An engineering project management auxiliary decision-making method based on domain knowledge enhancement

Abstract

The application provides an engineering project management auxiliary decision-making method based on domain knowledge enhancement, and belongs to the technical field of engineering project management. First, the construction plan network diagram and historical delay cause data are acquired, the plan diagram is mapped into an adjacency matrix, and the delay data is converted into a propagation probability table through Bayesian inference. Second, the betweenness centrality is calculated by using the adjacency matrix, and the node attribute is updated to generate a weighted topological graph. Third, the SIR cascade failure simulation model is constructed based on the topological graph and the probability table, the delay diffusion process in the network is simulated, and an influence matrix is generated. Finally, the high-risk target nodes are identified according to the influence matrix, the time buffer value is calculated, and accurate engineering management auxiliary decision-making information is generated. The application can accurately identify the invisible key nodes in the complex engineering network which are prone to cause global paralysis, and automatically quantitatively generate the time buffer strategy against risks, thereby significantly improving the robustness of the engineering progress plan.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of engineering project management, and particularly relates to an engineering project management auxiliary decision-making method based on domain knowledge enhancement. BACKGROUND

[0002] The engineering project management auxiliary decision-making method based on domain knowledge enhancement provides intelligent means for progress control and risk response of complex projects by fusing historical experience data of the engineering industry and modern computing technology. In scenes such as large-scale infrastructure construction and high-end equipment research and development, which have a very high degree of process coupling, the method can effectively assist managers in predicting schedule risks and optimizing resource allocation, and has a broad application prospect.

[0003] In the prior art, engineering project management auxiliary decision-making usually adopts the critical path method or the planning review technique to construct a project progress network, and combines historical engineering data to statistically obtain the probability distribution of the duration of each task. In order to evaluate risks, some existing solutions use the Monte Carlo simulation technique to calculate the completion probability of the total project duration on the premise that the duration fluctuations of each task are independent or only follow a simple linear dependence, thereby prompting managers to pay attention to the task nodes on the critical path.

[0004] However, the above method mainly focuses on the longest path in the time dimension, and often ignores the chain reaction that may be triggered by the delay of some non-critical nodes in a pivotal position in the topology structure of the progress network. The existing static or simple linear simulation cannot accurately reflect the cascading diffusion effect and dynamic decay law of risks in complex process logic, making it difficult to quantitatively evaluate the damage degree of local delay to the global network and develop accurate buffer strategies accordingly. Therefore, there is a technical problem of insufficient engineering project management auxiliary decision-making in the prior art due to the inability to effectively identify and defend against invisible progress risks with high propagation properties in complex engineering networks. SUMMARY

[0005] The purpose of the present application is to provide an engineering project management auxiliary decision-making method based on domain knowledge enhancement to solve the problem of insufficient engineering project management auxiliary decision-making in the prior art.

[0006] To solve the above technical problems, in a first aspect, the present application provides an engineering project management auxiliary decision-making method based on domain knowledge enhancement, comprising:

[0007] obtaining a construction plan network diagram of a target engineering project and delay attribution data of projects of the same type as the target engineering project, the construction plan network diagram comprising a set of task nodes and logical dependency relationships between tasks, and the delay attribution data comprising delay propagation paths and delay probability information of each delayed task represented in a tree-like data structure;

[0008] mapping the task node set as graph nodes and the logical dependency relationship as directed edges to construct an adjacency matrix by graph theory mapping on the construction plan network diagram;

[0009] Based on the delay occurrence frequency of each cascade path in the delay attribution data, the delay conditional probability between different task nodes is calculated through Bayesian inference, and a propagation probability table is generated with the delay conditional probability as the core parameter;

[0010] The betweenness centrality of each graph node is calculated according to the adjacency matrix, and the node attribute value of the associated graph node in the adjacency matrix is updated according to the betweenness centrality, and a topology graph is generated;

[0011] Based on the topology graph, the propagation probability table and the SIR state mapping rule, a cascade failure simulation model is constructed, the construction state corresponding to the graph node in the topology graph is mapped into the object in the SIR model, and the cascade failure simulation model is used for multiple rounds of random simulation. The delay conditional probability in the propagation probability table is set as the infection rate parameter to simulate the diffusion process from the infected type to the susceptible type. The delay diffusion range and decay speed are calculated to generate an impact matrix composed of the initial risk source and the corresponding delay diffusion range and decay speed.

[0012] Extract the target node with a failure impact degree greater than a preset threshold in the impact matrix, and calculate the time buffer value according to the delay coefficient preset according to the task process type according to the delay diffusion characteristics of the target node in the simulation diffusion process of the cascade failure simulation model. Generate management auxiliary decision information for the target engineering project.

[0013] Optionally, by graph theory mapping on the construction plan network diagram, the task node set is mapped as graph nodes and the logical dependency relationship is mapped as directed edges to construct an adjacency matrix, including:

[0014] Traverse the task node set in the construction plan network diagram, take each independent task in the task node set as a graph node, and assign a sequence index to each graph node;

[0015] Extract the pre-graph node and the post-graph node with the front and rear constraint attribute in the logical dependency relationship, and construct a directed edge in the vector connection relationship of the pre-graph node pointing to the post-graph node;

[0016] An initial matrix is constructed according to the total number of graph nodes, the mapping coordinates of the sequence index between the graph nodes are determined through the vector connection relationship of the directed edge, and the matrix elements located at the mapping coordinates in the initial matrix are set as numerical values representing the connected state. Output the adjacency matrix.

[0017] Optionally, the betweenness centrality of each graph node is calculated according to the adjacency matrix, and the node attribute value of the associated graph node in the adjacency matrix is updated according to the betweenness centrality, and a topology graph is generated, including:

[0018] According to the connection relationship represented by the matrix elements in the adjacency matrix, the shortest connection path between any two graph nodes is determined, and the betweenness centrality of each graph node is obtained by calculating the path cumulative value of each graph node covered by the shortest connection path as a transit node;

[0019] The betweenness centrality is converted into a weight parameter, and the weight parameter is associated to the corresponding graph node as the node attribute value of the graph node, and a topology graph is constructed in combination with the adjacency matrix.

[0020] Optionally, based on the delay occurrence frequency of each cascading path in the delay attribution data, the delay conditional probability between different task nodes is calculated by Bayesian inference, and a propagation probability table is generated with the delay conditional probability as the core parameter, including:

[0021] The parent task node and the child task node existing in the cascading relationship are extracted from the delay transmission path included in the delay attribution data, the prior occurrence probability of the parent task node and the joint occurrence probability of the parent task node and the child task node occurring delay are determined by counting the delay occurrence frequency of the corresponding task in the historical samples of the same type of project;

[0022] The ratio of the joint occurrence probability and the prior occurrence probability is calculated to obtain the delay conditional probability;

[0023] The mapping relationship between the parent task node and the child task node is established, and the propagation probability table is constructed based on the mapping relationship and the delay conditional probability between different task nodes.

[0024] Optionally, based on the topology graph, the propagation probability table and the SIR state mapping rule, a cascading failure simulation model is constructed, and the construction state of the graph node in the topology graph is mapped to the object in the SIR model, including:

[0025] The SIR model including susceptible type, infected type and removed type is established, the graph node in the topology graph in the execution stage is mapped to the susceptible type, the graph node with delay is mapped to the infected type, and the graph node without delay or completed is mapped to the removed type;

[0026] The directed edge of the topology graph is traversed, the delay conditional probability corresponding to the directed edge is extracted from the propagation probability table, and the delay conditional probability is set as the infection rate parameter for state evolution from the infected type to the susceptible type.

[0027] Optionally, a plurality of rounds of random simulation are performed using the cascading failure simulation model, the delay conditional probability in the propagation probability table is set as the infection rate parameter to simulate the diffusion process from the infected type to the susceptible type, and the delay diffusion range and the decay speed are counted to generate an influence matrix composed of the initial risk source and the corresponding delay diffusion range and decay speed, including:

[0028] establish a loop simulation sequence including a preset total number of iterations, select a graph node set as an infection type as an initial risk source in each simulation sequence, and initialize a time step counter;

[0029] In each discrete time step, retrieve a successor graph node pointed to by the graph node of the infection type in a susceptible type to generate a random probability value, compare the random probability value with an infection rate parameter of a corresponding directed edge, update a state of the successor graph node to the infection type when the random probability value meets a trigger condition, and update a state of the graph node originally in the infection type to a removal type, and continue iteration until there is no graph node in the infection type in the cascading failure simulation model;

[0030] In each simulation sequence, obtain a delay spread range and a decay speed by statistically counting a total number of graph nodes finally in the removal type and a cumulative value of the time step counter;

[0031] statistically average all delay spread ranges and decay speeds obtained in the preset total number of iterations, define a node index of the initial risk source as a row, and define an average spread range and an average decay speed as a column index to construct an influence matrix.

