Typical construction period-based capital project collaborative monitoring method and system

Abstract

The invention relates to the technical field of data processing, and discloses a capital project collaborative monitoring method and system based on a typical construction period. The method comprises the following steps: performing index correlation analysis processing on a multi-dimensional collaborative planning system to obtain a collaborative monitoring index system with a hierarchical structure and a data acquisition scheme; based on the data acquisition scheme, performing data acquisition according to the collaborative monitoring index system to obtain target acquisition data, and performing dynamic analysis processing on the target acquisition data to obtain a project monitoring data set; performing multi-level response strategy analysis processing on the project monitoring data set through a decision tree algorithm to obtain a collaborative response mechanism; and performing effect evaluation processing on an execution result of the collaborative response mechanism through an iterative optimization algorithm to obtain a construction period reference parameter set and a monitoring early warning rule. According to the invention, the efficiency and accuracy of capital project collaborative monitoring based on the typical construction period are improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method and system for collaborative monitoring of capital projects based on typical construction periods. Background Art

[0002] As capital projects continue to expand in scale and complexity, project management faces numerous challenges, including multi-party collaboration, resource allocation, and progress control. Currently, the industry is widely adopting information technology for project monitoring, including using project management software to develop plans, using IoT technology to collect field data, and applying data analysis methods to assess progress. These technologies have significantly improved project management and provided data support for project decision-making.

[0003] However, existing project monitoring methods have the following shortcomings: First, project monitoring often relies on experience and judgment, lacks in-depth mining and analysis of historical data, and is difficult to establish scientific project duration benchmark parameters; second, the collaboration efficiency of all parties involved is low, and information sharing is not timely, resulting in delayed problem discovery and slow response; third, the monitoring indicator system is not perfect, making it difficult to achieve multi-dimensional comprehensive evaluation, and the early warning mechanism is relatively simple, making it impossible to make accurate predictions; finally, the resource allocation plan lacks an optimization mechanism, and resource allocation imbalance often occurs. Summary of the Invention

[0004] The present application provides a method and system for collaborative monitoring of capital projects based on typical construction periods, which are used to improve the accuracy of collaborative monitoring of capital projects based on typical construction periods.

[0005] In the first aspect, the present application provides a collaborative monitoring method for capital projects based on typical construction periods, and the collaborative monitoring method for capital projects based on typical construction periods includes: performing feature extraction and cluster analysis on historical project data to obtain typical project construction period benchmark parameters; performing multi-dimensional plan decomposition processing on the typical project construction period benchmark parameters to obtain a multi-dimensional collaborative planning system; performing indicator correlation analysis processing on the multi-dimensional collaborative planning system to obtain a collaborative monitoring indicator system and data collection scheme with a hierarchical structure; based on the data collection scheme, data is collected according to the collaborative monitoring indicator system to obtain target collection data, and the target collection data is dynamically analyzed to obtain a project monitoring data set; multi-level response strategy analysis processing is performed on the project monitoring data set through a decision tree algorithm to obtain a collaborative response mechanism; and the execution results of the collaborative response mechanism are evaluated and processed through an iterative optimization algorithm to obtain a construction period benchmark parameter set and monitoring and early warning rules.

[0006] In a second aspect, the present application provides a capital project collaborative monitoring system based on typical construction periods, the capital project collaborative monitoring system based on typical construction periods comprising: The extraction module is used to perform feature extraction and cluster analysis on historical project data to obtain the typical project duration benchmark parameters; A decomposition module is used to perform multi-dimensional plan decomposition processing on the typical construction period benchmark parameters of the project to obtain a multi-dimensional collaborative planning system; An analysis module is used to perform indicator correlation analysis on the multi-dimensional collaborative planning system to obtain a collaborative monitoring indicator system and data collection plan with a hierarchical structure; A collection module, configured to collect data based on the data collection scheme and the collaborative monitoring indicator system to obtain target collected data, and dynamically analyze and process the target collected data to obtain a project monitoring data set; A processing module, configured to perform multi-level response strategy analysis and processing on the project monitoring data set using a decision tree algorithm to obtain a coordinated response mechanism; The evaluation module is used to evaluate the execution results of the collaborative response mechanism through an iterative optimization algorithm to obtain a construction period benchmark parameter set and monitoring and early warning rules.

[0007] In the technical solution provided by this application, by extracting features and performing cluster analysis on historical project data, scientific and reasonable typical project duration benchmark parameters are obtained, providing a reliable reference basis for duration control. By performing multi-dimensional plan decomposition on the typical project duration benchmark parameters, a multi-dimensional collaborative planning system including progress, resources and interface relationships is formed, realizing the systematic decomposition and collaborative management of project plans. On this basis, by performing indicator correlation analysis on the multi-dimensional collaborative planning system, a hierarchical collaborative monitoring indicator system and data collection scheme are constructed to ensure the comprehensiveness and accuracy of monitoring data. The target data collected based on the data collection scheme is dynamically analyzed and processed to obtain a monitoring data set reflecting the project operation status, providing data support for anomaly identification and early warning. A decision tree algorithm is used to perform multi-level response strategy analysis on the project monitoring data set, establishing a scientific collaborative response mechanism, and improving the timeliness and effectiveness of problem handling. Finally, the execution results of the collaborative response mechanism are evaluated and processed through an iterative optimization algorithm, realizing the dynamic update of the duration benchmark parameters and monitoring and early warning rules, and ensuring the continuous improvement of the monitoring system. This method organically combines data analysis, early warning and prediction, collaborative response and other technologies to form a closed-loop monitoring and management system, significantly improving the construction period control level and collaborative management efficiency of capital projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 This is a schematic diagram of an embodiment of a collaborative monitoring method for capital projects based on typical construction periods in an embodiment of the present application; Figure 2 This is a schematic diagram of an embodiment of a capital project collaborative monitoring system based on a typical construction period in an embodiment of the present application. DETAILED DESCRIPTION

[0010] The embodiments of the present application provide a method and system for collaborative monitoring of capital projects based on typical construction periods. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0011] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In one embodiment of the present application, a collaborative monitoring method for capital projects based on typical construction periods includes: Step S101: Perform feature extraction and cluster analysis on historical project data to obtain typical project duration benchmark parameters; Step S102: Perform multi-dimensional planning decomposition processing on the typical project duration benchmark parameters to obtain a multi-dimensional collaborative planning system; Step S103: performing indicator correlation analysis on the multi-dimensional collaborative planning system to obtain a collaborative monitoring indicator system and data collection plan with a hierarchical structure; Step S104: Based on the data collection plan, data is collected according to the collaborative monitoring indicator system to obtain target collected data, and the target collected data is dynamically analyzed and processed to obtain a project monitoring data set; Step S105: Perform multi-level response strategy analysis and processing on the project monitoring data set using a decision tree algorithm to obtain a coordinated response mechanism; Step S106: The execution result of the collaborative response mechanism is evaluated and processed through an iterative optimization algorithm to obtain a construction period benchmark parameter set and monitoring and early warning rules.

