Intelligent engineering construction progress monitoring and optimizing method

Abstract

The invention relates to the technical field of engineering informatization, in particular to an intelligent engineering construction progress monitoring and optimizing method. According to the invention, by obtaining the building construction drawings, the real-time images, the historical construction data, the network information and the external environment data, comprehensive monitoring of the construction progress is realized. The method comprises the steps of determining building partitions, analyzing construction conditions through image recognition, establishing and correcting a construction progress prediction model, and finally generating a multi-objective optimized construction progress plan and a visual report and a dynamically optimized construction strategy. According to the method, various technologies are effectively integrated, intelligentization of construction progress management is realized, the level of engineering project management is improved, the construction period is shortened, the cost is reduced, and powerful technical support is provided for efficiency improvement and cost control of the building industry.

Description

Technical Field

[0001] This invention relates to the field of engineering information technology, and in particular to an intelligent method for monitoring and optimizing engineering construction progress. Background Technology

[0002] With the rapid development of the construction industry and the continuous expansion of project scale, monitoring and optimizing construction progress has become a crucial aspect of project management. Traditional construction progress management methods mainly rely on manual observation and experience-based judgment, which suffers from problems such as strong subjectivity, poor real-time performance, and low accuracy. In recent years, with the advancement of information technology, some new methods for monitoring and optimizing construction progress have gradually emerged.

[0003] Currently, more advanced construction progress monitoring methods typically employ electronic data acquisition systems, such as barcode scanning and RFID tags, to record various data at the construction site. These methods improve the efficiency and accuracy of data collection to some extent. However, these systems often only collect discrete, single-point data, making it difficult to comprehensively reflect the complexities of the construction site.

[0004] In terms of construction schedule optimization, existing technologies mainly focus on resource allocation and critical path analysis. Some researchers have proposed intelligent optimization methods based on genetic algorithms and ant colony optimization to solve project scheduling problems under resource constraints. These methods can theoretically achieve good optimization results, but in practical applications, they often face problems such as overly simplified models and difficulty in adapting to complex and changing construction environments.

[0005] Furthermore, most existing construction progress monitoring and optimization systems are fragmented, with the monitoring and optimization systems often operating independently. This makes it difficult for optimization decisions to respond promptly to actual on-site conditions, significantly reducing the effectiveness of optimization. At the same time, existing systems generally lack in-depth mining and utilization of historical data, hindering the accumulation and transfer of experience.

[0006] Therefore, there is an urgent need for a method that can comprehensively, in real-time, and accurately monitor construction progress and intelligently optimize based on the monitoring results. This method should be able to adapt to complex and ever-changing construction environments, achieve seamless integration of monitoring and optimization, and possess the ability to learn and continuously improve itself. Summary of the Invention

[0007] This invention addresses the aforementioned technical problems. It provides an intelligent method for monitoring and optimizing construction progress, integrating advanced technologies such as multi-source data acquisition, image recognition, machine learning, and multi-objective optimization to achieve comprehensive perception, accurate prediction, and intelligent optimization of construction progress.

[0008] This invention proposes an intelligent method for monitoring and optimizing engineering construction progress, including: The acquisition steps include: Obtain architectural construction drawings and building functional division information; Acquire real-time image data of the construction site; Obtain historical construction data and actual construction progress data; Acquire online information data and external environment data; The processing steps include: Based on the architectural construction drawings and architectural functional division information, the multiple zones of the building and the structural composition of each zone are determined; Based on the real-time image data, the current construction status of each construction area is analyzed using image recognition technology; Based on the historical construction data and actual construction progress data, a construction progress prediction model is established. Based on the online information data and external environment data, the construction progress prediction model is revised. Output steps, including: Generate a multi-objective optimized construction schedule; Output a visualized construction progress monitoring report; Generate dynamically optimized construction strategy recommendations.

[0009] Preferably, determining the multiple zones of the building and the structural composition of each zone specifically includes: The building is divided into multiple zones, each zone labeled as... ,in The integer is positive; determine the structural composition within each partition, and label the structural composition as... ,in A positive integer, representing the first... The number of structural components within each partition; For each structural component, a corresponding construction sequence is determined, and the construction sequence is marked as follows: ,in A positive integer, representing the first... The first partition The number of construction procedures for each structure; Each construction process is further subdivided into multiple process steps, and these process steps are marked as follows: ,in , A positive integer, representing the first... The first partition The first structure The number of steps in each process.

[0010] Preferably, the analysis of the current construction status of each construction area using image recognition technology specifically includes: Image recognition algorithms are used to identify the construction status of each construction area at the construction site; Calculate the nominal total labor, measured total labor, and cumulative total labor for each construction area; Calculate the actual number of days to complete the project based on the nominal total engineer, the measured total engineer, and the cumulative total engineer. By comparing the actual number of days to be completed with the planned number of days to determine whether the construction site is ahead of schedule, behind schedule, or in a normal schedule.