[0032] Optionally, a target node with a failure influence degree greater than a preset threshold in the influence matrix is extracted, and a time buffer value is calculated according to a delay coefficient preset according to a task process type based on a delay spread characteristic of the target node in a simulation of a diffusion process of the cascading failure simulation model to generate management auxiliary decision information of the target engineering project, including:

[0033] Each initial risk source in the influence matrix is extracted, a delay spread range and a decay speed corresponding to each initial risk source are obtained, a failure influence degree is calculated by weighted calculation of the delay spread range and the decay speed, and an initial risk source with a failure influence degree greater than a preset threshold is determined as a target node;

[0034] A task process type corresponding to the target node is obtained, and a delay coefficient preset according to the task process type is matched, wherein the delay coefficient is set in positive correlation with process complexity and recovery difficulty;

[0035] The decay speed corresponding to the target node is multiplied by the preset delay coefficient to obtain a time buffer value;

[0036] A time buffer value and a task belonging relationship of the target node in a construction plan network diagram are established, and management auxiliary decision information including the task belonging relationship and the time buffer value is output.

[0037] In a second aspect, the application provides an engineering project management auxiliary decision system based on domain knowledge enhancement, including:

[0038] An acquisition module is configured to acquire a construction plan network diagram of a target engineering project and delay causation data of projects of the same type as the target engineering project, the construction plan network diagram comprising a set of task nodes and logical dependency relationships between tasks, and the delay causation data comprising delay propagation paths and delay probability information of each delayed task represented in a tree-like data structure;

[0039] A mapping module is configured to map the set of task nodes into graph nodes and the logical dependency relationships into directed edges by graph theory mapping on the construction plan network diagram to construct an adjacency matrix;

[0040] A calculation module is configured to calculate delay conditional probabilities between different task nodes by Bayesian inference based on delay occurrence frequencies of each cascading path in the delay causation data, and generate a propagation probability table with the delay conditional probabilities as core parameters;

[0041] A generation module is configured to calculate betweenness centrality of each graph node according to the adjacency matrix, update node attribute values of associated graph nodes in the adjacency matrix according to the betweenness centrality, and generate a topological graph;

[0042] The generation module is further configured to construct a cascading failure simulation model based on the topological graph, the propagation probability table, and SIR state mapping rules, map construction states of graph nodes in the topological graph into objects in the SIR model, and perform multiple rounds of random simulation using the cascading failure simulation model, set the delay conditional probabilities in the propagation probability table as infection rate parameters to simulate a diffusion process of evolution from an infected type to a susceptible type, and generate an influence matrix composed of an initial risk source and corresponding delay diffusion range and decay speed by counting the delay diffusion range and decay speed.

[0043] The generation module is further configured to extract target nodes with a failure influence degree greater than a preset threshold in the influence matrix, calculate a time buffer value according to a delay coefficient preset according to a task process type based on delay diffusion characteristics of the target nodes in the diffusion process simulated by the cascading failure simulation model, and generate management auxiliary decision information for the target engineering project.

[0044] In a third aspect, the present application provides an electronic device, comprising:

[0045] a memory configured to store a computer program;

[0046] a processor configured to implement the steps of the engineering project management auxiliary decision method based on domain knowledge enhancement according to the first aspect when the computer program is executed.

[0047] In a fourth aspect, the application provides a computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and the computer program, when executed by a processor, can implement the steps of the project management auxiliary decision method based on domain knowledge enhancement according to the first aspect.

[0048] The project management auxiliary decision method based on domain knowledge enhancement provided by the application can accurately identify the hub nodes that have an important influence on global connectivity from the complex process network topology structure by constructing an adjacency matrix of the engineering project and calculating the betweenness centrality, and overcomes the limitation of the traditional method that only focuses on the critical path and ignores the structural risk. Secondly, in combination with the propagation probability table generated by Bayesian inference and the SIR cascade failure simulation model, the application can dynamically simulate the nonlinear diffusion process and decay trend of the delay risk among the task nodes, so as to quantitatively evaluate the chain damage effect of local delay on the whole project.

[0049] Finally, the targeted time buffer strategy generated based on the simulation results above realizes the effective defense against the high propagation attribute hidden risk. Therefore, the application effectively solves the problem of insufficient robustness of the project management auxiliary decision in the prior art due to the inability to accurately quantify the cascade failure risk in the complex engineering network.

[0050] Further, the application realizes the conversion of the complex engineering progress network into a graph theory topology model that can be efficiently processed by computer algorithms by accurately mapping the engineering tasks and logical dependency relationship into an adjacency matrix including sequence index and vector connection features.

[0051] This matrix construction based on mapping coordinates can completely retain the complex time sequence constraints and coupling structure among the task nodes, and provides a standardized mathematical basis and data interface for subsequent analysis of network hub characteristics by betweenness centrality and cascade failure simulation. Therefore, the application provides key data structure support for solving the problem in the prior art that it is difficult to accurately identify and simulate the hidden risk propagation in the complex network due to the lack of fine topology model. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0053] Figure 1 A flowchart of a project management auxiliary decision method based on domain knowledge enhancement provided by the embodiments of the application is shown in the figure.

[0054] Figure 2 A tree structure diagram of delay attribution data provided for an embodiment of the present application;

[0055] Figure 3 A structure diagram of a topology graph provided for an embodiment of the present application;

[0056] Figure 4 A structure diagram of an engineering project management auxiliary decision system based on domain knowledge enhancement provided for an embodiment of the present application;

[0057] Figure 5 A hardware structure diagram of an electronic device provided for an embodiment of the present application. DETAILED DESCRIPTION

[0058] In the field of engineering project management, although the prior art has realized basic prediction of project duration by means of critical path method and other means, it still faces severe challenges in dealing with complex engineering with high coupling degree. Traditional static analysis or simple linear simulation mainly focuses on the longest path in the time dimension, ignoring the cascading effect that may be caused by non-critical nodes in the network topology.

[0059] This means that once a minor task at a structural hub position is delayed, the existing decision model often cannot accurately capture how the risk spreads non-linearly in the process network like a virus, and it is also difficult to quantify the damage it causes to the whole. This blind spot for invisible structural risks makes it difficult for managers to develop accurate defense strategies, and engineering projects often evolve from local minor delays to global progress paralysis, and there is an urgent need for an auxiliary decision-making method that can penetrate the internal structure of the network topology and dynamically simulate the risk propagation mechanism.

[0060] To solve the above problems, the present application proposes an engineering project management auxiliary decision method based on domain knowledge enhancement, the core of which is to organically integrate graph theory topology analysis and SIR cascading failure simulation model. Specifically, the present application first accurately identifies those hub nodes that are crucial to global connectivity although not on the critical path by constructing the adjacency matrix of the engineering network and calculating the betweenness centrality. Then, combined with the historical delay propagation probability based on Bayesian inference, the SIR model is used to dynamically simulate the diffusion path and decay trend of delay risk in the complex network, and a quantitative influence matrix is generated. Finally, according to the simulation results, time buffer values are automatically calculated and configured for high-risk nodes.

[0061] This method breaks through the limitations of traditional linear analysis, and through the dual mechanisms of structure recognition and dynamic simulation, the decision can not only see the explicit progress lag, but also understand the invisible cascading risk, thereby solving the problem of insufficient decision robustness caused by the inability to defend the structural risk of the network in the prior art, and significantly improving the predictability and risk resistance of engineering management.

[0062] For those skilled in the technical field, the present application will be further described in detail below in combination with the drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the present application.

[0063] To solve the problems in the prior art, the embodiments of the present application provide an engineering project management auxiliary decision-making method, device, equipment, computer storage medium and computer program product based on domain knowledge enhancement. First, the engineering project management auxiliary decision-making method based on domain knowledge enhancement provided by the embodiments of the present application is introduced.

[0064] Figure 1 The flowchart of the engineering project management auxiliary decision-making method based on domain knowledge enhancement provided by an embodiment of the present application is shown. As shown in Figure 1 The method comprises:

[0065] S101, acquiring the construction plan network diagram of the target engineering project and the delay cause data of the same type project as the target engineering project, the construction plan network diagram comprising a task node set and a logical dependency relationship between tasks, and the delay cause data comprising a delay propagation path and delay probability information of each delay task represented in a tree-like data structure.

