Road and bridge construction progress real-time monitoring system

Abstract

The present application relates to the technical field of real-time monitoring, in particular to a real-time monitoring system for road and bridge construction progress, which comprises a dynamic relationship analysis module, an event response decision module, a structure health monitoring module, a risk management and prediction module, a progress prediction and adjustment module, a resource optimization configuration module, an environmental impact assessment module and an overall progress management module.In the present application, the dynamic relationship in construction is captured through the graph neural network, the graph attention network and the dynamic graph convolution network, the construction model is updated in real time, the accuracy of progress analysis is ensured, the event response decision module quickly optimizes the construction plan through the predictive control algorithm, the strategy flexibility is improved, the structure health monitoring and risk management module combines the Markov decision process and the Monte Carlo tree search, effectively prevents risks and ensures safety, the progress prediction, resource optimization and environmental impact assessment module improves resource efficiency and environmental adaptability, and the application of comprehensive evaluation method realizes efficient and high-quality construction management.

Description

Technical Field

[0001] This invention relates to the field of real-time monitoring technology, and in particular to a real-time monitoring system for road and bridge construction progress. Background Technology

[0002] Real-time monitoring technology primarily utilizes automation and information technology to track and manage the progress of road and bridge construction projects in real time. This technology encompasses the entire process of data acquisition, transmission, processing, and display, aiming to improve the efficiency and accuracy of construction management through real-time data analysis and feedback. Real-time monitoring technology enables project managers to obtain detailed information on project progress in a timely manner, allowing for rapid and accurate decision-making to ensure the project proceeds as planned while minimizing resource waste and delay risks.

[0003] The real-time monitoring system for road and bridge construction progress is a monitoring system specifically designed for road and bridge construction projects. Its purpose is to achieve real-time tracking and management of construction progress by collecting data from the construction site. This system aims to help project managers adjust construction plans and resource allocation in a timely manner, optimize construction processes, and thus improve the efficiency and quality of project construction by providing accurate construction progress information. By achieving real-time monitoring of construction progress, the system helps to promptly identify and resolve problems that arise during construction, ensuring that projects are completed on time and achieving a comprehensive effect of reducing costs, improving safety, and enhancing project quality.

[0004] Traditional road and bridge construction monitoring systems have shortcomings in real-time data processing, dynamic monitoring, risk management, and resource optimization. These systems rely on manual judgment and lack scientific data analysis capabilities, impacting construction efficiency and cost control. They fail to fully utilize real-time data, resulting in slow responses to monitoring and adjustments to construction progress. They are unable to respond promptly to unexpected events and changes during construction, increasing the risk of project delays and cost overruns. Structural health monitoring relies heavily on periodic inspections, lacking real-time early warnings and making it difficult to identify structural risks in a timely manner. Risk management lacks effective tools, and response measures are not precise enough, increasing construction uncertainty. Furthermore, environmental impact assessments often fail to adequately consider the environmental impact of construction, leading to environmental problems and public concern. Summary of the Invention

[0005] This application provides a real-time monitoring system for road and bridge construction progress, addressing the shortcomings of traditional road and bridge construction monitoring systems in areas such as real-time data processing, dynamic monitoring, risk management, and resource optimization. Traditional systems rely on manual judgment and lack scientific data analysis capabilities, leading to resource allocation deviating from optimal levels and impacting construction efficiency and cost control. Furthermore, the system fails to fully utilize real-time data, resulting in slow responses to monitoring and adjustments to construction progress and an inability to respond promptly to unexpected events and changes during construction, increasing the risk of project delays and cost overruns. Structural health monitoring relies heavily on periodic inspections, lacking real-time early warnings and making it difficult to identify structural risks in a timely manner. Risk management lacks effective tools, and response measures are not precise enough, increasing construction uncertainty. Finally, the environmental impact assessment does not adequately consider the environmental impact of construction, raising environmental issues and public concerns.

[0006] In view of the above problems, this application provides a real-time monitoring system for road and bridge construction progress.

[0007] This application provides a real-time monitoring system for road and bridge construction progress, wherein the system includes a dynamic relationship analysis module, an event response decision module, a structural health monitoring module, a risk management and prediction module, a progress prediction and adjustment module, a resource optimization allocation module, an environmental impact assessment module, and an overall progress management module; The dynamic relationship analysis module is based on real-time construction progress data and uses a graph neural network algorithm to capture the relationship between resource allocation, task priority and progress changes by constructing a dynamic graph representation of the construction project. It uses a graph attention network to enhance the influence analysis of key nodes and combines a dynamic graph convolutional network to dynamically update the graph structure according to time changes, thereby performing dynamic relationship modeling of construction progress and generating a dynamic relationship analysis graph. The event response decision module is based on a dynamic relationship analysis graph and adopts an event-driven strategy and model predictive control theory. For real-time events, including task completion and resource changes, it predicts and optimizes the construction plan for future time periods through rolling time domain optimization, uses an event triggering mechanism to determine the optimization timing, adjusts the construction strategy, and generates detailed construction plans. The structural health monitoring module is based on a detailed construction plan and uses a graph convolutional network algorithm to analyze structural vibration data. It captures the spatial relationship between structures through spectral convolution, analyzes time series data in combination with a spatiotemporal graph network model, monitors the health status of the structure at the construction site in real time, predicts and assesses potential structural damage, and generates structural health status results. The risk management and prediction module, based on the structural health status results, uses a continuous-time Markov decision process algorithm to dynamically calculate the state transition probability and combines it with a Monte Carlo tree search algorithm to perform random simulation identification and assessment of construction risks, formulate countermeasures, and generate risk management strategies. The progress prediction and adjustment module is based on risk management strategy, uses long short-term memory network algorithm to analyze the time series data of construction progress, capture long-term dependencies, and combines it with Gaussian process regression algorithm to simulate the randomness of prediction, optimize the construction plan according to the prediction results, and generate the adjusted construction plan. The resource optimization and allocation module, based on the adjusted construction plan, uses a mixed-integer linear programming algorithm to mathematically model and optimize the resource allocation problem. Combined with a genetic algorithm, it captures the optimal solution in the search space and optimizes resource allocation to generate a resource allocation scheme. The environmental impact assessment module is based on the resource allocation scheme and uses the kernel ridge regression algorithm to perform high-dimensional data analysis on the impact of changes in the construction environment, including weather conditions, on the construction progress. It is combined with the Gaussian process regression algorithm to generate environmental impact analysis results. The overall progress management module, based on the environmental impact analysis results, adopts a comprehensive evaluation method, and manages and adjusts the progress of the construction project through multi-indicator decision analysis and performance evaluation. It also formulates comprehensive management strategies and generates the final construction management plan.

[0008] Preferably, the dynamic relationship analysis diagram includes resource nodes, task nodes, schedule nodes, and relationship strength indicators between nodes; the detailed construction plan includes a resource allocation table, a priority task list, and a schedule adjustment strategy; the structural health status results include vibration frequency data, damage probability scores, and structural health indicators; the risk management strategy includes risk level classification, prediction model results, and an overview of response strategies; the adjusted construction plan includes activity priority adjustments, resource reallocation schemes, and a revised timetable; the resource allocation scheme includes a resource allocation diagram, cost-benefit analysis, and an optimized allocation list; the environmental impact analysis results include weather condition impact assessment, correlation analysis between environmental variables and construction progress, and quantification of impact degree; and the final construction management scheme includes a schedule management plan, resource management guidelines, and a risk response framework.

[0009] Preferably, the dynamic relationship analysis module includes a resource allocation submodule, a task priority submodule, and a progress change analysis submodule; The resource allocation submodule is based on real-time construction progress data. It adopts a graph neural network algorithm, uses the PyTorch framework to build a graph attention network model, defines the model structure, adds graph convolutional layers to capture the resource allocation relationship between nodes, uses the attention mechanism to analyze the influence of key resource nodes, and generates a dynamic resource influence graph. The task priority submodule is based on the resource dynamic influence graph, adopts the graph attention network algorithm, uses the PyTorch framework to build a dynamic graph convolutional network model, defines the model structure, adds dynamic graph convolutional layers to match the time sensitivity of tasks, applies the graph attention mechanism to strengthen task dependency analysis, and generates a task priority structure graph. The progress change analysis submodule is based on the task priority structure graph, adopts a dynamic graph convolutional network algorithm, uses the PyTorch framework for dynamic graph updates, defines an update mechanism, adjusts the graph structure according to timestamps, applies graph convolutional layers to analyze the impact of progress changes on resources and tasks, and generates a dynamic relationship analysis graph.

[0010] Preferably, the event response decision module includes an event capture submodule, a decision optimization submodule, and an emergency response submodule; The event capture submodule is based on a dynamic relationship analysis graph, adopts an event-driven strategy and model predictive control theory, uses Python to program the event capture logic, defines event capture rules, monitors task completion and resource change events, sets event trigger thresholds, applies rolling time domain optimization to predict event impact, and generates an event impact analysis graph. The decision optimization submodule is based on the event impact analysis diagram, adopts the rolling time domain optimization algorithm, uses Python programming to predict and optimize the construction plan, defines the optimization strategy, including adjusting the construction strategy parameters, setting the optimization cycle and objective function, using the event triggering mechanism to determine the optimization timing, and generating the optimized construction strategy diagram. The emergency response submodule is based on the optimized construction strategy diagram, adopts a decision optimization algorithm, uses Python programming to formulate emergency response strategies, defines emergency response rules, assesses the impact of emergencies, adjusts construction plans and resource allocation, and generates detailed construction scheme diagrams.

