Intelligent construction management system and method

Abstract

The application discloses an intelligent construction management system and method, belongs to the technical field of building engineering management, and comprises a construction task relationship network of a construction project; an influence factor characteristic vector of a construction task is acquired, and an initial delay probability of the construction task is generated; based on the construction task relationship network, a connection path between the construction tasks is analyzed, the path quantity, the path length and the number of preceding construction tasks of the construction task of the connection path are acquired, a construction task influence analysis model is established, and a construction task influence factor is generated; an iterative analysis model is established according to the initial delay probability of the construction task and the construction task influence factor, and a comprehensive risk value of the construction task is generated; and the construction task is managed according to the comprehensive risk value of the construction task; the application realizes dynamic quantitative evaluation of task risks, has the advantages that key bridge tasks can be accurately identified, risk propagation paths can be quantified, and the precision of project management is improved.

Description

Technical Field

[0001] This invention belongs to the field of construction engineering management technology, and in particular relates to an intelligent construction management system and method. Background Technology

[0002] As the construction industry transforms towards digitalization and intelligentization, project management for large-scale construction projects faces increasingly complex challenges. Construction projects typically involve numerous interconnected tasks with strict logical dependencies. Delays in any single task can trigger a chain reaction, impacting the entire project schedule. In practical project management, the topology of task networks is complex and varied, with different tasks playing different roles. Some tasks, while not on the traditional critical path, may act as crucial bridges connecting multiple sub-networks, and their delay risks are easily overlooked by existing methods.

[0003] Currently, project management in construction projects primarily employs traditional methods such as the Critical Path Method (CPM) and Program Review and Evaluation Technique (PRP). Some systems use Gantt charts and milestone nodes for schedule control, estimating task durations based on historical data; others use simple risk assessment matrices to conduct independent risk assessments for individual tasks. In terms of information management, existing technologies largely focus on schedule tracking and resource allocation, lacking in-depth analysis of the complex relationships between tasks. The CPM only focuses on the sequence of tasks on the critical path, failing to fully consider the structural importance of tasks within the network. It cannot identify task nodes that, while not on the critical path, play a crucial bridging role; problems at these nodes can disrupt the connectivity of the entire task network. Existing risk assessment methods are mostly based on independent analysis of single tasks, ignoring the cascading effects of risk propagation through task dependencies. For example, the risk of delays in preceding tasks can spread along directed edges to subsequent tasks, creating a cascading impact. Most management systems lack the ability to quantitatively model the dynamic risk propagation process, making it impossible to accurately predict the impact of local risks on the overall project schedule. This makes it difficult for project managers to distinguish between high-impact and low-impact tasks and to develop targeted risk response strategies. Furthermore, existing methods fail to combine the topological characteristics of the task network with external influencing factors, making it impossible to dynamically track risk propagation paths and iteratively calculate comprehensive risk values, resulting in deviations between risk assessment results and actual engineering conditions. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent construction management system and method that solves the aforementioned problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent construction management method, specifically comprising the following steps:

[0006] Constructing a construction task relationship network for construction projects; the construction task relationship network refers to a project planning quantification model based on directed graph theory, specifically by abstracting construction tasks as nodes and the logical dependencies between tasks as directed edges, thereby constructing a computable and analyzable complex network topology.

[0007] Obtain the feature vector of the influencing factors of the construction task, input the feature vector of the influencing factors of the construction task into the machine learning model, and generate the initial delay probability of the construction task;

[0008] Based on the construction task relationship network, the connection paths between construction tasks are parsed out, and the number of paths, path lengths, and the number of preceding construction tasks of each construction task are obtained.

[0009] Based on the number of connecting paths, path length, and the number of preceding construction tasks, a construction task impact analysis model is established to generate construction task impact factors.

[0010] An iterative analysis model is established based on the initial delay probability of the construction task and the impact factor of the construction task to generate a comprehensive risk value for the construction task; wherein, the initial delay probability of the construction task is the initial comprehensive risk value of the construction task in the iterative analysis model.

[0011] The construction task is managed based on its comprehensive risk value.

[0012] Based on the above technical solutions, the present invention also provides the following optional technical solutions:

[0013] Further technical solution: The method for generating the initial delay probability of the construction task specifically includes:

[0014] Through the formula: ;

[0015] Initial delay probability for generating construction tasks ;

[0016] In the formula, This indicates a construction task. The eigenvectors of influencing factors at time t For machine learning models.

[0017] Further technical solution: The method for generating the construction task influence factor specifically includes the following steps:

[0018] Generate the bridge factor for the construction task based on the number and length of the connecting paths;

[0019] Generate a PageRank value based on the number of prerequisite construction tasks for the construction task;

[0020] An impact analysis model for construction tasks is established based on the bridge factor and PageRank value of the construction task, generating the influence factor of the construction task.

[0021] Further technical solution: The specific method for generating the bridge factor in the construction task includes:

[0022] Through the formula: ;

[0023] Generate bridge factors for construction tasks ;

[0024] In the formula, This represents the relationship between construction task nodes in the construction task relationship network. To the construction task node The total number of shortest paths, This represents the total number of shortest paths. After completing the construction task node The number of paths, where n is the number of construction task nodes in the construction task relationship network; the formula is subject to the following conditions: s≠t, s≠i, t≠i; these conditions mean that the construction task nodes s, t, and i in the construction task relationship network are not the same construction task nodes.

[0025] Further technical solutions: The specific methods for generating the PageRank value include:

[0026] Through the formula: ;

[0027] Generate PageRank value ;

[0028] In the formula, d represents the damping coefficient, and n represents the number of construction task nodes in the construction task relationship network. This refers to all nodes that point to the construction task. A collection of prerequisite construction task nodes. This represents the collection of prerequisite construction task nodes. The j-th pre-construction task node, where m represents the set of pre-construction task nodes. The number of pre-construction task nodes. This represents the collection of prerequisite construction task nodes. Pre-construction task nodes PageRank value, This represents the collection of prerequisite construction task nodes. Pre-construction task nodes Out-degree refers to the out-degree of the preceding construction task node in the construction task relationship network. The number of all directed edges that are radiated outwards.

[0029] Further technical solution: The expression of the construction task impact analysis model is specifically as follows: ;

[0030] In the expression, This represents the influence factor of the construction task. This represents the bridge construction factor. The normalized value, This represents the PageRank value. The normalized value, , All are weighting coefficients, and .