[0066] The construction plan network diagram refers to the logical topology structure representing the whole process of the implementation of the engineering project in the form of graph theory, which can specifically include a task node set defining specific work content and a logical dependency relationship defining the construction sequence or constraint conditions between tasks. The delay cause data refers to the structured data extracted from the historical same type project for describing the mechanism and propagation law of the delay event, which records the complete link of the delay risk from the source task to the subsequent task, the delay occurrence frequency of each level of the path, and the statistical probability of each link.

[0067] In the implementation process of the scheme, first, the construction plan network diagram is acquired by reading the electronic progress management file or database record of the target engineering project, and all the tasks to be executed and their preset process logic included in the project are parsed, such as the tight-precedence and tight-succession relationship that a task must be started after another task is completed.

[0068] Secondly, by searching the company's internal historical project knowledge base, past project cases similar to the target project in terms of construction type, scale and process are selected, and the schedule deviation records in these cases are extracted in a structured manner to obtain a dataset including the source of risk, the sequence of affected work processes and the frequency of corresponding delays as delay attribution factors.

[0069] For example, taking the construction of a type A infrastructure project as an example, firstly, the construction plan network diagram of the project is exported from the project management database. Assume that the task node set includes foundation excavation task N1, concrete pouring task N2, and equipment installation task N3, with the logical dependency relationship represented as a singly linked list structure where N1 points to N2 and N2 points to N3. Simultaneously, delay attribution data for similar projects is obtained from the historical database. This data is represented as a fault tree structure with material supply delays as the root node. This tree structure records the path information of how material supply delays cause delays in N2, which then propagate to N3, as well as the frequency of specific delays occurring along this propagation path in historical statistics.

[0070] Figure 2 This illustration shows a tree structure diagram of delay attribution data provided in one embodiment of this application. For example... Figure 2 As shown, this structure is interconnected with the construction schedule network diagram of the target project. The construction schedule network diagram includes specific task nodes such as the N1 foundation excavation and the N5 main steel beam hoisting. The diagram details how different types of risk sources, such as external environment, supply chain, and technology management, propagate along the process logic, and marks the frequency of delays at each node and the corresponding propagation probability values ​​in historical project samples.

[0071] For example, risks stemming from supply chain factors first impact the affected node N4, namely the critical steel structure delivery task, which is marked in the figure with an initial propagation probability of 0.95. Subsequently, this delay propagates further to the core affected node N5 with a specific probability, and eventually spreads to subsequent end-stage processes, including the laying of profiled steel sheets at N6 and the construction of curtain wall embedded parts at N7.

[0072] S102. By performing graph theory mapping on the construction plan network graph, the task node set is mapped to graph nodes and logical dependencies are mapped to directed edges to construct an adjacency matrix.

[0073] Optionally, step S102, which involves mapping the task node set to graph nodes and logical dependencies to directed edges in order to construct an adjacency matrix by performing graph theory mapping on the construction plan network graph, may specifically include:

[0074] S1021. Traverse the task node set in the construction plan network graph, treat each independent task in the task node set as a graph node, and assign a sequence index to each graph node.

[0075] In the implementation process, firstly, the task node set in the whole construction plan network diagram is traversed to identify each independent construction operation item included therein. Secondly, for each identified independent task, it is defined as a graph node in the graph theory model, and a non-repeated integer is sequentially assigned to each graph node as a sequence index according to a preset rule. The sequence index is subsequently used as a unique addressing reference for mapping the row coordinates and column indicators in the influence matrix, thereby ensuring a one-to-one correspondence between the cascading failure simulation results in the matrix space and the engineering task entities.

[0076] For example, it is assumed that the identified basic excavation task N1, concrete pouring task N2 and equipment installation task N3 are respectively assigned sequence indexes 0, 1 and 2 in the order of task start time or the size order of task ID. At this time, the graph node N1 corresponds to the index 0, the graph node N2 corresponds to the index 1, and the graph node N3 corresponds to the index 2.

[0077] S1022, the pre-node and the post-node with the front and back constraint properties in the logical dependency relationship are extracted, and a directed edge is constructed in a vector connection relationship in which the pre-node points to the post-node.

[0078] In the implementation process, firstly, the logical dependency relationship data in the construction plan network diagram is parsed to identify each group of task pairs with constraint association. Secondly, for each group of task pairs, it is determined which task must be completed first and defined as a pre-node, and which task must be started later and defined as a post-node. Finally, based on the determination result, a directed edge is constructed between the two graph nodes, the direction of the directed edge is from the pre-node to the post-node, thereby forming a determined vector connection relationship.

[0079] For example, it is assumed that for the logical dependency relationship N1 pointing to N2, N1 is identified as a pre-node and N2 is identified as a post-node, and a directed edge from the index 0 pointing to the index 1 is constructed. Similarly, for the relationship of N2 pointing to N3, a directed edge from the index 1 pointing to the index 2 is constructed.

[0080] S1023, an initial matrix is constructed according to the total number of graph nodes, the mapping coordinates of the sequence indexes between the graph nodes are determined through the vector connection relationship of the directed edges, and the matrix elements at the mapping coordinates in the initial matrix are set as values representing a connected state, and an adjacency matrix is output.

[0081] The value of the connected state refers to a specific value defined in advance to represent the existence of a direct directed connection between two nodes, which is usually set to 1, and the value is usually 0 in the case of no connection.

[0082] In the process of implementing the scheme, first, an initial square matrix of a corresponding dimension is constructed according to the total number of graph nodes, and all elements of the initial matrix are initially set to 0. The row and column size of the initial matrix is determined by the total number of graph nodes, representing the full task dimension of the engineering network. Then, all directed edges generated in step S1022 are traversed, and for each directed edge, the sequence index of its starting point is obtained as the row coordinate, and the sequence index of its ending point is obtained as the column coordinate, so as to determine the mapping coordinates of the directed edge in the matrix. Through the coordinate mapping mechanism, the standardized definition of all matrix structures in this application is established, that is, the matrix row represents the node as the source of risk propagation, that is, the initial risk source, and the matrix column represents the affected node or the delay propagation index of the corresponding node. Finally, the element value at the mapping coordinates in the initial matrix is modified to the value 1 representing the connected state, thereby completing the construction of the adjacency matrix.

[0083] For example, first, assuming that there is a directed edge from index 0 to index 1, and a directed edge from index 1 to index 2, and the total number of graph nodes is 3, a 3x3 initial matrix is constructed. Second, according to the mapping coordinates (0, 1) and (1, 2), the element of the first row and the second column of the matrix is set to 1, and the element of the second row and the third column is set to 1, and the remaining positions remain 0. The finally output adjacency matrix is a sparse matrix, and the non-zero elements accurately describe the serial dependence structure between project tasks. The adjacency matrix is shown in Table 1 as follows:

[0084] Table 1: Adjacency matrix example

[0085]

[0086] Table 1 shows the adjacency matrix structure constructed for the A-type infrastructure project in this embodiment. The row header of the matrix represents the pre-task node, and the column header represents the post-task node. Based on the sequence index mapping and the directed constraint relationship of N1 pointing to N2 and N2 pointing to N3 in the construction logic, the element values at coordinates (0, 1) and (1, 2) in the matrix are set to 1, which clearly identifies the connected state; the values of the remaining coordinate positions are all 0, indicating no direct logical connection.

[0087] This embodiment accurately converts the complex engineering construction logic into a digital adjacency matrix model that can be efficiently processed by a computer. This mapping mechanism based on sequence index and vector connection not only preserves the strict time sequence constraints between tasks, but also effectively solves the technical problem that traditional engineering management data is difficult to be directly analyzed by intelligent algorithms.

[0088] S103, based on the delay occurrence frequency of each cascade path in the delay attribution data, the delay conditional probability between different task nodes is calculated through Bayesian inference, and a propagation probability table is generated with the delay conditional probability as the core parameter.

[0089] Optionally, step S103 can specifically include:

[0090] S1031, by extracting the parent task node and the child task node with a cascade relationship from the delay transmission path included in the delay attribution data, determining the prior occurrence probability of the parent task node and the joint occurrence probability of the parent task node and the child task node occurring delay simultaneously by counting the delay occurrence frequency of the corresponding task in the historical same-type project sample.

[0091] The cascade relationship refers to the causal association that the abnormal state of the previous node directly leads to the abnormality of the subsequent node in the risk transmission chain, wherein the previous node triggering the abnormality is referred to as the parent task node, and the subsequent node affected by the abnormality is referred to as the child task node. The prior occurrence probability refers to the basic probability value obtained according to the proportion of the frequency of the occurrence of delay of the parent task node in the total sample number in the historical statistical data. The joint occurrence probability refers to the probability value that the delay frequency of the parent task node and the delay frequency of the child task node are observed to occur simultaneously in the same project case in the historical statistical data.