[0012] It is understandable that the execution subject of this application can be a capital project collaborative monitoring system based on typical construction periods, or a terminal or a server, which is not specifically limited here. The embodiment of this application is described by taking the server as the execution subject as an example.

[0013] Specifically, historical project data, including project size (building area, equipment quantity, etc.), investment amount (total investment amount, sub-item investment, etc.), and construction period (start date, completion date, etc.), were cleaned and processed to obtain a standardized historical dataset by removing outliers and filling in missing values. This standardized historical dataset was then segmented into time series, dividing the projects into preparation, construction, and closing phases. Key nodes were defined for each phase, and their time points were recorded to form time series data. The time series data for these key nodes were categorized and statistically analyzed by project type (e.g., industrial construction, civil construction, etc.). Eigenvalues ​​such as the average duration and duration variance of each project at different phases were calculated to generate a distribution of duration characteristics. Principal component analysis (PCA) was used to reduce the dimensionality of the duration distribution characteristics and extract the core factors influencing the duration. For example, in a specific industrial construction project, PCA revealed that equipment procurement cycle time, foundation construction time, and equipment installation and commissioning time were the primary factors influencing the total duration, with contribution rates of 35%, 28%, and 22%, respectively.

[0014] Correlations between core project duration influencing factors were calculated to establish a correlation matrix between the factors. Parameter calibration was then used to determine the weight coefficients for each factor. Projects were classified using a hierarchical clustering algorithm to generate a project type feature library. For example, industrial buildings with an investment of over 1 billion yuan and a construction area of ​​over 50,000 square meters were classified as large industrial buildings. Typical project duration benchmark parameters included: equipment procurement cycle of 12-15 months, foundation construction time of 6-8 months, and equipment installation and commissioning time of 4-6 months. Based on these typical project duration benchmark parameters, a multi-dimensional plan decomposition was performed. First, an overall project plan framework was constructed, and the work content and timelines for each phase were determined, resulting in a key node schedule. A resource requirements analysis was conducted to calculate the required resources, such as manpower, equipment, and materials, for each phase. A resource allocation list was compiled and time-balanced to generate a resource allocation plan. Collaboration interfaces within the plan were identified, and the coordination relationships between the various stakeholders were determined. A plan collaboration node table was generated, ultimately resulting in a multi-dimensional collaborative planning system.

[0015] Conduct indicator correlation analysis on the multi-dimensional collaborative planning system, divide the monitoring dimensions into progress, quality, cost, safety, and other dimensions, and set specific indicator parameters for each dimension. Use hierarchical classification to divide indicators into primary and secondary indicators, calculate the correlation between indicators, and establish an indicator impact relationship matrix. Determine the data source and collection frequency based on the characteristics of the indicators, formulate data collection specifications, and form a collaborative monitoring indicator system and data collection plan. Collect project operation data based on the data collection plan, preprocess and standardize the collected data to obtain a standardized data stream. Extract time series features, use neural network algorithms for data training, set warning thresholds, and predict project status. The neural network adopts a three-layer structure: the input layer contains monitoring indicator data, the hidden layer performs feature learning, and the output layer predicts project status. Identify abnormal patterns, establish warning discrimination rules, and ultimately form a project monitoring data set.

[0016] A decision tree algorithm was used to analyze the project monitoring dataset and construct a multi-level response strategy. Each node in the decision tree represents a judgment condition, such as a schedule delay exceeding 5% or a cost overrun exceeding 3%. Branch nodes correspond to different response measures. Resource requirements were analyzed based on the response decision path, and deployment strategies and priorities were formulated to form a resource allocation plan and establish a coordinated response mechanism. The effectiveness of the coordinated response mechanism was evaluated. Project data after the response was collected, and deviations from the target were calculated to generate performance evaluation indicators. An iterative optimization algorithm was used to continuously adjust parameters and update the construction period baseline parameters and early warning rules. The iterative optimization algorithm used a gradient descent method, and the optimization objective function was: F = w1 × (actual construction period - planned construction period)² + w2 × (actual cost - planned cost)² + w3 × (quality compliance rate - target compliance rate)², where w1, w2, and w3 are weight coefficients.

[0017] For example, a large-scale industrial construction project, based on historical data analysis, established baseline parameters for equipment procurement: 14 months for equipment procurement, 7 months for foundation construction, and 5 months for equipment installation. During project execution, collaborative monitoring revealed a 15-day delay in equipment procurement and a 2% cost overrun. Decision tree analysis identified response plans, including adjusting procurement strategies and increasing construction teams. After implementing these measures, the schedule delay was reduced to 7 days, and the cost overrun was kept below 1%. Through iterative optimization, the updated baseline parameters for the equipment procurement cycle were adjusted to 14.5 months, and the schedule delay threshold in the early warning rule was adjusted from 10 days to 7 days.

