Building construction carbon emission dynamic optimization method and system based on Revit and AI linkage

Abstract

The invention discloses a building construction carbon emission dynamic optimization method and system based on Revit and AI linkage. The method comprises the following steps: step 1, data integration; step 2, model training; step 3, logistics simulation; step 4, collaborative optimization; 5, dynamic early warning is carried out; step 6, generating a thermodynamic diagram; step 7, integrating paths; 8, iterative optimization is carried out; and 9, multi-scene analysis is carried out. According to the method, real-time variables are dynamically captured through an Azure AutoML time sequence model, and the prediction error is reduced; automatic management is realized by integrating a code-free tool chain, and the cost is reduced; double objective functions are innovated to synchronously optimize transportation and environmental protection benefits; a full-life-cycle closed-loop system is constructed to improve the dynamic adjustment capability; and the generation of visual decision support efficiency is improved, and revolutionary technical support is provided for green construction.

Description

Technical Field

[0001] The present invention relates to the field of digital management of building construction, and in particular to a method and system for dynamic optimization of carbon emissions from building construction based on the linkage of Revit and AI. Background Art

[0002] Construction carbon emissions refer to greenhouse gas emissions, such as carbon dioxide, generated directly or indirectly by various construction activities during the construction process. These emissions occur throughout the entire construction lifecycle, from site preparation, foundation construction, main structure construction, to interior and decoration.

[0003] Existing construction carbon emission management has significant flaws: traditional methods rely on manual experience or static BIM plug-ins (such as Autodesk Insight), and cannot dynamically respond to real-time factors such as construction progress and weather changes; AI prediction solutions require customized development of algorithm models, which places high demands on the programming capabilities of construction companies; logistics and carbon emission optimization are separated, and there is a lack of collaborative optimization mechanism. To this end, a dynamic optimization method and system for construction carbon emissions based on the linkage of Revit and AI is proposed. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a dynamic optimization method and system for carbon emissions in construction based on the linkage of Revit and AI to solve the problems raised in the above background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a dynamic optimization method for construction carbon emissions based on the linkage of Revit and AI, comprising the following steps: Step 1: Data integration: The Dynamo plug-in for Revit automatically extracts component material usage and construction phase schedules from the BIM model, and uses Excel to match the public carbon emission factor library to calculate the carbon emission baseline value; Step 2: Model training: Import historical data into the Microsoft Azure AutoML platform for preprocessing and training regression models to predict future carbon emissions. The historical data includes date, construction stage, material usage, weather, and actual carbon emissions. Step 3: Logistics Simulation: Import the Revit site model into Pathfinder software, set yard rules, and simulate to generate the optimal material transportation path; Step 4: Collaborative optimization: Based on the logistics simulation results, a dual-objective optimization function was set up in the Azure AutoML platform to simultaneously minimize transportation distance and interference in carbon emission-sensitive areas; Step 5: Dynamic warning: When the predicted carbon emissions exceed the threshold, an early warning mark is automatically triggered in the Revit model, and adjustments to the construction plan are recommended; Step 6: Heat map generation: Based on the optimized carbon emission data, a carbon emission heat map is generated in the Revit model to visually display the emission intensity of different areas; Step 7: Path integration: Embed the optimal transportation route generated by Pathfinder into Revit drawings to form a construction instruction manual with route annotations; Step 8: Iterative optimization: Dynamically update Azure AutoML model parameters based on real-time construction data feedback for continuous optimization; Step 9: Multi-scenario analysis: By adjusting the construction schedule and weather parameters, we simulate carbon emission changes under different scenarios and generate comparative analysis reports; Carbon Emission Factor Library: Internationally recognized databases (such as IPCC reports and ISO standards) are used and regularly updated to ensure data accuracy and timeliness. Regression Model Selection: A random forest regression model is selected, and metrics such as R² and MSE are used to evaluate model performance and ensure prediction accuracy. Yard Rule Implementation: Yard rules are set using Pathfinder software and adjusted based on actual construction site conditions to improve simulation accuracy. Dual-Objective Optimization Function: A weighted summation method is used to dynamically adjust weight coefficients to balance transportation distance and interference from carbon emission-sensitive areas. Warning Marker Implementation: Warning areas are marked in red in the Revit model, and the Revit API is used to automatically trigger and update warning markers. Heat Map Gradient Color Scale: A heat map is generated using a gradient of red, yellow, and green to visually display the carbon emission intensity of different areas. This method is more complete in terms of carbon emission calculation, model prediction, logistics simulation, collaborative optimization, dynamic early warning and visualization. It adopts an internationally recognized carbon emission factor library and updates it regularly, which improves the accuracy of carbon emission calculation; selects a random forest regression model and uses R² and MSE to evaluate performance, which improves prediction accuracy; sets and adjusts yard rules through Pathfinder software, which enhances simulation accuracy; uses the weighted summation method to dynamically adjust weight coefficients, balancing transportation distance and carbon emissions; uses red mark early warning in the Revit model and automatically triggers it through the API, which improves the practicality of the method; uses red, yellow and green gradient color levels to generate heat maps, which enhances the visualization effect. These improvements will help improve the efficiency and effectiveness of dynamic optimization of carbon emissions in construction and promote the widespread application and promotion of the method.