[0092] In the process of implementing the scheme, firstly, the tree structure in the delay attribution data is deeply analyzed, all delay transmission paths are traversed, and each pair of task pairs with direct causal connection is identified. Secondly, for each pair of identified parent task node and child task node, the proportion of the actual frequency of the occurrence of delay of the parent task node in the total sample number in all historical same-type project samples is counted by searching the historical project database, and is determined as the prior occurrence probability. At the same time, the proportion of the cumulative frequency of the parent task node and the child task node being marked as delay state simultaneously in these samples in the total sample number is counted, and is determined as the joint occurrence probability.

[0093] For example, it is assumed that in 100 historical projects, the concrete pouring task N2, i.e. the parent task node, occurs delay in 40 cases, and among the 40 cases, the subsequent equipment installation task N3, i.e. the child task node, also occurs delay in 30 cases. At this time, it is identified that N2 and N3 have a cascade relationship, and it is assumed that the prior occurrence probability of N2 is calculated as 0.4, and the joint occurrence probability of N2 and N3 occurring delay simultaneously is calculated as 0.3.

[0094] S1032, calculating the ratio of the joint occurrence probability and the prior occurrence probability to obtain the delay conditional probability.

[0095] The delay conditional probability refers to the possibility of the child task node occurring delay immediately under the premise condition that the parent task node has occurred delay, and it is the core output of the Bayesian inference, and is used to quantify the infection intensity of the risk on the specific path.

[0096] In the implementation process, the joint occurrence probability obtained in step S1031 is taken as the numerator, the prior occurrence probability is taken as the denominator, and a division operation is performed according to the basic principle of Bayes formula. The calculation process eliminates the interference of the delay of the child task node due to other reasons when the parent task node does not delay, and purely extracts the risk intensity transmitted from the parent task node to the child task node. Through this calculation, the delay condition probability value describing the risk flow from the parent node to the child node is obtained.

[0097] For example, assuming that the joint occurrence probability is 0.3 and the prior occurrence probability is 0.4, the delay condition probability of N2 causing N3 delay is 0.75. This means that once the concrete pouring task N2 delays, the equipment installation task N3 has a 75% chance of also delaying, which accurately quantifies the risk susceptibility of this process link.

[0098] S1033, a mapping relationship between the parent task node and the child task node is established, and a propagation probability table is constructed based on the mapping relationship and the delay condition probability between different task nodes.

[0099] The propagation probability table is a structured data storage object that stores each pair of task nodes in a cascading relationship and its corresponding delay condition probability in the form of key-value pairs or matrices as a lookup table for determining state transitions in the subsequent cascading failure simulation model.

[0100] In the implementation process, first, all calculated delay condition probabilities are traversed to extract the corresponding parent task node ID and child task node ID to establish a unique mapping relationship key. Secondly, the calculated delay condition probability value is written into the preset data structure as the value, so that the propagation probability table is constructed.

[0101] For example, assuming that the generated propagation probability table includes a key record: the key is “N2->N3”, and the value is 0.75. When the simulation is performed subsequently, once the model detects that N2 enters the delay state, it will consult this table, read the parameter 0.75, and set it as the infection rate parameter to determine whether N3 is also marked as delayed, thereby realizing the effective conversion of historical experience data to dynamic simulation parameters. The propagation probability table is shown in Table 2 as follows:

[0102] Table 2: Propagation probability table

[0103]

[0104] Table 2 shows the propagation probability table generated based on Bayesian inference logic in this embodiment. This table clarifies the risk transmission intensity calculated based on the frequency of historical delays. Taking the data in the first row of the table as an example, for the cascading relationship of N2 concrete pouring leading to N3 equipment installation, the table records its joint occurrence probability of 0.30 and the prior occurrence probability of the parent node of 0.40, and calculates the delay conditional probability of 0.75, which serves as the core parameter for simulation evolution. This embodiment transforms massive, discrete historical delay frequency data into a precise probabilistic model describing the intensity of risk transmission. By clearly defining the data basis as historical sample statistics, it effectively solves the problems pointed out in the search comments regarding unclear Bayesian inference implementation processes and unclear data sources. Simultaneously, by defining the probability value as an infection rate parameter rather than a threshold, logical contradictions are eliminated, providing dynamic evolution parameters that conform to objective laws for subsequent cascading failure simulations. This effectively solves the technical problem of large risk prediction deviations caused by the lack of quantitative basis in traditional experience-based judgments.

[0105] S104. Calculate the betweenness centrality of each graph node based on the adjacency matrix, and update the node attribute values ​​of the associated graph nodes in the adjacency matrix based on the betweenness centrality to generate a topology graph.

[0106] Optionally, step S104, which calculates the betweenness centrality of each graph node based on the adjacency matrix and updates the node attributes of the associated graph nodes in the adjacency matrix based on the betweenness centrality, can specifically include:

[0107] S1041. Based on the connectivity relationships represented by the matrix elements in the adjacency matrix, determine the shortest connected path between any two graph nodes. Calculate the betweenness centrality of each graph node by calculating the cumulative value of the paths covered by the shortest connected path that serve as intermediate nodes for each graph node.

[0108] The shortest connected path, in a network structure defined by an adjacency matrix, is the sequence of paths that connects any two graph nodes with the fewest number of edges or the lowest total weight. It represents the most efficient logical flow channel between engineering tasks. Betweenness centrality, in graph theory, is an indicator used to measure the pivotal role of a node in a network. Specifically, it is defined as the proportion of paths passing through that node out of all possible shortest connected paths in the network. The higher the value of this indicator, the more crucial the bridging role of that task node in the overall project schedule control; delays there can easily disrupt multiple work processes.

[0109] During the implementation of the scheme, the adjacency matrix output in step S102 is searched using either the interpolation method or the Brandes algorithm. First, all possible starting nodes in the matrix are traversed. and termination node Then, calculate the set of all shortest connected paths between them. Next, for each graph node in the network... , statistics in all the above shortest path, how many paths must pass through the node , get the corresponding path cumulative value. Finally, the path cumulative value as the numerator, the to the shortest path total number as the denominator, add up all the node pair calculation results, so as to get the betweenness centrality of each graph node .

[0110] For example, assume that in addition to the main chain N1->N2->N3, there is an auxiliary task N4 pointing to N2. At this time, the calculation found that from N1 to N3 must pass through N2, and from N4 to N3 must also pass through N2. This means that N2 is at the intersection of two logical chains, and the algorithm counts that N2 is covered by multiple shortest paths, so N2 is given a higher betweenness centrality value, for example, 0.8, while N1 and N4 at the edge have a lower betweenness centrality, for example, 0.0.

[0111] S1042, convert the betweenness centrality to a weight parameter, and associate the weight parameter to the corresponding graph node as a node attribute value, and construct a topology graph combined with the adjacency matrix.

[0112] In the implementation process, first, the betweenness centrality of all graph nodes calculated in step S1041 is normalized, that is, the maximum and minimum values are found, and the value of each node is mapped to the interval of 0 to 1 to obtain the standardized weight parameter. Next, the network connection structure defined by the adjacency matrix is kept unchanged, and an attribute list corresponding one-to-one to the graph node sequence index is established. The calculated weight parameter is written into the list as a node attribute value, so that each graph node carries an identity tag reflecting its structural importance. Finally, the adjacency matrix and the attribute list are integrated and encapsulated to construct an output complete topology graph object.

[0113] For example, assume that the betweenness centrality of N2 is normalized as a weight parameter 0.9, and the parameter is associated to the N2 graph node as a node attribute value. The finally generated topology graph not only represents "N2 is connected to N3", but also represents "N2 is a key hub with a node attribute value of 0.9".

[0114] Figure 3 A topology graph diagram provided by an embodiment of the present application is shown. As Figure 3 ​As shown, the circular elements in the figure represent various independent construction task nodes in the project, such as N1 or N3, etc. The directed solid arrows connecting these circles clearly define the strict logical sequence and workflow direction between tasks, for example, the arrow from N1 to N3 indicates that the former is the predecessor of the latter. The text inside the circle further indicates the specific number of the node, the task type attribute, and the structural weight parameter calculated based on the betweenness centrality, that is, the node attribute value, for example, the hub node N3 is marked with a high weight value of 0.9, which intuitively shows its status as a key connection point in the overall network structure.