[0011] Preferably, the establishment of the construction progress prediction model specifically includes: Based on historical construction data, extract the time parameters and influencing factors for each process; Using the Markov prediction algorithm, process state transition equations are established; Based on the dependencies between processes, an overall construction progress prediction model is constructed; Improve the accuracy of prediction models through model validation and parameter tuning.

[0012] Preferably, the generation of the multi-objective optimized construction schedule specifically includes: Establish a multi-objective optimization model that includes the following objectives: Minimize the total project duration; Balance the progress of each process; To meet resource and technological constraints; Consider the dependencies between processes; The model is solved using a multi-objective optimization algorithm; Generate the Pareto optimal solution set; The final optimized construction schedule is selected based on the decision-maker's preferences.

[0013] As a preferred option, a dynamic optimization step is also included: Real-time comparison of actual construction progress with planned progress; Identify areas and processes with abnormal progress; Analyze the causes of time loss and establish a model of the relationship between time loss and influencing factors; Based on the time loss analysis results, the optimization strategy is dynamically adjusted; Update the construction schedule.

[0014] Preferably, the reasons for the analysis time loss specifically include: Calculate the baseline time and actual time consumed for each process step; Extract the time loss index of each process step; Construct a first coordinate system and plot the relationship between the time loss index and the reference time as a curve; Analyze the curve trend and identify key influencing factors; Establish a mathematical model of the relationship between time loss and various influencing factors.

[0015] Preferably, the output of the visualized construction progress monitoring report specifically includes: Construct a second coordinate system, with the horizontal axis representing the timestamp and the vertical axis representing the progress information; Plot the actual construction progress curve and the planned progress curve in the second coordinate system; Identify areas with abnormal progress; Generate a schedule deviation statistics report; Output the analysis results of key influencing factors.

[0016] As a preferred option, the step of establishing an optimization strategy library is also included: Identify the key factors affecting construction progress; For each key factor, develop multiple optional optimization strategies; Define the applicable conditions and expected results for each optimization strategy; Establish the correspondence between optimization strategies and factors affecting construction progress; Based on the real-time monitoring of the construction status, corresponding optimization strategies can be triggered automatically or manually.

[0017] As a preferred option, iterative optimization steps are also included: Based on the preliminary optimization results, an optimized sequence is generated; During the construction process, actual construction data was continuously collected; Construct a third coordinate system to compare the optimized predicted progress with the actual progress. Calculate the deviation between the predicted schedule and the actual schedule; If the deviation exceeds the preset threshold, a second optimization will be performed; Adjust and optimize parameters, and update optimization strategies; Repeat the above steps until the desired optimization effect is achieved.

[0018] The beneficial effects of this invention are mainly reflected in the following aspects: First, this invention achieves comprehensive perception of the construction site through multi-dimensional data collection and processing. In particular, by introducing image recognition technology, the system can automatically analyze the real-time situation of the construction site, greatly improving the comprehensiveness and real-time nature of data collection. This solves the problems of incomplete and untimely data collection in traditional methods.

[0019] Secondly, this invention establishes a construction progress prediction model based on historical and real-time data. By introducing Markov prediction algorithms and Bayesian networks, this model can more accurately predict the completion time of each process, providing a reliable basis for subsequent optimization decisions. This overcomes the shortcomings of existing prediction models that are too simple and have low prediction accuracy.

[0020] Furthermore, this invention proposes a multi-objective optimization model that simultaneously considers multiple objectives such as total project duration, resource balance, and technical constraints. By employing a multi-objective genetic algorithm, the system can obtain a set of Pareto optimal solutions, providing decision-makers with more choices. This solves the problem that existing optimization methods often only consider a single objective and struggle to balance multiple needs.

[0021] More importantly, this invention achieves closed-loop feedback between monitoring and optimization. The system can dynamically adjust optimization strategies based on real-time monitoring results and continuously improve optimization effectiveness through iterative optimization. This dynamic optimization mechanism greatly enhances the system's adaptability to complex and ever-changing construction environments, solving the problems of fragmented monitoring and optimization, and the difficulty in implementing optimization results in existing systems.

[0022] Finally, this invention establishes an optimization strategy base and a knowledge base, enabling the accumulation and transfer of experience. This not only improves the management efficiency of individual projects but also provides enterprises with the possibility of accumulating management experience and enhancing their core competitiveness over the long term.