[0031] Further technical solutions: The method for generating the comprehensive risk value of the construction task specifically includes:

[0032] Based on the initial delay probability of the construction task and the impact factor of the construction task, a comprehensive risk probability of the construction task not being delayed is generated;

[0033] An iterative analysis model is established based on the comprehensive risk probability of the construction task not being delayed, and a comprehensive risk value for the construction task is generated.

[0034] Further technical solutions: The specific method for generating the comprehensive risk probability of the construction task not being delayed includes:

[0035] Through the formula: ;

[0036] Generate construction tasks without delays - comprehensive risk probability ;

[0037] In the formula, This represents the construction task node. The overall risk value of the construction task after the kth iteration. This represents the construction task node. Pre-construction task nodes The overall risk value of the construction task after the k-th iteration. This represents the construction task node. Construction task influence factors This represents the propagation attenuation factor. This refers to all nodes that point to the construction task. The set of prerequisite construction task nodes, where m represents the set of prerequisite construction task nodes. The number of pre-construction task nodes.

[0038] A further technical solution: The expression of the iterative analysis model is specifically as follows: ;

[0039] In the expression, This represents the overall risk value of the construction task. This represents the overall risk probability of the construction task not being delayed.

[0040] An intelligent construction management system, which executes the aforementioned intelligent construction management method, specifically includes:

[0041] The network construction unit is used to construct the construction task relationship network of a construction project. The construction task relationship network refers to a project planning quantification model based on directed graph theory. Specifically, it abstracts construction tasks into nodes and the logical dependencies between tasks into directed edges, thereby constructing a computable and analyzable complex network topology.

[0042] The initial delay probability analysis unit is used to obtain the feature vector of the influencing factors of the construction task, and input the feature vector of the influencing factors of the construction task into the machine learning model to generate the initial delay probability of the construction task.

[0043] The data acquisition unit is used to parse out the connection paths between construction tasks based on the construction task relationship network, and to obtain the number of paths, path length, and number of preceding construction tasks of each construction task.

[0044] The impact analysis unit is used to establish an impact analysis model for construction tasks based on the number of connecting paths, path length, and the number of preceding construction tasks, and to generate impact factors for construction tasks.

[0045] The risk comprehensive analysis unit is used to establish an iterative analysis model based on the initial delay probability of the construction task and the influence factors of the construction task, and generate a comprehensive risk value for the construction task; wherein, the initial delay probability of the construction task is the initial comprehensive risk value of the construction task in the iterative analysis model.

[0046] The management unit is used to manage construction tasks based on the comprehensive risk value of the construction task.

[0047] This invention provides an intelligent construction management system and method, which has the following advantages compared with the prior art:

[0048] This invention constructs a construction task relationship network based on directed graph theory, generates initial delay probabilities by combining a machine learning model, and establishes an influence factor and iterative risk analysis model using network topology features. This enables dynamic quantitative assessment of task risks and has the advantages of accurately identifying key bridge tasks, quantifying risk propagation paths, and improving the accuracy of project management. Attached Figure Description

[0049] Figure 1This is a flowchart illustrating an intelligent construction management method provided by the present invention.

[0050] Figure 2 This is a structural diagram of an intelligent construction management system provided by the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0052] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0053] Please see Figure 1 An intelligent construction management method provided in one embodiment of the present invention includes the following steps:

[0054] Step S10: Construct the construction task relationship network of the construction project; the construction task relationship network refers to the project planning quantification model based on directed graph theory. Specifically, it abstracts construction tasks into nodes and the logical dependencies between tasks into directed edges, thereby constructing a computable and analyzable complex network topology.

[0055] Step S20: Obtain the feature vector of influencing factors of the construction task, and input the feature vector of influencing factors of the construction task into the machine learning model to generate the initial delay probability of the construction task; wherein, the feature vector of influencing factors includes future rainfall, number of workers on site, failure rate of core equipment, material inventory level, etc.; the machine learning model includes, but is not limited to, logistic regression, decision tree, etc.; this machine learning model is existing technology and will not be described in detail here.

[0056] Step S30: Based on the construction task relationship network, parse out the connection paths between construction tasks, and obtain the number of paths, path length, and number of prerequisite construction tasks for each construction task.

[0057] Step S40: Based on the number of connecting paths, path length, and the number of preceding construction tasks, establish a construction task impact analysis model and generate construction task impact factors;

[0058] Step S50: Establish an iterative analysis model based on the initial delay probability of the construction task and the influence factor of the construction task to generate a comprehensive risk value for the construction task; wherein, the initial delay probability of the construction task is the initial comprehensive risk value of the construction task in the iterative analysis model.

[0059] Step S60: Manage the construction task based on the comprehensive risk value of the construction task;

[0060] The influencing factor feature vector refers to a set of multi-dimensional variables describing the impact of construction task delays. It can be constructed using data such as future rainfall data obtained from meteorological monitoring systems, the number of on-site workers recorded by human resource management systems, the failure rate of core equipment in equipment maintenance logs, and material inventory levels provided by inventory management platforms. Specifically, this data can be integrated into a feature vector through database interfaces or manual input. In practical applications, machine learning models are used to generate the initial probability of construction task delays. This can be implemented using algorithms such as random forests or neural networks. For example, the model can be trained using historical project data to predict the likelihood of task delays. This process primarily aims to transform the influencing factor feature vector into a quantitative risk indicator.

[0061] The number and length of connected paths refer to the topological characteristics of task dependencies. The number of paths can be calculated using a breadth-first search algorithm, or the path length can be determined using Dijkstra's algorithm, thereby quantifying the tightness of connections and the depth of dependencies between tasks.

[0062] The construction task impact analysis model is used to generate construction task impact factors, which can be calculated based on the centrality index of the task in the network. For example, degree centrality can be used to reflect the number of direct connections of the task, or proximity centrality can be used to measure the average distance of the task to other nodes, thereby assessing the risk amplification effect of the task on the overall network.

[0063] Iterative analysis models simulate the dynamic process of risk propagation along task dependencies. They can be implemented using a recursive calculation framework, such as gradually updating the current task risk based on the risk values ​​of preceding tasks and propagation weights, thereby generating a comprehensive risk value for the construction task.

