators. The supply chain layer includes material price fluctuation coefficient and transportation efficiency indicators. The construction layer includes assembly speed and on-site management cost indicators. When calculating the weights using the analytic hierarchy process, a judgment matrix for each influencing factor is first constructed through expert scoring. The consistency of the judgment matrix is ​​then checked. After the check is passed, the weight values ​​of each factor are calculated. Among them, the weight value of component standardization rate is 0.25, the weight value of material price fluctuation coefficient is 0.20, the weight value of transportation distance is 0.15, the weight value of assembly efficiency is 0.20, and the weight value of construction time is 0.20.

[0013] Furthermore, the cost prediction sub-model in S3 is a three-layer backpropagation neural network model. The input layer of this model has twelve neurons, which correspond to twelve core cost influencing factors, the hidden layer has twenty neurons, and the output layer has one neuron. The output is the predicted cost value of prefabricated buildings.

[0014] The cost prediction sub-model is trained using at least one hundred sets of historical prefabricated building project data. After at least one thousand iterations, the mean square error of the model is no greater than 0.001. The scheme optimization sub-model is an improved particle swarm optimization algorithm model, with a population size of fifty, an iteration count of eighty, a self-awareness learning factor of 1.5, and a social awareness learning factor of 1.7. The optimization solution of the cost scheme is completed with the dual objective functions of lowest cost and shortest construction period.

[0015] Furthermore, the multi-dimensional cost data in S4 includes total project cost, unit area cost, and dynamic cost. The dynamic cost is the cost value calculated considering the time value of money. When using the approximation of ideal solution ranking method, the multi-dimensional cost data is first standardized, and then a weighted normalized decision matrix is ​​constructed by combining the weight values ​​of cost influencing factors. After calculating the positive and negative ideal solutions, the closeness of each candidate scheme to the positive and negative ideal solutions is calculated, and the closeness value is used as the comprehensive scoring result.

[0016] The cost composition analysis in the visualized decision report includes the cost proportion of each link in the production, transportation and on-site assembly of prefabricated components, and the sensitivity analysis is the cost change rate analysis when the material price fluctuates within a range of plus or minus 5%.

[0017] An intelligent cost comparison system for prefabricated buildings is disclosed. The intelligent comparison system includes a data acquisition module, a data preprocessing module, a cost influencing factor analysis module, an intelligent comparison model construction module, and a comparison output module. The data acquisition module is used to connect to various data sources throughout the entire life cycle of prefabricated buildings, realizing the unified collection and storage of multi-dimensional basic data on prefabricated building project design, materials, production, transportation, and construction. The data sources include building information modeling design software, building material price database, prefabrication plant production management system, logistics management platform, and construction site management platform.

[0018] Furthermore, the data preprocessing module is signal-connected to the data acquisition module and is used to receive multi-dimensional basic data transmitted by the data acquisition module. It sequentially performs outlier removal, standardization, and normalization preprocessing operations. The module has built-in three sigma criterion outlier removal program, Z-score standardization program, and minimum-maximum normalization program. After processing, it outputs standardized and normalized model input data.

[0019] Furthermore, the cost influencing factor analysis module is signal-connected to the data preprocessing module to receive the model input data. It has a built-in three-level cost influencing factor index system construction program and analytic hierarchy process (AHP) weight calculation program. First, it generates a three-level index system of component feature layer, supply chain layer, and construction layer based on the cost characteristics of prefabricated buildings. Then, it constructs a judgment matrix through expert scoring and completes consistency verification. Finally, it outputs the weight values ​​of each cost influencing factor.

[0020] Furthermore, the intelligent comparison model construction module is signal-connected to the cost influencing factor analysis module, and is used to receive model input data and cost influencing factor weight values. It has a built-in backpropagation neural network training program and an improved particle swarm optimization program to construct a cost prediction sub-model and a scheme optimization sub-model, forming a twin-architecture intelligent comparison model. The module supports manual tuning and iterative updates of model parameters and outputs the trained intelligent comparison model.