[0011] Preferably, the structural health monitoring module includes a data analysis submodule, a vibration monitoring submodule, and a health assessment submodule; The data analysis submodule is based on a detailed construction plan, uses a graph convolutional network algorithm, and utilizes the PyTorch framework for model construction. It constructs an adjacency matrix to represent the spatial relationships between structural elements, sets graph convolutional layer parameters, matches the spatial distribution characteristics of the structure, captures the interactions between structures, extracts spatial features, and generates a spatial relationship feature map. The vibration monitoring submodule is based on a spatial relationship feature map, utilizes a spatiotemporal graph network model and combines it with a long short-term memory network, and is implemented using the TensorFlow framework. This includes adjusting the hidden layer state to 128 and setting the number of layers to 2, analyzing the structural vibration data, capturing the dependencies in the time series, and generating a time series dependency graph. The health assessment submodule is based on a time-series dependency graph and applies a convolutional neural network to monitor and assess the structural health status in real time. It uses the Keras framework for model construction, including 64 filters and ReLU activation function, to extract features and identify damage from vibration data and generate structural health status results.

[0012] Preferably, the risk management and prediction module includes a risk assessment submodule, an early warning issuance submodule, and a strategy formulation submodule; The risk assessment submodule, based on the structural health status results, executes a continuous-time Markov decision process algorithm, dynamically calculates state transition probabilities using Python's SciPy library, simulates risk events in the construction project, combines a Monte Carlo tree search algorithm, and performs random simulations using Python's NumPy library to identify and assess construction risks and generate risk assessment results. The early warning issuance submodule is based on the risk assessment results and adopts an early warning strategy. By analyzing the risk assessment results, it identifies key risk points and risk levels, sets thresholds to activate the early warning mechanism according to the risk points and levels, and generates early warning notifications by sending email and SMS notifications to the project management team and stakeholders. The strategy formulation submodule is based on early warning notifications, adopts a decision analysis framework, processes risk data through Python's Pandas library, analyzes the impact of risk occurrence, formulates countermeasures, and generates risk management strategies based on information from early warning results and risk assessment results.

[0013] Preferably, the schedule prediction and adjustment module includes a schedule prediction submodule, an impact assessment submodule, and a schedule adjustment submodule; The progress prediction submodule is based on a risk management strategy, employs a long short-term memory network algorithm, and is implemented using the TensorFlow framework. It designs an LSTM layer to analyze the time series data of construction progress, captures long-term dependencies in the data, introduces a Gaussian process regression algorithm, uses the Sklearn library, sets the radial basis function as the kernel function, and adjusts the scale parameter of the kernel to enhance the recognition ability and modeling ability of Gaussian processes, and generates progress prediction results. The impact assessment submodule uses Python to perform impact assessment analysis based on the schedule prediction results. It assesses the impact of the prediction results on the construction plan by calculating the variance and covariance of the prediction data, and uses the NumPy library to perform mathematical operations to predict the construction plan and generate impact assessment results. The plan adjustment submodule adjusts the construction plan based on the impact assessment results and uses decision support tools. It uses the Pandas library in Python to process construction progress data, combines key indicators from the impact assessment results, and optimizes the construction plan by writing optimization scripts, taking into account the trade-off between resource utilization and construction delays, and generates the adjusted construction plan.

[0014] Preferably, the resource optimization and configuration module includes a resource demand analysis submodule, an optimization algorithm submodule, and a configuration scheme generation submodule; The resource demand analysis submodule is based on the adjusted construction plan and uses data analysis methods. It analyzes the resource demand of the project stage through the Pandas library of Python, predicts the resource demand based on the timeline of the construction plan and the resource consumption rate of the tasks, and generates resource demand analysis results. The optimization algorithm submodule is based on the resource demand analysis results. It combines mixed-integer linear programming and genetic algorithm, and uses Python's PuLP library to build a mathematical model for resource allocation. The mixed-integer linear programming algorithm is used to define the constraints and objective function of the resource allocation problem. The genetic algorithm implemented by the DEAP library captures the optimal solution in the resource allocation scheme and generates a resource optimization strategy. The configuration scheme generation submodule is based on resource optimization strategies. It uses project management software, including Microsoft Project, to transform the optimization scheme into a resource allocation plan, plan the allocation of resources on the project timeline, including staff allocation plans and material procurement schedules, and generate a resource allocation scheme.

[0015] Preferably, the environmental impact assessment module includes an environmental data analysis submodule, an impact prediction submodule, and a result generation submodule; The environmental data analysis submodule is based on the resource allocation scheme, executes a combination of kernel ridge regression algorithm and Gaussian process regression algorithm, uses Python's Sklearn library to set the alpha parameter to 1.0, uses radial basis function as kernel, optimizes scale parameter to match the characteristics of construction environment data, analyzes environmental factors of weather conditions, and generates environmental data feature analysis results. The impact prediction submodule uses the results of environmental data feature analysis and applies a Gaussian process regression model to predict the impact of changes in the construction environment, reflecting the randomness of weather condition changes. By analyzing the relationship between construction environment data and construction progress, it quantifies the potential impact of environmental changes on construction progress and generates ecological impact assessment results. The results generation submodule is based on the ecological impact assessment results. It uses data visualization technology and the Matplotlib library to show the impact of environmental changes on the construction progress, converts the prediction results into graphical form, and generates environmental impact analysis results.

[0016] Preferably, the overall progress management module includes a comprehensive analysis submodule, a management decision-making submodule, and a plan execution submodule; The comprehensive analysis submodule, based on the environmental impact analysis results, performs a comprehensive evaluation method to conduct a multi-dimensional analysis of the overall progress of the construction project, including multi-indicator decision analysis and performance evaluation. Implemented through Python, it comprehensively refers to multiple performance indicators such as cost, time, and quality, and evaluates the progress performance of the construction project through weight allocation and scoring, generating comprehensive progress analysis results. The management decision submodule manages and adjusts the project schedule based on the comprehensive schedule analysis results and uses a decision support framework. Through decision logic, it formulates management measures and adjustment strategies by referring to the deviation between the real-time status of the project and the target, and generates construction schedule adjustment decisions. The implementation submodule of the solution is based on construction schedule adjustment decisions, and uses the project management tool Microsoft Project to adjust the project timeline and resource allocation, match changes in construction schedule, monitor and control the project execution process, and generate the final construction management solution.

[0017] One or more technical solutions provided in this application have at least the following technical effects or advantages: By applying graph neural networks and graph attention networks, the complex dynamic relationships between resource allocation, task priority, and schedule changes in construction projects can be accurately captured. Furthermore, the construction model is updated in real time through dynamic graph convolutional networks, ensuring the timeliness and accuracy of construction schedule analysis. The introduction of an event response decision module enables the system to react quickly to real-time events, optimizing future construction plans through predictive control algorithms and improving the flexibility and efficiency of construction strategies. The implementation of a structural health monitoring module effectively prevents potential structural risks and ensures construction safety through real-time monitoring and analysis. The risk management and prediction module utilizes Markov decision processes and Monte Carlo tree search to provide a scientific basis for the identification and management of construction risks, reducing the impact of unforeseen risks. The schedule prediction and adjustment module combines long short-term memory networks and Gaussian process regression to optimize construction plans, enhancing the flexibility of project management and its ability to respond to future changes. The application of resource optimization and environmental impact assessment modules improves resource utilization efficiency and environmental adaptability, ensuring that the dual goals of construction schedule and environmental protection are met. The comprehensive evaluation method of the overall schedule management module provides comprehensive decision support for construction management, enabling efficient and high-quality project management.

[0018] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0019] Figure 1 This invention presents a module diagram of a real-time monitoring system for road and bridge construction progress. Figure 2 The present invention provides a system framework diagram for a real-time monitoring system for road and bridge construction progress. Figure 3 A schematic diagram illustrating the specific process of the dynamic relationship analysis module in the real-time monitoring system for road and bridge construction progress proposed in this invention. Figure 4 This invention presents a schematic diagram illustrating the specific process of the event response decision module in a real-time monitoring system for road and bridge construction progress. Figure 5 This invention presents a schematic diagram illustrating the specific process of the structural health monitoring module in a real-time monitoring system for road and bridge construction progress. Figure 6 This invention presents a schematic diagram illustrating the specific process of the risk management and prediction module in a real-time monitoring system for road and bridge construction progress. Figure 7 This invention provides a schematic diagram of the progress prediction and adjustment module of the real-time monitoring system for road and bridge construction progress. Figure 8 This invention presents a schematic diagram illustrating the specific process of the resource optimization and allocation module in the real-time monitoring system for road and bridge construction progress. Figure 9 This invention presents a schematic diagram illustrating the specific process of the environmental impact assessment module of the real-time monitoring system for road and bridge construction progress. Figure 10 This invention presents a schematic diagram illustrating the overall progress management module of the real-time monitoring system for road and bridge construction progress. Detailed Implementation

[0020] This application provides a real-time monitoring system for road and bridge construction progress, addressing the shortcomings of traditional road and bridge construction monitoring systems in areas such as real-time data processing, dynamic monitoring, risk management, and resource optimization. Traditional systems rely on manual judgment and lack scientific data analysis capabilities, leading to resource allocation deviating from optimal levels and impacting construction efficiency and cost control. Furthermore, the system fails to fully utilize real-time data, resulting in slow responses to monitoring and adjustments to construction progress and an inability to respond promptly to unexpected events and changes during construction, increasing the risk of project delays and cost overruns. Structural health monitoring relies heavily on periodic inspections, lacking real-time early warnings and making it difficult to identify structural risks in a timely manner. Risk management lacks effective tools, and response measures are not precise enough, increasing construction uncertainty. Finally, the environmental impact assessment does not adequately consider the environmental impact of construction, raising environmental issues and public concerns.