[0064] This application effectively addresses the problems of insufficient analysis of the importance of task network structure, neglect of the chain effect of risk propagation, and lack of dynamic risk propagation quantification capabilities in construction project management through the aforementioned technical means. Specifically, the construction of the construction task relationship network enables accurate modeling of task dependencies, avoiding the shortcomings of traditional critical path methods that only focus on the critical path while ignoring structural importance; the dynamic acquisition of influencing factor feature vectors combined with machine learning models achieves real-time quantification of the impact of external condition changes on task delays; the parsing of connection path parameters provides structured input for task influence analysis; the influence analysis model generates influence factors through network topology features, comprehensively assessing the risk propagation potential of tasks in the network; and the iterative analysis model quantifies the diffusion effect of local risks on the overall project based on the chain-like transmission of risks through dependency relationships. Therefore, this embodiment systematically integrates task network structure analysis and risk propagation modeling, enabling risk management to shift from passive response to proactive intervention. It allows for the early deployment of resources or adjustment of plans for high-risk tasks, thereby preventing local risks from spreading through the task dependency network and affecting the overall project schedule.

[0065] In implementing intelligent construction management methods, the first step is to construct a network of construction tasks for the construction project. This network, based on directed graph theory, forms a quantitative model of the project plan. Construction tasks are abstracted as network nodes, and the logical dependencies between tasks are abstracted as directed edges, thus constructing a computationally calculable and analyzable complex network topology. This allows for the precise capture of the location attributes of construction tasks within the network, identifying construction task nodes that, while not on the critical path, play a crucial bridging role, overcoming the shortcomings of traditional critical path methods that focus solely on the critical path while neglecting structural importance. Furthermore, feature vectors representing influencing factors of construction tasks are obtained, including dynamic environmental and resource variables such as future rainfall, the number of on-site workers, the failure rate of core equipment, and material inventory levels. These feature vectors are then input into a machine learning model to generate initial delay probabilities. For example, in a high-rise residential construction project, the feature vectors representing influencing factors for the main structure construction task include seven-day rainfall forecasts, real-time on-site worker attendance records, tower crane operating status monitoring information, and steel reinforcement material inventory levels. These are input into a logistic regression model to generate initial delay probabilities. This process avoids the static estimation that traditional methods rely solely on historical data, and can more accurately reflect the immediate impact of changes in external conditions on mission delays.

[0066] Based on the construction task relationship network, the connection paths between construction tasks are analyzed, obtaining the number of paths, path lengths, and the number of preceding construction tasks for each construction task. These parameters quantify the connection tightness and dependency depth between tasks according to the network topology. The number and length of paths directly reflect the potential of a task as a risk propagation hub, while the number of preceding tasks indicates the degree to which a task is influenced by upstream factors. Specifically, in the aforementioned high-rise residential project, the preceding construction tasks for interior decoration include wall construction and water and electricity installation. The number of paths, path lengths, and the number of preceding construction tasks for each connection path are analyzed. Subsequently, based on the number and length of paths and the number of preceding construction tasks, a construction task impact analysis model is established to generate a construction task impact factor. This impact factor comprehensively considers the structural importance of tasks in the network and can comprehensively assess the risk amplification effect of tasks on the overall network. Its unique feature is that it captures the unique role of tasks as bridges for information or risk propagation and the cumulative influence within dependency chains.

[0067] An iterative analysis model is established based on the initial delay probability and impact factors of the construction task to generate a comprehensive risk value for the construction task. The initial delay probability serves as the initial value for the iterative analysis model. During the iteration process, the current task risk is dynamically updated based on the risk values ​​of preceding tasks and impact factors, simulating the decay and diffusion of risk along dependent paths. This effectively quantifies the chain reaction of risk propagation, overcoming the limitation of traditional independent risk assessments that ignore the dynamic impact between tasks. Finally, construction tasks are managed based on the comprehensive risk value, shifting risk management from passive response to proactive intervention, enabling the early deployment of resources or adjustment of plans for high-risk tasks.

[0068] This method systematically addresses the problems in construction project management, such as insufficient analysis of the importance of task network structures, neglect of the chain effect of risk propagation, and lack of dynamic risk propagation quantification capabilities, by integrating task network structure analysis and risk propagation modeling. Specifically, the constructed construction task relationship network provides the mathematical basis for quantifying the complex dependencies between tasks, the generation mechanism of influence factors comprehensively assesses the structural importance of tasks, and the iterative analysis model realizes the dynamic quantification of risk propagation.

[0069] As one implementation method, this approach can accurately identify high-risk task nodes, providing a scientific basis for project management and thus preventing local risks from spreading through the task dependency network and affecting the overall project progress.

[0070] In some of the embodiments described above in this application, an initial delay probability for the construction task is proposed to quantify the task risk. However, in its implementation, the generation of the initial delay probability lacks a clear mathematical expression and a calculable quantification mechanism, making it difficult to accurately capture the changes in dynamic influencing factors in the risk propagation iterative analysis. This makes the calculation of the comprehensive risk value of the subsequent task dependent on the fuzzy model output, which cannot effectively support the accurate assessment of the overall project risk.

[0071] In response, this invention further proposes a method for generating the initial delay probability of the construction task, specifically including:

[0072] Through the formula: ;

[0073] Initial delay probability for generating construction tasks ;

[0074] In the formula, This indicates a construction task. The eigenvectors of influencing factors at time t For machine learning models;

[0075] Among them, the initial delay probability of the construction task This refers to the construction task. A quantitative indicator of the likelihood of delays can be expressed as a probability value, with the aim of providing a basic input for risk assessment.

[0076] The influencing factor eigenvector refers to the feature vector describing the construction task at time t. The multidimensional data set of the status can be realized in vector form, which includes parameters such as future rainfall, number of on-site workers, failure rate of core equipment, and material inventory level. The purpose is to integrate dynamic environmental factors to reflect real-time construction conditions.

[0077] A machine learning model is an algorithmic model that maps input features to output probabilities. It can be implemented using machine learning algorithms such as logistic regression, decision trees, random forests, or neural networks. The goal is to capture the complex relationships between features through nonlinear processing and improve prediction accuracy.

[0078] Specifically, after obtaining the feature vectors of the influencing factors of the construction task at time t, the system inputs them into a pre-trained machine learning model. This model performs the transformation from features to probabilities based on the nonlinear mapping relationship established by historical construction data and outputs the initial delay probability. This process ensures the real-time quantitative processing of dynamic construction environment data, avoids the lag defect of static historical data in traditional methods, and effectively captures the interaction between multiple factors through the nonlinear capability of the model, providing accurate and verifiable initial input for iterative risk analysis.

[0079] As a specific implementation method, the influencing factor feature vector may specifically include the predicted rainfall for the next 24 hours, the current number of workers on site, the operating status indicators of key equipment, and the inventory level of major materials.