[0023] In summary, the intelligent engineering construction progress monitoring and optimization method provided by this invention, through the organic combination and synergistic effect of various technical links, achieves comprehensive perception, accurate prediction, intelligent optimization and continuous improvement of construction progress management, and provides strong technical support for improving the level of engineering project management, shortening the construction period and reducing costs. Attached Figure Description

[0024] Figure 1 This is the overall method flow of the present invention; Figure 2 The detailed process of obtaining the present invention is as follows; Figure 3 The detailed process flow of the present invention is as follows; Figure 4 The detailed flow of the output steps of this invention; Figure 5 This invention provides a flowchart for analyzing the current situation of the construction area. Figure 6 The process for establishing a construction progress prediction model according to the present invention; Figure 7 This invention provides a process for generating multi-objective optimized construction schedule plans. Detailed Implementation

[0025] Please refer to the attached document. Figure 1-7 This invention provides an intelligent method for monitoring and optimizing engineering construction progress, which can effectively improve the efficiency and quality of engineering construction. The invention will be described in detail below with reference to specific embodiments.

[0026] The present invention provides an intelligent engineering construction progress monitoring and optimization method, which includes an acquisition step, a processing step, and an output step.

[0027] The first step in the acquisition process is to obtain the building construction drawings and functional layout information. This information forms the foundation of the entire method, including the building's floor plans, elevations, sections, and the division of various functional areas. For example, for a multi-story office building, this might include detailed information on different functional areas such as office areas, meeting rooms, and rest areas.

[0028] Secondly, this method acquires real-time image data of the construction site using cameras installed at the site. Preferably, these cameras are high-definition cameras with a resolution of at least 1080p to ensure image quality. The cameras are deployed to cover all key areas of the construction site, typically with 1-2 cameras per 100 square meters.

[0029] Furthermore, this method acquires historical construction data and actual construction progress data. Historical construction data can come from the company's past construction records of similar projects, including information such as the duration of each process and resource consumption. Actual construction progress data is obtained through daily reporting by on-site management personnel or from automated data acquisition systems to track the current project's progress.

[0030] Finally, this method also acquires online information data and external environmental data. Online information may include factors affecting construction, such as material price fluctuations and labor market changes. External environmental data mainly refers to environmental factors that may affect construction, such as weather forecasts and air quality. This data is typically obtained in real time from relevant websites or meteorological departments via API interfaces.

[0031] In this process, the method first determines the multiple zones of the building and the structural composition of each zone based on the architectural construction drawings and functional zoning information. This process is usually assisted by professional Building Information Modeling (BIM) software. For example, a 20-story office building may be divided into multiple zones such as foundation, main structure, exterior walls, and interior decoration, and each zone contains multiple structural components.

[0032] Next, this method analyzes the current construction status of each construction area based on real-time image data using image recognition technology. Deep learning computer vision algorithms, such as convolutional neural networks (CNNs), are employed to identify various elements at the construction site, such as workers, machinery, and material stacks. By analyzing the quantity and distribution of these elements, the construction status and progress of each area can be determined.

[0033] Then, this method establishes a construction progress prediction model based on historical construction data and actual construction progress data. Machine learning algorithms, such as Support Vector Machine (SVM) or Random Forest, are used to train a model capable of predicting the completion time of each construction process. The model's input includes factors such as process type, resource input, and weather conditions, and the output is the estimated completion time.

[0034] Finally, this method refines the construction progress prediction model based on online information data and external environmental data. For example, if continuous rainfall is forecast for the next week, the model will automatically adjust the estimated completion time of outdoor work procedures.

[0035] In the output step, this method first generates a multi-objective optimized construction schedule. A multi-objective genetic algorithm (MOGA) is used here, simultaneously considering multiple objectives such as shortest construction period, lowest cost, and optimal resource utilization. The mathematical model of the algorithm can be expressed as: , in, Let the objective function vector be... For the first One optimization objective, Let be the decision variable vector, representing the start time and resource allocation of each process.

[0036] Next, this method outputs a visualized construction progress monitoring report. This is typically in the form of a Gantt chart or network diagram, visually displaying the planned and actual progress of each process. For processes that are lagging behind schedule, the system will mark them in red and provide warning messages.

[0037] Finally, this method generates dynamically optimized construction strategy recommendations. Based on the current construction status and prediction results, the system provides a series of optimization suggestions, such as increasing manpower input for a certain process or adjusting the construction sequence of certain processes. These suggestions are presented in an easy-to-understand text format using Natural Language Generation (NLG) technology.

[0038] The method of this invention employs a multi-level coding system when determining the multiple zones of a building and the structural composition of each zone. This coding system can accurately locate each specific construction step in the building, providing a solid foundation for subsequent progress monitoring and optimization.