[0006] Preferably, the component material usage data extracted in step 1 includes geometric dimensions, material density and construction loss rate parameters; Component material usage data is automatically extracted using Revit's Dynamo plug-in. The specific process is as follows: Dynamo reads the geometric dimensions of each component in the BIM model, matches the corresponding material density from a pre-set material library based on the component type, and calculates the actual material usage based on the construction loss rate parameter. This highly automated process ensures efficient and accurate data extraction. This method achieves dynamic optimization of construction carbon emissions through the linkage of Revit and AI, which has significant advantages. First, it optimizes material transportation routes through AI prediction models, effectively reducing carbon emissions and transportation costs; second, it combines iterative optimization with real-time construction data feedback to ensure that the optimization effect is sustained and effective; finally, through intuitive heat map display and early warning mechanism, construction teams can easily identify and adjust high-emission areas to achieve more environmentally friendly and efficient construction. In addition, this method improves the accuracy of carbon emission calculations and provides a scientific basis for construction decision-making.

[0007] Preferably, the historical data preprocessing in step 2 includes missing value filling, outlier detection and time series alignment operations, wherein missing value filling adopts linear interpolation, outlier detection is based on isolation forest algorithm, and time series alignment is achieved by dynamic time warping; For historical data preprocessing, linear interpolation is used to fill missing values ​​because it is computationally simple and can effectively maintain data trends. The isolation forest algorithm detects outliers by constructing a random forest to isolate abnormal points. Parameter settings require adjusting the sampling size based on the data distribution. Dynamic time warping achieves time series alignment, resolving the problem of uneven time series lengths through local optimal matching. After preprocessing, data quality needs to be evaluated, using cross-validation to check interpolation effects, secondary outlier screening, and time series alignment error analysis. The process is automated through Python scripts to ensure repeatability and stability across datasets. This preprocessing solution significantly improves data availability: linear interpolation repairs missing fragments while retaining time series characteristics, isolation forest accurately identifies noise data to avoid model bias, dynamic time warping solves the problem of aligning multi-source heterogeneous data, and triple processing forms a closed-loop quality control, so that subsequent carbon emission calculations, transportation route optimization and other links can be carried out based on clean data. Automated scripts reduce manual operation risks and improve processing efficiency. Parameter adjustability enhances the universality of the method. The final generated construction instruction manual has greater decision-making value due to its solid data foundation, and the dynamic early warning mechanism responds more accurately due to noise filtering, forming a positive cycle of "data cleaning, model optimization and solution implementation", providing reliable technical support for building carbon emission reduction.

[0008] Preferably, the yard rules in step 3 include minimum transportation distance constraints, high-risk area avoidance rules, and transportation vehicle capacity restrictions; In Pathfinder, the Dijkstra algorithm is used to calculate the shortest path, integrating three constraints: minimum transport distance constraint: using edge weights to represent distance, the algorithm automatically selects the path with the smallest total weight; high-risk area avoidance: marking high-risk nodes in the software, and the algorithm automatically avoids related nodes and connecting edges; and transport capacity constraint: setting an upper limit on the total weight of the path, and the algorithm only returns paths that meet the capacity constraint. The triple constraints of the yard rules significantly optimize the logistics simulation process: the shortest path algorithm reduces transportation time and carbon emissions; avoids high-risk areas (such as geologically unstable areas) to reduce construction safety hazards; and capacity limit constraints avoid overloading, reduce transportation frequency and tool wear. This design achieves a multi-objective balance among transportation efficiency, safety, and cost through the linkage of algorithms and rules, providing basic support for the dynamic optimization of carbon emissions in construction.

[0009] Preferably, the dual-objective optimization function in step 4 adopts a weighted summation method, and the weight coefficient is dynamically adjusted according to the construction priority; The dual-objective optimization function adopts a weighted summation method combined with a mechanism for dynamically adjusting weight coefficients, which has significant advantages in the dynamic optimization of carbon emissions in construction. First, this method transforms multi-objective problems into single-objective problems, making it easier to solve them using traditional algorithms and improving construction efficiency. Second, by dynamically adjusting the weight coefficients, the optimization function can be more closely aligned with actual construction needs, balancing transportation efficiency and carbon emission control, and helping to reduce carbon emissions during construction. In addition, the dynamic adjustment mechanism of the weight coefficients enables the construction team to adjust according to changing project needs and priorities, providing flexible guidance for construction management. Finally, although weight allocation may be subjective, the use of intelligent optimization methods to automatically adjust the weight coefficients can make the weight allocation more reasonable and enhance the scientific nature of decision-making. This method is simple, easy to implement, and has wide applicability, providing an effective strategy for optimizing carbon emissions in construction.