[0080] Through the above technical solution, the generation of the initial delay probability has a clear mathematical expression and a calculable quantification mechanism, which can accurately capture the changes in dynamic influencing factors and provide reliable input for the calculation of the comprehensive risk value of subsequent tasks, thereby effectively supporting the accurate assessment of the overall project risk.

[0081] In practical applications, some of the embodiments described above in this application propose a construction task influence factor to quantify the criticality of a task in the network structure. However, in its implementation, the generation of the influence factor only depends on basic parameters such as the number of paths, path length, and the number of preceding construction tasks. It lacks refined modeling of the task bridging role and network centrality, which makes it impossible to effectively identify task nodes that are not on the critical path but have high bridging value, thereby affecting the accuracy of risk propagation analysis.

[0082] In response, this invention further proposes a method for generating the construction task influence factor, specifically including the following steps:

[0083] Step S41: Generate the bridge factor for the construction task based on the number and length of the connecting paths;

[0084] Step S42: Generate a PageRank value based on the number of prerequisite construction tasks for the construction task;

[0085] Step S43: Establish an impact analysis model for the construction task based on the bridge factor and PageRank value, and generate the influence factor of the construction task;

[0086] Among them, the bridge construction factor refers to the indicator that quantifies the importance of a task node as a connection hub in the task network. It can be implemented by statistically analyzing the frequency of task nodes being traversed in all shortest paths, or it can be implemented based on the betweenness centrality calculation method in network flow theory.

[0087] PageRank can be understood as an indicator of the authority of a task node in the global network structure. It can be implemented by iteratively calculating the weight propagation process of task nodes, or by evaluating network centrality through a random walk algorithm.

[0088] The construction task impact analysis model is a mathematical framework that integrates multi-dimensional network features to generate a comprehensive impact index, which can be implemented using a linear weighted fusion method.

[0089] Specifically, the solution in this application quantifies the connectivity of task nodes as network bridges by analyzing the number and length of connection paths in step S41. A large number of paths reflects that the task covers multiple sets of dependencies, while a short path length indicates that the task is located on an efficient propagation path. Step S42 iteratively calculates the PageRank value based on the number of preceding construction tasks, transforming the number of directly preceding tasks into a network centrality indicator, and reflects the hierarchical dependencies between tasks through damping coefficients and out-degree normalization. Step S43 weightedly fuses the normalized construction task bridge factor and PageRank value, so that the bridge factor strengthens the pivotal role of the task in the multi-path risk propagation, and the PageRank value reflects the authority of the task in the global network. The two complement each other to ensure that the influence factor can both identify high-value bridge nodes on non-critical paths and assess the core position of the task in the risk propagation chain, thereby providing structured support for iterative risk analysis.

[0090] As a specific implementation method, the solution of this application is implemented as follows: In the construction task relationship network, for a certain bridge construction task node, the system first counts its mediation frequency in the shortest path of all task pairs to generate a construction task bridge factor; then iteratively calculates the PageRank value based on the number of preceding tasks and the network topology; finally, the normalized construction task bridge factor and the PageRank value are merged according to a preset weight ratio to generate the construction task influence factor of the task node.

[0091] Through the above technical solutions, this application can refine the bridging role and network centrality of the modeling task, effectively identify task nodes that are not on the critical path but have high bridging value, improve the accuracy of risk propagation analysis, and provide a more reliable decision-making basis for risk management of construction tasks.

[0092] In practical applications, some embodiments of this application propose a construction task bridge factor to quantify the bridging role of tasks in a network. However, in its implementation, since it is only roughly estimated based on the number and length of the connecting paths, it cannot accurately capture the criticality of task nodes as intermediaries in the task-dependent paths. This results in the inability to identify core nodes that are not on the critical path but connect different task subnets, thus distorting the calculation of the construction task influence factor and affecting the reliability of risk propagation analysis.

[0093] In response, this invention further proposes a specific method for generating the bridge factor in the construction task, including:

[0094] Through the formula: ;

[0095] Generate bridge factors for construction tasks ;

[0096] In the formula, This represents the relationship between construction task nodes in the construction task relationship network. To the construction task node The total number of shortest paths, This represents the total number of shortest paths. After completing the construction task node The number of paths, where n is the number of construction task nodes in the construction task relationship network; the formula is subject to the following conditions: s≠t, s≠i, t≠i; these conditions mean that construction task nodes s, t, and i in the construction task relationship network are not the same construction task node.

[0097] Among them, the construction task bridge factor refers to the key indicator that quantifies the role of task nodes as intermediaries in connecting different task subnets in the construction task relationship network. It can be implemented by using the calculation method based on the shortest path intermediate centrality, with the aim of accurately identifying the core nodes that have a bridging role.

[0098] The total number of shortest paths refers to the total number of all shortest paths from task nodes to task nodes in the task relationship network, which aims to reflect the global topology of task dependencies.

[0099] The number of paths passing through a specific node refers to the number of paths that pass through task nodes out of the total number of shortest paths. This can be achieved by recording the nodes that the path passes through when calculating the shortest path, with the aim of directly measuring the criticality of the node in the path propagation.

[0100] The total number of nodes, n, refers to the total number of construction task nodes in the construction task relationship network. It can be obtained by traversing the task relationship network. The purpose is to define the summation range to ensure the integrity of the calculation.

[0101] Specifically, this technical solution calculates the proportion of shortest paths between all valid node pairs satisfying the conditions s≠t, s≠i, and t≠i that pass through a specific task node, and then sums these proportions across all valid node pairs to accurately quantify the bridging role of task nodes. First, all valid node pairs (s,t) in the construction task relationship network are determined. Then, the bridging role of task nodes is calculated. To the construction task node Total number of shortest paths and the construction task nodes therein Path count To obtain the construction task nodes The proportion of importance is summed over all valid node pairs to obtain the bridge construction task factor. This calculation method avoids the one-sidedness of relying solely on path length or number, and can comprehensively assess the structural importance of nodes in the network, providing a reliable basis for the influence factors of construction tasks.

[0102] As a specific implementation method, the solution of this application is implemented as follows: The system constructs a construction task relationship network containing multiple construction task nodes, and uses an adjacency list to represent the logical dependencies between tasks; for each task node, the system traverses all other task node pairs (s,t), uses a breadth-first search algorithm to calculate the total number of shortest paths, and counts the number of paths passing through construction task nodes. The number of paths; then calculate the path ratio of each node pair and sum them to obtain the construction task node. The system calculates the bridge factor value for each construction task; this value is then used to calculate the influence factor of the construction task. During this process, the system can employ parallel computing techniques to accelerate the calculation of the shortest path, improving the efficiency of handling large-scale construction task networks.