[0018] In the embodiment of the present application, by performing feature extraction and cluster analysis on historical project data, scientific and reasonable typical project duration benchmark parameters are obtained, which provides a reliable reference basis for duration control, and by performing multi-dimensional plan decomposition processing on the typical project duration benchmark parameters, a multi-dimensional collaborative planning system including progress, resources and interface relationships is formed, realizing the systematic decomposition and collaborative management of project plans. On this basis, by performing indicator correlation analysis processing on the multi-dimensional collaborative planning system, a hierarchical collaborative monitoring indicator system and data collection scheme are constructed to ensure the comprehensiveness and accuracy of monitoring data. The target data collected based on the data collection scheme is dynamically analyzed and processed to obtain a monitoring data set reflecting the project operation status, providing data support for anomaly identification and early warning. The decision tree algorithm is used to perform multi-level response strategy analysis on the project monitoring data set, establish a scientific collaborative response mechanism, and improve the timeliness and effectiveness of problem handling. Finally, the execution results of the collaborative response mechanism are evaluated and processed through an iterative optimization algorithm, realizing the dynamic update of the duration benchmark parameters and monitoring and early warning rules, and ensuring the continuous improvement of the monitoring system. This method organically combines data analysis, early warning and prediction, collaborative response and other technologies to form a closed-loop monitoring and management system, significantly improving the construction period control level and collaborative management efficiency of capital projects.

[0019] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Data cleaning and processing of project scale, investment amount, and construction period in historical project data are performed to obtain standardized historical data sets, and time series segmentation processing is performed on the standardized historical data sets to obtain time series data of key project nodes; (2) Classify and statistically process the time series data of key project nodes according to project types to obtain the project duration distribution characteristics. Then, perform feature dimensionality reduction on the project duration distribution characteristics using the principal component analysis algorithm to obtain the core duration influencing factors. (3) Calculate the correlation of the core construction period influencing factors to obtain the construction period influencing factor weight matrix, and calibrate the parameters of the construction period influencing factor weight matrix to obtain the construction period benchmark parameter table; (4) The construction period benchmark parameter table is processed by hierarchical clustering algorithm to classify the projects and obtain the project type feature library. The benchmark parameters are extracted based on the project type feature library to obtain the standard construction period parameter set. (5) Perform statistical analysis on the standard construction period parameter set to obtain the key factors affecting the construction period, and perform weight calculation on the key factors affecting the construction period to obtain the typical construction period benchmark parameters of the project.

[0020] Specifically, historical project data was cleaned. This data includes three main categories: project scale (such as building area, equipment quantity, and device size), investment amount (total project investment, sub-project investment, and equipment investment), and construction period (start date, completion date, and duration of each phase). Data cleaning involves removing outliers (values ​​outside the normal range), filling in missing values ​​(using the average value of similar projects), and unifying data formats (using a unified time format and unit conversion), resulting in a standardized, ready-to-use historical dataset. This standardized historical dataset was then segmented into time series, dividing the entire project cycle into three main phases: project preparation (including preliminary approval and design), construction implementation (including civil engineering and installation), and final acceptance. Key milestones were set for each phase, such as design completion, main structure completion, and equipment installation completion. These milestones were recorded to form time series data for key project nodes.

[0021] Time series data for key project nodes were categorized and statistically analyzed by project type (e.g., petrochemical, metallurgical, and power projects). Statistical characteristics such as average duration, standard deviation, and distribution range of each project at different stages were calculated to generate project-specific duration distribution characteristics. Principal component analysis (PCA) was used to reduce the dimensionality of the duration distribution characteristics. PCA first calculated eigenvalues ​​and eigenvectors. Eigenvectors corresponding to eigenvalues ​​with a cumulative contribution rate exceeding 85% were selected as principal components. Ultimately, core factors influencing the duration were identified, such as equipment procurement cycle, construction difficulty coefficient, and climate impact coefficient. Correlations between these core duration-influencing factors were calculated using the Pearson correlation coefficient method. A factor correlation matrix was constructed, and initial weights for each factor were determined using expert scoring. These weights were then adjusted based on the correlation analysis results to generate a weight matrix for the duration-influencing factors. Parameters in the weight matrix were calibrated, and regression analysis was used to determine the impact of each factor on the duration. A duration prediction equation was established, resulting in a baseline duration parameter table.

[0022] A hierarchical clustering algorithm is used to classify the projects in the duration benchmark parameter table. Hierarchical clustering first calculates the Euclidean distance between projects. Then, the shortest distance method is used to gradually merge the closest categories until the preset number of categories is reached, resulting in a project type feature library. Based on the project type feature library, the duration characteristic parameters of each type of project are extracted, such as the standard duration for each stage and the duration fluctuation range, to form a standard duration parameter set. Statistical analysis is performed on the standard duration parameter set, calculating the mean, variance, distribution characteristics, and other statistical quantities of each parameter to identify key factors that significantly affect the duration. The hierarchical analysis method is used to calculate the weights of key factors affecting the duration. First, a judgment matrix is ​​constructed, and after a consistency test, the weight values ​​of each factor are obtained, ultimately forming the typical project duration benchmark parameters.

[0023] Taking a large petrochemical project as an example, standardized data from 100 historical projects was obtained through data cleaning. After time series segmentation, eight key milestones were identified, including design completion, civil construction completion, and equipment installation. Principal component analysis revealed that the cumulative contribution of three factors, namely equipment delivery lead time, foundation construction duration, and equipment installation duration, reached 87%, identifying them as core project duration influencing factors. Correlation analysis revealed a correlation coefficient of 0.82 between equipment delivery lead time and total project duration, 0.75 between foundation construction duration and total project duration, and 0.68 between equipment installation duration and total project duration. Hierarchical clustering was used to categorize projects into three groups: large (investment over 5 billion yuan), medium (investment between 1 billion and 5 billion yuan), and small (investment under 1 billion yuan). Project duration parameters were then extracted for each type. The resulting benchmark parameters for typical project durations for large petrochemical projects were: design period of 12 months (with a fluctuation range of ±1.5 months), civil construction duration of 18 months (with a fluctuation range of ±2 months), equipment installation duration of 15 months (with a fluctuation range of ±2 months), and total project duration of 45 months.

[0024] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) The typical construction period benchmark parameters and the project typical construction period benchmark parameters are processed at the planning level to obtain the overall project planning framework, and the overall project planning framework is decomposed into milestone nodes to obtain the key node schedule; (2) Perform resource demand analysis on the key node schedule to obtain a resource allocation list, and perform time-series balancing on the resource allocation list to obtain a resource allocation plan; (3) Perform interface identification processing on the resource allocation plan to obtain a plan coordination node table, and perform dependency analysis on the plan coordination node table to obtain a multi-dimensional collaborative planning system.