[0021] Application Overview Existing technologies for traditional road and bridge construction monitoring systems have shortcomings in real-time data processing, dynamic monitoring, risk management, and resource optimization. These systems rely on manual judgment and lack scientific data analysis capabilities, leading to resource allocation that deviates from optimal levels and impacting construction efficiency and cost control. They fail to fully utilize real-time data, resulting in slow responses to monitoring and adjustments to construction progress and an inability to respond promptly to unexpected events and changes during construction, increasing the risk of project delays and cost overruns. Structural health monitoring relies heavily on periodic inspections, lacking real-time early warnings and making it difficult to identify structural risks in a timely manner. Risk management lacks effective tools, and response measures are not precise enough, increasing construction uncertainty. Furthermore, environmental impact assessments often fail to adequately consider the environmental impact of construction, leading to environmental problems and public concern.

[0022] To address the aforementioned technical problems, the overall approach of the technical solution provided in this application is as follows: like Figure 1 , 2 As shown, this application provides a real-time monitoring system for road and bridge construction progress, wherein the system includes a dynamic relationship analysis module, an event response decision module, a structural health monitoring module, a risk management and prediction module, a progress prediction and adjustment module, a resource optimization allocation module, an environmental impact assessment module, and an overall progress management module; The dynamic relationship analysis module is based on real-time construction progress data and uses graph neural network algorithm to capture the relationship between resource allocation, task priority and progress changes by constructing a dynamic graph representation of the construction project. It uses graph attention network to enhance the influence analysis of key nodes and combines dynamic graph convolutional network to dynamically update the graph structure according to time changes, perform dynamic relationship modeling of construction progress, and generate dynamic relationship analysis graph. The event response decision module is based on a dynamic relationship analysis graph and adopts an event-driven strategy and model predictive control theory. For real-time events, including task completion and resource changes, it predicts and optimizes the construction plan for future time periods through rolling time-domain optimization, uses an event triggering mechanism to determine the optimization timing, adjusts the construction strategy, and generates detailed construction plans. The structural health monitoring module is based on a detailed construction plan and uses a graph convolutional network algorithm to analyze structural vibration data. It captures the spatial relationship between structures through spectral graph convolution, analyzes time series data in combination with a spatiotemporal graph network model, monitors the health status of the structure at the construction site in real time, predicts and assesses potential structural damage, and generates structural health status results. The risk management and prediction module, based on the structural health status results, adopts the continuous-time Markov decision process algorithm, and uses the dynamic calculation of state transition probabilities, combined with the Monte Carlo tree search algorithm, to conduct random simulation to identify and assess construction risks, formulate countermeasures, and generate risk management strategies. The schedule forecasting and adjustment module is based on risk management strategies. It uses a long short-term memory network algorithm to analyze the time series data of construction progress, capture long-term dependencies, and combine it with a Gaussian process regression algorithm to simulate the randomness of forecasting. Based on the forecast results, it optimizes the construction plan and generates an adjusted construction plan. Based on the adjusted construction plan, the resource optimization and allocation module uses a mixed-integer linear programming algorithm to mathematically model and optimize the resource allocation problem. Combined with a genetic algorithm, it captures the optimal solution in the search space and optimizes the resource allocation to generate a resource allocation scheme. The environmental impact assessment module is based on the resource allocation scheme and uses the kernel ridge regression algorithm to perform high-dimensional data analysis on the impact of changes in the construction environment, including weather conditions, on the construction progress. It is combined with the Gaussian process regression algorithm to generate environmental impact analysis results. The overall progress management module, based on the results of environmental impact analysis, adopts a comprehensive evaluation method, and manages and adjusts the progress of construction projects through multi-indicator decision analysis and performance evaluation. It also formulates comprehensive management strategies and generates the final construction management plan.

[0023] The dynamic relationship analysis diagram includes resource nodes, task nodes, schedule nodes, and relationship strength indicators between nodes. The detailed construction plan includes a resource allocation table, a priority task list, and schedule adjustment strategies. The structural health status results include vibration frequency data, damage probability scores, and structural health indicators. The risk management strategy includes risk level classification, prediction model results, and an overview of response strategies. The adjusted construction plan includes activity priority adjustments, resource reallocation schemes, and a revised timetable. The resource allocation scheme includes a resource allocation diagram, cost-benefit analysis, and an optimized allocation list. The environmental impact analysis results include weather condition impact assessment, correlation analysis between environmental variables and construction schedule, and quantification of impact degree. The final construction management scheme includes a schedule management plan, resource management guidelines, and a risk response framework.

[0024] In the dynamic relationship analysis module, through real-time data acquisition of construction progress and graph neural network algorithms, the module achieves efficient modeling of the dynamic relationships of construction projects. The key data format is time-series data, recording the start and end times of construction tasks, resource allocation, and task dependencies. The module first constructs a dynamic graph representation, where nodes represent construction tasks, edges represent dependencies between tasks, and edge weights represent resource allocation and task priority. The Graph Attention Network (GAT) algorithm is used to enhance the influence analysis of key nodes, i.e., key construction tasks. By assigning different weights to each node through an attention mechanism, the model's ability to identify key construction stages is strengthened. The Dynamic Graph Convolutional Network (DGCN) algorithm dynamically updates the graph structure according to time changes, involving time-series analysis of the features of graph nodes and the connection states of edges to reflect real-time changes in construction progress. The module can effectively capture the complex relationships between changes in construction progress, resource allocation, and task priority, and generate a dynamic relationship analysis graph. This graph intuitively displays the dynamic changes in construction progress and key task nodes, providing accurate data support for construction management.

[0025] In the event response decision-making module, dynamic relationship analysis graphs and event-driven strategy and model predictive control theory are used to predict and optimize the construction plan in real time. The module processes data including real-time event data such as task completion status and resource allocation changes. Utilizing an event-driven model, the module can respond quickly to new events, such as task completion or resource changes. Through rolling time-domain optimization techniques, it dynamically adjusts the construction plan for future time periods. An event triggering mechanism determines the timing of optimization, ensuring timely adjustments to the construction strategy and generating detailed construction plans. These plans take into account the impact of real-time events on the construction schedule, optimizing resource allocation and task scheduling, thereby improving construction efficiency and responsiveness.

[0026] In the structural health monitoring module, by refining construction plans and employing graph convolutional network algorithms, the module enables real-time monitoring and assessment of the structural health status at the construction site. The data format is structural vibration data, including vibration frequency and intensity information during construction. The module uses spectral graph convolutional networks for in-depth analysis of the vibration data, enhancing the monitoring capability of structural health status by capturing the spatial relationships between structures. Combined with a spatiotemporal graph network model, the module can analyze the time-series characteristics of vibration data, achieving real-time monitoring of the structural health status at the construction site. The module not only predicts and assesses potential structural damage but also generates structural health status results, providing strong data support for construction safety.

[0027] In the risk management and prediction module, construction risks are identified and assessed using structural health status results and the Continuous Time Markov Decision Process (CTMDP) algorithm. The data format covers the results of structural health monitoring and risk factor data during construction. The module dynamically calculates state transition probabilities using the CTMDP algorithm and combines it with the Monte Carlo Tree Search (MCTS) algorithm to perform stochastic simulations to identify potential risks during construction. Risk management strategies are then formulated, including countermeasures and prevention strategies, to reduce construction risks. The module not only effectively manages and predicts construction risks but also generates risk management strategies, providing scientific decision support for risk management in construction projects.

[0028] In the schedule prediction and adjustment module, a Long Short-Term Memory (LSTM) network algorithm is used to analyze the time-series data of construction progress. The data format is the progress of construction activities corresponding to timestamps, including start date, end date, and progress percentage. The LSTM network, through a hierarchical structure including an input layer, multiple hidden layers, and an output layer, captures long-term dependencies in the construction progress data. The gating mechanism of the LSTM unit controls the inflow and outflow of information. Model parameters are optimized using a backpropagation algorithm to minimize the difference between predicted and actual progress values. A Gaussian Process Regression (GPR) algorithm is combined to simulate the randomness of the prediction results. The GPR algorithm maps the input space to a high-dimensional feature space through a kernel function, using features to simulate the randomness of construction progress, generating prediction intervals. Dynamic programming is then used to optimize the construction plan, adjusting the construction sequence, resource allocation, and time arrangement to generate an adjusted construction plan. This module provides scientific decision support for construction management by accurately simulating and predicting the randomness of construction progress, effectively improving the adaptability and flexibility of the construction plan.

[0029] In the resource optimization and allocation module, the Mixed Integer Linear Programming (MILP) algorithm is used to mathematically model and optimize the resource allocation problem of the adjusted construction plan. The data format includes resource type, quantity, cost, and resource requirements for construction activities. The MILP algorithm defines an objective function that aims to minimize the total cost of resource allocation while satisfying the resource requirements and constraints of construction activities, including resource availability, time window limits, and budget constraints. During the optimization process, the branch and bound method is used to search for the optimal solution. This method combines branch and bound with cutting plane techniques, iteratively narrowing the search range until an optimal or near-optimal resource allocation scheme is found. Iteratively, a genetic algorithm is combined, simulating natural selection and genetic mechanisms such as crossover, mutation, and selection operations to search for the global optimum in the solution space. The parameters of the genetic algorithm, such as population size, crossover rate, and mutation rate, are finely tuned to ensure convergence and solution quality. The combined use of the MILP and genetic algorithms not only optimizes resource allocation but also enhances the diversity and robustness of the solutions. The generated resource allocation scheme ensures the economy and efficiency of the construction plan.