[0103] Through the above scheme, this application can accurately capture the criticality of task nodes as intermediaries in the task dependency path, effectively identify core nodes that are not on the critical path but connect different task subnets, thereby ensuring the accuracy of the construction task influence factor calculation and improving the reliability of risk propagation analysis.

[0104] Specifically, in some of the embodiments described above in this application, a PageRank value is proposed to quantify the structural importance of construction tasks in the network. However, in its implementation, the method of generating PageRank values ​​based solely on the number of preceding construction tasks fails to fully consider the out-degree of the task and the influence of the global network structure, resulting in inaccurate calculation of the construction task influence factor and an inability to effectively capture the propagation path and extent of risk in the task-dependent network.

[0105] In response, this invention further proposes a method for generating the PageRank value, specifically including:

[0106] Through the formula: ;

[0107] Generate PageRank value ;

[0108] In the formula, d represents the damping coefficient, and n represents the number of construction task nodes in the construction task relationship network. This refers to all nodes that point to the construction task. A collection of prerequisite construction task nodes. This represents the collection of prerequisite construction task nodes. The j-th pre-construction task node, where m represents the set of pre-construction task nodes. The number of pre-construction task nodes. This represents the collection of prerequisite construction task nodes. Pre-construction task nodes PageRank value, This represents the collection of prerequisite construction task nodes. Pre-construction task nodes Out-degree refers to the out-degree of the preceding construction task node in the construction task relationship network. The number of all directed edges that radiate outwards;

[0109] The damping coefficient d can be understood as a parameter that controls the probability of random jumps. Its value can be dynamically adjusted through empirical data or optimization algorithms. Its purpose is to balance the global random jumps and local structural dependencies in the task network and avoid local biases in the calculation process.

[0110] n represents the total number of construction task nodes in the construction task relationship network. Its value is determined by the task scale of the specific project. Specifically, it can be obtained by traversing the node set of the construction task relationship network. Its purpose is to provide a uniform initial distribution basis for the PageRank value and eliminate the influence of node size on the calculation.

[0111] This can be understood as all nodes that directly point to the current construction task node. The set of preceding task nodes is determined by the topology of the task relationship network, which can be determined by parsing the logical dependencies between tasks. Its purpose is to limit the scope of the directly preceding tasks that affect the current task and ensure that the calculation is strictly based on the actual dependencies.

[0112] Indicates the prerequisite construction task node The out-degree, i.e., the number of subsequent tasks connected to this node, can be determined by statistically analyzing the construction task relationship network. The purpose of obtaining the number of directed edges is to quantify the degree of decay of the task's influence during the propagation process, so that the influence of highly connected tasks can be reasonably distributed.

[0113] Specifically, the scheme in this application uses the damping coefficient d and the out-of-range The synergistic effect of these factors enables precise quantification of the importance of nodes in the construction task network. The first term in the formula... The first term represents the probability of randomly jumping to any node, ensuring that all nodes have basic importance; the second term, d, multiplied by the summation term, represents the transition probability based on the network structure, where each preceding task node... Its contribution to subsequent tasks was measured by its degree. Normalization ensures that the influence decays proportionally to the out-degree as it propagates along task-dependent paths. This design comprehensively considers the global position and local connectivity characteristics of tasks in the network, avoiding evaluation bias caused by calculating PageRank values ​​solely based on the number of preceding tasks, and providing a mathematical basis for accurately capturing risk propagation paths.

[0114] Through the above technical solutions, this application can more accurately quantify the structural importance of construction tasks in the network, effectively capture the propagation path and degree of risks in task-dependent networks, thereby improving the calculation accuracy of the construction task influence factor and providing a reliable basis for subsequent risk assessment.

[0115] In some of the embodiments described above in this application, a construction task impact analysis model is proposed to generate construction task impact factors. However, in its implementation, the specific expression of the model is not defined, which leads to the bridge factor and PageRank value being unable to be effectively integrated due to differences in dimensions and the lack of importance weights. This affects the accuracy and adaptability of the impact factor calculation, thereby weakening the ability to identify key nodes in the task network.

[0116] In response, this invention further proposes the following expression for the construction task impact analysis model: ;

[0117] In the expression, This represents the influence factor of the construction task. This represents the bridge construction factor. The normalized value, This represents the PageRank value. The normalized value, , All are weighting coefficients, and ;

[0118] Specifically, the construction task influence factor refers to the technical indicator used to quantify the comprehensive influence of construction tasks in the network. It can be achieved by weighting the normalized bridge factor and the PageRank value. Its purpose is to provide a unified standard for evaluating the influence of tasks.

[0119] The normalized value of the bridge factor in the construction task refers to the result of transforming the original bridge factor to a standard interval through linear transformation. It can be achieved by methods such as minimum-maximum normalization or probability scaling. Its purpose is to eliminate the dimensional differences of the original bridge factor caused by the discreteness of path count.

[0120] The normalized value of PageRank refers to the result of transforming the original PageRank value to a standard range through standardization. This can be achieved by dividing by the maximum value or by Z-score transformation, with the aim of making PageRank values ​​comparable to other metrics.

[0121] The weighting coefficient is a technical parameter used to adjust the contribution ratio of the bridge factor and PageRank value in the influence factor. It can be set by expert experience or optimized based on historical data. Its purpose is to flexibly adjust the evaluation focus according to the specific engineering network characteristics.

[0122] In detail, the scheme in this application eliminates the dimensional differences between the bridge factor and the PageRank value through normalization, enabling them to be compared and integrated on a unified scale. Since the bridge factor reflects the importance of a task node as a bridge between different subnets, while the PageRank value reflects the cumulative influence of a task node in the dependency chain, the normalized indicators are combined linearly with weights to form a comprehensive influence factor. Furthermore, the introduction of weight coefficients allows for dynamic adjustment of the contribution ratio of both based on the specific network characteristics of the construction project. For example, the weight of the bridge factor can be increased in scenarios with highly interconnected task networks, while the weight of the PageRank value can be increased in scenarios with distinct task hierarchies, thus ensuring that the influence factor accurately captures the true importance of task nodes in the network. The design of this expression achieves comparability and synergy among influence indicators of different dimensions, avoiding the problem of a single indicator dominating the results.