[0039] Specifically, this method first divides the building into multiple zones, each zone using a positive integer. Mark, where This represents the total number of zones. For example, for a mixed-use building, possible zones include: 1-foundation, 2-main structure, 3-exterior walls, 4-interior decoration, 5-mechanical and electrical installation, etc.

[0040] Within each partition, this method further determines the structural composition using positive integers. Mark, where Indicates the first The number of structural components within a zone. For example, for the main structural zone, structural components may include: 1-column, 2-beam, 3-floor slab, 4-staircase, etc.

[0041] For each structural component, this method determines the corresponding construction sequence, using positive integers. Mark, where Indicates the first The first partition The number of construction steps for a structure. Taking a column as an example, the construction steps may include: 1-reinforcement binding, 2-formwork installation, 3-concrete pouring, 4-curing, etc.

[0042] Finally, this method subdivides each construction process into multiple process steps, using positive integers. Mark, where Indicates the first The first partition The first structure The number of steps in a process. For example, the concrete pouring process may include the following steps: 1- Hopper placement, 2- Pumping concrete, 3- Vibration, 4- Smoothing.

[0043] Through this multi-level coding system, this method can precisely pinpoint each specific construction step in a building. For example, the code (2,1,3,2) might represent "the pumping concrete step in the column concrete pouring process of the main structural zoning section." This refined division greatly facilitates subsequent progress monitoring and optimization.

[0044] The method of this invention employs a series of advanced computer vision algorithms and data processing technologies when analyzing the current construction status of each construction area through image recognition technology.

[0045] First, this method utilizes image recognition algorithms to identify the construction status of each construction area at the construction site. Here, deep learning-based object detection algorithms are employed, such as Faster R-CNN or YOLO (You Only Look Once). These algorithms can accurately locate and classify various construction elements in images, such as workers, machinery, and material stacks. For example, the system can identify the number of workers in an image, their locations, and whether they are wearing safety helmets. For machinery, the system can identify the equipment type (e.g., excavators, cranes) and its operating status.

[0046] Next, this method calculates the nominal total work, measured total work, and cumulative total work for each construction area. The nominal total work refers to the amount of work that should be completed according to the construction plan; the measured total work refers to the amount of work actually completed; and the cumulative total work refers to the total amount of work done from the start of the project to the current point in time. These values ​​are calculated by analyzing image recognition results and combining them with pre-set work volume standards. For example, if the system identifies five workers painting walls in a certain area, the measured total work for that area can be calculated based on the standard work efficiency of each worker.

[0047] Then, this method calculates the actual number of days to completion based on the nominal total engineer, the measured total engineer, and the cumulative total engineer. The calculation formula is as follows: Actual completion days , The average daily workload can be determined using historical data or industry standards.

[0048] Finally, this method compares the actual number of days completed with the planned number of days completed to determine whether the construction site is ahead of schedule, behind schedule, or in a normal schedule state. The judgment criteria can be set as follows: If the actual completion time is more than 5% less than the planned completion time, it is considered to be ahead of schedule. If the actual completion time is more than 5% longer than the planned completion time, it is considered a delay in the construction period. Other situations are considered to be within the normal construction period; This judgment method takes into account a certain error range, avoiding the problem of frequently changing the judgment result due to small fluctuations.

[0049] Through the above steps, this method can accurately and in real time grasp the progress of each construction area, providing reliable data support for subsequent optimization decisions.

[0050] The method of this invention employs a series of advanced data analysis and machine learning techniques when establishing the construction progress prediction model. This prediction model is the core component of the entire method, capable of accurately predicting the completion time of each process, providing an important basis for optimizing the construction progress.

[0051] First, this method extracts time parameters and influencing factors for each process based on historical construction data. This historical data includes detailed construction records of similar past projects, covering information such as the actual duration, resource input, and environmental conditions of various processes. In a preferred embodiment of this invention, data mining techniques are used to extract valuable features from this raw data. For example, for the concrete pouring process, features that might be extracted include the pouring area, concrete strength grade, weather conditions, and the construction team's experience.

[0052] Next, this method utilizes a Markov prediction algorithm to establish the process state transition equation. The advantage of the Markov model lies in its ability to consider the dependencies and stochasticity between processes. In the embodiments of this invention, each process is considered a state, and the transitions between processes are considered state transitions. The state transition equation can be expressed as: , in, express The status of the process at any given moment. Indicates from state Transition to state The probability of transitions. These transition probabilities can be obtained by analyzing historical data.

[0053] Then, this method constructs an overall construction progress prediction model based on the dependencies between work processes. Preferably, a Bayesian network model is used here, which can effectively represent complex probabilistic dependencies. In a Bayesian network, each node represents a work process, and the edges between nodes represent the dependencies between work processes. By calculating the joint probability distribution, the completion time of the entire project can be predicted.