[0025] Specifically, typical duration benchmark parameters (including standard duration values ​​such as design, procurement, and construction periods) are compared and analyzed with typical project duration benchmark parameters (including specific duration parameters for various projects) to establish a hierarchical project plan structure. This plan hierarchy comprises an overall plan layer (overall project duration plan), a specialized plan layer (design, procurement, construction, and other specialized plans), and a control plan layer (monthly, weekly, and other specific implementation plans), forming the overall project plan framework. The overall project plan framework is broken down into milestone nodes, dividing the total project duration into several work packages based on key milestones. Milestone nodes include design completion, equipment ordering, main structure topping-out, single-unit commissioning, and integrated commissioning. Each node is assigned control elements such as planned completion time and deliverables. By analyzing the temporal relationships between milestone nodes, a key node schedule is compiled, clearly defining the planned start time, planned completion time, and duration of each node.

[0026] Based on the key node schedule, a resource demand analysis is performed to calculate the quantity and usage time of human resources (such as designers and construction personnel), equipment resources (such as gantry cranes and crawler cranes), and material resources (such as steel and concrete) required for each work package. Resource demand data is derived from technical documents such as the bill of quantities and construction plan. This data is calculated based on construction process requirements and productivity standards to form a resource allocation list. The resource allocation list is then time-balanced, first identifying peak and trough periods of resource utilization and calculating resource utilization. When resources are over-concentrated or idle, balanced resource use is achieved by adjusting work sequences, optimizing construction processes, and rationally arranging work hours. After time-balancing, a resource allocation plan is developed, containing detailed usage plans and allocation strategies for each resource.

[0027] Interfaces are identified for resource allocation plans, and the work interfaces between various parties involved (such as owners, designers, contractors, and suppliers) are organized. Interface types include time interfaces (time points for work handover), technical interfaces (delivery of technical documents), and management interfaces (communication and coordination methods). The identified interface information is organized into a plan coordination node table, clarifying the responsible parties, coordination content, time requirements, and other elements of each interface. Dependency analysis is performed on the plan coordination node table, and network planning techniques are used to determine the logical relationships between tasks. Dependency types include finish-start (the next task can only begin after the previous task is completed), start-start (two tasks begin simultaneously), and finish-finish (two tasks are completed simultaneously). By analyzing dependencies, critical paths are identified, key priorities for project duration control are determined, and ultimately a multi-dimensional collaborative planning system is formed.

[0028] Taking a large-scale power project as an example, a three-tiered planning system was constructed through hierarchical planning: an overall plan encompassing a 36-month total project duration; specialized plans encompassing 12 months for design, 18 months for procurement, and 24 months for construction; and a control plan broken down into monthly and weekly schedules. Milestone decomposition identified 15 key milestones, including completion of preliminary design (4 months), completion of construction drawings (12 months), ordering of major equipment (6 months), arrival of major equipment (18 months), completion of civil construction (20 months), completion of equipment installation (30 months), and completion of commissioning (36 months). Resource demand analysis revealed peak requirements of 80 designers, 600 construction workers, and 25 units of large machinery. Time balancing reduced the peak demand for construction workers from 600 to 450, and the peak demand for machinery from 25 to 18 units. Interface identification revealed 54 collaborative nodes, including 20 between design and construction and 15 between construction and commissioning. Dependency analysis identified three critical paths throughout the project: the design critical path (12 months), the equipment critical path (18 months), and the construction critical path (24 months), and corresponding control measures and coordination mechanisms were developed accordingly.

[0029] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) The multi-dimensional collaborative planning system is divided into monitoring dimensions to obtain a monitoring indicator parameter set, and the monitoring indicator parameter set is hierarchically classified to obtain an indicator classification table; (2) Calculate the correlation between indicators in the indicator classification table to obtain the indicator impact relationship matrix, and identify the data source of the indicator impact relationship matrix to obtain the data collection element table; (3) The data collection frequency setting process is performed on the data collection element table to obtain the data collection specification, and the data format of the data collection specification is standardized to obtain the collaborative monitoring indicator system and data collection plan.

[0030] Specifically, the multi-dimensional collaborative planning system is divided into five monitoring dimensions, with monitoring indicators categorized into five categories: progress (project progress, workload completion, etc.), quality (construction quality, material quality, etc.), cost (investment control, expense expenditure, etc.), safety (safety incidents, hidden danger rectification, etc.), and resource (manpower allocation, equipment utilization, etc.). Specific monitoring parameters are set for each dimension. For example, the progress dimension includes parameters such as plan completion rate, schedule deviation rate, and milestone completion rate; the quality dimension includes parameters such as first-time acceptance rate, number of quality defects, and rework rate, thus forming a monitoring indicator parameter set. This monitoring indicator parameter set is hierarchically categorized into primary, secondary, and tertiary indicators. Primary indicators are overall indicators for the five dimensions, such as the overall progress completion rate; secondary indicators are key control indicators for each dimension, such as the design progress completion rate and the construction progress completion rate; and tertiary indicators are specific assessment indicators, such as the construction drawing design completion rate and the main structure completion rate. This hierarchical categorization clarifies the calculation methods, assessment standards, and control requirements for each level of indicators, forming an indicator grading table. Correlations were calculated for each indicator in the indicator grading table, using grey correlation analysis to determine the degree of mutual influence between indicators. First, a reference series and a comparison series were determined. The historical data for each indicator was standardized, and the correlation coefficient was calculated to determine the correlation value. A higher correlation value indicates a stronger association between indicators. Indicators with correlations greater than 0.8 showed significant correlations. The results were organized into an indicator influence relationship matrix, where each element represents the strength of the association between two indicators.

[0031] Identify data sources for the indicator impact relationship matrix and sort out the data acquisition methods for each indicator. Data sources include project management platforms, on-site collection equipment, manual reporting, and other channels. Determine the primary and backup data sources for each indicator, clarify the responsible unit for data collection, collection methods, and data format requirements, and form a data collection element table. The data collection element table records in detail the data items, data types, value ranges, units of measurement, and other information for each indicator. Set the collection frequency for each indicator based on the data collection element table, and determine different collection cycles based on the indicator's importance, rate of change, and control requirements. Real-time collection indicators (such as safety monitoring data) are collected once a minute, high-frequency collection indicators (such as construction progress) are collected once a day, and routine collection indicators (such as cost data) are collected once a week to form data collection specifications. The collection specifications clearly stipulate the time nodes, collection methods, quality requirements, and other contents of data collection.