[0030] In the environmental impact assessment module, the impact of changes in the construction environment is analyzed using Kernel Ridge Regression (KRR) and Gaussian Process Regression (GPR) algorithms. The KRR algorithm incorporates ridge regression to process high-dimensional data and sets regularization parameters to control model complexity and reduce the risk of overfitting. The selection of the kernel function (e.g., a Gaussian kernel) and parameter optimization process ensure the algorithm can effectively handle nonlinear relationships. The GPR algorithm, used in conjunction with the KRR algorithm, provides probabilistic modeling of the relationship between environmental change factors (such as weather conditions) and construction progress by setting the kernel function and optimizing hyperparameters. This generates environmental impact analysis results, revealing the potential impact of specific environmental factors on construction progress, providing a basis for iterative adjustments to the construction plan, ensuring that construction activities can continue under adverse environmental conditions while minimizing negative environmental impacts.

[0031] In the overall schedule management module, a comprehensive evaluation method is used for multi-indicator decision analysis and performance evaluation to manage and adjust the schedule of construction projects. This comprehensive evaluation method includes setting up an evaluation indicator system, such as schedule, cost, quality, and safety, and using a weighted allocation method to determine the relative importance of each indicator. Data collection and analysis utilizes statistical methods (such as weighted average and standard deviation analysis) to comprehensively evaluate the collected data, identifying key issues and improvement opportunities in schedule management. Based on the evaluation results, a comprehensive management strategy is developed, including adjusting workflows, optimizing resource allocation, and enhancing monitoring and control measures. This results in a final construction management plan, which details the implementation steps, expected goals, and monitoring mechanisms to ensure the smooth completion of the construction project as planned, while improving efficiency and quality and reducing costs and risks.

[0032] Specifically, such as Figure 2 , 3 As shown, the dynamic relationship analysis module includes a resource allocation submodule, a task priority submodule, and a progress change analysis submodule; The resource allocation submodule is based on real-time construction progress data. It adopts a graph neural network algorithm and uses the PyTorch framework to build a graph attention network model. It defines the model structure, adds graph convolutional layers to capture the resource allocation relationship between nodes, uses the attention mechanism to analyze the influence of key resource nodes, and generates a dynamic resource influence graph. The task priority submodule is based on the resource dynamic influence graph, adopts the graph attention network algorithm, uses the PyTorch framework to build a dynamic graph convolutional network model, defines the model structure, adds dynamic graph convolutional layers to match the time sensitivity of tasks, applies the graph attention mechanism to strengthen task dependency analysis, and generates a task priority structure graph. The progress change analysis submodule is based on the task priority structure graph, adopts a dynamic graph convolutional network algorithm, uses the PyTorch framework for dynamic graph updates, defines an update mechanism, adjusts the graph structure according to timestamps, applies graph convolutional layers to analyze the impact of progress changes on resources and tasks, and generates a dynamic relationship analysis graph.

[0033] In the resource allocation submodule, a Graph Attention Network (GAT) model is constructed within the PyTorch framework using real-time construction progress data acquisition and graph neural network algorithms to refine the capture and analysis of resource allocation relationships. The real-time data format includes key information such as task identifier, resource type, allocation quantity, and task start and end times. The model structure design includes multiple graph convolutional layers, each designed to capture complex resource allocation relationships between nodes. By defining a specific graph structure, the model can understand the dependencies and resource flows between task nodes. The application of the attention mechanism allows the model to automatically emphasize key resource nodes that have a significant impact on the overall construction progress when analyzing resource allocation. Utilizing the APIs provided by PyTorch, the model is trained to process real-time data and generate a dynamic resource impact graph. This graph visually displays the dynamic changes in resource allocation and their impact on construction progress, assisting managers in optimizing resource allocation decisions.

[0034] In the task prioritization submodule, a Dynamic Graph Convolutional Network (DGCN) model is constructed within the PyTorch framework using a dynamic resource impact graph and a graph attention network algorithm to refine the analysis and structure graph generation of task priorities. The dynamic resource impact graph serves as input data, encompassing the dependencies between tasks and resource allocation. This submodule defines a dynamic graph convolutional network model, utilizing dynamic graph convolutional layers to match the temporal sensitivity and priority settings of tasks. The graph attention mechanism is applied in this process to enhance the analysis of task dependencies, particularly considering task execution order and resource availability. The task priority information learned by the model is used to generate a task priority structure graph, which details the priority relationships between different tasks, providing a scientific basis for construction management and ensuring the rational arrangement of construction schedules and efficient utilization of resources.

[0035] In the schedule change analysis submodule, a dynamic graph convolutional network algorithm is employed within the PyTorch framework, using a task priority structure graph, to dynamically update the graph and analyze schedule changes. The task priority structure graph provides priority relationships and time-sensitive data between tasks. This submodule defines a graph update mechanism that adjusts the graph structure based on timestamps to reflect real-time schedule changes. Graph convolutional layers are used to analyze the impact of schedule changes on resource allocation and task priorities. The model can dynamically update the graph structure reflecting construction progress, and the generated dynamic relationship analysis graph details the interactions between schedule changes, resource allocation, and task priorities, providing real-time and dynamic data support for construction schedule management and adjustment.

[0036] Based on the real-time monitoring system for road and bridge construction progress, the data items considered include task identifiers (e.g., T1, T2), resource types (e.g., manpower, machinery), allocated quantities (e.g., 10 people, 2 machines), and task start and end times (e.g., 2023-03-01 to 2023-03-10). The simulated values ​​are: task T1 allocated 10 manpower from 2023-03-01 to 2023-03-05; and task T2 allocated 2 machines from 2023-03-02 to 2023-03-10. Through the operation of the above sub-modules, the system can effectively capture the resource allocation, task priority, and progress changes at the construction site. The generated dynamic relationship analysis diagram, resource dynamic impact diagram, and task priority structure diagram provide detailed data support and decision-making basis for construction management, ensuring the smooth progress of construction.

[0037] Specifically, such as Figure 2 , 4 As shown, the event response decision module includes an event capture submodule, a decision optimization submodule, and an emergency response submodule; The event capture submodule is based on dynamic relationship analysis graphs, adopts event-driven strategies and model predictive control theory, uses Python to program the event capture logic, defines event capture rules, monitors task completion and resource change events, sets event trigger thresholds, applies rolling time-domain optimization to predict event impact, and generates event impact analysis graphs. The decision optimization submodule is based on the event impact analysis diagram, adopts the rolling time domain optimization algorithm, uses Python programming to predict and optimize the construction plan, defines the optimization strategy, including adjusting the construction strategy parameters, setting the optimization cycle and objective function, using the event triggering mechanism to determine the optimization timing, and generates the optimized construction strategy diagram. The emergency response submodule is based on the optimized construction strategy diagram, adopts a decision optimization algorithm, uses Python programming to formulate emergency response strategies, defines emergency response rules, assesses the impact of emergencies, adjusts construction plans and resource allocation, and generates detailed construction plan diagrams.

[0038] In the event capture submodule, real-time monitoring and capture of key events in the construction schedule are implemented using Python programming through dynamic relationship analysis graphs and event-driven strategy and model predictive control theory. The data format processed by this submodule includes task completion status, resource allocation, and changes, with each data item timestamped for real-time updates and tracking. The event capture logic programming first defines a set of event capture rules based on thresholds set for specific stages of construction tasks or resource changes, such as reaching a certain percentage of task completion or significant changes in resource allocation. By setting event trigger thresholds, the submodule can automatically activate the event capture mechanism when specific conditions are met. It employs rolling time-domain optimization techniques to predict and analyze captured events and their potential impacts, involving simulations and impact assessments of future construction progress. Python control flow statements and data processing libraries (such as Pandas) are used to process the event data, implementing the event monitoring logic through conditional statements and loop structures. The submodule's output is an event impact analysis graph, which details the potential impact of captured events on the construction schedule, helping the management team identify key impact points and adjust the construction plan accordingly.

[0039] In the decision optimization submodule, Python is used to predict and optimize the construction plan through event impact analysis graphs and rolling time-domain optimization algorithms. Based on the event impact analysis graphs, this submodule dynamically adjusts and optimizes the construction schedule, including adjusting construction strategy parameters, setting optimization cycles, and constructing objective functions. This submodule can determine the optimal optimization timing based on event triggering mechanisms. During the optimization process, Python's numerical computing libraries (such as NumPy) and optimization tools (such as SciPy) are used to execute rolling time-domain optimization algorithms, dynamically adjusting the construction plan to cope with the impact of real-time events. This submodule generates an optimized construction strategy graph, showing the adjusted construction plan and resource allocation, effectively improving the adaptability and efficiency of the construction schedule.

[0040] In the emergency response submodule, emergency response strategies are formulated using Python programming through optimized construction strategy diagrams and decision optimization algorithms. Based on the optimized construction strategy diagrams, this submodule assesses the impact of emergencies on construction progress and resource allocation, and adjusts the construction plan accordingly. It defines a set of emergency response rules, including the setting of response time windows and the selection of adjustment strategies, to quickly and effectively respond to anticipated construction interruptions or resource shortages. Different types of emergencies are classified and processed through Python's logic control and decision tree structure. Conditional judgments and loop traversal are used to dynamically adjust the construction plan and resource allocation. The generated detailed construction scheme diagram provides the construction site with a set of specific emergency response measures and adjusted construction plans, ensuring the continuity of construction progress and optimal resource allocation.

[0041] Taking a real-time monitoring system for road and bridge construction progress as an example, data items include task number (e.g., T1, T2), task status (e.g., completion rate 70%), resource type (e.g., manpower, machinery), and resource changes (e.g., manpower increased by 5). A simulated numerical example: the completion rate of task T1 is updated from 60% to 70%, while manpower increases from 15 to 20. Through the operation of the above sub-modules, the system can capture events in real time and predict their impact, optimize construction strategies to cope with resource changes, and finally generate detailed construction plan diagrams, showing the adjusted construction progress and resource allocation in response to events, providing effective decision support for construction management.