[0010] Preferably, the threshold in step 5 is determined by a Monte Carlo simulation method, taking into account the uncertainty of the construction progress and weather fluctuation factors; The Monte Carlo simulation method is an uncertainty calculation method based on random sampling. Its basic principle is to simulate actual problems by generating a large number of random samples, and use statistical methods to analyze these samples to obtain an estimate of the solution or result of the problem. In the determination of carbon emission thresholds in construction, the specific implementation method is as follows: Define input variables: clarify the key variables that affect carbon emissions, such as construction progress, weather conditions, etc., and determine the probability distribution type (such as normal distribution, uniform distribution, etc.) and parameters of these variables; Generate random samples: Use a computer to generate random samples that conform to the probability distribution of each variable, and simulate carbon emissions under different construction progress and weather conditions; Calculate carbon emissions: For each random sample, calculate the corresponding carbon emissions, and determine whether the carbon emissions exceed the standard based on the preset threshold; Statistical analysis results: Perform statistical analysis on a large number of simulation results (such as tens of thousands of times), calculate the probability of carbon emissions exceeding the standard, and determine a reasonable carbon emission threshold based on the probability distribution; By considering the uncertainty of construction progress and weather fluctuations, the Monte Carlo simulation method can more accurately predict carbon emissions under different conditions, thereby determining a more reasonable carbon emission threshold; the Monte Carlo simulation method is based on statistical analysis of a large number of random samples, which can reduce the uncertainty of a single prediction result and improve the reliability and replicability of the method; by simulating carbon emissions under different construction progress and weather conditions, the Monte Carlo simulation method can provide support for decisions such as construction plan adjustments and the formulation of carbon emission control measures; the Monte Carlo simulation method can flexibly handle various uncertainties and adjust simulation parameters and model structures according to actual conditions to adapt to different construction scenarios and needs.

[0011] Preferably, the heat map in step 6 adopts red, yellow and green gradient color scales, and a second-level warning is triggered when the carbon emission intensity in the red area exceeds the baseline value by 20%; The heat map visually displays carbon emission intensity through a gradient of red, yellow, and green. Red represents high-emission areas where carbon emission intensity exceeds the baseline value by 20%, yellow represents medium-emission areas, and green represents low-emission areas. When a red area appears, the system automatically triggers a secondary warning and highlights the problem area in the Revit model. Detailed warning information and optimization suggestions are also provided, allowing the construction team to quickly locate the problem and take countermeasures. The heat map, with its intuitive color gradation display, enables the construction team to quickly understand the carbon emissions profile of different areas, enabling more precise planning and adjustment of construction plans. An early warning mechanism promptly alerts the team when carbon emissions exceed standards, facilitating the formulation and implementation of environmentally friendly construction decisions. This not only helps reduce overall carbon emissions and enhance the project's environmental image, but also improves construction efficiency and reduces unnecessary resource waste.

[0012] Preferably, the construction instruction manual in step seven includes a three-dimensional view of the path, carbon emission numerical annotations, and optimization suggestion texts; In addition to 3D path annotations, the construction manual integrates a QR code for the bill of materials, a Gantt chart of transportation schedules, and safety warning signs. Path annotations use gradient arrows superimposed with satellite terrain, and the accuracy of turning point radiuses is improved. Carbon emission data is embedded in the path view as a heat map, and numerical annotations include real-time dynamic intervals. The manual is automatically generated in PDF format via a Revit plug-in, supporting layer switches for on-site adjustments. Users can modify path parameters directly in the Revit interface, triggering manual updates to ensure that construction operations are synchronized with the design plan. The three-dimensional path view makes complex construction routes intuitive and traceable, reducing cognitive errors among construction workers; the carbon emission heat map helps identify high-emission sections on site in real time, and mechanical configuration can be quickly adjusted in combination with the optimization suggestion text; the bill of materials QR code realizes digital material traceability, the transportation Gantt chart is accurate to hourly scheduling, the automated generation mechanism reduces the workload of manual drawing, and the layer control function adapts to the needs of different construction scenarios. Most importantly, Revit's native modification synchronization function ensures seamless connection between manual and design changes, avoiding information transmission gaps.