[0032] Data collection specifications are standardized, with unified data formats, units of measurement, and accuracy requirements. For example, time data should be standardized using the "year-month-day hour:minute:second" format, percentage data should be rounded to two decimal places, and monetary amounts should be in units of 10,000 yuan. This standardization ensures data consistency and comparability, ultimately forming a complete collaborative monitoring indicator system and data collection plan.

[0033] Taking a petrochemical project as an example, a monitoring indicator parameter set consisting of 72 specific indicators was determined through monitoring dimension classification. These indicators include 20 indicators for progress, 15 for quality, 12 for cost, 13 for safety, and 12 for resources. The hierarchical indicator classification table contains 5 first-level indicators, 18 second-level indicators, and 49 third-level indicators. Grey correlation analysis revealed a correlation of 0.92 between equipment arrival rate and construction completion rate, 0.88 between construction personnel availability and construction progress, and 0.85 between quality defect rectification rate and first-time acceptance rate. Data source identification revealed that 45% of indicator data came from the project management platform, 35% from on-site data collection equipment, and 20% required manual reporting. The collection frequency was set as follows: 10 indicators require real-time collection, 25 indicators daily, and 37 indicators weekly. Standardization unified eight data formats and 12 units of measurement, creating a comprehensive data collection standard.

[0034] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Based on the data collection plan, data is collected according to the collaborative monitoring indicator system to obtain the target collection data; (2) Perform data preprocessing and standardization on the target collected data to obtain a normalized data stream, and perform time series feature extraction on the normalized data stream to obtain a dynamic feature sequence; (3) The dynamic feature sequence is trained and processed by a neural network to obtain a warning threshold parameter set, and the warning threshold parameter set is processed by indicator prediction calculation to obtain a state prediction result; (4) Perform abnormal pattern recognition processing on the state prediction results to obtain early warning discrimination rules, and perform correlation analysis on the early warning discrimination rules to obtain the project monitoring data set.

[0035] Specifically, data collection is performed for indicators within the collaborative monitoring indicator system based on the frequency and method specified in the data collection plan. Real-time indicators are recorded every minute by data collection equipment, high-frequency indicators are collected daily, and routine indicators are collected periodically. Collected data includes progress data (such as construction drawing design progress, equipment procurement progress, and actual construction progress), quality data (such as first-time acceptance rate and number of quality defects), cost data (such as actual investment and cost expenditure), safety data (such as number of safety incidents and hazard rectification status), and resource data (such as staff attendance rate and equipment utilization rate), forming the target data. Data preprocessing is performed on the target data, including data cleaning (eliminating outliers and erroneous data), missing value processing (using adjacent value filling or mean filling), and data alignment (standardizing sampling time). Standardization is then performed to convert data of different dimensions to the same scale. For example, maximum and minimum value normalization is used to map data to the 0-1 range, resulting in a normalized data stream. Time series features are extracted from normalized data streams to calculate statistical features such as the sliding mean (reflecting data trends), standard deviation (reflecting data fluctuations), and rate of change (reflecting the speed of data changes). Time series pattern features are extracted in combination with project stage features to form a dynamic feature sequence.

[0036] A neural network algorithm is used to train data on dynamic feature sequences. The neural network structure consists of an input layer (for each indicator's characteristic data), a hidden layer (for data feature learning), and an output layer (for early warning status determination). The training dataset contains both normal and abnormal data samples from historical projects. The network weights are adjusted using a backpropagation algorithm, enabling the network to develop early warning capabilities. After training, a set of early warning threshold parameters is generated for each indicator, including a level 1 warning threshold (indicating attention), a level 2 warning threshold (requiring intervention), and a level 3 warning threshold (requiring action). Indicator prediction is then calculated based on the warning threshold parameter set. Based on current data and historical trends, indicator changes over a period of time are predicted to produce a status prediction result. Anomaly pattern recognition is performed on the status prediction results, and the predicted values ​​are compared with the warning thresholds to identify anomalies exceeding the threshold range. Anomaly patterns include sudden anomalies (sudden changes in indicators within a short period of time), gradual anomalies (continuous deviations from target values), and periodic anomalies (unusual fluctuations in indicators). Based on the characteristics and impact of the anomaly patterns, corresponding early warning identification rules are developed. Correlation analysis is performed on the warning identification rules to study the warning correlations between different indicators. A warning transmission mechanism is established, ultimately forming a complete project monitoring dataset.

[0037] Taking a nuclear power plant construction project as an example, according to the data collection plan, data for 72 monitoring indicators were collected daily, including 35 progress indicators, 15 quality indicators, 12 cost indicators, and 10 safety indicators. Data preprocessing removed outliers, accounting for 3% of the total data volume, and filled in 1.5% of missing values. After normalization, the 10-day moving average, standard deviation, and rate of change of each indicator were calculated. A three-layer neural network was used for training, with 72 nodes in the input layer (corresponding to the 72 indicators), 120 nodes in the hidden layer, and 3 nodes in the output layer (corresponding to the three warning levels). Training was performed using 1,000 historical data samples, including 850 normal samples and 150 abnormal samples. The network converged after 200 training rounds. The warning thresholds for the main structure construction progress indicators were set as follows: Level 1 (progress deviation of 5%-10%), Level 2 (progress deviation of 10%-15%), and Level 3 (progress deviation exceeding 15%). Forecast calculations indicate that a construction section will experience a 12% progress deviation over the next 15 days, triggering a Level 2 alert. Anomaly pattern recognition revealed this to be a gradual anomaly, significantly correlated with labor shortages and low equipment availability, with correlation coefficients of 0.85 and 0.82, respectively. This necessitates timely adjustments to resource allocation plans.