[0115] The embodiment not only restores the connection skeleton of the engineering network, but also accurately identifies those invisible key nodes lurking on non-critical paths by quantifying the structural hub degree of each task node. The generated topology graph provides a high-dimensional data model with structural weight for subsequent simulation, effectively solving the technical defects of traditional methods that only rely on experience to judge task importance while ignoring network structure vulnerabilities.

[0116] S105, based on the topology graph, the propagation probability table, and the SIR state mapping rule, constructing a cascading failure simulation model, mapping the construction state corresponding to the graph node in the topology graph to the object in the SIR model, and using the cascading failure simulation model to perform multiple rounds of random simulation, setting the delay condition probability in the propagation probability table as the infection rate parameter to simulate the diffusion process and count the delay diffusion range and decay speed to generate the influence matrix.

[0117] Optionally, step S105 can specifically include:

[0118] S1051, establishing an SIR model including susceptible type, infected type, and removed type, mapping the graph nodes in the topology graph that are in the on-schedule execution phase to the susceptible type, mapping the graph nodes that have occurred delay to the infected type, and mapping the graph nodes that have no delay or have completed work to the removed type.

[0119] The SIR model is a classic dynamic model in epidemiology, which is usually used to describe the propagation mechanism of viruses in the population. In the engineering management scenario of the present application, the model is innovatively applied, in which the susceptible type refers to the task node whose current construction progress is normal but faces the risk of being "infected" due to the existence of logical dependency; the infected type refers to the abnormal task node that has occurred delay and has the ability to propagate the delay risk to subsequent task nodes; the removed type refers to the stable task node that has completed the work or eliminated the delay risk through acceleration measures and no longer participates in the risk propagation.

[0120] In the implementation process of the scheme, the environmental parameters of the cascading failure simulation model are initialized, and all graph nodes in the topology graph are mapped to three state spaces of the SIR model according to their real-time construction states. Specifically, for those task nodes that have not started or are being carried out according to the plan and have a pre-dependent relationship, the state mark thereof is set to the susceptible type, that is, the S state; for those task nodes for which it is monitored that the actual progress lags behind the planned progress, the state mark thereof is set to the infected type, that is, the I state, and these nodes are the risk sources in the initial stage of the simulation; for those task nodes that have completed acceptance or are at the end of the non-critical path and have no subsequent impact, the state mark thereof is set to the removed type, that is, the R state.

[0121] For example, it is assumed that the concrete pouring task N2 currently lags behind by 3 days due to material shortage, and the subsequent equipment installation task N3 has not started. In the initialization, N2 is marked as the infected type, N3 is marked as the susceptible type, and the completed foundation excavation task N1 is marked as the removed type.

[0122] S1052, traverse the directed edges of the topology graph, extract the delay condition probability corresponding to the directed edge from the propagation probability table, set the delay condition probability as the infection rate parameter for state evolution from the infected type to the susceptible type, and construct the cascading failure simulation model according to the infection rate parameter.

[0123] The infection rate parameter refers to a dynamic index for indicating the diffusion intensity of the delay risk from the infected type object to the susceptible type object in the simulation running process. The parameter is directly derived from the delay condition probability obtained by Bayesian inference of historical data, and is used to calculate the possibility of infection occurrence in each discrete time step.

[0124] In the implementation process of the scheme, first, all directed edge connection relationships defined in the topology graph are traversed. For each directed edge, the parent node at the starting point and the child node at the end point are used as index keys to query the propagation probability table generated in step S103. The queried delay condition probability is extracted and directly assigned to the directed edge as the infection rate parameter.

[0125] For example, it is assumed that the directed edge from N2 to N3 is traversed, and the corresponding delay condition probability is found to be 0.75 in the propagation probability table. Therefore, 0.75 is set as the infection rate parameter of the edge. This means that in the subsequent simulation, a random number between 0 and 1 will be generated, and if it is less than or equal to 0.75, it is determined that N3 is infected by N2, that is, a delay occurs; if the random number is greater than 0.75, it is determined that N3 is immune, for example, by working overtime to catch up with the schedule.

[0126] The embodiment creatively introduces a biological virus transmission mechanism into the field of engineering management. This interdisciplinary simulation modeling not only visually simulates the nonlinear diffusion process of delay risk from point to plane, but also strictly follows objective laws by introducing historical probability parameters, enabling accurate pre-visualization of the chain reaction path after the outbreak of different risk sources.

[0127] Optionally, the process of using the cascading failure simulation model in step S105 to perform multiple rounds of random simulation to simulate the diffusion process using the delay propagation probability of the task nodes in the propagation probability table as the infection rate parameter and to generate the influence matrix by counting the delay propagation range and decay speed can specifically include:

[0128] S1053, a loop simulation sequence including a preset total number of iterations is established, and a graph node set as an infection type is selected as an initial risk source in each simulation sequence, and a time step counter is initialized.

[0129] The loop simulation sequence refers to a set of multiple repeated independent experiments set by the Monte Carlo method to eliminate accidental errors caused by a single random simulation, and each experiment starts from the same initial state. The initial risk source refers to the first graph node with a delay that is artificially designated or randomly selected at the beginning of each simulation experiment, which is the zeroth patient triggering subsequent cascading reactions. The preset total number of iterations is shown in Table 3:

[0130] Table 3: Comparison of preset iteration total number

[0131]

[0132] As shown in Table 3, the selection of the preset total number of iterations directly affects the accuracy and efficiency of the simulation. In the standard accuracy evaluation scenario, 1000 iterations are usually selected, at which time the simulation time is moderate and the result error is controlled within 5%, which can meet the decision-making needs of most engineering projects.

[0133] During the implementation of the scheme, first, a large enough integer is set as the preset total number of iterations N to ensure that the statistical results have statistical significance. Secondly, for each graph node in the topology graph, it is designated in turn as the initial risk source to start a complete loop simulation. Finally, before the start of each simulation, the time step counter is reset to zero, and all nodes except the initial risk source are reset to the susceptible type.

[0134] For example, to evaluate the risk damage of task N2, it is assumed that the preset total number of iterations N is set to 1000 according to Table 3. In the first simulation sequence, the state of N2 is forcibly set to the infection type, N1 and N3 are set to the susceptible type, and the counter is set to 0.

[0135] S1054、In each discrete time step, retrieve the successor graph nodes pointed by the graph node in infection type which is in susceptible type to generate a random probability value, compare the random probability value with the corresponding infection rate parameter in the propagation probability table, update the state of the successor graph node to infection type when the random probability value meets the trigger condition, and update the state of the graph node originally in infection type to removal type, and continue iteration until there is no graph node in infection type in the cascading failure simulation model.

[0136] In each discrete time step, first, the algorithm scans all the nodes currently in infection type, and for each such node, finds all the successor nodes it points to which are in susceptible type. Second, for each successor node, a random number between 0 and 1 is generated. The random number is compared with the trigger threshold, i.e. the delay condition probability, set in S1052. If the random number is less than the threshold, it is determined that the trigger condition is met, and the state of the successor node is updated to infection type at the next time.

[0137] At the same time, in order to simulate the process of delay being processed or risk energy dissipation, the state of the current node in infection type is updated to removal type, indicating that it has completed the mission of propagation. This process continues to circulate until there is no node in the network in infection type, and the simulation is automatically terminated.

[0138] For example, assume that in the first time step, N2 is in infection state and N3 is in susceptible state. A random number 0.6 is generated, which is less than the threshold 0.75. Therefore, it is determined that N3 is infected. In the second time step, N3 becomes in infection state and N2 becomes in removal state. Thereafter, if N3 has no successor node, the simulation ends.

[0139] S1055、In each simulation sequence, the delay diffusion range and the decay speed are obtained by counting the total number of graph nodes finally in removal type and the cumulative value of the time step counter.

[0140] The delay diffusion range refers to the total number of nodes finally marked as removal type in the network after a single simulation ends, which quantifies the breadth of the risk spread. The decay speed refers to the total number of time steps experienced by the simulation from the beginning to the end. The smaller the value, the faster the risk spreads or the faster it is blocked.

[0141] In the implementation process, first, the entire topology graph is traversed, and the number of nodes in the final state of removal type is counted, which is recorded as the delay diffusion range of this simulation. Second, the final value of the time step counter is read, which is recorded as the decay speed of this simulation. For example, assume that in the first simulation, N2 and N3 finally become in removal state, then the delay diffusion range is 2. The simulation ends after 2 time steps, so the decay speed is recorded as 2.

[0142] S1056, statistically average all the delay spread ranges and attenuation speeds obtained within the preset total number of iterations, and map the operation result to the matrix coordinate position corresponding to the initial risk source to generate an influence matrix.