[0054] Finally, this method improves the accuracy of the prediction model through model validation and parameter tuning. Specifically, historical data can be divided into training and test sets. The model is built using the training set, and then its performance is validated on the test set. Commonly used performance metrics include mean absolute error (MAE) and root mean square error (RMSE). If the prediction error exceeds a preset threshold (e.g., 5% of the actual project duration), parameter tuning is required. Tuning methods can employ algorithms such as grid search or Bayesian optimization.

[0055] Through the above steps, the method of the present invention can establish an accurate and reliable construction progress prediction model, providing a scientific basis for subsequent optimization decisions.

[0056] The method of this invention employs complex mathematical models and advanced optimization algorithms when generating a multi-objective optimized construction schedule. The goal of this step is to find an optimal solution that balances multiple objectives while satisfying various constraints.

[0057] First, this method establishes an optimization model that includes multiple objectives. In one embodiment of the invention, the following four main objectives are considered: 1. Minimize total project duration: This is the most intuitive goal, which can be expressed as the total number of days from the start to the end of the project; 2. Balance the progress of each process: This goal aims to avoid a situation where some processes are progressing too quickly while others lag behind; 3. Meet resource and technical constraints: This includes limitations on human resources and equipment resources, as well as technical specifications. 4. Consider the dependencies between processes: Some processes can only begin after other processes have been completed.

[0058] Minimize satisfies , , , in, It is the objective function vector. arrive These correspond to the four objectives mentioned above. and Representing inequalities and equality constraints, and These are the upper and lower bounds of the decision variable.

[0059] Next, this method uses a multi-objective optimization algorithm to solve the model. Preferably, a multi-objective genetic algorithm (MOGA) is used. The advantage of genetic algorithms lies in their ability to effectively handle large-scale, nonlinear optimization problems. In MOGA, each individual represents a possible construction schedule, and new solutions are generated through operations such as crossover and mutation. Non-dominated sorting and crowding calculation are used to maintain the diversity of the population.

[0060] This method then generates a Pareto optimal solution set. A Pareto optimal solution is one that cannot improve any other objective without compromising the others. This solution set provides decision-makers with a series of alternative solutions.

[0061] Finally, based on the decision-maker's preferences, the final optimized construction schedule is selected. In one embodiment of the invention, the Analytic Hierarchy Process (AHP) can be used to quantify the decision-maker's preferences for each objective, and then the solution that best matches the preferences is selected as the final solution.

[0062] Using this method, the present invention can generate an optimized construction schedule that satisfies multiple objective requirements and conforms to actual constraints.

[0063] The method of the present invention also includes a dynamic optimization step, which enables the entire system to respond promptly to changes during the construction process, maintaining the effectiveness and feasibility of the plan.

[0064] First, this method compares the actual construction progress with the planned progress in real time. This requires a real-time data acquisition system, which can be implemented using Internet of Things (IoT) technology. For example, RFID tags can be used to track the usage of materials and equipment, and smart safety helmets can be used to collect data on worker location and working hours. After processing, this data can be used to obtain the actual progress of each process. In a preferred embodiment of the invention, a progress deviation threshold is set; if the deviation between the actual progress and the planned progress exceeds ±10%, subsequent optimization steps will be triggered.

[0065] Next, this method identifies regions and processes with schedule anomalies. Anomaly detection algorithms, such as Isolation Forest or Local Outlier Factor (LOF), can be used here. These algorithms can effectively identify data points that significantly differ from the normal pattern. In this method, schedule anomalies are not only considered for individual processes, but the correlations between processes are also analyzed to identify potential chain reactions.

[0066] Then, this method analyzes the causes of time loss and establishes a model of the relationship between time loss and influencing factors. This step employs data mining and machine learning techniques. First, various factors that may affect construction progress are collected, such as weather conditions, material supply, and personnel changes. Then, correlation analysis and feature importance assessment are used to identify the factors with the greatest impact on time loss. Finally, a regression model is established to quantify the impact of these factors on time loss. The model can be represented as: , in, It's a matter of time loss. These are the various influencing factors. These are the corresponding coefficients. This is the error term. Based on the time loss analysis results, this method dynamically adjusts the optimization strategy. A reinforcement learning algorithm is used here, specifically Q-learning or a deep Q-network (DQN). The system takes the current construction state as input, possible adjustment measures as the action space, and then finds the optimal adjustment strategy through repeated trials and learning.