[0013] Preferably, the scenario analysis module in step eight supports parameter sensitivity analysis, which can quantify the marginal benefits of different optimization strategies on carbon emissions; Parameter sensitivity analysis can be implemented using analytical, numerical, and experimental methods. Analytical methods are based on mathematical derivation and are applicable to simple linear systems. Numerical methods use numerical simulations to solve problems and are applicable to complex and nonlinear systems. Experimental methods obtain data through field measurements and are highly reliable but costly. Quantifying marginal benefits involves establishing carbon emission and cost calculation models, combined with scenario and sensitivity analysis, to evaluate the actual effects of different optimization strategies on carbon emissions. Parameter sensitivity analysis and quantified marginal benefits are of significant value in carbon emission management in construction. Through parameter sensitivity analysis, users can identify parameters that have a significant impact on carbon emissions, providing a scientific basis for formulating emission reduction strategies. Quantified marginal benefits enable users to simulate the effects of different optimization strategies and select the optimal emission reduction plan. These functions help improve the scientific nature of construction decision-making and the effectiveness of carbon emission management.

[0014] The system for dynamic optimization of carbon emissions from construction based on the linkage of Revit and AI adopts the above-mentioned dynamic optimization method for carbon emissions from construction based on the linkage of Revit and AI, including: Data integration module: The output of the data integration module is connected to the input of the Azure AutoML platform to transmit component material usage and carbon emission baseline values; Model training module: deployed on the Azure cloud platform. The output of the model training module is connected to the input of the logistics simulation module to transmit predicted carbon emission data; Logistics simulation module: running Pathfinder software, the output of the logistics simulation module is connected to the input of the collaborative optimization module to transmit transportation routes and carbon emission data; Collaborative optimization module: a built-in dual-objective optimization algorithm, the output ends of which are connected to the dynamic early warning module and the iterative optimization module respectively; Dynamic early warning module: interacts with the BIM model through the Revit API, and the output end of the dynamic early warning module is connected to the heat map generation module; Heat map generation module: generates a visualization layer based on the optimized data, and the output end of the heat map generation module is embedded in the Revit display interface; Path integration module: superimposes the simulation path with the BIM model, and the output end of the path integration module is connected to the construction instruction manual generation module; Iterative optimization module: connects to the model training module through a feedback loop to achieve online parameter updates; Multi-scenario analysis module: The multi-scenario analysis module is connected in parallel with the model training module to support parameterized simulation analysis; User interaction interface: integrates the operation entrances of each module, supports parameter setting and result display; Among them, each module realizes data flow through a standardized API interface to form a closed-loop optimization system.

[0015] In summary, compared with the existing technology, the present invention provides a dynamic optimization method and system for construction carbon emissions based on Revit and AI linkage, which has the following beneficial effects: This invention uses a time series regression model trained on the Azure AutoML platform to dynamically capture real-time variables such as construction progress and weather changes, reducing prediction errors compared to traditional static models and providing a decision-making basis for precise emission reduction. Integrating the Revit Dynamo plug-in, the code-free AI platform, and Pathfinder simulation software allows construction companies to achieve full-process automation management without a programming team, reducing deployment costs and significantly improving accessibility. Innovatively introduce a dual-objective function in logistics route planning to simultaneously optimize transportation distance and interference with carbon emission-sensitive areas, achieving dual improvements in construction efficiency and environmental benefits; From BIM model data extraction, AI prediction and early warning, to integrated path guidance, a closed-loop system of "monitoring, early warning, optimization, and feedback" is formed, allowing carbon emissions management to run throughout the entire construction lifecycle, with improved dynamic adjustment capabilities compared to existing solutions. The generated carbon emission heat map and three-dimensional path annotation manual transform complex data into intuitive decision-making basis, improve management efficiency, and provide revolutionary technical support for green construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a step-by-step diagram of the invention's method for dynamically optimizing carbon emissions in construction based on the linkage between Revit and AI.

[0017] Figure 2 This is a system diagram of the invention's dynamic optimization of carbon emissions in construction based on the linkage between Revit and AI. DETAILED DESCRIPTION