[0143] The influence matrix refers to a two-dimensional data structure, which defines the node sequence index as the initial risk source as the row coordinate, and the statistical average of the delay spread range and the attenuation speed as the column index, and is used to macroscopically show the potential damage of each node as a risk source.

[0144] In the process of implementing the scheme, when all the simulations of the preset total number N of iterations are completed for a certain initial risk source, the delay spread range values of the N records are accumulated and divided by the preset total number N of iterations to obtain the average spread range. Similarly, the average attenuation speed is calculated. Finally, the two average values are filled into the position corresponding to the initial risk source in the influence matrix. For example, the calculated average spread range is filled into the first column corresponding to the row index of the influence matrix, and the average attenuation speed is filled into the second column corresponding to the row index.

[0145] For example, it is assumed that after 1000 simulations, the delay caused by N2 sometimes spreads to N3 and sometimes does not spread. Statistics show that the average spread range is 1.75, which means that on average 1.75 tasks are involved each time, and the average attenuation speed is 1.8. The vector [1.75, 1.8] is written into the row corresponding to N2 in the influence matrix.

[0146] The embodiment constructs a risk evolution laboratory capable of thousands of virtual trial and error. This simulation mechanism based on multiple iterations effectively eliminates the random deviation of a single simulation, and in combination with the precision control strategy shown in Table 3, an influence matrix with high confidence level including average spread range and attenuation speed is accurately output. The matrix quantifies the potential damage of each task node, solving the technical difficulties of traditional methods in evaluating risk uncertainty and long tail effect.

[0147] S106, extracting a target node with a failure influence degree greater than a preset threshold from the influence matrix, and generating management auxiliary decision information of the target engineering project according to the delay spread characteristics of the target node in the cascading failure simulation model simulation diffusion process and in combination with the time buffer value calculated according to the delay coefficient preset according to the task process type.

[0148] Optionally, step S106 can specifically include:

[0149] S1061, extracting the delay spread range and attenuation speed corresponding to each initial risk source in the influence matrix, performing weighted calculation on the delay spread range and attenuation speed to obtain the failure influence degree, and determining the initial risk source with the failure influence degree greater than the preset threshold as the target node.

[0150] The failure impact degree is a comprehensive quantitative index for evaluating the damage potential of the entire project network caused by the delay of a specific task node. It is usually composed of two dimensions: delay propagation range and attenuation speed, and the higher the value, the stronger the risk damage of the node. The preset threshold is a risk judgment dividing line set according to the sensitivity of the project to the delay of the whole project, which is used to divide the ordinary risk nodes and high-risk nodes into two categories. The preset threshold is shown in Table 4.

[0151] Table 4: Preset threshold table

[0152]

[0153] As shown in Table 4, the setting of the preset threshold T directly determines the sensitivity of the auxiliary decision. For aggressive projects with extremely tight schedules, the threshold is set to 2.5, meaning only those nodes that can trigger large-scale cascading reactions will be locked; for conservative projects that do not allow errors, the threshold is reduced to 0.8.

[0154] In the implementation process, first, the influence matrix generated in step S1056 is traversed to extract the initial risk source corresponding to each row and its associated average delay propagation range and average attenuation speed. The influence matrix is defined as a row index of the node sequence of the initial risk source, and the statistical average of the delay propagation range and the attenuation speed is defined as a column index. Then, two weight factors corresponding to the breadth of the propagation range and the persistence of the attenuation speed are set according to engineering management experience, and the failure impact degree of each initial risk source is calculated using a weighted sum formula. Finally, the threshold judgment is introduced by Table 4, and all nodes with scores higher than the standard are selected as target nodes.

[0155] For example, assuming that the robust strategy is selected according to Table 4, the preset threshold T is 1.5. For task N2, the average propagation range is 1.75 and the average attenuation speed is 1.8 extracted from the influence matrix. The weight of both is 0.5, then the failure impact degree of N2 is 1.775. Comparing 1.775 with the preset threshold T=1.5, it can be determined that N2 is a target node that needs to be controlled.

[0156] S1062, obtain the task process type corresponding to the target node, and match the preset delay coefficient according to the task process type, wherein the delay coefficient is set in positive correlation with the process complexity and recovery difficulty; perform multiplication operation on the attenuation speed corresponding to the target node through the delay coefficient, to obtain the time buffer value.

[0157] The preset delay coefficient is a multiplier factor used to correct the deviation between the theoretical simulation value and the actual construction environment, which reflects the amplification multiple of the recovery difficulty or uncertainty of different process types of tasks when facing delay. Specifically, the delay coefficient is positively correlated with the process complexity and recovery difficulty, that is, the more complex the process, the more coordination interfaces or the stronger the hard physical constraints of the task, the higher the set value of the delay coefficient.

[0158] The time buffer value is the free time difference or safety margin that is suggested to be additionally added after the planned duration of the target node, which is usually measured in days, and is intended to physically block the further cascade diffusion of risks. The preset delay coefficient is set according to the physical properties and process complexity of the task, as shown in Table 5 below:

[0159] Table 5: Preset delay coefficient table

[0160]

[0161] As shown in Table 5, the preset delay coefficient is differentiated according to the physical properties and process complexity of the task. The coefficient of the standardized operation is low, while the coefficient of the complex system installation is high due to the involvement of multi-specialty coordination.

[0162] During the implementation of the scheme, for each determined target node, first, its task process type belonging to the construction plan network diagram is identified. Second, its average decay speed shown in the simulation is extracted. Third, the preset delay coefficient table is queried to match the corresponding delay coefficient K according to the task process type to which the node belongs. Finally, multiplication operation is performed to multiply the decay speed and the delay coefficient K, and the calculation result is the recommended time buffer value.

[0163] For example, assuming that the target node N2 belongs to the concrete pouring task, corresponding to the structured construction type in Table 5, the delay coefficient K is selected as 1.5. The average decay speed of N2 is 1.8, which is multiplied by the coefficient K = 1.5 to calculate 2.7. According to the engineering management habit, the recommended time buffer value of N2 is finally determined as 3 days by rounding up. This means that 3 days of buffer period is suggested to be reserved after the completion of N2 and before the start of N3.

[0164] S1063, establish the time buffer value and the task belonging relationship of the target node in the construction plan network diagram, and output the management auxiliary decision-making information including the task belonging relationship and the time buffer value.

[0165] Task attribution refers to the process of reanalyzing abstract graph node IDs or matrix indexes into actual task names, WBS codes, and responsible departments in project management software, ensuring that computer results can be understood by people. Management assistance decision information refers to the visual report or data instruction presented to the project manager, including risk positioning, cause analysis, and specific adjustment suggestions.

[0166] During the implementation of the scheme, first, the sequence index mapping table established in step S102 is used to reverse lookup the corresponding original task information with the index value of the target node as the key. Second, the calculated time buffer value is bound with the task information to generate a complete decision suggestion record, which is output through a man-machine interaction interface or a data interface.

[0167] For example, assume that the task details corresponding to index 1 are found, and it is confirmed that it is a concrete pouring task N2. Then generate decision information: detect that the concrete pouring task N2 has a high-level cascading failure risk, and suggest adding a 3-day time buffer at this task node to block the potential delay risk of the subsequent equipment installation task N3.

[0168] This embodiment realizes a precise decision-making closed loop from macro risk quantification to micro strategy landing. This method abandons the one-size-fits-all mode of increasing buffers in traditional management, but based on domain knowledge rules, combined with the failure influence degree obtained by simulation, accurately locates the vulnerable points in the network, and allocates time buffer resources as needed. This not only effectively reduces the invalid redundancy of the overall project duration, but also ensures that critical defense resources are allocated to the sharpest point that can block cascading risks, significantly improving the risk resistance efficiency of engineering plans.