[0123] As a specific embodiment, the solution of this application is implemented as follows: Consider a construction task relationship network for a large commercial complex construction project, which includes task nodes for multiple construction stages. For a specific task node, firstly, its original bridge factor value and PageRank value are calculated, and then the min-max normalization method is applied to transform both to the [0,1] interval. Based on the project network characteristic analysis, the weight coefficients α and β are set to 0.65 and β to 0.35, respectively, and the influence factor of the task node is calculated. This influence factor value can be used to identify key task nodes in the network, such as those bridge tasks connecting the structural construction and electromechanical installation subnets, thereby strengthening resource allocation and risk monitoring in a targeted manner.

[0124] The above technical solutions effectively solve the evaluation bias problem caused by the difference in dimensions and the lack of importance weights in the bridge factor and PageRank value, improve the accuracy and adaptability of the construction task influence factor calculation, thereby enhancing the ability to identify key nodes in the task network and providing a reliable foundation for subsequent risk propagation analysis.

[0125] Specifically, in some of the embodiments described above in this application, a method for generating a comprehensive risk value for construction tasks is proposed to assess project risks. However, in its implementation, there is a lack of a precise modeling mechanism for the dynamic process of risk propagation, which makes it impossible to effectively capture the effect of risk chain propagation between tasks, thereby affecting the accuracy of the comprehensive risk assessment.

[0126] In response, this invention further proposes a method for generating the comprehensive risk value of the construction task, specifically including:

[0127] Step S51: Generate the comprehensive risk probability of the construction task not being delayed based on the initial delay probability and the construction task influence factor;

[0128] Step S52: Establish an iterative analysis model based on the comprehensive risk probability of the construction task not being delayed, and generate a comprehensive risk value for the construction task;

[0129] Among them, the comprehensive risk probability of construction task not being delayed refers to the probability state representation of the task being completed on time. Specifically, it can be calculated jointly based on the task's own delay risk and the network topology correlation effect. Its purpose is to quantify the probability distribution of the task completion state and provide an iterative mathematical basis for the risk propagation process.

[0130] The iterative analysis model can be understood as a computational framework that simulates risk evolution through multiple rounds of state updates. Its purpose is to accurately characterize the temporal characteristics and cumulative effects of risk propagation from upstream tasks to downstream tasks, avoiding the shortcomings of traditional static assessment methods in handling risk chain reactions.

[0131] Specifically, the solution in this application couples the initial delay probability of the construction task with the construction task influence factor in step S51. The initial delay probability represents the inherent delay risk of the task itself, while the construction task influence factor quantifies the structural importance of the task in the network topology. Based on this, step S52 constructs an iterative analysis model based on the generated comprehensive risk probability of the construction task not being delayed. This model is updated progressively through multiple rounds of calculation. This iterative process dynamically simulates the propagation behavior of risk along the task-dependent path. Each iteration uses the risk state of the previous round as input, accurately calculating the attenuation coefficient and amplification effect along the risk propagation path, thus forming a complete dynamic representation of the risk chain reaction. Since the comprehensive risk probability of the construction task not being delayed is directly related to the task completion status, its iterative update mechanism can effectively reflect the convergence characteristics and steady-state distribution of risk propagation, thereby generating an accurate comprehensive risk value for the construction task.

[0132] As a specific implementation method, the solution of this application is implemented as follows: In the progress management of a super high-rise building project, for the core tube concrete pouring task, the system first obtains the initial delay probability and influence factor of the task. The initial delay probability is generated by a machine learning model based on real-time meteorological data and equipment status feature vectors, and the influence factor is calculated by the bridge factor and PageRank value of the task relationship network. Subsequently, the system generates a comprehensive risk probability of no delay in the construction task based on the two, and establishes an iterative analysis model based on this. During the iteration process, the system continuously calculates the risk propagation impact of upstream formwork and rebar binding tasks, and through multiple rounds of status updates until the risk value converges, finally generating a comprehensive risk value for the concrete pouring task, providing a decision-making basis for the allocation of construction resources.

[0133] Through the above scheme, this application achieves accurate modeling of the dynamic process of risk propagation, effectively captures the effect of risk chain propagation between tasks, and improves the accuracy of comprehensive risk assessment.

[0134] Specifically, in some of the solutions mentioned above in this application, a comprehensive risk probability of construction tasks not being delayed is proposed to establish an iterative analysis model to generate a comprehensive risk value for construction tasks. However, in this process, the generation method lacks precise quantification of the risk propagation mechanism between tasks and fails to combine the dynamic interaction of the risk status of the preceding tasks, the influence factors of the tasks, and the propagation attenuation factors. This results in the inability to accurately simulate the chain propagation effect of risks through dependencies, making it difficult for the comprehensive risk assessment to reflect the real impact of local risks on the overall project schedule, thereby affecting the reliability of management decisions.

[0135] In response, this invention further proposes a method for generating the comprehensive risk probability of the construction task not being delayed, specifically including:

[0136] Through the formula: ;

[0137] Generate construction tasks without delays - comprehensive risk probability ;

[0138] In the formula, This represents the construction task node. The overall risk value of the construction task after the kth iteration. This represents the construction task node. Pre-construction task nodes The overall risk value of the construction task after the k-th iteration. This represents the construction task node. Construction task influence factors This represents the propagation attenuation factor. This refers to all nodes that point to the construction task. The set of prerequisite construction task nodes, where m represents the set of prerequisite construction task nodes. The number of pre-construction task nodes;

[0139] Among them, the comprehensive risk probability of construction tasks not being delayed refers to a dynamic indicator that quantifies the possibility of tasks being completed on time in the task dependency network. It can be implemented using the conditional probability model in probability theory, with the aim of providing real-time basic data for risk assessment.

[0140] This refers to the construction task node. The comprehensive risk value after the kth iteration can be determined by a risk assessment algorithm based on historical risk data and current status of the task, with the aim of dynamically reflecting the evolution of task risk.

[0141] This refers to the prerequisite construction task node. The comprehensive risk value after the kth iteration can be inherited from the result of the previous iteration, with the aim of capturing the dynamic changes in the risk level of the preceding task.

[0142] It refers to the construction of task influence factors, which aim to quantify the criticality of a task in the network;

[0143] This refers to the propagation attenuation factor, which can be set to a positive number less than 1. Its purpose is to adjust the degree of attenuation of the risk propagation intensity and avoid the excessive spread of risk.