[0018] This invention provides a technical solution, a dynamic optimization method for carbon emissions from construction based on the linkage between Revit and AI, please refer to Figure 1 , including the following steps: Step 1: Data integration: The Dynamo plug-in for Revit automatically extracts component material usage and construction phase schedules from the BIM model, and uses Excel to match the public carbon emission factor library to calculate the carbon emission baseline value; Step 2: Model training: Import historical data into the Microsoft Azure AutoML platform for preprocessing and training regression models to predict future carbon emissions. The historical data includes date, construction stage, material usage, weather, and actual carbon emissions. Step 3: Logistics Simulation: Import the Revit site model into Pathfinder software, set yard rules, and simulate to generate the optimal material transportation path; Step 4: Collaborative optimization: Based on the logistics simulation results, a dual-objective optimization function was set up in the Azure AutoML platform to simultaneously minimize transportation distance and interference in carbon emission-sensitive areas; Step 5: Dynamic warning: When the predicted carbon emissions exceed the threshold, an early warning mark is automatically triggered in the Revit model, and adjustments to the construction plan are recommended; Step 6: Heat map generation: Based on the optimized carbon emission data, a carbon emission heat map is generated in the Revit model to visually display the emission intensity of different areas; Step 7: Path integration: Embed the optimal transportation route generated by Pathfinder into Revit drawings to form a construction instruction manual with route annotations; Step 8: Iterative optimization: Dynamically update Azure AutoML model parameters based on real-time construction data feedback for continuous optimization; Step 9: Multi-scenario analysis: By adjusting the construction schedule and weather parameters, we simulate carbon emission changes under different scenarios and generate comparative analysis reports; Carbon Emission Factor Library: Internationally recognized databases (such as IPCC reports and ISO standards) are used and regularly updated to ensure data accuracy and timeliness. Regression Model Selection: A random forest regression model is selected, and metrics such as R² and MSE are used to evaluate model performance and ensure prediction accuracy. Yard Rule Implementation: Yard rules are set using Pathfinder software and adjusted based on actual construction site conditions to improve simulation accuracy. Dual-Objective Optimization Function: A weighted summation method is used to dynamically adjust weight coefficients to balance transportation distance and interference from carbon emission-sensitive areas. Warning Marker Implementation: Warning areas are marked in red in the Revit model, and the Revit API is used to automatically trigger and update warning markers. Heat Map Gradient Color Scale: A heat map is generated using a gradient of red, yellow, and green to visually display the carbon emission intensity of different areas. This method is more complete in terms of carbon emission calculation, model prediction, logistics simulation, collaborative optimization, dynamic early warning and visualization. It adopts an internationally recognized carbon emission factor library and updates it regularly, which improves the accuracy of carbon emission calculation; selects a random forest regression model and uses R² and MSE to evaluate performance, which improves prediction accuracy; sets and adjusts yard rules through Pathfinder software, which enhances simulation accuracy; uses the weighted summation method to dynamically adjust weight coefficients, balancing transportation distance and carbon emissions; uses red mark early warning in the Revit model and automatically triggers it through the API, which improves the practicality of the method; uses red, yellow and green gradient color levels to generate heat maps, which enhances the visualization effect. These improvements will help improve the efficiency and effectiveness of dynamic optimization of carbon emissions in construction and promote the widespread application and promotion of the method.

[0019] See also Figure 1 ,The component material usage data extracted in step 1 include geometric dimensions, material density and construction loss rate parameters; Component material usage data is automatically extracted using Revit's Dynamo plug-in. The specific process is as follows: Dynamo reads the geometric dimensions of each component in the BIM model, matches the corresponding material density from a pre-set material library based on the component type, and calculates the actual material usage based on the construction loss rate parameter. This highly automated process ensures efficient and accurate data extraction. This method achieves dynamic optimization of construction carbon emissions through the linkage of Revit and AI, which has significant advantages. First, it optimizes material transportation routes through AI prediction models, effectively reducing carbon emissions and transportation costs; second, it combines iterative optimization with real-time construction data feedback to ensure that the optimization effect is sustained and effective; finally, through intuitive heat map display and early warning mechanism, construction teams can easily identify and adjust high-emission areas to achieve more environmentally friendly and efficient construction. In addition, this method improves the accuracy of carbon emission calculations and provides a scientific basis for construction decision-making.

[0020] See also Figure 1 ,The historical data preprocessing in step 2 includes missing value filling, outlier detection and time series alignment operations, where missing value filling adopts linear interpolation, outlier detection is based on the isolation forest algorithm, and time series alignment is ,achieved through dynamic time warping; For historical data preprocessing, linear interpolation is used to fill missing values ​​because it is computationally simple and can effectively maintain data trends. The isolation forest algorithm detects outliers by constructing a random forest to isolate abnormal points. Parameter settings require adjusting the sampling size based on the data distribution. Dynamic time warping achieves time series alignment, resolving the problem of uneven time series lengths through local optimal matching. After preprocessing, data quality needs to be evaluated, using cross-validation to check interpolation effects, secondary outlier screening, and time series alignment error analysis. The process is automated through Python scripts to ensure repeatability and stability across datasets. This preprocessing solution significantly improves data availability: linear interpolation repairs missing fragments while retaining time series characteristics, isolation forest accurately identifies noise data to avoid model bias, dynamic time warping solves the problem of aligning multi-source heterogeneous data, and triple processing forms a closed-loop quality control, so that subsequent carbon emission calculations, transportation route optimization and other links can be carried out based on clean data. Automated scripts reduce manual operation risks and improve processing efficiency. Parameter adjustability enhances the universality of the method. The final generated construction instruction manual has greater decision-making value due to its solid data foundation, and the dynamic early warning mechanism responds more accurately due to noise filtering, forming a positive cycle of "data cleaning, model optimization and solution implementation", providing reliable technical support for building carbon emission reduction.