[0042] Specifically, such as Figure 2 , 5 As shown, the structural health monitoring module includes a data analysis submodule, a vibration monitoring submodule, and a health assessment submodule; The data analysis submodule is based on a detailed construction plan. It adopts a graph convolutional network algorithm and uses the PyTorch framework to build the model. It constructs an adjacency matrix to represent the spatial relationship between structural elements, sets the parameters of the graph convolutional layer, matches the spatial distribution characteristics of the structure, captures the interaction between structures and extracts spatial features, and generates a spatial relationship feature map. The vibration monitoring submodule is based on a spatial relationship feature map, utilizes a spatiotemporal graph network model and combines it with a long short-term memory network, and is implemented using the TensorFlow framework. This includes adjusting the hidden layer state to 128 and setting the number of layers to 2, analyzing the structural vibration data, capturing the dependencies in the time series, and generating a time series dependency graph. The health assessment submodule is based on a time-series dependency graph and applies a convolutional neural network for real-time monitoring and assessment of structural health status. The model is built using the Keras framework, which includes 64 filters and ReLU activation function. It performs feature extraction and damage identification on vibration data to generate structural health status results.

[0043] In the data analysis submodule, the construction plan is refined and modeled using the Graph Convolutional Network (GCN) algorithm and the PyTorch framework. The data format primarily involves the spatial coordinates and characteristic information of structural elements, such as their location, type, and connection status. This information is used to construct an adjacency matrix, representing the spatial relationships between structural elements. In the PyTorch framework, the GCN is first initialized, setting parameters for the GCN layers, such as input feature dimension, output feature dimension, and activation function, to ensure the network matches the spatial distribution characteristics of the structure. By defining a forward propagation function, graph convolution operations are performed using the adjacency matrix and node features to capture the interactions between structures and extract spatial features. The network progressively extracts higher-order spatial features through multiple layers of graph convolution, generating a spatial relationship feature map. This feature map details the complex spatial relationships between structural elements, providing fundamental data for vibration monitoring and health assessment. The generation of the spatial relationship feature map enhances the depth of understanding of structural spatial relationships, making subsequent monitoring and assessment more accurate. It effectively captures important spatial feature information within the structure, providing crucial data support for structural health monitoring.

[0044] In the vibration monitoring submodule, a spatiotemporal graph network model combined with a Long Short-Term Memory (LSTM) network and the TensorFlow framework is used to process structural vibration data based on spatial relationship feature maps, reflecting the vibration characteristics of the structure over time. The spatiotemporal graph network is constructed using the TensorFlow framework. First, model parameters are set, including a hidden layer state dimension of 128 and a network layer count of 2. By defining the spatiotemporal graph network model and integrating the LSTM network, the model can capture dependencies in the time series. During model training, structural vibration data is input. The spatiotemporal graph network captures the spatial dependencies between structural elements, while the LSTM captures the temporal dependencies. The combination of these two methods generates a time-series dependency graph. This dependency graph details the complex relationships between structural vibrations in time and space, providing crucial information for structural health assessment. The generated time-series dependency graph reveals the time-series dependencies of structural vibrations, leading to a deeper understanding of structural vibration behavior and providing an accurate analytical foundation for iterative health assessment and damage identification.

[0045] In the health assessment submodule, real-time monitoring and evaluation of structural health status are performed using a Convolutional Neural Network (CNN) and the Keras framework. The data format is a time-series dependency graph, showcasing the characteristics of structural vibration data. A CNN is constructed using the Keras framework, with the model construction including setting the number of filters to 64 and selecting ReLU as the activation function. The CNN model structure is defined, including convolutional layers for feature extraction and damage identification. During model training, the time-series dependency graph is input, and the convolutional layers extract features from the vibration data through filters to identify the structural health status. Through multi-layer feature extraction and analysis of the network, structural health status results are generated. This result provides direct evidence for real-time monitoring and health assessment of structures. Based on the extracted features and identified patterns, the health status of the structure is accurately assessed, providing a scientific basis for maintenance and repair.

[0046] In the real-time monitoring system for road and bridge construction progress, the three sub-modules mentioned above are comprehensively applied to monitor a bridge under construction. The data analysis sub-module processes detailed construction plan data, including the location coordinates and connection status of the bridge's main structural elements such as piers and bridge decks. For example, pier data might be denoted as (X: 100, Y: 200, Z: 50, type: support, connection status: connected to bridge deck). The vibration monitoring sub-module processes structural vibration data in time-series format, with simulated values ​​including the vibration intensity at various measuring points on the bridge at different time points, such as a bridge deck vibration intensity of 0.5 at time point 1. The health assessment sub-module ultimately generates a detailed structural health status result, indicating the overall health condition of the bridge, including the location and extent of identified potential damage. This provides the construction team with timely maintenance and repair recommendations, ensuring the smooth progress of construction and the safe construction of the bridge.

[0047] Specifically, such as Figure 2 , 6 As shown, the risk management and forecasting module includes a risk assessment submodule, an early warning issuance submodule, and a strategy formulation submodule; The risk assessment submodule, based on the structural health status results, executes a continuous-time Markov decision process algorithm, dynamically calculates the state transition probability using Python's SciPy library, simulates risk events in the construction project, combines a Monte Carlo tree search algorithm, and performs random simulations using Python's NumPy library to identify and assess construction risks and generate risk assessment results. The early warning issuance submodule is based on the risk assessment results and adopts an early warning strategy. By analyzing the risk assessment results, it identifies key risk points and risk levels, sets thresholds to activate the early warning mechanism based on the risk points and levels, and generates early warning notifications by sending email and SMS notifications to the project management team and stakeholders. The strategy formulation submodule is based on early warning notifications, adopts a decision analysis framework, uses Python's Pandas library to process risk data, analyzes the impact of risk occurrence, formulates countermeasures, and generates risk management strategies based on information from early warning results and risk assessment results.

[0048] In the risk assessment submodule, a continuous-time Markov decision process (CTMDP) algorithm is executed based on structural health status results, and the state transition probability is dynamically calculated using Python's SciPy library. This submodule is based on structural health monitoring data, including vibration levels, deformation indices, and crack development, with each data point timestamped to track changes in structural condition. By defining a CTMDP model, the submodule can simulate anticipated risk events in the construction project, such as crack propagation and overload effects. The calculation of state transition probabilities relies on the numerical analysis capabilities of the SciPy library, particularly in solving sparse matrices and linear systems. Combined with the Monte Carlo Tree Search (MCTS) algorithm, stochastic simulations are performed using Python's NumPy library. This involves generating a large number of random samples to simulate the occurrence of different risk events and their impact on the project. Through calculation and simulation, the submodule identifies and assesses construction risks. The generated risk assessment results detail the probability, estimated impact, and severity of each identified risk, providing a scientific basis for subsequent risk management.

[0049] In the early warning issuance submodule, an early warning strategy is adopted based on the risk assessment results. By analyzing the risk assessment results, key risk points and risk levels are identified. The data format processed by this submodule includes risk type, probability assessment, and impact degree. Each risk data point is based on the assessment results of the previous submodule. By setting different risk levels and corresponding thresholds, the submodule can activate the early warning mechanism when the risk reaches a specific level. After the early warning mechanism is activated, the submodule automatically sends email and SMS notifications to the project management team and stakeholders, using Python's email and SMS APIs for automated communication. The notifications include descriptions of key risk points, risk levels, and recommended response measures, ensuring that all stakeholders are promptly informed of the risk situation and are prepared to take appropriate countermeasures. The generated early warning notifications not only enhance the real-time nature of project risk management but also improve the preparedness for responding to risks.

[0050] In the strategy formulation submodule, based on early warning notifications, a decision analysis framework is employed. The Pandas library in Python is used to process and analyze risk data to formulate response measures. This submodule utilizes risk assessment results and data from early warning notifications, including risk level, scope of impact, and recommended response measures. Through comprehensive data analysis, targeted risk management strategies are developed. The Pandas library is used for efficient processing and analysis of large amounts of risk data, including data cleaning, grouping, and aggregation operations to extract key risk information and trends. Based on the analysis results, the submodule generates risk management strategies, detailing response measures for different risk levels, such as resource reallocation, adjustments to construction methods, and emergency repair plans. The strategies aim to minimize the impact of risk events on the construction project and ensure that the project maintains stable progress even when facing potential risks.

[0051] In the real-time monitoring system for road and bridge construction progress, structural health monitoring data includes vibration levels (e.g., 0.5 mm / s), deformation indices (e.g., 2 cm offset), and crack widths (e.g., 0.1 cm). Simulated values ​​represent a specific bridge structure at a specific monitoring point where the vibration level increases from 0.3 mm / s to 0.5 mm / s, the deformation index increases from 1.5 cm to 2 cm, and the crack width remains constant. Through the operation of the aforementioned sub-modules, the system can identify changes in structural health as potential risks, automatically issue early warnings, and generate management strategies for these risks, such as increasing the monitoring frequency in the area and implementing emergency reinforcement, providing timely and effective risk response solutions for construction management.