[0038] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) The project monitoring data set is divided into early warning levels to obtain a hierarchical response strategy table, and the decision rules are extracted from the hierarchical response strategy table to obtain a response decision path; (2) Perform resource demand analysis on the response decision path to obtain a resource allocation strategy, and perform priority calculation on the resource allocation strategy to obtain a resource allocation plan; (3) A collaborative mechanism is constructed for the resource allocation plan to obtain a collaborative response mechanism.

[0039] Specifically, the project monitoring dataset was divided into three levels based on warning levels: Level 1 (minor indicator anomalies requiring attention), Level 2 (significant indicator anomalies requiring intervention), and Level 3 (serious indicator anomalies requiring action). Response strategies were defined for each warning level, including response timelines (within 24 hours for Level 1, 12 hours for Level 2, and 4 hours for Level 3), response measures (enhanced monitoring for Level 1, implementation adjustments and optimization for Level 2, and activation of the emergency plan for Level 3), and responsible individuals (project managers for Level 1, responsible leaders for Level 2, and responsible general managers for Level 3). This resulted in a hierarchical response strategy table. A decision tree algorithm was applied to the hierarchical response strategy table to extract decision rules. Decision tree nodes contain judgment conditions (e.g., schedule delay exceeding 10%, cost overrun exceeding 5%, or number of quality defects exceeding 3), and branch paths correspond to different action measures. By analyzing historical case studies, effective decision rules were extracted and categorized according to factors such as warning level, response type, and action measures, resulting in a response decision path. This path defines the complete process from warning triggering to problem resolution.

[0040] A resource demand analysis is conducted based on the response decision path, calculating the quantity of various resources required for the response measures. Resource types include human resources (such as technical personnel and construction personnel), equipment resources (such as construction machinery and testing equipment), material resources (such as building materials and spare parts), and financial resources. Based on the workload and implementation cycle of the response measures, combined with the production and utilization efficiency of resources, resource demand is calculated to form a resource allocation strategy. The resource allocation strategy includes a resource input plan, utilization plan, and withdrawal schedule. Priority is calculated for the resource allocation strategy, using evaluation metrics such as the impact on schedule (number of days of delay), cost impact (amount of cost overrun), quality impact (level of quality defects), and safety impact (level of safety risk). The analytic hierarchy process (AHP) is used to determine the weight of each metric, calculate the overall score of different allocation plans, and rank them by score to produce a resource allocation plan. The resource allocation plan details the specific allocation quantity, utilization time, and deployment sequence of each resource.

[0041] Based on the resource allocation plan, a collaborative mechanism is established, including a coordination organizational structure (including decision-making, management, and execution levels), a communication and coordination mechanism (including regular meetings, reporting, and liaison systems), and an assessment and reward and punishment mechanism (including performance assessment standards and reward and punishment measures) to form a complete collaborative response mechanism. This collaborative response mechanism ensures effective cooperation among all participants in the early warning process.

[0042] For example, a chemical plant construction project revealed anomalies in the progress of major equipment installation. Alert level classification revealed a 15-day delay and a 12% deviation rate, placing it at level 2. Decision rule extraction indicated that when equipment installation delays exceed 10 days and the deviation rate exceeds 10%, additional construction teams and adjustments to construction processes are necessary. Resource requirement analysis revealed that the necessary measures would require the addition of 30 welders, two cranes, and five fixtures, with a construction period of 20 days. Priority calculation revealed that this issue had a schedule impact of 0.8 (weight 0.4), a cost impact of 0.6 (weight 0.3), a quality impact of 0.7 (weight 0.2), and a safety impact of 0.5 (weight 0.1), resulting in an overall score of 0.69, ranking second among all pending issues. The final decision was to redeploy welders from other projects, temporarily rent cranes from an equipment rental company, and relocate fixtures from the spare parts warehouse. A three-tiered coordination mechanism was established: the project manager chaired daily coordination meetings, the M&E construction manager reported progress every four hours, and the site supervisor inspected construction every two hours. After 10 days of implementation, schedule delays were reduced to five days, the deviation rate dropped to 4%, and the early warning was lifted.

[0043] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Collect and process the effect data of the processing results of the coordinated response mechanism to obtain a response effect data set, and perform deviation analysis on the response effect data set to obtain effect evaluation indicators; (2) Performing target comparison processing on the effect evaluation indicators to obtain the improvement direction parameters, and performing baseline parameter update processing on the improvement direction parameters to obtain candidate baseline parameters; (3) Perform rule extraction processing on the candidate benchmark parameters to obtain the warning rule set, and verify the warning rule set to obtain the construction period benchmark parameters and monitoring warning rules.

[0044] Specifically, data on the effectiveness of the coordinated response mechanism's results is collected, including various indicators before and after the implementation of the response measures, such as progress indicators (such as schedule deviation and completion rate), quality indicators (such as first-time acceptance rate and number of defects), cost indicators (such as investment deviation and expenditure), and resource indicators (such as staff utilization and equipment utilization). The collected data is organized in a time series, recording the changes in indicators at each time point to form a response effect dataset. Deviation analysis is performed on this response effect dataset, calculating the change and rate of change in the indicators before and after the implementation of the response measures. The absolute deviation value is calculated by calculating the difference between the actual value and the target value, and the relative deviation rate is calculated by calculating the ratio of the deviation value to the target value. Statistical analysis of the deviation data is performed, calculating statistical quantities such as average deviation, maximum deviation, and cumulative deviation, and analyzing the deviation trends to obtain effect evaluation indicators. These evaluation indicators reflect the effectiveness and implementation efficiency of the response measures.

[0045] Compare and analyze the effect evaluation indicators with the project goals. The target values ​​include control requirements such as the planned construction period, quality standards, and cost limits. Through comparative analysis, identify indicators that are significantly different from the targets, and analyze the reasons for the differences, such as unreasonable setting of construction period benchmark parameters, overly loose warning thresholds, and unsatisfactory response measures. Based on the analysis results, determine the direction and content that need to be improved, and form improvement direction parameters. The improvement direction parameters clarify the specific content that needs to be adjusted, such as construction period benchmark parameters and warning rules. The improvement direction parameters are updated with baseline parameters, and the original parameters are revised based on the response effect data. The baseline parameter update adopts the sliding weighted average method, which comprehensively considers the historical baseline values ​​and the current response results to determine the new parameter values. Through iterative calculation, a set of candidate baseline parameters are obtained, including construction period parameters, quality control parameters, cost management parameters, etc. for each stage.