[0021] See also Figure 1 ,The yard rules in step three include minimum transportation distance constraints, ,high-risk area avoidance rules, and transportation vehicle capacity ,limitations; In Pathfinder, the Dijkstra algorithm is used to calculate the shortest path, integrating three constraints: minimum transport distance constraint: using edge weights to represent distance, the algorithm automatically selects the path with the smallest total weight; high-risk area avoidance: marking high-risk nodes in the software, and the algorithm automatically avoids related nodes and connecting edges; and transport capacity constraint: setting an upper limit on the total weight of the path, and the algorithm only returns paths that meet the capacity constraint. The triple constraints of the yard rules significantly optimize the logistics simulation process: the shortest path algorithm reduces transportation time and carbon emissions; avoids high-risk areas (such as geologically unstable areas) to reduce construction safety hazards; and capacity limit constraints avoid overloading, reduce transportation frequency and tool wear. This design achieves a multi-objective balance among transportation efficiency, safety, and cost through the linkage of algorithms and rules, providing basic support for the dynamic optimization of carbon emissions in construction.

[0022] See also Figure 1 ,The dual-objective optimization function in step 4 adopts the weighted sum method, and the weight coefficient is dynamically adjusted according to the ,construction priority; The dual-objective optimization function adopts a weighted summation method combined with a mechanism for dynamically adjusting weight coefficients, which has significant advantages in the dynamic optimization of carbon emissions in construction. First, this method transforms multi-objective problems into single-objective problems, making it easier to solve them using traditional algorithms and improving construction efficiency. Second, by dynamically adjusting the weight coefficients, the optimization function can be more closely aligned with actual construction needs, balancing transportation efficiency and carbon emission control, and helping to reduce carbon emissions during construction. In addition, the dynamic adjustment mechanism of the weight coefficients enables the construction team to adjust according to changing project needs and priorities, providing flexible guidance for construction management. Finally, although weight allocation may be subjective, the use of intelligent optimization methods to automatically adjust the weight coefficients can make the weight allocation more reasonable and enhance the scientific nature of decision-making. This method is simple, easy to implement, and has wide applicability, providing an effective strategy for optimizing carbon emissions in construction.

[0023] See also Figure 1 ,The threshold in step five is determined by Monte Carlo simulation method, ,taking into account the uncertainty of construction progress and weather ,fluctuations; The Monte Carlo simulation method is an uncertainty calculation method based on random sampling. Its basic principle is to simulate actual problems by generating a large number of random samples, and use statistical methods to analyze these samples to obtain an estimate of the solution or result of the problem. In the determination of carbon emission thresholds in construction, the specific implementation method is as follows: Define input variables: clarify the key variables that affect carbon emissions, such as construction progress, weather conditions, etc., and determine the probability distribution type (such as normal distribution, uniform distribution, etc.) and parameters of these variables; Generate random samples: Use a computer to generate random samples that conform to the probability distribution of each variable, and simulate carbon emissions under different construction progress and weather conditions; Calculate carbon emissions: For each random sample, calculate the corresponding carbon emissions, and determine whether the carbon emissions exceed the standard based on the preset threshold; Statistical analysis results: Perform statistical analysis on a large number of simulation results (such as tens of thousands of times), calculate the probability of carbon emissions exceeding the standard, and determine a reasonable carbon emission threshold based on the probability distribution; By considering the uncertainty of construction progress and weather fluctuations, the Monte Carlo simulation method can more accurately predict carbon emissions under different conditions, thereby determining a more reasonable carbon emission threshold; the Monte Carlo simulation method is based on statistical analysis of a large number of random samples, which can reduce the uncertainty of a single prediction result and improve the reliability and replicability of the method; by simulating carbon emissions under different construction progress and weather conditions, the Monte Carlo simulation method can provide support for decisions such as construction plan adjustments and the formulation of carbon emission control measures; the Monte Carlo simulation method can flexibly handle various uncertainties and adjust simulation parameters and model structures according to actual conditions to adapt to different construction scenarios and needs.