[0067] Finally, this method updates the construction schedule. This includes not only adjusting the start and end times of each process, but may also involve resource reallocation and adjustments to the construction sequence. The updated schedule needs to meet the multi-objective optimization requirements mentioned earlier; therefore, the multi-objective optimization algorithm will be rerun here, but with the current situation as the new constraints.

[0068] Through this dynamic optimization mechanism, the method of the present invention can keep the construction schedule in an optimal state at all times and effectively cope with various uncertainties in the construction process.

[0069] The method of this invention employs a series of sophisticated data processing and analysis techniques when analyzing the causes of time loss. This step is crucial for understanding efficiency bottlenecks in the construction process and identifying areas for improvement.

[0070] First, this method calculates the baseline time and actual time consumed for each process step. The baseline time can be obtained from historical data or industry standards, representing the time required to complete the process step under ideal conditions. The actual time consumed is obtained through an on-site data acquisition system. In one embodiment of this invention, an automated time recording system based on computer vision is used, which can be accurate to the minute level.

[0071] Next, this method extracts the time loss index for each process step. The time loss index is defined as: , This indicator visually reflects the efficiency level of each process step. For example, if the TII of a certain step is 209%, it means that it took 20% longer than expected.

[0072] Then, this method constructs a first coordinate system, plotting the relationship between the time loss index and the baseline time as a curve. In this coordinate system, the horizontal axis represents the baseline time, and the vertical axis represents the time loss index. Each process step is represented as a point on the graph. This visualization method can intuitively show the efficiency of process steps with different complexities (represented by the baseline time).

[0073] Next, this method analyzes the curve trend to identify key influencing factors. Regression analysis methods, such as multinomial regression or spline regression, can be used here to fit a trend line. The shape of the trend line can reveal important information, such as: If the trend line is rising, it indicates that complex processes are more prone to time loss.

[0074] If the trend line is trending downwards, it may indicate that simpler process steps are actually less efficient and require close attention.

[0075] If the trend line is U-shaped, it indicates that the process steps with medium complexity are the most efficient.

[0076] In addition, this method will identify points that deviate significantly from the trend line, which may represent special cases or outliers that require further investigation.

[0077] Finally, this method establishes a mathematical model of the relationship between time loss and various influencing factors. Multiple linear regression or more complex machine learning models, such as random forests or support vector machine regression, can be used here. The general form of the model can be expressed as: , in, arrive These represent various possible influencing factors, such as weather conditions, worker skill level, and equipment condition. By analyzing the coefficients or importance scores of each factor in the model, it is possible to determine which factors have the greatest impact on time loss.

[0078] Through the above steps, the method of this invention can deeply analyze the time loss during the construction process, providing accurate data support and scientific decision-making basis for subsequent optimization. This refined analysis method helps improve construction efficiency and reduce unnecessary time waste.

[0079] The method of this invention employs a series of advanced data visualization technologies when outputting a visualized construction progress monitoring report. This visualized report can intuitively and clearly display the construction progress, providing decision support for project managers.

[0080] First, this method constructs a second coordinate system, where the horizontal axis represents the timestamp and the vertical axis represents progress information. This coordinate system provides the basic framework for subsequent data display. In a preferred embodiment of the invention, the timestamp can be accurate to the hour to meet the needs of refined management. The progress information can be expressed as a percentage, reflecting the completion status of each process.

[0081] Next, this method plots the actual construction progress curve and the planned progress curve in the second coordinate system. The comparison of these two curves can intuitively reflect the actual construction progress. Preferably, the actual progress curve can be represented by a solid line, and the planned progress curve by a dashed line for easy differentiation. Furthermore, different colors can be used to represent different work processes or construction areas to enhance the information content of the charts.

[0082] Then, this method identifies areas of abnormal progress in the chart. Here, "abnormal" refers to areas where there is a significant difference between actual and planned progress. In one embodiment of the invention, if the actual progress at a certain point in time is more than 20% behind or ahead of the planned progress, it will be marked as abnormal. These abnormal areas can be highlighted or marked with special symbols (such as exclamation marks) to attract the attention of managers.

[0083] Next, this method generates a schedule deviation statistics report. This report not only includes the overall schedule deviation but also breaks it down to the specific deviations for each process and construction area. For example, it can list the top 5 processes that are ahead of schedule and the top 5 processes that are behind schedule, helping managers quickly pinpoint the problems. Preferably, these statistics can be presented in the form of bar charts or radar charts to make the information more intuitive.

[0084] Finally, this method outputs the results of the key influencing factor analysis. This part is based on the time loss analysis mentioned earlier. The report can use pie charts to show the weight distribution of each influencing factor and line graphs to show the change of a factor (such as weather) over time and its impact on schedule. In addition, predictive analyses can be provided, such as which processes might be affected by the weather forecast for the coming week.