[0046] Rules are extracted from candidate benchmark parameters, and early warning judgment conditions are set based on parameter characteristics. A decision tree algorithm is used to establish early warning rules, dividing the parameter value range into normal, concern, and warning intervals, and determining the processing requirements for each interval. Through rule induction and optimization, an early warning rule set is formed. The early warning rule set includes a complete set of early warning trigger conditions and response measures. The early warning rule set is validated using typical case data. The accuracy and applicability of the rules are evaluated by comparing and analyzing the degree of conformity between the rule judgment results and the actual situation. Based on the validation results, the rules are optimized and adjusted, and the final construction period benchmark parameters and monitoring and early warning rules are determined.

[0047] Taking a petrochemical plant construction project as an example, the effectiveness of addressing schedule delays during the equipment installation phase was evaluated. Response data showed that after implementing measures such as increasing construction teams and adjusting construction processes, a 15-day schedule delay was reduced to 5 days within 10 days, and the schedule deviation rate dropped from 12% to 4%. Deviation analysis revealed a schedule recovery rate of 1 day per day (recovering one day of delay per day), exceeding the original target of 0.8 days per day. The increased resource investment cost was 520,000 yuan, within budget. The quality acceptance rate remained at 98%, meeting requirements. Comparison with project goals revealed that the original baseline parameters for the equipment installation phase were too tight and needed to be adjusted. The warning threshold was also set too high, preventing timely detection of issues. Improvement plans included adjusting the equipment installation phase baseline from 60 to 65 days, lowering the first-level warning threshold from 10% to 8%, and the second-level warning threshold from 15% to 12%. Using a sliding weighted average calculation, the new baseline parameters were determined: a 65-day equipment installation phase duration, with a tolerance of ±5 days. The warning rules were set as follows: a 5-8% schedule deviation triggers a Level 1 warning, an 8-12% schedule deviation triggers a Level 2 warning, and a 12% schedule deviation triggers a Level 3 warning. The new parameters and rules were validated and applied to equipment installation work for three subsequent subprojects. The deviation detection rate increased to 95%, and the warning accuracy rate reached 92%. The validation results demonstrate that the adjusted parameters and rules are more reasonable and effective.

[0048] The above describes the collaborative monitoring method of capital projects based on typical construction periods in the embodiment of the present application. The following describes the collaborative monitoring system of capital projects based on typical construction periods in the embodiment of the present application. Figure 2 In one embodiment of the present application, a capital project collaborative monitoring system based on typical construction periods includes: Extraction module 201, used to perform feature extraction and cluster analysis on historical project data to obtain typical project duration benchmark parameters; Decomposition module 202, configured to perform multi-dimensional plan decomposition processing on the typical project duration benchmark parameters to obtain a multi-dimensional collaborative plan system; An analysis module 203 is used to perform an indicator correlation analysis on the multi-dimensional collaborative planning system to obtain a collaborative monitoring indicator system and a data collection plan with a hierarchical structure; The collection module 204 is used to collect data based on the data collection scheme and the collaborative monitoring indicator system to obtain target collected data, and dynamically analyze and process the target collected data to obtain a project monitoring data set; Processing module 205, configured to perform multi-level response strategy analysis and processing on the project monitoring data set using a decision tree algorithm to obtain a coordinated response mechanism; The evaluation module 206 is used to perform effect evaluation on the execution result of the collaborative response mechanism through an iterative optimization algorithm to obtain a construction period benchmark parameter set and monitoring and early warning rules.

[0049] Through the collaborative efforts of these components, and through feature extraction and cluster analysis of historical project data, scientifically reasonable benchmark parameters for typical project durations are derived, providing a reliable reference for duration control. Furthermore, through multidimensional plan decomposition of these typical project duration benchmark parameters, a multidimensional collaborative planning system, encompassing schedule, resource, and interface relationships, is formed, achieving systematic decomposition and collaborative management of project plans. Furthermore, through indicator correlation analysis within the multidimensional collaborative planning system, a hierarchical collaborative monitoring indicator system and data collection scheme are constructed, ensuring the comprehensiveness and accuracy of monitoring data. Dynamic analysis and processing of target data collected by the data collection scheme yield a monitoring dataset reflecting the project's operational status, providing data support for anomaly identification and early warning. A decision tree algorithm is employed to analyze multi-level response strategies within the project monitoring dataset, establishing a scientific collaborative response mechanism and improving the timeliness and effectiveness of problem resolution. Finally, an iterative optimization algorithm is employed to evaluate the effectiveness of the collaborative response mechanism's execution, enabling dynamic updating of duration benchmark parameters and monitoring and early warning rules, ensuring the continuous improvement of the monitoring system. This method organically combines data analysis, early warning and prediction, collaborative response and other technologies to form a closed-loop monitoring and management system, significantly improving the construction period control level and collaborative management efficiency of capital projects.

[0050] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A collaborative monitoring method for capital projects based on typical construction periods, characterized in that: The collaborative monitoring method for capital projects based on typical construction periods includes: Perform feature extraction and cluster analysis on historical project data to obtain typical project duration benchmark parameters; Perform multi-dimensional planning decomposition on the typical project duration benchmark parameters to obtain a multi-dimensional collaborative planning system; Performing indicator correlation analysis on the multi-dimensional collaborative planning system to obtain a collaborative monitoring indicator system and data collection plan with a hierarchical structure; Based on the data collection scheme, data is collected according to the collaborative monitoring indicator system to obtain target collected data, and the target collected data is dynamically analyzed and processed to obtain a project monitoring data set; Performing multi-level response strategy analysis and processing on the project monitoring data set using a decision tree algorithm to obtain a coordinated response mechanism; The execution results of the collaborative response mechanism are evaluated and processed through an iterative optimization algorithm to obtain a construction period benchmark parameter set and monitoring and early warning rules.