[0024] See also Figure 1 ,The heat map in step six uses red, yellow and green gradients. ,When the carbon emission intensity in the red area exceeds the baseline value by 20%, ,a secondary warning is triggered; The heat map visually displays carbon emission intensity through a gradient of red, yellow, and green. Red represents high-emission areas where carbon emission intensity exceeds the baseline value by 20%, yellow represents medium-emission areas, and green represents low-emission areas. When a red area appears, the system automatically triggers a secondary warning and highlights the problem area in the Revit model. Detailed warning information and optimization suggestions are also provided, allowing the construction team to quickly locate the problem and take countermeasures. The heat map, with its intuitive color gradation display, enables the construction team to quickly understand the carbon emissions profile of different areas, enabling more precise planning and adjustment of construction plans. An early warning mechanism promptly alerts the team when carbon emissions exceed standards, facilitating the formulation and implementation of environmentally friendly construction decisions. This not only helps reduce overall carbon emissions and enhance the project's environmental image, but also improves construction efficiency and reduces unnecessary resource waste.

[0025] See also Figure 1 ,The construction instruction manual in step seven includes a three-dimensional ,view of the path, carbon emission numerical annotations, and ,optimization suggestion text; In addition to 3D path annotations, the construction manual integrates a QR code for the bill of materials, a Gantt chart of transportation schedules, and safety warning signs. Path annotations use gradient arrows superimposed with satellite terrain, and the accuracy of turning point radiuses is improved. Carbon emission data is embedded in the path view as a heat map, and numerical annotations include real-time dynamic intervals. The manual is automatically generated in PDF format via a Revit plug-in, supporting layer switches for on-site adjustments. Users can modify path parameters directly in the Revit interface, triggering manual updates to ensure that construction operations are synchronized with the design plan. The three-dimensional path view makes complex construction routes intuitive and traceable, reducing cognitive errors among construction workers; the carbon emission heat map helps identify high-emission sections on site in real time, and mechanical configuration can be quickly adjusted in combination with the optimization suggestion text; the bill of materials QR code realizes digital material traceability, the transportation Gantt chart is accurate to hourly scheduling, the automated generation mechanism reduces the workload of manual drawing, and the layer control function adapts to the needs of different construction scenarios. Most importantly, Revit's native modification synchronization function ensures seamless connection between manual and design changes, avoiding information transmission gaps.

[0026] See also Figure 1 ,The scenario analysis module in step eight supports parameter sensitivity analysis, which can quantify the marginal benefits of different optimization ,strategies on carbon emissions; Parameter sensitivity analysis can be implemented using analytical, numerical, and experimental methods. Analytical methods are based on mathematical derivation and are applicable to simple linear systems. Numerical methods use numerical simulations to solve problems and are applicable to complex and nonlinear systems. Experimental methods obtain data through field measurements and are highly reliable but costly. Quantifying marginal benefits involves establishing carbon emission and cost calculation models, combined with scenario and sensitivity analysis, to evaluate the actual effects of different optimization strategies on carbon emissions. Parameter sensitivity analysis and quantified marginal benefits are of significant value in carbon emission management in construction. Through parameter sensitivity analysis, users can identify parameters that have a significant impact on carbon emissions, providing a scientific basis for formulating emission reduction strategies. Quantified marginal benefits enable users to simulate the effects of different optimization strategies and select the optimal emission reduction plan. These functions help improve the scientific nature of construction decision-making and the effectiveness of carbon emission management.

[0027] The system for dynamic optimization of carbon emissions from construction based on the linkage of Revit and AI adopts the above-mentioned dynamic optimization method for carbon emissions from construction based on the linkage of Revit and AI. Figure 2 ,include: Data integration module: The output of the data integration module is connected to the input of the Azure AutoML platform to transmit component material usage and carbon emission baseline values; Model training module: Deployed on the Azure cloud platform, the output of the model training module is connected to the input of the logistics simulation module to transmit predicted carbon emission data; Logistics simulation module: Run the Pathfinder software. The output of the logistics simulation module is connected to the input of the collaborative optimization module to transmit transportation routes and carbon emission data. Collaborative optimization module: It has a built-in dual-objective optimization algorithm. The output of the collaborative optimization module is connected to the dynamic early warning module and the iterative optimization module respectively. Dynamic warning module: interacts with the BIM model through the Revit API, and the output of the dynamic warning module is connected to the heat map generation module; Heat map generation module: Generates visualization layers based on optimized data. The output of the heat map generation module is embedded in the Revit display interface. Path integration module: superimposes the simulation path with the BIM model. The output of the path integration module is connected to the construction instruction manual generation module. Iterative optimization module: connects to the model training module through a feedback loop to achieve online parameter updates; Multi-scenario analysis module: The multi-scenario analysis module is connected in parallel with the model training module to support parameterized simulation analysis; User interaction interface: integrates the operation entrances of each module, supports parameter setting and result display; Among them, each module realizes data flow through a standardized API interface to form a closed-loop optimization system.