[0021] Furthermore, the comparison output module is signal-connected to the intelligent comparison model construction module, and is used to receive the cost scheme data to be compared and input it into the trained intelligent comparison model. It has built-in multi-dimensional cost calculation program, approximation ideal solution ranking method comprehensive scoring program and visualization report generation program to complete the calculation of total cost, unit area cost and dynamic cost of each scheme, output comprehensive scoring results and optimal cost scheme, and generate a visualization decision report including cost composition ratio and sensitivity analysis. The module also supports real-time monitoring and result feedback of the comparison process.

[0022] The beneficial effects of this invention are:

[0023] 1. This invention constructs a three-tiered cost influencing factor index system covering component feature layer, supply chain layer, and construction layer. Combined with the analytic hierarchy process (AHP), it achieves scientific quantification and weighting of multi-dimensional cost influencing factors such as component standardization rate, transportation radius, and assembly efficiency. This completely abandons the traditional model that relies on the experience judgment of cost estimators, significantly reducing the subjective error of the comparison results. At the same time, the full life cycle cost analysis system, designed for the special cost structure of prefabricated buildings, realizes integrated analysis of costs throughout the entire process of design, production, transportation, and construction. It solves the adaptation problem of traditional cost software's cast-in-place thinking and meets the core needs of dynamic cost analysis of the entire chain of prefabricated buildings. The multi-factor quantitative analysis combined with the intelligent prediction model further controls the cost error rate within ±3%, which significantly improves the comparison accuracy compared to the ±10% error rate of traditional manual methods.

[0024] 2. This invention automates the entire cost scheme comparison process through automated multi-source data acquisition and preprocessing, intelligent model rapid calculation, and TOPSIS comprehensive scoring. It reduces the comparison cycle for a single prefabricated building project from the traditional 3-5 days to within 2 hours, improving efficiency by over 90%. Simultaneously, the constructed twin-architecture intelligent comparison model forms an intelligent optimization mechanism based on project characteristics. The scheme optimization sub-model achieves accurate solutions with the dual objective functions of lowest cost and shortest construction period. Furthermore, the model supports real-time updates of parameters such as material prices and transportation rates. The algorithm modules are pluggable and iterative, adapting to the differentiated needs of prefabricated buildings in different climate zones and building types. This provides scientific and accurate digital support for cost scheme decisions in prefabricated building projects, effectively meeting the rapid iteration needs of prefabricated building scheme decisions. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart of the method of the present invention;

[0027] Figure 2 This is a schematic diagram of the method architecture of the present invention;

[0028] Figure 3 This is a flowchart of the system of the present invention;

[0029] Figure 4 This part is the structural block diagram of the intelligent selection system of the present invention;

[0030] Figure 5 This is the lower half of the structural block diagram of the intelligent comparison and selection system of the present invention;

[0031] Figure 6 This is a schematic diagram of the twin architecture of the intelligent selection model of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some 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 are within the scope of protection of the present invention.

[0033] Example 1: Please refer to Figure 1 - Figure 6 As shown in the figure, this embodiment is a method and system for intelligent comparison of engineering cost schemes for prefabricated buildings, including the following steps:

[0034] S1. Multi-dimensional basic data collection and preprocessing: Based on the pre-defined area, collect multi-dimensional basic data of the entire life cycle of the target prefabricated shear wall residential project. Mark the prefabricated components of standard floors and non-standard floors as design parameter data; mark the recorded data of concrete, steel bars and prefabricated auxiliary materials as material cost data; summarize and mark the daily production capacity of the prefabrication plant, steam curing energy consumption and production line balance rate as production process data; summarize and mark the transportation distance of prefabricated components, dump truck transportation unit price and transportation efficiency as transportation logistics data; summarize and mark the relevant recorded data of personnel and construction machinery required for assembly as construction management data.

[0035] The obtained data is processed using formulas ,in, Represented as a single basic data value, This is expressed as the sample mean of the data set. This is represented by the sample standard deviation of the data set. When the deviation of a data value from the mean exceeds three times the standard deviation, it is judged as an outlier and removed. In this embodiment, after calculation, there are no outliers that meet this condition, and all basic data are retained.