[0085] Through this multi-dimensional and highly visualized report, the method of this invention can help project managers quickly grasp the construction progress, identify and solve problems in a timely manner, and effectively improve project management efficiency.

[0086] The method of this invention also includes the step of establishing an optimization strategy library. This strategy library is the "brain" of the entire system and can provide corresponding optimization suggestions based on different construction conditions.

[0087] First, this method identifies key factors affecting construction progress. This step is based on the preceding time loss analysis. Preferably, statistical methods such as principal component analysis (PCA) or factor analysis can be used to extract the most critical factors from numerous possible influencing factors. In one embodiment of the invention, the identified key factors may include weather conditions, human resource allocation, equipment utilization rate, and the timeliness of material supply.

[0088] Next, this method develops several optional optimization strategies for each key factor. These strategies are based on industry best practices and expert experience. For example, for weather factors, possible strategies include: 1. Adjust outdoor work hours to avoid inclement weather; 2. Add temporary protective measures, such as building rain shelters; 3. Prioritize indoor work to improve overall work efficiency.

[0089] Possible strategies for human resource allocation include: 1. Increase the number of skilled workers; 2. Adjust working hours and implement a shift system; 3. Provide additional training to improve workers' skill levels.

[0090] Next, this method sets the applicable conditions and expected effects for each optimization strategy. This step is to ensure the strategy's relevance and effectiveness. The applicable conditions can be described using a series of parameters, such as: , , in, Indicates the first The applicable conditions for each strategy arrive These are the various parameters describing the construction status. The expected results can be expressed using quantitative indicators, such as: where, Indicates the first The expected effect of the strategy , , These represent the expected impacts on schedule, cost, and quality, respectively. Next, this method establishes the correspondence between optimization strategies and factors influencing construction schedule. This can be represented by a matrix: , in, Indicates the first The strategy for the first The intensity of the influence of each factor.

[0091] Finally, this method automatically or manually triggers corresponding optimization strategies based on real-time monitoring of the construction status. In automatic mode, the system matches the current situation with the applicable conditions of each strategy and selects the most suitable strategy. In manual mode, the system recommends several optional strategies to the project manager, who then makes the final decision.

[0092] By establishing such a dynamic and intelligent optimization strategy library, the method of this invention can provide timely and effective solutions to various problems encountered in the construction process, greatly improving the efficiency and quality of project management.

[0093] The method of this invention also includes an iterative optimization step. This step enables the entire system to continuously learn and improve, adapting to complex and ever-changing construction environments.

[0094] First, this method generates an optimization sequence based on preliminary optimization results. This optimization sequence contains a series of optimization measures ordered by priority. In one embodiment of the invention, the optimization sequence can be represented as: , in, Indicates the first The optimization measures include specific implementation details, expected results, and execution priorities.

[0095] Next, this method continuously collects actual construction data during the construction process. This requires a comprehensive data acquisition system, including but not limited to: Real-time images from the on-site camera; Workers' movement trajectory data; Equipment usage data; Material consumption data; Environmental monitoring data (such as temperature, humidity, PM2.5, etc.); Then, this method constructs a third coordinate system to compare the optimized predicted schedule with the actual schedule. This coordinate system is similar to the second coordinate system mentioned earlier, but it adds a predicted schedule curve. By comparing these three curves (planned schedule, predicted schedule, and actual schedule), the effectiveness of the optimization measures can be intuitively evaluated.

[0096] Next, this method calculates the deviation between the predicted and actual schedules. The root mean square error (RMSE) can be used as an evaluation metric here. , If the deviation exceeds a preset threshold, this method will perform secondary optimization. In a preferred embodiment of the invention, this threshold is set to 1096. That is, if the RMSE is greater than 0.1, secondary optimization will be triggered. In secondary optimization, this method will adjust the optimization parameters and update the optimization strategy. Here, a Bayesian optimization algorithm can be used, which can effectively find the optimal solution in a high-dimensional parameter space. The objective function of Bayesian optimization can be defined as: in, This is the actual progress. It is a projected progress. It represents the number of data points.

[0097] , in, The next step is to optimize the parameter vector. Finally, this method repeats the above steps until the desired optimization effect is achieved. This process can be viewed as a reinforcement learning process, where the system gradually improves its optimization performance through continuous trial and learning.

[0098] Preferably, this method also records the results of each optimization and the corresponding construction conditions, establishing a knowledge base. This knowledge base can be used for optimization in future projects, enabling the accumulation and transfer of experience.

[0099] Through this iterative optimization mechanism, the method of the present invention can continuously improve the optimization effect, adapt to various complex construction situations, and provide continuous and effective support for project management.