[0028] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0029] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic optimization method for carbon emissions from construction based on the linkage between Revit and AI, characterized by: The steps include: Step 1: Data integration: The Dynamo plug-in for Revit automatically extracts component material usage and construction phase schedules from the BIM model, and uses Excel to match the public carbon emission factor library to calculate the carbon emission baseline value; Step 2: Model training: Import historical data into the Microsoft Azure AutoML platform for historical data preprocessing and training regression models to predict future carbon emissions; Step 3: Logistics Simulation: Import the Revit site model into Pathfinder software, set yard rules, and simulate to generate the optimal material transportation path; Step 4: Collaborative optimization: Based on the logistics simulation results, a dual-objective optimization function was set up in the Azure AutoML platform to simultaneously minimize transportation distance and interference in carbon emission-sensitive areas; Step 5: Dynamic warning: When the predicted carbon emissions exceed the threshold, an early warning mark is automatically triggered in the Revit model, and adjustments to the construction plan are recommended; Step 6: Heat map generation: Based on the optimized carbon emission data, a carbon emission heat map is generated in the Revit model to visually display the emission intensity of different areas; Step 7: Path integration: Embed the optimal transportation route generated by Pathfinder into Revit drawings to form a construction instruction manual with route annotations; Step 8: Iterative optimization: Dynamically update Azure AutoML model parameters based on real-time construction data feedback for continuous optimization; Step 9: Multi-scenario analysis: By adjusting the construction progress and weather parameters, we simulate the changes in carbon emissions under different scenarios and generate a comparative analysis report.

2. The method for dynamic optimization of construction carbon emissions based on Revit and AI linkage according to claim 1 is characterized by: The component material usage data extracted in step 1 includes geometric dimensions, material density and construction loss rate parameters.

3. The method for dynamic optimization of construction carbon emissions based on Revit and AI linkage according to claim 1 is characterized by: The historical data preprocessing in step 2 includes missing value filling, outlier detection and time series alignment operations.

4. The method for dynamic optimization of construction carbon emissions based on Revit and AI linkage according to claim 1 is characterized by: The yard rules in step three include minimum transportation distance constraints, high-risk area avoidance rules, and transportation vehicle capacity restrictions.

5. The method for dynamic optimization of construction carbon emissions based on Revit and AI linkage according to claim 1 is characterized by: The dual-objective optimization function in step 4 adopts a weighted summation method, and the weight coefficient is dynamically adjusted according to the construction priority.

6. The method for dynamic optimization of construction carbon emissions based on Revit and AI linkage according to claim 1 is characterized by: The threshold in step five is determined by a Monte Carlo simulation method, taking into account the uncertainty of the construction progress and weather fluctuations.

7. The method for dynamic optimization of construction carbon emissions based on Revit and AI linkage according to claim 1 is characterized by: The heat map in step 6 uses red, yellow and green gradient color levels, and a second-level warning is triggered when the carbon emission intensity in the red area exceeds the baseline value by 20%.

8. The method for dynamic optimization of construction carbon emissions based on Revit and AI linkage according to claim 1 is characterized by: The construction instruction manual in step seven includes a three-dimensional view of the path, carbon emission numerical annotations, and optimization suggestion texts.

9. The method for dynamic optimization of construction carbon emissions based on Revit and AI linkage according to claim 1 is characterized by: The scenario analysis module in step eight supports parameter sensitivity analysis.

10. A system for dynamic optimization of carbon emissions from construction based on the linkage of Revit and AI, using the method for dynamic optimization of carbon emissions from construction based on the linkage of Revit and AI as described in any one of claims 1 to 9, characterized in that: include: Data integration module: The output of the data integration module is connected to the input of the Azure AutoML platform to transmit component material usage and carbon emission baseline values; Model training module: deployed on the Azure cloud platform. The output of the model training module is connected to the input of the logistics simulation module to transmit predicted carbon emission data; Logistics simulation module: running Pathfinder software, the output of the logistics simulation module is connected to the input of the collaborative optimization module to transmit transportation routes and carbon emission data; Collaborative optimization module: a built-in dual-objective optimization algorithm, the output ends of which are connected to the dynamic early warning module and the iterative optimization module respectively; Dynamic early warning module: interacts with the BIM model through the Revit API, and the output end of the dynamic early warning module is connected to the heat map generation module; Heat map generation module: generates a visualization layer based on the optimized data, and the output end of the heat map generation module is embedded in the Revit display interface; Path integration module: superimposes the simulation path with the BIM model, and the output end of the path integration module is connected to the construction instruction manual generation module; Iterative optimization module: connects the model training module through a feedback loop; Multi-scenario analysis module: The multi-scenario analysis module is connected in parallel with the model training module to support parameterized simulation analysis; User interaction interface: integrates the operation entrances of each module, supports parameter setting and result display.

Images