[0036] According to the formula ,in, Represented as standardized data values, Represented as the raw data value of a single basic data value. This is expressed as the sample mean of the data set. This is represented as the sample standard deviation of the data set, which transforms the original data into standardized data with a mean of 0 and a standard deviation of 1, thus eliminating the influence of dimensions. In this embodiment, data with different dimensions such as the number of components, transportation distance, and labor cost are standardized. For example, the standardized value of a transportation distance of 25km is 0.32 after calculation.

[0037] According to the formula ,in, Represented as normalized data values, This is represented as the standardized data value. Represented as the minimum value of standardized data. The value is represented as the maximum value of the standardized data. The standardized data is mapped to the 0-1 interval to obtain the model input data. In this embodiment, after normalization, all data are in the 0-1 interval, thus completing the generation of the model input data.

[0038] S2: Construction and weight calculation of a three-level cost influencing factor index system. A three-level cost influencing factor index system is constructed, including the component feature layer, the supply chain layer, and the construction layer. The component feature layer selects the component standardization rate and component integration degree; the supply chain layer selects the material price fluctuation coefficient and transportation efficiency; and the construction layer selects the assembly speed and on-site management cost. At the same time, transportation distance and construction period are included as core related indicators in the system.

[0039] A judgment matrix was constructed based on expert scoring. Five experts in prefabricated building engineering cost were invited to score the importance of five core factors: component standardization rate, material price fluctuation coefficient, transportation distance, assembly efficiency, and construction period, thus constructing a 5th-order judgment matrix. ,in, Represented as the i-th factor relative to the i-th factor The importance ratio of each factor The possible values ​​are 1, or Possible values ;

[0040] Consistency check, according to the formula ,in, Expressed as the consistency ratio, As a consistency indicator, It is represented by the largest eigenvalue of the judgment matrix. Represented as the number of factors, Represented as the average random consistency index, a 5th order matrix. The possible value is 1.12, when When the value is less than 0.1, the consistency check of the judgment matrix is ​​passed, and the weight calculation result is valid; in this embodiment, the calculated value is... The possible value is 5.12. The possible value is 0.03. =0.027<0.1, inspection passed;

[0041] Weight calculation is performed by solving the eigenvectors of the judgment matrix using the eigenvalue method. After normalization, the weight values ​​of each factor are obtained: component standardization rate 0.25, material price fluctuation coefficient 0.20, transportation distance 0.15, assembly efficiency 0.20, and construction period 0.20.

[0042] S3: Intelligent comparison model construction of twin architecture. The intelligent comparison model of twin architecture is constructed using the model input data generated by S1 and the cost influencing factor weights obtained by S2 as inputs. It is divided into a cost prediction sub-model and a scheme optimization sub-model.

[0043] Construct a three-layer backpropagation neural network with 12 input neurons, 20 hidden neurons, and 1 output neuron. The ReLU activation function is used, and the formula is as follows: ,in, Represented as the neuron output value, This is represented as the neuron's input value. If positive input values ​​are retained, negative input values ​​are set to 0 to solve the gradient vanishing problem.

[0044] Historical data of prefabricated building projects stored in the intelligent comparison system are retrieved as training samples and trained iteratively 1,000 times. The formula for the mean square error of the model is as follows: ,in, This is expressed as the sample size. Represented as actual cost value, This represents the model's predicted cost value, which measures the mean squared error between the model's predicted and actual values. A smaller value indicates higher model accuracy. A preset iteration threshold is used, which can be 0.001. In this embodiment, after one thousand iterations... =0.0008, meeting the training requirements, the cost prediction sub-model is now complete;

[0045] Based on population size 50, iteration count 80, and self-awareness learning factor Social cognitive learning factors Inertial weight The speed update formula is: The position update formula is: ;in, Represented as The velocity of particle i in the d-th dimension Represented as inertia weight, , Represented as a random number between 0 and 1 Let represent the optimal position of particle i. This represents the globally optimal position of the population. Let be the position of particle i in the d-th dimension of the t-th generation. By iteratively updating the velocity and position, we search for the optimal solution that satisfies the dual objective function.