[0144] This refers to all nodes pointing to the construction task node. The set of prerequisite construction task nodes can be obtained by parsing the construction task relationship network, with the aim of identifying all prerequisite tasks that affect the current task;

[0145] m refers to the number of nodes in the set of preceding construction task nodes, which can be determined by counting, with the aim of defining the calculation range of the product term.

[0146] Specifically, the solution in this application dynamically couples the current task's own risk status with the risk propagation contribution of previous tasks, constructing an iterative generation mechanism for the comprehensive risk probability of a construction task not being delayed. The formula... The item directly represents the current construction task node. The impact of the risk state after the k-th iteration on the probability of no delay is considered to ensure that the risk assessment reflects the real-time evolution of the task's risk. Simultaneously, the product term... The combined effect of all risks from preceding tasks propagating to the current task through dependencies was quantified, among which... Dynamically capture changes in the risk level of preceding tasks. Amplifying the risk propagation effect of high-impact missions This allows for precise control over the attenuation of risk propagation. This design enables accurate modeling of the risk propagation mechanism, gradually converging through an iterative process to accurately simulate the chain reaction of risk propagation in task-dependent networks.

[0147] As a specific implementation method, the solution of this application is implemented as follows: In a large-scale bridge construction project, considering the concrete pouring task, its prerequisite tasks include rebar tying and formwork installation. During the iterative analysis, the system obtains the comprehensive risk value of the concrete pouring task in the previous iteration, as well as the comprehensive risk values ​​of the two prerequisite tasks in the previous iteration. The system calls the construction task influence factor and the preset propagation attenuation factor of the concrete pouring task, and generates the non-delayed comprehensive risk probability of the concrete pouring task by calculating the product of the complement of the current task's own risk state and the risk propagation contribution of all prerequisite tasks. This process dynamically integrates the risk level of prerequisite tasks, the structural importance of tasks in the network, and the risk propagation attenuation characteristics, accurately quantifying the propagation effect of risk through dependencies.

[0148] Through the above technical solution, this application achieves precise quantification of the risk propagation mechanism between tasks. It can combine the dynamic interaction of the risk status of the preceding task, the task influence factor and the propagation attenuation factor to accurately simulate the chain propagation effect of risk through the dependency relationship. This enables the comprehensive risk assessment to truly reflect the impact of local risks on the overall project schedule, thereby improving the reliability of construction task management decisions.

[0149] In practical applications, some of the embodiments described above in this application propose an iterative analysis model to generate a comprehensive risk value for construction tasks. However, in this process, the specific mathematical expression of the iterative analysis model is not clearly defined, resulting in a lack of precise mathematical basis for the quantitative analysis of risk propagation. This makes it impossible to effectively capture the dynamic propagation effect of risk in the task dependency network, and thus difficult to accurately assess the chain reaction of local task delays on the overall project schedule.

[0150] In response, this invention further proposes the following specific expression for the iterative analysis model: ;

[0151] In the expression, This represents the overall risk value of the construction task. This represents the overall risk probability of the construction task not being delayed;

[0152] Among them, the comprehensive risk value of the construction task refers to the risk value of the construction task in the (k+1)th iteration. The overall probability of a delay can be expressed as a probability value, ranging from 0 to 1, with the aim of quantifying the overall risk level of a task delay.

[0153] The overall risk probability of a construction task not being delayed refers to the probability of a construction task being delayed in the (k+1)th iteration. The overall probability of no delay can be expressed as a probability value, ranging from 0 to 1. Its purpose is to serve as an intermediate variable in risk propagation calculations, and to accurately derive the overall risk value through its complement relationship.

[0154] Specifically, the solution in this application establishes a complementary relationship between the comprehensive risk probability of the construction task not being delayed and the comprehensive risk value of the construction task, thereby achieving precise quantification of risk propagation. During the iteration process, the comprehensive risk probability of the construction task not being delayed is first calculated based on the risk status of the preceding tasks. Then, it is directly converted into a comprehensive risk value through the expression of the iterative analysis model. This allows the risk assessment to dynamically reflect the propagation effect of the risks of the preceding tasks, thus updating the task risk status in each iteration and forming a continuous risk propagation chain.

[0155] As a specific implementation method, the solution of this application is implemented as follows: Considering the construction task During the iteration process, the overall risk probability of the construction task not being delayed is calculated by the risk status of the preceding task; subsequently, this value is used to determine the overall risk value of the construction task. For example, when the overall risk probability of the construction task not being delayed is high, the overall risk value of the construction task is correspondingly low, indicating that the risk of the task being delayed is small.

[0156] Through the above scheme, this application achieves accurate quantification of the comprehensive risk value of construction tasks, effectively captures the dynamic propagation effect of risks in the task dependency network, and thus can accurately assess the chain reaction of local task delays on the overall project schedule.

[0157] Please see Figure 2 The present invention also proposes an intelligent construction management system, which is used to execute the above-described intelligent construction management method, specifically including:

[0158] The network construction unit 10 is used to construct the construction task relationship network of the construction project. The construction task relationship network refers to the project planning quantification model based on directed graph theory. Specifically, it abstracts construction tasks into nodes and the logical dependencies between tasks into directed edges, thereby constructing a computable and analyzable complex network topology.

[0159] The initial delay probability analysis unit 20 is used to obtain the feature vector of the influencing factors of the construction task, and input the feature vector of the influencing factors of the construction task into the machine learning model to generate the initial delay probability of the construction task.

[0160] The data acquisition unit 30 is used to parse out the connection paths between construction tasks based on the construction task relationship network, and to obtain the number of paths, path length, and number of preceding construction tasks of each construction task.

[0161] The influence analysis unit 40 is used to establish a construction task influence analysis model and generate construction task influence factors based on the number of connecting paths, path length, and the number of preceding construction tasks of the construction task.

[0162] The risk comprehensive analysis unit 50 is used to establish an iterative analysis model based on the initial delay probability of the construction task and the influence factor of the construction task, and generate a comprehensive risk value for the construction task; wherein, the initial delay probability of the construction task is the initial comprehensive risk value of the construction task in the iterative analysis model.

[0163] Management unit 60 is used to manage the construction task based on the comprehensive risk value of the construction task.

[0164] The present invention further proposes that the influence analysis unit 40 specifically includes:

[0165] The bridge factor generation module for construction tasks is used to generate bridge factors for construction tasks based on the number and length of connecting paths.

[0166] The PageRank value generation module is used to generate a PageRank value based on the number of preceding construction tasks of a construction task.

[0167] The comprehensive analysis module is used to establish an impact analysis model for construction tasks based on the bridge factor and PageRank value of the construction task, and to generate the impact factor of the construction task.