[0169] Figure 4 A structural schematic diagram of one specific implementation mode of an engineering project management assistance decision system based on domain knowledge enhancement provided by the embodiment of the present application is shown in Figure 4 The system can include:

[0170] 410 An acquisition module is configured to acquire a construction plan network diagram of a target engineering project and delay cause data of a same-type project as the target engineering project, the construction plan network diagram including a task node set and a logical dependency relationship between tasks, and the delay cause data including delay transmission paths and delay probability information of each delay task represented in a tree-shaped data structure;

[0171] 420 A mapping module is configured to map the task node set to graph nodes and the logical dependency relationship to directed edges to construct an adjacency matrix by graph theory mapping on the construction plan network diagram;

[0172] 430, a calculation module, configured to calculate delay conditional probability between different task nodes by Bayesian inference based on delay occurrence frequency of each cascading path in the delay attribution data, and generate a propagation probability table with the delay conditional probability as a core parameter;

[0173] 440, a generation module, configured to calculate betweenness centrality of each graph node according to the adjacency matrix, and update node attribute values of associated graph nodes in the adjacency matrix according to the betweenness centrality, and generate the topological graph;

[0174] 440, the generation module is further configured to construct a cascading failure simulation model based on the topological graph, the propagation probability table and the SIR state mapping rule, map construction states of graph nodes in the topological graph to objects in the SIR model, and perform multi-round random simulation by using the cascading failure simulation model, set the delay conditional probability in the propagation probability table as an infection rate parameter, simulate a diffusion process of evolution from an infected type to a susceptible type, and generate an influence matrix composed of an initial risk source and corresponding delay diffusion range and decay speed by counting the delay diffusion range and the decay speed;

[0175] 440, the generation module is further configured to extract a target node with a failure influence degree greater than a preset threshold in the influence matrix, calculate a time buffer value by combining a delay coefficient preset according to a task process type according to delay diffusion characteristics of the target node in the diffusion process simulated by the cascading failure simulation model, and generate management auxiliary decision information of the target engineering project.

[0176] The engineering project management auxiliary decision system based on domain knowledge enhancement provided in the embodiments of the present application is used to implement the engineering project management auxiliary decision method based on domain knowledge enhancement described above, and therefore the specific implementation of the engineering project management auxiliary decision system based on domain knowledge enhancement can be seen from the embodiment part of the engineering project management auxiliary decision method based on domain knowledge enhancement described above. The specific implementation can be referred to the description of the corresponding embodiment part, and will not be repeated here.

[0177] Figure 5 A hardware structure schematic diagram of an electronic device provided by an embodiment of the present application is shown.

[0178] The electronic device can include a processor 510 and a memory 520 storing computer program instructions.

[0179] Specifically, the processor 510 described above can include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or can be configured as one or more integrated circuits that implement one or more embodiments of the present application.

[0180] The memory 520 can include mass storage for data or instructions. As an example and not by way of limitation, the memory 520 can include a Hard Disk Drive (HDD), a floppy disk drive, a flash memory, a Compact Disc (CD) or other optical disk, a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 520 can include removable or non-removable (or fixed) media, where appropriate. Where appropriate, the memory 520 can be internal or external to the integrated gateway disaster recovery appliance. In particular embodiments, the memory 520 is non-volatile, solid-state memory.

[0181] The memory can include read-only memory (ROM), random access memory (RAM), magnetic disk storage mediums, optical storage mediums, flash memory devices, electrical, optical, or other physical / tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software that, when executed (by one or more processors), is operable to

[0182] The processor 510 implements any one of the above-mentioned engineering project management auxiliary decision-making methods based on domain knowledge enhancement by reading and executing the computer program instructions stored in the memory 520.

[0183] In one example, the electronic device can further include a communication interface 530 and a bus 540. Wherein, as shown in the figure, the processor 510, the memory 520, the communication interface 530 are connected through the bus 540 and complete the communication between each other. Figure 5

[0184] The communication interface 530 is mainly used to realize the communication between each module, device, unit and / or equipment in the embodiments of the application.

[0185] ​Bus 540 includes hardware, software, or both, to couple components of the online data traffic metering device to each other and to couple components to other components within the online data traffic metering device. While bus 540 is shown for the sake of clarity as a single bus, it can include one or more buses operating together. Bus 540 can be implemented using any suitable type of bus or buses, including, but not limited to, an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand™ interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or any other suitable bus or interconnect, or a combination of two or more of these. Where appropriate, bus 540 can be implemented as a system-wide interconnect or a combination of busses.

[0186] The electronic device can execute the project management auxiliary decision method based on domain knowledge enhancement in the embodiments of the application, so as to realize the project management auxiliary decision method based on domain knowledge enhancement described in combination with the drawings.

[0187] In addition, in combination with the project management auxiliary decision method based on domain knowledge enhancement in the above embodiments, the embodiments of the application can provide a computer readable storage medium to realize. The computer readable storage medium has computer program instructions stored thereon; the computer program instructions are executed by the processor to realize any one of the project management auxiliary decision methods based on domain knowledge enhancement in the above embodiments.

[0188] It needs to be clear that the application is not limited to the specific configurations and processes described above and shown in the drawings. For the sake of simplicity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between steps, after understanding the spirit of the application.

[0189] The functions shown in the above-described structural block diagrams can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application specific integrated circuits (ASICs), appropriate firmware, plug-ins, functional cards, and the like. When implemented in software, the elements of the present application are program or code segments that are used to perform the required tasks. The program or code segments can be stored in a machine-readable medium or transmitted through a data signal carried in a carrier wave over a transmission medium or communication link. A "machine-readable medium" includes any medium that can store or transport information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memories, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, and the like. The code segments can be downloaded via a computer network such as the Internet, an intranet, and the like.

[0190] It is also important to note that the examples described herein can be implemented in a variety of systems, including and not limited to a digital circuit, an analog circuit, a state machine, a stored-program processor, a programmable logic controller, an application specific integrated circuit, and / or a system-on-chip. One of ordinary skill in the art will recognize that an implementation of an embodiment of the present application can be a

[0191] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. Alternatively, computer program implemented steps can be

[0192] The above provides a kind of engineering project management auxiliary decision-making method based on field knowledge enhancement provided in the present application.The principle and implementation mode of the present application are described in this paper with specific examples, the above example is only used to help understand the method of the present application and its core idea.It should be pointed out that, for ordinary skilled person in the art, without departing from the principle of the present application, some improvements and modifications can be made to the present application, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A method for engineering project management decision support based on domain knowledge enhancement, characterized in that, The method comprises: obtaining a construction plan network diagram of a target engineering project and delay cause data of projects of the same type as the target engineering project, the construction plan network diagram comprising a set of task nodes and logical dependency relationships between tasks, and the delay cause data comprising delay propagation paths and delay probability information of each delayed task represented in a tree data structure; mapping the set of task nodes to graph nodes and the logical dependency relationships to directed edges by graph theory mapping on the construction plan network diagram to construct an adjacency matrix; calculating delay conditional probabilities between different task nodes by Bayesian inference based on delay occurrence frequencies of cascading paths at each level in the delay cause data, and generating a propagation probability table with the delay conditional probabilities as core parameters; calculating betweenness centrality of each graph node according to the adjacency matrix, and updating node attribute values of associated graph nodes in the adjacency matrix according to the betweenness centrality to generate a topology graph; constructing a cascading failure simulation model based on the topology graph, the propagation probability table, and SIR state mapping rules, mapping construction states of graph nodes in the topology graph to objects in the SIR model, and performing multiple rounds of random simulation using the cascading failure simulation model, setting delay conditional probabilities in the propagation probability table as infection rate parameters to simulate a diffusion process of evolution from an infected type to a susceptible type, and calculating a delay diffusion range and a decay speed to generate an impact matrix composed of an initial risk source and corresponding delay diffusion range and decay speed; extracting target nodes with a failure impact degree greater than a preset threshold in the impact matrix, and generating management auxiliary decision information for the target engineering project according to delay diffusion characteristics of the target nodes in the diffusion process simulated by the cascading failure simulation model, and combining a delay coefficient calculated according to a task process type to calculate a time buffer value; the method comprises: extracting parent task nodes and child task nodes having a cascading relationship from the delay propagation paths included in the delay cause data, determining a prior occurrence probability of the parent task nodes and a joint occurrence probability of the parent task nodes and the child task nodes occurring delay at the same time by counting delay occurrence frequencies of corresponding tasks in historical samples of projects of the same type; calculating the delay conditional probability by calculating a ratio of the joint occurrence probability and the prior occurrence probability; establishing a mapping relationship between the parent task nodes and the child task nodes, and constructing the propagation probability table based on the mapping relationship and the delay conditional probabilities between different task nodes.

2. The method of claim 1, wherein, the method comprises: Traverse the set of task nodes in the construction plan network diagram, take each independent task in the set of task nodes as a graph node, and assign a sequence index to each graph node; Extract the pre-requisite graph node and the post-requisite graph node with the front and rear constraint attribute in the logical dependency relationship, and construct a directed edge in the vector connection relationship of the pre-requisite graph node pointing to the post-requisite graph node; Construct an initial matrix according to the total number of graph nodes, determine the mapping coordinates of the sequence index between the graph nodes through the vector connection relationship of the directed edge, and set the matrix element at the mapping coordinates in the initial matrix to a numerical value representing a connected state, and output the adjacency matrix.