[0046] With the dual objectives of lowest cost and shortest construction period as the goals, and a weighting ratio of 0.6:0.4, a comprehensive objective function is constructed. ,in, Represented as the normalized cost value, Expressed as the normalized value of the construction period, solve... The minimum value is the optimal solution, and the solution optimization sub-model is now complete.

[0047] S4: Comparison and Output of Potential Schemes. Three cost schemes for this prefabricated shear wall residential project are selected as potential schemes for comparison. The scheme data is as follows:

[0048] Option A: Component standardization rate 85%, prefabrication locally, transportation radius 25km, planned construction period 300 days;

[0049] Option B: Component standardization rate 60%, procurement of low-cost materials from other regions, concrete unit price 380 yuan / cubic meter, transportation radius 150km, planned construction period 300 days;

[0050] Option C: Component standardization rate of 40%, rapid on-site construction, 10% reduction in construction period (270 days), 15% increase in labor costs (402.5 yuan / person / day), and the data is not limited to these.

[0051] Input the data from the three options into the intelligent comparison model to calculate the total cost and unit area cost. Simultaneously, calculate the dynamic cost considering the time value of money. The dynamic cost formula is as follows: ,in, Represented as dynamic cost, Represented as static total cost, Represented as present value factor, This is expressed as the benchmark rate of return, which can take a value of 6%. The figures represent the construction period. Based on this, Option A has a static total cost of 28.6 million yuan, a unit area cost of 4,200 yuan / ㎡, and a dynamic cost of 27.896 million yuan; Option B has a static total cost of 29.5 million yuan, a unit area cost of 4,350 yuan / ㎡, and a dynamic cost of 28.763 million yuan; Option C has a static total cost of 31 million yuan, a unit area cost of 4,580 yuan / ㎡, and a dynamic cost of 30.528 million yuan.

[0052] The approximation of the ideal solution ranking method, the comprehensive score, and the normalized decision matrix are given by the formula: ,in, Represented as normalized values, Let be the cost index value of the i-th scheme, and m be the number of schemes;

[0053] The weighted normalized decision matrix is ​​given by the following formula: ,in, Represented as a weighted normalized value, Let the weight of the j-th indicator be denoted as ; calculate the positive ideal solution. Negative ideal solution ;

[0054] The formula for calculating closeness is: ,in, Let be the distance from the solution to the ideal solution. This represents the distance from the proposed solution to the negative ideal solution. The value is represented as proximity, and can be between 0 and 1. The larger the value, the better the solution. Based on the data provided above, the proximity values ​​are 0.82 for solution A, 0.65 for solution B, and 0.41 for solution C. Solution A has the highest proximity and is the optimal cost solution. Specifically, these values ​​are summarized to generate a visual decision report.

[0055] Example 2: The intelligent comparison system for prefabricated building engineering cost schemes constructed in this example includes a data acquisition module, a data preprocessing module, a cost influencing factor analysis module, an intelligent comparison model construction module, and a comparison output module. The modules are connected by signals and work together to complete the intelligent comparison operation.

[0056] The data acquisition module interfaces with Revit building information modeling software, a national building materials price database, a prefabrication plant production management system, a logistics and transportation management platform, and a construction site smart management platform. It develops a unified data interface to achieve automatic collection and batch storage of multi-dimensional basic data in design, materials, production, transportation, and construction. In this embodiment, the module completes the collection of all basic data for the aforementioned 100,000 square meter project within 5 minutes, with a data collection accuracy of 100%. It also supports manual data supplementation and modification.

[0057] The data preprocessing module is connected to the data acquisition module by signal. It has built-in programmed algorithms for outlier removal using the three sigma criterion, standardization using the Z-score method, and normalization using the minimum-maximum method. After receiving the basic data transmitted by the data acquisition module, it automatically performs preprocessing operations. The operation process is visualized, and the data processing progress and results can be viewed in real time. In this embodiment, the module completes the preprocessing of all data within 1 minute, generates model input data, and transmits it to the cost influencing factor analysis module.