[0100] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent engineering construction progress monitoring and optimization, characterized in that, include: The acquisition steps include: Obtain architectural construction drawings and building functional division information; Acquire real-time image data of the construction site; Obtain historical construction data and actual construction progress data; Acquire online information data and external environment data; The processing steps include: Based on the architectural construction drawings and architectural functional division information, the multiple zones of the building and the structural composition of each zone are determined; Based on the real-time image data, the current construction status of each construction area is analyzed using image recognition technology; Based on the historical construction data and actual construction progress data, a construction progress prediction model is established. Based on the online information data and external environment data, the construction progress prediction model is revised. Output steps, including: Generate a multi-objective optimized construction schedule; Output a visualized construction progress monitoring report; Generate dynamically optimized construction strategy recommendations.

2. The method according to claim 1, characterized in that, The determination of the multiple zones of the building and the structural composition of each zone specifically includes: The building is divided into multiple zones, each zone labeled as... ,in The integer is positive; determine the structural composition within each partition, and label the structural composition as... ,in A positive integer, representing the first... The number of structural components within each partition; For each structural component, a corresponding construction sequence is determined, and the construction sequence is marked as follows: ,in A positive integer, representing the first... The first partition The number of construction procedures for each structure; Each construction process is further subdivided into multiple process steps, and these process steps are marked as follows: ,in , A positive integer, representing the first... The first partition The first structure The number of steps in each process.

3. The method according to claim 1, characterized in that, The analysis of the current construction status of each construction area using image recognition technology specifically includes: Image recognition algorithms are used to identify the construction status of each construction area at the construction site; Calculate the nominal total labor, measured total labor, and cumulative total labor for each construction area; Calculate the actual number of days to complete the project based on the nominal total engineer, the measured total engineer, and the cumulative total engineer. By comparing the actual number of days to be completed with the planned number of days to determine whether the construction site is ahead of schedule, behind schedule, or in a normal schedule.

4. The method according to claim 1, characterized in that, The establishment of the construction progress prediction model specifically includes: Based on historical construction data, extract the time parameters and influencing factors for each process; Using the Markov prediction algorithm, process state transition equations are established; Based on the dependencies between processes, an overall construction progress prediction model is constructed; Improve the accuracy of prediction models through model validation and parameter tuning.

5. The method according to claim 1, characterized in that, The generation of the multi-objective optimized construction schedule specifically includes: Establish a multi-objective optimization model that includes the following objectives: Minimize the total project duration; Balance the progress of each process; To meet resource and technological constraints; Consider the dependencies between processes; The model is solved using a multi-objective optimization algorithm; Generate the Pareto optimal solution set; The final optimized construction schedule is selected based on the decision-maker's preferences.

6. The method according to claim 1, characterized in that, It also includes dynamic optimization steps: Real-time comparison of actual construction progress with planned progress; Identify areas and processes with abnormal progress; Analyze the causes of time loss and establish a model of the relationship between time loss and influencing factors; Based on the time loss analysis results, the optimization strategy is dynamically adjusted; Update the construction schedule.

7. The method according to claim 6, characterized in that, The specific reasons for the time loss in the analysis include: Calculate the baseline time and actual time consumed for each process step; Extract the time loss index of each process step; Construct a first coordinate system and plot the relationship between the time loss index and the reference time as a curve; Analyze the curve trend and identify key influencing factors; Establish a mathematical model of the relationship between time loss and various influencing factors.

8. The method according to claim 1, characterized in that, The output of the visualized construction progress monitoring report specifically includes: Construct a second coordinate system, with the horizontal axis representing the timestamp and the vertical axis representing the progress information; Plot the actual construction progress curve and the planned progress curve in the second coordinate system; Identify areas with abnormal progress; Generate a schedule deviation statistics report; Output the analysis results of key influencing factors.

9. The method according to claim 1, characterized in that, It also includes the step of establishing an optimization strategy library: Identify the key factors affecting construction progress; For each key factor, develop multiple optional optimization strategies; Define the applicable conditions and expected results for each optimization strategy; Establish the correspondence between optimization strategies and factors affecting construction progress; Based on the real-time monitoring of the construction status, corresponding optimization strategies can be triggered automatically or manually.

10. The method according to claim 1, characterized in that, It also includes iterative optimization steps: Based on the preliminary optimization results, an optimized sequence is generated; During the construction process, actual construction data was continuously collected; Construct a third coordinate system to compare the optimized predicted progress with the actual progress. Calculate the deviation between the predicted schedule and the actual schedule; If the deviation exceeds the preset threshold, a second optimization will be performed; Adjust and optimize parameters, and update optimization strategies; Repeat the above steps until the desired optimization effect is achieved.

Images