[0052] Specifically, such as Figure 2 , 7 As shown, the schedule forecasting and adjustment module includes a schedule forecasting submodule, an impact assessment submodule, and a schedule adjustment submodule; The schedule prediction submodule is based on risk management strategies and adopts a long short-term memory network algorithm. It is implemented through the TensorFlow framework, designs LSTM layers to analyze the time series data of construction schedule, captures long-term dependencies in the data, introduces Gaussian process regression algorithm, uses the Sklearn library, sets radial basis functions as kernel functions, and adjusts the scale parameters of the kernel to improve the recognition ability and modeling ability of Gaussian processes, and generates schedule prediction results. The impact assessment submodule uses Python to perform impact assessment analysis based on the schedule forecast results. It assesses the impact of the forecast results on the construction plan by calculating the variance and covariance of the forecast data, and uses the NumPy library to perform mathematical operations to predict the construction plan and generate impact assessment results. The planning adjustment submodule adjusts the construction plan based on the impact assessment results and uses decision support tools. It uses the Pandas library in Python to process construction progress data, combines key indicators from the impact assessment results, and optimizes the construction plan by writing optimization scripts, taking into account the trade-off between resource utilization and schedule delays, and generates the adjusted construction plan.

[0053] In the schedule prediction submodule, the Long Short-Term Memory (LSTM) network algorithm and the Gaussian Process Regression (GPR) algorithm, combined with the TensorFlow and Sklearn libraries, are used to refine and analyze the time series data of construction progress. The data processed includes time series data such as date, construction activity, and percentage of completion, reflecting the historical and current status of construction progress. The LSTM model is built using the TensorFlow framework, and LSTM layers are designed to analyze the time series data. By setting network structure parameters such as the number of layers and the number of neurons per layer, as well as training parameters such as learning rate and batch size, long-term dependencies in the data are captured. The GPR algorithm is introduced and implemented using the Sklearn library. The radial basis function (RBF) is set as the kernel function, and the scale parameter of the kernel is adjusted to enhance the model's ability to fit and predict construction progress data. The generated schedule prediction results describe in detail the future construction progress predicted based on historical data, providing reliable decision support for project management. The generated schedule prediction results not only demonstrate the predictability of future construction progress but also provide data support for the formulation of risk management strategies, assisting the project management team in effectively predicting and planning resource allocation and optimizing construction plans.

[0054] In the impact assessment submodule, based on the schedule forecast results, impact assessment analysis is performed using Python. The data format used is statistical data of the schedule forecast results, including the predicted completion date, the duration of the activity, and its estimated range of variation. By calculating the variance and covariance of the forecast data, mathematical operations are performed using the NumPy library to assess the impact of the forecast results on the construction schedule. The process analyzes the uncertainty and risks of the forecast in detail, generating impact assessment results. This provides a quantitative basis for adjusting the construction schedule, reveals the potential impact of forecast deviations on project schedule, and enables the project team to better understand risks and uncertainties, taking preventative measures to mitigate potential negative impacts.

[0055] In the planning adjustment submodule, the construction plan is adjusted using decision support tools based on the impact assessment results. The processed data includes construction schedule data and key indicators from the impact assessment results, such as the estimated completion time of tasks, resource allocation and the predictability of changes. The Pandas library in Python is used to process the data. Combining the impact assessment results, optimization scripts are written to consider the trade-offs between resource utilization and schedule delays, thus optimizing the construction plan. This process generates an adjusted construction plan that details resource reallocation, task rearrangement, and adjustments to expected completion times, providing the project team with specific execution plans. The generated adjusted construction plan helps the project management team effectively address uncertainties and risks in the forecasts, ensuring that the construction schedule meets project objectives and improving the flexibility and efficiency of project management.

[0056] In the real-time monitoring system for road and bridge construction progress, the above three sub-modules are comprehensively applied to monitor a bridge project under construction. The progress prediction sub-module processes time-series data including daily construction activity records from the project's inception to the present. Simulated data items include "Date: 2024-03-01, Activity: Pier Pouring, Completion Percentage: 40%", predicting future construction progress, such as the estimated completion date of pier pouring as 2024-03-10. The impact assessment sub-module analyzes the prediction results and assesses their impact on the overall construction plan, such as the impact of a delayed predicted completion date on the overall project schedule. The plan adjustment sub-module adjusts the construction plan based on the assessment results, such as reallocating resources to ensure the pier pouring activity is completed as planned. This ensures real-time monitoring and effective management of construction progress, improves construction efficiency, and ensures the project proceeds according to schedule.

[0057] Specifically, such as Figure 2 , 8 As shown, the resource optimization and allocation module includes a resource demand analysis submodule, an optimization algorithm submodule, and a configuration scheme generation submodule; The resource demand analysis submodule is based on the adjusted construction plan. It uses data analysis methods and Python's Pandas library to analyze the resource demand of the project phase. Based on the construction plan timeline and the resource consumption rate of the tasks, it predicts the resource demand and generates resource demand analysis results. The optimization algorithm submodule, based on the resource demand analysis results, combines mixed-integer linear programming and genetic algorithms. It uses Python's PuLP library to construct a mathematical model for resource allocation. The mixed-integer linear programming algorithm is used to define the constraints and objective function of the resource allocation problem. The genetic algorithm implemented using the DEAP library captures the optimal solution in the resource allocation scheme and generates a resource optimization strategy. The configuration scheme generation submodule is based on resource optimization strategies. It uses project management software, including Microsoft Project, to transform optimization schemes into resource allocation plans, plan the allocation of resources on the project timeline, including staff allocation plans and material procurement schedules, and generate resource allocation schemes.

[0058] In the resource requirements analysis submodule, the Pandas library in Python is used to accurately predict resource requirements for each stage of the project, based on a revised construction plan and data analysis methods. The construction plan provides the project timeline and details of each task, including expected start and end dates, task duration, and resource consumption rate for each task. The data is in tabular format, listing each task and its associated resource requirements. The submodule first imports the construction plan data, uses Pandas for data cleaning and preprocessing (e.g., removing duplicates, filling in missing values), and calculates the total resource requirements across the entire project timeline based on the resource consumption rate and duration of each task. By summing and aligning the resource requirements for each task over time, the submodule generates a time series forecast, displaying resource requirements at different points in time throughout the project cycle. This involves complex data operations such as grouping, summarizing, and time series analysis. The generated resource requirements analysis results are presented in charts or tables, providing the project management team with clear resource requirements forecasts and helping them make more accurate resource planning and procurement decisions before project execution.

[0059] In the optimization algorithm submodule, based on the resource demand analysis results, combined with mixed-integer linear programming (MILP) and genetic algorithms, the mathematical model of the resource allocation problem is constructed and the optimal solution is found using Python's PuLP and DEAP libraries. MILP is used to define the constraints (such as total resource limits and task resource requirements) and objective functions (such as minimizing cost or time) of the resource allocation problem. Using the PuLP library, the submodule constructs an optimization model for resource allocation, defining variables, constraints, and the objective function. Then, the genetic algorithm is implemented using the DEAP library to search the solution space of the MILP model to find the optimal resource allocation scheme that satisfies all constraints. The genetic algorithm simulates the process of natural selection, including selection, crossover, and mutation operations, to iteratively improve the solution. This process involves extensive parameter tuning and performance testing to ensure that an efficient and feasible resource allocation strategy is found. The generated resource optimization strategy details the optimal allocation plan for various resources, reducing resource waste and improving the efficiency and economy of project execution.

[0060] In the resource allocation scheme generation submodule, based on resource optimization strategies, project management software such as Microsoft Project is used to transform the optimized scheme into a specific resource allocation plan. This submodule accepts the resource allocation scheme generated by the optimization algorithm submodule as input, including the resource type, quantity, and allocation time for each task. Using Microsoft Project or similar project management software, the submodule plans the resource allocation scheme in detail onto the project timeline, creating staff allocation plans and material procurement schedules. The submodule considers the actual execution conditions of the project, such as resource availability, supply chain constraints, and staff working hours, to ensure the practicality and feasibility of the configuration scheme. The generated resource allocation scheme is presented in the form of a project plan document, detailing how resources are effectively allocated and utilized throughout the project lifecycle, ensuring the project proceeds smoothly according to schedule while optimizing cost and time efficiency.

[0061] In the real-time monitoring system for road and bridge construction progress, detailed data items are considered, including the name of each task, the expected start and end dates, duration, and the manpower and material requirements. Simulated numerical examples: Task T1 (pier construction) starts on January 1, 2024, lasts for 30 days, and requires 20 workers and 100 tons of cement; Task T2 (paving) starts on February 1, 2024, lasts for 15 days, and requires 15 workers and 50 tons of asphalt. Through the operation of sub-modules, the system can predict the resource requirements of the entire project, optimize resource allocation, and generate detailed resource configuration plans, such as the allocation of specific personnel to various tasks and the material procurement schedule, providing a precise decision support tool for project management.

[0062] Specifically, such as Figure 2 , 9 As shown, the environmental impact assessment module includes an environmental data analysis submodule, an impact prediction submodule, and a results generation submodule. The environmental data analysis submodule is based on the resource allocation scheme. It combines the kernel ridge regression algorithm and the Gaussian process regression algorithm. Using the Sklearn library in Python, the alpha parameter is set to 1.0. The radial basis function is used as the kernel to optimize the scale parameter to match the characteristics of the construction environment data. The environmental factors of weather conditions are analyzed to generate environmental data feature analysis results. The impact prediction submodule uses Gaussian process regression model to predict the impact of changes in the construction environment based on the results of environmental data feature analysis, reflecting the randomness of weather condition changes. By analyzing the relationship between construction environmental data and construction progress, it quantifies the potential impact of environmental changes on construction progress and generates ecological impact assessment results. The results generation submodule, based on the ecological impact assessment results, uses data visualization technology and the Matplotlib library to show the impact of environmental changes on construction progress, converting the prediction results into graphical form and generating environmental impact analysis results.