[0058] The cost influencing factor analysis module has a built-in program for constructing a three-level cost influencing factor index system and a weight calculation program for the analytic hierarchy process (AHP). It has a preset three-level index system template that is common to prefabricated buildings and allows users to customize and adjust the indexes according to the project type (frame / shear wall structure). The AHP program supports online expert scoring, automatic generation of judgment matrices, and automatic calculation of consistency checks. In this embodiment, the module completes the construction of the index system and weight calculation within 3 minutes, outputs the weight values ​​of cost influencing factors, and transmits them to the intelligent comparison model construction module.

[0059] The intelligent selection model construction module has a built-in backpropagation neural network training program and an improved particle swarm optimization program. It presets the model parameters defined in claim 4 (12 neurons in the input layer, 20 neurons in the hidden layer, population size of 50, etc.) and supports users to manually adjust the model parameters according to project characteristics. The program has a built-in historical project database, which can automatically retrieve training samples to complete model training. The mean square error changes are displayed in real time during the training process, and a model training report is generated after training is completed. In this embodiment, the module completes the construction and training of the twin architecture intelligent selection model within 20 minutes. The model accuracy meets the requirements and is transmitted to the selection output module.

[0060] The comparison and selection output module has a built-in multi-dimensional cost calculation program, a comprehensive scoring program based on the approximation of ideal solution ranking method, and a visualization report generation program. It supports batch import of data for the options to be compared and automatically inputs the data into the intelligent comparison and selection model to complete multi-dimensional cost calculation. The comprehensive scoring program displays the scoring process and results of each option in real time and supports the export of scoring results. The generated decision report supports export in PDF, Excel and other formats. At the same time, this module has a comparison and selection process monitoring interface, which can view the running status of each stage in real time and supports feedback of results to the preceding modules for model parameter adjustment. In this embodiment, this module completes the comparison and scoring of three options to be compared within 5 minutes, outputs the optimal solution and generates a visualization decision report.

[0061] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

[0062] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0063] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for intelligently comparing and selecting engineering cost schemes for prefabricated buildings, characterized in that, Includes the following steps: S1. Collect multi-dimensional basic data of the entire life cycle of prefabricated building projects, and perform preprocessing operations to generate model input data; S2. Based on the cost composition characteristics of prefabricated buildings, a three-level cost influencing factor index system is constructed. The weight of each influencing factor in the index system is calculated by the analytic hierarchy process to obtain the weight value of the cost influencing factor. S3. Using the model input data and the weight values ​​of cost influencing factors as input, construct a twin-architecture intelligent comparison model. The intelligent comparison model includes a cost prediction sub-model and a scheme optimization sub-model, which respectively complete the cost prediction and cost scheme optimization of prefabricated buildings. S4. Input the cost data of the prefabricated building schemes to be compared into the intelligent comparison model, calculate the multi-dimensional cost data of each scheme, and comprehensively score each scheme to be compared by the approximation ideal solution ranking method. Output the optimal cost scheme based on the scoring results, and generate a visual decision report that includes cost composition analysis and sensitivity analysis.

2. The intelligent comparison method for engineering cost schemes of prefabricated buildings according to claim 1, characterized in that, The multi-dimensional basic data in S1 includes design parameter data, material cost data, production process data, transportation and logistics data, and construction management data. The preprocessing operation specifically involves using the three sigma criterion to remove outliers, using the Z-score method to standardize the data, and using the minimum-maximum method to normalize the data. After completing these operations in sequence, the standardized and normalized new data are obtained and marked as model input data.

3. The intelligent comparison and selection method for engineering cost schemes of prefabricated buildings according to claim 1, characterized in that, The three-level cost influencing factor index system in S2 includes the component characteristic layer, the supply chain layer, and the construction layer. When calculating the weights using the analytic hierarchy process, a judgment matrix for each influencing factor is first constructed through expert scoring. The consistency of the judgment matrix is ​​then checked. After the check is passed, the weight values ​​of each factor are calculated. Among them, the weight value of the component standardization rate is 0.25, the weight value of the material price fluctuation coefficient is 0.20, the weight value of the transportation distance is 0.15, the weight value of the assembly efficiency is 0.20, and the weight value of the construction work is 0.