3. The method of claim 2, wherein, According to the adjacency matrix, the betweenness centrality of each graph node is calculated, and the node attribute value of the associated graph node in the adjacency matrix is updated according to the betweenness centrality, and a topology graph is generated, including: According to the connected relationship represented by the matrix elements in the adjacency matrix, determine the shortest connected path between any two graph nodes, and calculate the path cumulative value of each graph node as a transit node covered by the shortest connected path to obtain the betweenness centrality of each graph node; Convert the betweenness centrality into a weight parameter, associate the weight parameter to the corresponding graph node as the node attribute value of the graph node, and construct the topology graph in combination with the adjacency matrix.

4. The method of claim 1, wherein, Based on the topology graph, the propagation probability table, and the SIR state mapping rule, a cascading failure simulation model is constructed, and the construction state of the graph node in the topology graph is mapped to the object in the SIR model, including: The SIR model including susceptible type, infected type and removed type is established, the graph node in the topology graph in the execution phase is mapped to the susceptible type, the graph node with delay is mapped to the infected type, and the graph node without delay or completed is mapped to the removed type; Traverse the directed edge of the topology graph, extract the delay condition probability corresponding to the directed edge from the propagation probability table, and set the delay condition probability as the infection rate parameter of the state evolution from the infected type to the susceptible type.

5. The method of claim 4, wherein, The cascading failure simulation model is used for multiple rounds of random simulation, the delay condition probability in the propagation probability table is set as the infection rate parameter, the diffusion process from the infected type to the susceptible type is simulated, the delay diffusion range and the decay speed are counted to generate an impact matrix composed of an initial risk source and the corresponding delay diffusion range and decay speed, including: A loop simulation sequence including a preset total number of iterations is established, and a graph node set as the infected type is selected as an initial risk source in each simulation sequence, and a time step counter is initialized; The construction plan network diagram is traversed, and each task node in the set of task nodes is taken as a graph node, and a sequence index is assigned to each graph node; The pre-requisite graph node and the post-requisite graph node with the front and rear constraint attribute in the logical dependency relationship are extracted, and a directed edge is constructed in the vector connection relationship of the pre-requisite graph node pointing to the post-requisite graph node; An initial matrix is constructed according to the total number of graph nodes, the mapping coordinates of the sequence index between the graph nodes are determined through the vector connection relationship of the directed edge, and the matrix element at the mapping coordinates in the initial matrix is set to a numerical value representing a connected state, and the adjacency matrix is output. According to the adjacency matrix, the betweenness centrality of each graph node is calculated, and the node attribute value of the associated graph node in the adjacency matrix is updated according to the betweenness centrality, and a topology graph is generated, including: According to the connected relationship represented by the matrix elements in the adjacency matrix, determine the shortest connected path between any two graph nodes, and calculate the path cumulative value of each graph node as a transit node covered by the shortest connected path to obtain the betweenness centrality of each graph node; Convert the betweenness centrality into a weight parameter, associate the weight parameter to the corresponding graph node as the node attribute value of the graph node, and construct the topology graph in combination with the adjacency matrix. Based on the topology graph, the propagation probability table, and the SIR state mapping rule, a cascading failure simulation model is constructed, and the construction state of the graph node in the topology graph is mapped to the object in the SIR model, including: The SIR model including susceptible type, infected type and removed type is established, the graph node in the topology graph in the execution phase is mapped to the susceptible type, the graph node with delay is mapped to the infected type, and the graph node without delay or completed is mapped to the removed type; Traverse the directed edge of the topology graph, extract the delay condition probability corresponding to the directed edge from the propagation probability table, and set the delay condition probability as the infection rate parameter of the state evolution from the infected type to the susceptible type. The cascading failure simulation model is used for multiple rounds of random simulation, the delay condition probability in the propagation probability table is set as the infection rate parameter, the diffusion process from the infected type to the susceptible type is simulated, the delay diffusion range and the decay speed are counted to generate an impact matrix composed of an initial risk source and the corresponding delay diffusion range and decay speed, including: A loop simulation sequence including a preset total number of iterations is established, and a graph node set as the infected type is selected as an initial risk source in each simulation sequence, and a time step counter is initialized; retrieving a successor graph node in the susceptible type pointed by a graph node in the infected type at each discrete time step to generate a random probability value, comparing the random probability value with the infection rate parameter of a corresponding directed edge, updating a state of the successor graph node to the infected type when the random probability value satisfies a trigger condition, and updating a state of the graph node originally in the infected type to the removed type, and continuing iteration until there is no graph node in the infected type in the cascading failure simulation model; in each of the simulation sequences, obtaining the delay diffusion range and the decay speed by counting a total number of graph nodes finally in the removed type and an accumulated value of the time step counter; statistically averaging all the delay diffusion ranges and the decay speeds obtained in the preset total number of iterations to construct the influence matrix, with a node index as an initial risk source defined as a row, and an average diffusion range and an average decay speed defined as a column index.

6. The method of claim 5, wherein, the target node in the influence matrix whose failure influence degree is greater than a preset threshold is extracted, and a time buffer value is calculated according to a delay coefficient preset according to a task process type based on delay diffusion characteristics of the target node in a simulation diffusion process of the cascading failure simulation model to generate management auxiliary decision information of the target engineering project, including: the delay diffusion range and the decay speed corresponding to each of the initial risk sources in the influence matrix are extracted, the delay diffusion range and the decay speed are weighted to obtain the failure influence degree, and the initial risk source whose failure influence degree is greater than the preset threshold is determined as the target node; the task process type corresponding to the target node is obtained, and a preset delay coefficient is matched according to the task process type, wherein the delay coefficient is set according to a positive correlation between process complexity and recovery difficulty; the decay speed corresponding to the target node is multiplied by the preset delay coefficient to obtain the time buffer value; a task attribution relationship between the time buffer value and the target node in the construction plan network graph is established, and the management auxiliary decision information including the task attribution relationship and the time buffer value is output.

7. A domain knowledge enhancement based engineering project management aided decision system, characterized in that, including: an acquisition module configured to acquire a construction plan network graph of a target engineering project and delay cause data of a same-type project as the target engineering project, the construction plan network graph including a task node set and a logical dependency relationship between tasks, and the delay cause data including a delay transmission path and delay probability information of each delay task represented in a tree-shaped data structure; a mapping module configured to map the task node set to graph nodes and the logical dependency relationship to directed edges by graph theory mapping on the construction plan network graph to construct an adjacency matrix; The computing module is configured to calculate delay conditional probabilities between different task nodes by Bayesian inference based on delay occurrence frequencies of each cascading path in the delay attribution data, and generate a propagation probability table with the delay conditional probabilities as core parameters; the computing module is further configured to extract parent task nodes and child task nodes having a cascading relationship from the delay transmission paths included in the delay attribution data, determine a prior occurrence probability of the parent task nodes and a joint occurrence probability of the parent task nodes and the child task nodes occurring delay at the same time by counting delay occurrence frequencies of corresponding tasks in historical samples of the same type of project, calculate the delay conditional probability by dividing the joint occurrence probability by the prior occurrence probability, establish a mapping relationship between the parent task nodes and the child task nodes, and construct the propagation probability table based on the mapping relationship and the delay conditional probabilities between different task nodes; The generating module is configured to calculate betweenness centrality of each graph node according to the adjacency matrix, and update node attribute values of associated graph nodes in the adjacency matrix according to the betweenness centrality, and generate a topology graph; The generating module is further configured to construct a cascading failure simulation model based on the topology graph, the propagation probability table, and a SIR state mapping rule, map construction states of graph nodes in the topology graph to objects in a SIR model, and perform multiple rounds of random simulation by using the cascading failure simulation model, set delay conditional probabilities in the propagation probability table as infection rate parameters, simulate a diffusion process of evolution from an infected type to a susceptible type, and count a delay diffusion range and a decay speed to generate an influence matrix composed of an initial risk source and corresponding delay diffusion range and decay speed; The generating module is further configured to extract target nodes having a failure influence degree greater than a preset threshold from the influence matrix, calculate a time buffer value according to a delay coefficient preset according to a task process type based on delay diffusion characteristics of the target nodes in a diffusion process simulated by the cascading failure simulation model, and generate management auxiliary decision information of the target engineering project.

8. An electronic device, comprising: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the engineering project management auxiliary decision method based on domain knowledge enhancement. The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the engineering project management auxiliary decision method based on domain knowledge enhancement. ​ 9. A computer-readable storage medium, characterized in that, ​

Images