[0063] In the environmental data analysis submodule, a combination of kernel ridge regression and Gaussian process regression algorithms is implemented through resource allocation schemes, using Python's Sklearn library for refined environmental data feature analysis. The data format for this submodule includes weather condition variables such as date, temperature, humidity, and wind speed, collected from meteorological stations around the construction site, aiming to capture changing factors in the construction environment. First, the kernel ridge regression algorithm is used, with the alpha parameter set to 1.0 and a radial basis function (RBF) as the kernel function. Regularization is used to control model complexity and prevent overfitting. The RBF also provides an effective way to capture nonlinear relationships between environmental variables. The Gaussian process regression algorithm iteratively optimizes the scale parameter to match the characteristics of the construction environment data. By considering the uncertainty of the observed data, probabilistic predictions about the impact of environmental factors can be provided, increasing the flexibility and adaptability of the model. The submodule can generate environmental data feature analysis results, presented in statistical reports or graphical representations, demonstrating how weather conditions affect construction activities, providing a scientific basis for subsequent environmental adaptability strategy development.

[0064] In the impact prediction submodule, based on the environmental data feature analysis results, a Gaussian process regression model is applied to predict the impact of changes in the construction environment. This submodule focuses on analyzing the relationship between construction environmental data and construction progress, especially how to quantify the potential impact of weather condition changes on construction progress. Through the Gaussian process regression model, the submodule can take into account the randomness of weather condition changes and provide a flexible framework for predicting the impact of environmental changes on construction activities. This includes constructing a training set containing construction environmental data, then using the data to train the model, and finally, the model can predict the estimated changes in construction progress under given environmental conditions. The generated ecological impact assessment results detail the risks and challenges faced in predicting construction progress under different weather conditions, providing important decision support information for the project management team and helping to prepare response strategies to minimize the impact of adverse weather conditions.

[0065] In the results generation submodule, based on the ecological impact assessment results, data visualization technology is employed, using Python's Matplotlib library to convert the impact of environmental changes on construction schedule into graphical form. The goal of this submodule is to transform complex data analysis results into intuitive and easy-to-understand charts and graphs, enabling the project management team and relevant stakeholders to quickly understand the specific impact of environmental changes on construction schedule. By plotting time series graphs, scatter plots, and heatmaps, the submodule displays the changing trends of construction schedule and potential risk areas under different environmental conditions. This not only increases the interpretability of the analysis results but also provides the project team with an intuitive tool to plan and adjust construction schedules, ensuring the project can continue to progress under adverse environmental conditions.

[0066] In the real-time monitoring system for road and bridge construction progress, detailed data items include date, average temperature (e.g., 15 degrees Celsius), relative humidity (e.g., 80%), and wind speed (e.g., 10 km / h). A simulated numerical example: during construction in March 2024, the temperature ranged from 10 to 20 degrees Celsius, humidity from 70% to 90%, and wind speed from 5 km / h to 15 km / h. Through the operation of the above sub-modules, the system can analyze how changes in environmental conditions affect construction progress. For example, rising temperatures are expected to accelerate the curing process of certain materials, while high humidity and strong winds are expected to cause construction delays. The generated environmental impact analysis results are displayed graphically, helping the project management team understand and adapt to environmental changes and develop effective construction adjustment strategies.

[0067] Specifically, such as Figure 2 , 10 As shown, the overall progress management module includes a comprehensive analysis submodule, a management decision-making submodule, and a plan execution submodule; The comprehensive analysis submodule, based on the environmental impact analysis results, performs a comprehensive evaluation method to conduct a multi-dimensional analysis of the overall progress of the construction project, including multi-indicator decision analysis and performance evaluation. Implemented through Python, it comprehensively refers to multiple performance indicators such as cost, time, and quality, and evaluates the progress performance of the construction project through weight allocation and scoring, generating comprehensive progress analysis results. The management decision-making submodule manages and adjusts the project schedule based on the comprehensive schedule analysis results and uses a decision support framework. Through decision logic, it formulates management measures and adjustment strategies by referring to the deviation between the real-time status of the project and the target, and generates construction schedule adjustment decisions. The implementation module is based on construction schedule adjustment decisions and uses the project management tool Microsoft Project to adjust the project timeline and resource allocation, match changes in construction schedule, monitor and control the project execution process, and generate the final construction management plan.

[0068] In the comprehensive analysis submodule, a comprehensive evaluation of the environmental impact analysis results is performed using multi-dimensional analysis and the Python programming language. The data format covers multiple performance indicators for the construction project, including cost, time, and quality. This includes comparisons between actual and budgeted construction costs, deviations from project completion dates and planned schedules, and the pass rate of quality inspection results. By defining weight allocation and a scoring mechanism, each performance indicator is quantitatively evaluated. Scientific computing libraries in Python, such as NumPy and Pandas, are used to process the data. A scoring algorithm is developed to synthesize the scores of each indicator according to predefined weights, generating a comprehensive progress analysis result. This comprehensive analysis meticulously considers key performance indicators in project management, ensuring that the evaluation results fully reflect the project's progress performance. The generated comprehensive progress analysis result details the project's performance in terms of cost, time, and quality, providing the project management team with a comprehensive view of the project schedule. This helps the team identify key issues and areas for improvement in the project, guiding project management decisions.

[0069] In the management decision-making submodule, based on the comprehensive schedule analysis results, a decision support framework is used to precisely manage and adjust the project schedule. The methods employed include the application of decision logic, referring to the deviation between the project's real-time status and objectives to formulate corresponding management measures and adjustment strategies. By analyzing the comprehensive schedule analysis results, deviations in project schedule, cost, and quality are identified, and decision logic scripts written in Python are used to automatically propose adjustment suggestions. This process ensures the timeliness and accuracy of project management decisions. The generated construction schedule adjustment decisions detail the adjustment measures and strategies, providing the project team with a clear direction for execution, helping the team effectively address challenges in project schedule and ensuring the achievement of project goals.

[0070] In the implementation submodule, based on construction schedule adjustment decisions, the project management tool Microsoft Project is used for specific operations. This involves adjusting the project timeline and resource allocation to match changes in construction progress, including detailed rescheduling of the project timeline, reallocation of resources, and adjustments to task dependencies. Utilizing the powerful project management capabilities of Microsoft Project, detailed adjustments to the project plan are made, and the project execution process is monitored and controlled. The resulting final construction management plan details the adjusted project plan, including new timelines, resource allocation schemes, and expected deliverables. This provides the project team with a concrete plan for implementing project adjustments, ensuring the project is executed smoothly according to the adjusted plan and achieves project goals.

[0071] The real-time monitoring system for road and bridge construction progress integrates the three sub-modules mentioned above to monitor and manage the construction project of a cross-river bridge. The comprehensive analysis sub-module processes data including comparisons between budgeted and actual construction costs (e.g., budget of 10 million, actual expenditure of 10.5 million), deviations from project completion time (planned completion date of December 2024, current predicted completion date of January 2025), and quality pass rates (e.g., inspection pass rate of 95%). Based on the comprehensive progress analysis results, the management decision-making sub-module proposes adjustment strategies, such as increasing human resources and adjusting the construction sequence, to shorten the construction period and reduce costs. The solution execution sub-module adjusts the project schedule using Microsoft Project, such as rearranging certain tasks on the critical path and optimizing resource allocation. The resulting construction management plan details the adjustment measures, ensuring the project can be successfully completed according to the new objectives, improving the efficiency and effectiveness of project management, ensuring construction progress and quality, and ultimately achieving the project goals.

[0072] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of the claims and their equivalents, this application intends to include such modifications and modifications.

Claims

1. A real-time monitoring system for road and bridge construction progress, characterized in that, The system includes a dynamic relationship analysis module, an event response decision-making module, a structural health monitoring module, a risk management and prediction module, a schedule prediction and adjustment module, a resource optimization and allocation module, an environmental impact assessment module, and an overall schedule management module. The dynamic relationship analysis module is based on real-time construction progress data and uses a graph neural network algorithm to capture the relationship between resource allocation, task priority and progress changes by constructing a dynamic graph representation of the construction project. It uses a graph attention network to enhance the influence analysis of key nodes and combines a dynamic graph convolutional network to dynamically update the graph structure according to time changes, thereby performing dynamic relationship modeling of construction progress and generating a dynamic relationship analysis graph. The event response decision module is based on a dynamic relationship analysis graph and adopts an event-driven strategy and model predictive control theory. For real-time events, including task completion and resource changes, it predicts and optimizes the construction plan for future time periods through rolling time domain optimization, uses an event triggering mechanism to determine the optimization timing, adjusts the construction strategy, and generates detailed construction plans. The structural health monitoring module is based on a detailed construction plan and uses a graph convolutional network algorithm to analyze structural vibration data. It captures the spatial relationship between structures through spectral convolution, analyzes time series data in combination with a spatiotemporal graph network model, monitors the health status of the structure at the construction site in real time, predicts and assesses potential structural damage, and generates structural health status results. The risk management and prediction module, based on the structural health status results, uses a continuous-time Markov decision process algorithm to dynamically calculate the state transition probability and combines it with a Monte Carlo tree search algorithm to perform random simulation identification and assessment of construction risks, formulate countermeasures, and generate risk management strategies. The progress prediction and adjustment module is based on risk management strategy, uses long short-term memory network algorithm to analyze the time series data of construction progress, capture long-term dependencies, and combines it with Gaussian process regression algorithm to simulate the randomness of prediction, optimize the construction plan according to the prediction results, and generate the adjusted construction plan. The resource optimization and allocation module, based on the adjusted construction plan, uses a mixed-integer linear programming algorithm to mathematically model and optimize the resource allocation problem. Combined with a genetic algorithm, it captures the optimal solution in the search space and optimizes resource allocation to generate a resource allocation scheme. The environmental impact assessment module is based on the resource allocation scheme and uses the kernel ridge regression algorithm to perform high-dimensional data analysis on the impact of changes in the construction environment, including weather conditions, on the construction progress. It is combined with the Gaussian process regression algorithm to generate environmental impact analysis results. The overall progress management module, based on the environmental impact analysis results, adopts a comprehensive evaluation method, and manages and adjusts the progress of the construction project through multi-indicator decision analysis and performance evaluation. It also formulates comprehensive management strategies and generates the final construction management plan.

2. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The dynamic relationship analysis diagram includes resource nodes, task nodes, schedule nodes, and relationship strength indicators between nodes. The detailed construction plan includes a resource allocation table, a priority task list, and a schedule adjustment strategy. The structural health status results include vibration frequency data, damage probability scores, and structural health indicators. The risk management strategy includes risk level classification, prediction model results, and an overview of response strategies. The adjusted construction plan includes activity priority adjustments, resource reallocation schemes, and a revised timetable. The resource allocation scheme includes a resource allocation diagram, cost-benefit analysis, and an optimized allocation list. The environmental impact analysis results include weather condition impact assessment, correlation analysis between environmental variables and construction progress, and quantification of impact degree. The final construction management scheme includes a schedule management plan, resource management guidelines, and a risk response framework.

3. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The dynamic relationship analysis module includes a resource allocation submodule, a task priority submodule, and a progress change analysis submodule. The resource allocation submodule is based on real-time construction progress data. It adopts a graph neural network algorithm, uses the PyTorch framework to build a graph attention network model, defines the model structure, adds graph convolutional layers to capture the resource allocation relationship between nodes, uses the attention mechanism to analyze the influence of key resource nodes, and generates a dynamic resource influence graph. The task priority submodule is based on the resource dynamic influence graph, adopts the graph attention network algorithm, uses the PyTorch framework to build a dynamic graph convolutional network model, defines the model structure, adds dynamic graph convolutional layers to match the time sensitivity of tasks, applies the graph attention mechanism to strengthen task dependency analysis, and generates a task priority structure graph. The progress change analysis submodule is based on the task priority structure graph, adopts a dynamic graph convolutional network algorithm, uses the PyTorch framework for dynamic graph updates, defines an update mechanism, adjusts the graph structure according to timestamps, applies graph convolutional layers to analyze the impact of progress changes on resources and tasks, and generates a dynamic relationship analysis graph.

4. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The event response decision module includes an event capture submodule, a decision optimization submodule, and an emergency response submodule. The event capture submodule is based on a dynamic relationship analysis graph, adopts an event-driven strategy and model predictive control theory, uses Python to program the event capture logic, defines event capture rules, monitors task completion and resource change events, sets event trigger thresholds, applies rolling time domain optimization to predict event impact, and generates an event impact analysis graph. The decision optimization submodule is based on the event impact analysis diagram, adopts the rolling time domain optimization algorithm, uses Python programming to predict and optimize the construction plan, defines the optimization strategy, including adjusting the construction strategy parameters, setting the optimization cycle and objective function, using the event triggering mechanism to determine the optimization timing, and generating the optimized construction strategy diagram. The emergency response submodule is based on the optimized construction strategy diagram, adopts a decision optimization algorithm, uses Python programming to formulate emergency response strategies, defines emergency response rules, assesses the impact of emergencies, adjusts construction plans and resource allocation, and generates detailed construction scheme diagrams.

5. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The structural health monitoring module includes a data analysis submodule, a vibration monitoring submodule, and a health assessment submodule. The data analysis submodule is based on a detailed construction plan, uses a graph convolutional network algorithm, and utilizes the PyTorch framework for model construction. It constructs an adjacency matrix to represent the spatial relationships between structural elements, sets graph convolutional layer parameters, matches the spatial distribution characteristics of the structure, captures the interactions between structures, extracts spatial features, and generates a spatial relationship feature map. The vibration monitoring submodule is based on a spatial relationship feature map, utilizes a spatiotemporal graph network model and combines it with a long short-term memory network, and is implemented using the TensorFlow framework. This includes adjusting the hidden layer state to 128 and setting the number of layers to 2, analyzing the structural vibration data, capturing the dependencies in the time series, and generating a time series dependency graph. The health assessment submodule is based on a time-series dependency graph and applies a convolutional neural network to monitor and assess the structural health status in real time. It uses the Keras framework for model construction, including 64 filters and ReLU activation function, to extract features and identify damage from vibration data and generate structural health status results.

6. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The risk management and prediction module includes a risk assessment submodule, an early warning issuance submodule, and a strategy formulation submodule. The risk assessment submodule, based on the structural health status results, executes a continuous-time Markov decision process algorithm, dynamically calculates state transition probabilities using Python's SciPy library, simulates risk events in the construction project, combines a Monte Carlo tree search algorithm, and performs random simulations using Python's NumPy library to identify and assess construction risks and generate risk assessment results. The early warning issuance submodule is based on the risk assessment results and adopts an early warning strategy. By analyzing the risk assessment results, it identifies key risk points and risk levels, sets thresholds to activate the early warning mechanism according to the risk points and levels, and generates early warning notifications by sending email and SMS notifications to the project management team and stakeholders. The strategy formulation submodule is based on early warning notifications, adopts a decision analysis framework, processes risk data through Python's Pandas library, analyzes the impact of risk occurrence, formulates countermeasures, and generates risk management strategies based on information from early warning results and risk assessment results.

7. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The schedule prediction and adjustment module includes a schedule prediction submodule, an impact assessment submodule, and a schedule adjustment submodule; The progress prediction submodule is based on a risk management strategy, employs a long short-term memory network algorithm, and is implemented using the TensorFlow framework. It designs an LSTM layer to analyze the time series data of construction progress, captures long-term dependencies in the data, introduces a Gaussian process regression algorithm, uses the Sklearn library, sets the radial basis function as the kernel function, and adjusts the scale parameter of the kernel to enhance the recognition ability and modeling ability of Gaussian processes, and generates progress prediction results. The impact assessment submodule uses Python to perform impact assessment analysis based on the schedule prediction results. It assesses the impact of the prediction results on the construction plan by calculating the variance and covariance of the prediction data, and uses the NumPy library to perform mathematical operations to predict the construction plan and generate impact assessment results. The plan adjustment submodule adjusts the construction plan based on the impact assessment results and uses decision support tools. It uses the Pandas library in Python to process construction progress data, combines key indicators from the impact assessment results, and optimizes the construction plan by writing optimization scripts, taking into account the trade-off between resource utilization and construction delays, and generates the adjusted construction plan.

8. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The resource optimization and allocation module includes a resource demand analysis submodule, an optimization algorithm submodule, and a configuration scheme generation submodule. The resource demand analysis submodule is based on the adjusted construction plan and uses data analysis methods. It analyzes the resource demand of the project stage through the Pandas library of Python, predicts the resource demand based on the timeline of the construction plan and the resource consumption rate of the tasks, and generates resource demand analysis results. The optimization algorithm submodule is based on the resource demand analysis results. It combines mixed-integer linear programming and genetic algorithm, and uses Python's PuLP library to build a mathematical model for resource allocation. The mixed-integer linear programming algorithm is used to define the constraints and objective function of the resource allocation problem. The genetic algorithm implemented by the DEAP library captures the optimal solution in the resource allocation scheme and generates a resource optimization strategy. The configuration scheme generation submodule is based on resource optimization strategies. It uses project management software, including Microsoft Project, to transform the optimization scheme into a resource allocation plan, plan the allocation of resources on the project timeline, including staff allocation plans and material procurement schedules, and generate a resource allocation scheme.

9. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The environmental impact assessment module includes an environmental data analysis submodule, an impact prediction submodule, and a result generation submodule. The environmental data analysis submodule is based on the resource allocation scheme, executes a combination of kernel ridge regression algorithm and Gaussian process regression algorithm, uses Python's Sklearn library to set the alpha parameter to 1.0, uses radial basis function as kernel, optimizes scale parameter to match the characteristics of construction environment data, analyzes environmental factors of weather conditions, and generates environmental data feature analysis results. The impact prediction submodule uses the results of environmental data feature analysis and applies a Gaussian process regression model to predict the impact of changes in the construction environment, reflecting the randomness of weather condition changes. By analyzing the relationship between construction environment data and construction progress, it quantifies the potential impact of environmental changes on construction progress and generates ecological impact assessment results. The results generation submodule is based on the ecological impact assessment results. It uses data visualization technology and the Matplotlib library to show the impact of environmental changes on the construction progress, converts the prediction results into graphical form, and generates environmental impact analysis results.

10. The real-time monitoring system for road and bridge construction progress according to claim 1, characterized in that, The overall progress management module includes a comprehensive analysis submodule, a management decision-making submodule, and a plan execution submodule; The comprehensive analysis submodule, based on the environmental impact analysis results, performs a comprehensive evaluation method to conduct a multi-dimensional analysis of the overall progress of the construction project, including multi-indicator decision analysis and performance evaluation. Implemented through Python, it comprehensively refers to multiple performance indicators such as cost, time, and quality, and evaluates the progress performance of the construction project through weight allocation and scoring, generating comprehensive progress analysis results. The management decision submodule manages and adjusts the project schedule based on the comprehensive schedule analysis results and uses a decision support framework. Through decision logic, it formulates management measures and adjustment strategies by referring to the deviation between the real-time status of the project and the target, and generates construction schedule adjustment decisions. The implementation submodule of the solution is based on construction schedule adjustment decisions, and uses the project management tool Microsoft Project to adjust the project timeline and resource allocation, match changes in construction schedule, monitor and control the project execution process, and generate the final construction management solution.

Images