20.

4. The intelligent comparison and selection method for engineering cost schemes of prefabricated buildings according to claim 1, characterized in that, The cost prediction sub-model in S3 is a three-layer backpropagation neural network model. The input layer of this model has twelve neurons, which correspond to twelve core cost influencing factors. The hidden layer has twenty neurons, and the output layer has one neuron. The output is the predicted cost value of prefabricated buildings. The cost prediction sub-model is trained using at least one hundred sets of historical prefabricated building project data. After at least one thousand iterations, the mean square error of the model is no greater than 0.

001. The scheme optimization sub-model is an improved particle swarm optimization algorithm model, with a population size of fifty, an iteration count of eighty, a self-awareness learning factor of 1.5, and a social awareness learning factor of 1.

7. The optimization solution of the cost scheme is completed with the dual objective functions of lowest cost and shortest construction period.

5. The intelligent comparison method for engineering cost schemes of prefabricated buildings according to claim 1, characterized in that, The multi-dimensional cost data in S4 includes total project cost, unit area cost, and dynamic cost. The dynamic cost is the cost value calculated considering the time value of money. When using the approximation of ideal solution ranking method, the multi-dimensional cost data is first standardized, and then a weighted normalized decision matrix is ​​constructed by combining the weight values ​​of cost influencing factors. After calculating the positive and negative ideal solutions, the closeness of each candidate scheme to the positive and negative ideal solutions is calculated, and the closeness value is used as the comprehensive scoring result.

6. An intelligent cost comparison system for prefabricated buildings, used in the intelligent cost comparison method for prefabricated buildings as described in any one of claims 1-5, characterized in that, The intelligent comparison system includes a data acquisition module, a data preprocessing module, a cost influencing factor analysis module, an intelligent comparison model construction module, and a comparison output module. The data acquisition module is used to connect to various data sources throughout the entire life cycle of prefabricated buildings, realizing the unified collection and storage of multi-dimensional basic data on the design, materials, production, transportation, and construction of prefabricated building projects. The data sources include building information modeling design software, building material price database, prefabrication plant production management system, logistics management platform, and construction site management platform.

7. The intelligent cost comparison system for prefabricated buildings according to claim 6, characterized in that, The module receives multi-dimensional basic data transmitted by the data acquisition module and sequentially performs outlier removal, standardization, and normalization preprocessing operations. The module has built-in three sigma criterion outlier removal program, Z-score standardization program, and minimum-maximum normalization program. After processing, it outputs standardized and normalized model input data.

8. The intelligent comparison system for engineering cost schemes of prefabricated buildings according to claim 6, characterized in that, Based on the model input data, the built-in three-level cost influencing factor index system construction program and the analytic hierarchy process (AHP) weight calculation program first generate a three-level index system of component feature layer, supply chain layer and construction layer based on the cost characteristics of prefabricated buildings. Then, the judgment matrix is ​​constructed through expert scoring and consistency verification is completed. Finally, the weight values ​​of each cost influencing factor are output.

9. The intelligent comparison system for engineering cost schemes of prefabricated buildings according to claim 6, characterized in that, The module acquires model input data and cost influencing factor weights, incorporates a backpropagation neural network training program and an improved particle swarm optimization program, and constructs a cost prediction sub-model and a scheme optimization sub-model to form a twin-architecture intelligent comparison model. The module supports manual tuning and iterative updates of model parameters and outputs the trained intelligent comparison model.

10. The intelligent comparison system for engineering cost schemes of prefabricated buildings according to claim 6, characterized in that, The system obtains the cost scheme data to be compared and inputs it into the trained intelligent comparison model. It has built-in multi-dimensional cost calculation program, approximation ideal solution ranking method comprehensive scoring program and visualization report generation program to complete the calculation of total cost, unit area cost and dynamic cost of each scheme, output comprehensive scoring results and optimal cost scheme, and generate a visualization decision report including cost composition ratio and sensitivity analysis. The module also supports real-time monitoring of the comparison process and result feedback.

Images