[0168] The present invention further proposes that the risk comprehensive analysis unit 50 specifically includes:

[0169] The non-delay probability analysis module is used to generate a comprehensive risk probability of non-delay for a construction task based on the initial delay probability and the construction task influence factor.

[0170] The construction task comprehensive risk value generation module is used to establish an iterative analysis model based on the comprehensive risk probability of the construction task not being delayed, and generate the comprehensive risk value of the construction task.

[0171] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart construction management method, characterized in that, The method specifically includes the following steps: Constructing a construction task relationship network for construction projects; the construction task relationship network refers to a project planning quantification model based on directed graph theory, specifically by abstracting construction tasks as nodes and the logical dependencies between tasks as directed edges, thereby constructing a computable and analyzable complex network topology. Obtain the feature vector of the influencing factors of the construction task, input the feature vector of the influencing factors of the construction task into the machine learning model, and generate the initial delay probability of the construction task; Based on the construction task relationship network, the connection paths between construction tasks are parsed out, and the number of paths, path lengths, and the number of preceding construction tasks of each construction task are obtained. Based on the number of connecting paths, path length, and the number of preceding construction tasks, a construction task impact analysis model is established to generate construction task impact factors. An iterative analysis model is established based on the initial delay probability of the construction task and the impact factor of the construction task to generate a comprehensive risk value for the construction task; wherein, the initial delay probability of the construction task is the initial comprehensive risk value of the construction task in the iterative analysis model. The construction task is managed based on its comprehensive risk value. The specific methods for generating the comprehensive risk value of the construction task include: Based on the initial delay probability of the construction task and the impact factor of the construction task, a comprehensive risk probability of the construction task not being delayed is generated; An iterative analysis model is established based on the comprehensive risk probability of the construction task not being delayed, and a comprehensive risk value of the construction task is generated. The specific methods for generating the comprehensive risk probability of the construction task not being delayed include: Through the formula: ； Generate construction tasks without delays - comprehensive risk probability ; In the formula, This represents the construction task node. The overall risk value of the construction task after the kth iteration. This represents the construction task node. Pre-construction task nodes The overall risk value of the construction task after the k-th iteration. This represents the construction task node. Construction task influence factors This represents the propagation attenuation factor. This refers to all nodes that point to the construction task. The set of prerequisite construction task nodes, where m represents the set of prerequisite construction task nodes. The number of pre-construction task nodes; The expression for the iterative analysis model is as follows: ； In the expression, This represents the overall risk value of the construction task. This represents the overall risk probability of the construction task not being delayed.

2. The intelligent construction management method according to claim 1, characterized in that, The method for generating the initial delay probability of the construction task specifically includes: Through the formula: ； Initial delay probability for generating construction tasks ; In the formula, This indicates a construction task. The eigenvectors of influencing factors at time t For machine learning models.

3. The intelligent construction management method according to claim 1, characterized in that, The method for generating the construction task influence factor specifically includes the following steps: Generate the bridge factor for the construction task based on the number and length of the connecting paths; Generate a PageRank value based on the number of prerequisite construction tasks for the construction task; An impact analysis model for construction tasks is established based on the bridge factor and PageRank value of the construction task, generating the influence factor of the construction task.

4. The intelligent construction management method according to claim 3, characterized in that, The specific methods for generating the bridge factors in the construction task include: Through the formula: ； Generate bridge factors for construction tasks ; In the formula, This represents the relationship between construction task nodes in the construction task relationship network. To the construction task node The total number of shortest paths, This represents the total number of shortest paths. After completing the construction task node The number of paths, where n is the number of construction task nodes in the construction task relationship network; the formula is subject to the following conditions: s≠t, s≠i, t≠i; these conditions mean that the construction task nodes s, t, and i in the construction task relationship network are not the same construction task nodes.

5. The intelligent construction management method according to claim 3, characterized in that, The specific methods for generating PageRank values ​​include: Through the formula: ； Generate PageRank value ; In the formula, d represents the damping coefficient, and n represents the number of construction task nodes in the construction task relationship network. This refers to all nodes that point to the construction task. A collection of prerequisite construction task nodes. This represents the collection of prerequisite construction task nodes. The j-th pre-construction task node, where m represents the set of pre-construction task nodes. The number of pre-construction task nodes. This represents the collection of prerequisite construction task nodes. Pre-construction task nodes PageRank value, This represents the collection of prerequisite construction task nodes. Pre-construction task nodes Out-degree refers to the out-degree of the preceding construction task node in the construction task relationship network. The number of all directed edges that are radiated outwards.

6. The intelligent construction management method according to claim 3, characterized in that, The specific expression of the construction task impact analysis model is as follows: ； In the expression, This represents the influence factor of the construction task. This represents the bridge construction factor. The normalized value, This represents the PageRank value. The normalized value, , All are weighting coefficients, and .

7. An intelligent construction management system, characterized in that, The system is used to execute the intelligent construction management method according to any one of claims 1-6, specifically including: The network construction unit is used to construct the construction task relationship network of a construction project. The construction task relationship network refers to a project planning quantification model based on directed graph theory. Specifically, it abstracts construction tasks into nodes and the logical dependencies between tasks into directed edges, thereby constructing a computable and analyzable complex network topology. The initial delay probability analysis unit is used to obtain the feature vector of the influencing factors of the construction task, and input the feature vector of the influencing factors of the construction task into the machine learning model to generate the initial delay probability of the construction task. The data acquisition unit is used to parse out the connection paths between construction tasks based on the construction task relationship network, and to obtain the number of paths, path length, and number of preceding construction tasks of each construction task. The impact analysis unit is used to establish an impact analysis model for construction tasks based on the number of connecting paths, path length, and the number of preceding construction tasks, and to generate impact factors for construction tasks. The risk comprehensive analysis unit is used to establish an iterative analysis model based on the initial delay probability of the construction task and the influence factors of the construction task, and generate a comprehensive risk value for the construction task; wherein, the initial delay probability of the construction task is the initial comprehensive risk value of the construction task in the iterative analysis model. The management unit is used to manage the construction task based on the comprehensive risk value of the construction task; The risk comprehensive analysis unit specifically includes: The non-delay probability analysis module is used to generate a comprehensive risk probability of non-delay for a construction task based on the initial delay probability and the construction task influence factor. The construction task comprehensive risk value generation module is used to establish an iterative analysis model based on the comprehensive risk probability of the construction task not being delayed, and generate the comprehensive risk value of the construction task.

Images