2. The collaborative monitoring method for capital projects based on typical construction periods according to claim 1, characterized in that: The feature extraction and cluster analysis of historical project data are performed to obtain typical project duration benchmark parameters, including: Performing data cleaning on the project scale, investment amount, and construction period in the historical project data to obtain a standardized historical data set, and performing time series segmentation processing on the standardized historical data set to obtain time series data of key project nodes; Classify and statistically process the time series data of the key nodes of the project according to the project type to obtain the project type duration distribution characteristics, and perform feature dimensionality reduction processing on the project type duration distribution characteristics using the principal component analysis algorithm to obtain the core duration influencing factors; Performing correlation calculation processing on the core construction period influencing factors to obtain a construction period influencing factor weight matrix, and performing parameter calibration processing on the construction period influencing factor weight matrix to obtain a construction period benchmark parameter table; Performing project classification processing on the construction period benchmark parameter table using a hierarchical clustering algorithm to obtain a project type feature library, and performing benchmark parameter extraction processing based on the project type feature library to obtain a standard construction period parameter set; Statistical analysis is performed on the standard construction period parameter set to obtain key factors affecting the construction period, and weight calculation is performed on the key factors affecting the construction period to obtain typical project construction period benchmark parameters.

3. The collaborative monitoring method for capital projects based on typical construction periods according to claim 1, characterized in that: The multi-dimensional planning decomposition process of the typical project duration benchmark parameters is performed to obtain a multi-dimensional collaborative planning system, including: Performing a plan hierarchy construction process on the typical construction period benchmark parameters and the project typical construction period benchmark parameters to obtain a project overall plan framework, and performing a milestone node decomposition process on the project overall plan framework to obtain a key node plan table; Performing resource demand analysis on the key node schedule to obtain a resource allocation list, and performing time-series balancing on the resource allocation list to obtain a resource allocation plan; Interface identification processing is performed on the resource allocation plan to obtain a plan collaboration node table, and dependency analysis processing is performed on the plan collaboration node table to obtain a multi-dimensional collaboration plan system.

4. The collaborative monitoring method for capital projects based on typical construction periods according to claim 1, characterized in that: The multi-dimensional collaborative planning system is subjected to an indicator correlation analysis process to obtain a collaborative monitoring indicator system and a data collection scheme having a hierarchical structure, including: Performing monitoring dimension division processing on the multi-dimensional collaborative planning system to obtain a monitoring indicator parameter set, and performing hierarchical classification processing on the monitoring indicator parameter set to obtain an indicator classification table; Performing correlation calculation processing on the indicator grading table to obtain an indicator impact relationship matrix, and performing data source identification processing on the indicator impact relationship matrix to obtain a data collection element table; The data collection element table is processed for collection frequency setting to obtain data collection specifications, and the data collection specifications are processed for data format standardization to obtain a collaborative monitoring indicator system and a data collection plan.

5. The collaborative monitoring method for capital projects based on typical construction periods according to claim 1, characterized in that: The data collection scheme is based on the collaborative monitoring indicator system to collect data to obtain target collected data, and the target collected data is dynamically analyzed and processed to obtain a project monitoring data set, including: Based on the data collection scheme, data is collected according to the collaborative monitoring indicator system to obtain target collection data; Performing data preprocessing and standardization on the target collected data to obtain a normalized data stream, and performing time series feature extraction on the normalized data stream to obtain a dynamic feature sequence; Performing data training processing on the dynamic feature sequence through a neural network to obtain a warning threshold parameter set, and performing indicator prediction calculation processing on the warning threshold parameter set to obtain a state prediction result; An abnormal pattern recognition process is performed on the state prediction result to obtain an early warning discrimination rule, and an association analysis process is performed on the early warning discrimination rule to obtain a project monitoring data set.

6. The collaborative monitoring method for capital projects based on typical construction periods according to claim 1, characterized in that: The project monitoring data set is subjected to multi-level response strategy analysis and processing by a decision tree algorithm to obtain a coordinated response mechanism, including: Performing early warning level classification processing on the project monitoring data set to obtain a hierarchical response strategy table, and performing decision rule extraction processing on the hierarchical response strategy table to obtain a response decision path; Performing resource demand analysis on the response decision path to obtain a resource allocation strategy, and performing priority calculation on the resource allocation strategy to obtain a resource allocation plan; A collaborative mechanism construction process is performed on the resource allocation scheme to obtain a collaborative response mechanism.

7. The collaborative monitoring method for capital projects based on typical construction periods according to claim 1, characterized in that: The execution results of the collaborative response mechanism are evaluated and processed through an iterative optimization algorithm to obtain a construction period benchmark parameter set and monitoring and early warning rules, including: Performing effect data collection processing on the processing results of the collaborative response mechanism to obtain a response effect data set, and performing deviation analysis processing on the response effect data set to obtain an effect evaluation index; Performing target comparison processing on the effect evaluation index to obtain an improved direction parameter, and performing benchmark parameter update processing on the improved direction parameter to obtain a candidate benchmark parameter; A rule extraction process is performed on the candidate benchmark parameters to obtain an early warning rule set, and a verification process is performed on the early warning rule set to obtain a construction period benchmark parameter and monitoring early warning rules.

8. A capital project collaborative monitoring system based on typical construction periods, for implementing the capital project collaborative monitoring method based on typical construction periods as claimed in any one of claims 1 to 7, characterized in that: The typical construction period-based capital project collaborative monitoring system includes: The extraction module is used to perform feature extraction and cluster analysis on historical project data to obtain the typical project duration benchmark parameters; A decomposition module is used to perform multi-dimensional plan decomposition processing on the typical construction period benchmark parameters of the project to obtain a multi-dimensional collaborative planning system; An analysis module is used to perform indicator correlation analysis on the multi-dimensional collaborative planning system to obtain a collaborative monitoring indicator system and data collection plan with a hierarchical structure; A collection module, configured to collect data based on the data collection scheme and the collaborative monitoring indicator system to obtain target collected data, and dynamically analyze and process the target collected data to obtain a project monitoring data set; A processing module, configured to perform multi-level response strategy analysis and processing on the project monitoring data set using a decision tree algorithm to obtain a coordinated response mechanism; The evaluation module is used to evaluate the execution results of the collaborative response mechanism through an iterative optimization algorithm to obtain a construction period benchmark parameter set and monitoring and early warning rules.

Images