Target deviation dynamic identification method for different stages of reservoir EPC project

Abstract

The invention relates to the technical field of project management analysis, in particular to a method for dynamically identifying target deviations in different stages of a reservoir EPC project, which comprises the following steps of: constructing a three-stage associated multiple regression model, and configuring a dynamic influence coefficient for each EPC stage node; acquiring operation data of each stage in real time, performing stage weight distribution on the data by utilizing the dynamic influence coefficient, and generating a node state vector comprising a stage identifier; and S3, establishing a Markov transfer chain according to the stage identifier of the node state vector, calculating a deviation transfer probability between adjacent stages, and when the deviation transfer probability exceeds a warning threshold, triggering a reverse stage compensation instruction and updating the dynamic influence coefficient of S1 to form closed-loop regulation and control. According to the method, under the scenes of large topographic relief and complex conduction paths in reservoir projects, the method has higher environmental adaptability and real-time performance, and the stability and robustness of deviation chain diagnosis are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of project management analysis, and in particular to a method for dynamically identifying target deviations at different stages of a reservoir EPC project. Background Art

[0002] As a major infrastructure project, reservoir projects generally adopt the EPC general contracting model that integrates design, procurement and construction to improve overall progress coordination and resource allocation efficiency. Under this model, the target connection relationship between different stages is highly coupled. Deviations in any stage may cause a cumulative amplification effect on subsequent stages, thereby affecting the quality, safety and schedule control of the entire project.

[0003] Existing technologies often rely on static schedules and single-stage data monitoring to manage project deviations, lacking cross-stage information integration mechanisms and dynamic response capabilities. This prevents timely feedback of design change information into procurement and construction plans. Furthermore, delays or quality issues arising during procurement and construction lack a traceable source identification mechanism, resulting in delayed response and difficulty in determining responsibility, impacting the efficiency and accuracy of target closed-loop control. Summary of the Invention

[0004] The present invention provides a target deviation dynamic identification method for different stages of reservoir EPC projects. The target deviation identification method combines multi-stage key parameters, identifies deviation transmission paths, and has dynamic feedback adjustment capabilities. It is suitable for realizing coordinated regulation and risk suppression across the entire project process and across links in complex reservoir projects.

[0005] A method for dynamically identifying target deviations at different stages of a reservoir EPC project includes the following steps: S1, stage target association modeling: Based on the BIM parameters of the design phase, the material list of the procurement phase, and the process plan of the construction phase, a three-phase correlation multivariate regression model is constructed, and a dynamic influence coefficient is configured for each EPC phase node. The dynamic influence coefficient is calculated and generated based on the cross-phase deviation transmission law in historical data; S2, multi-stage data fusion and state quantization: Real-time collection of operational data for each stage, including material arrival time differences at procurement nodes, mechanical efficiency values ​​at construction nodes, and parameter change frequencies at design nodes, is performed. The data is weighted according to the dynamic influence coefficients, generating a node state vector including a stage identifier. S3: Cross-stage deviation propagation identification and closed-loop control: A Markov transfer chain is established based on the stage identifier of the node state vector, and the deviation transfer probability between adjacent stages is calculated. When the transfer probability of design→procurement or procurement→construction exceeds the warning threshold, the reverse stage compensation instruction is triggered and the dynamic influence coefficient of S1 is updated to form a closed-loop control.

[0006] Optionally, the S1 specifically includes: S11, extracting a set of physical parameters of components in the BIM model in the design phase, including concrete strength grade, steel bar specifications, and anti-seepage coefficient, and generating a design feature matrix with a unique component ID; S12, matching the material technical parameters in the purchase material list with the design feature matrix, establishing a component ID → material code association mapping table, and calculating the impact weight of material parameter deviation on design parameters based on historical supply data, recorded as the purchase impact factor ; S13: Analyze the process constraints in the construction process plan, identify the key construction paths that are strongly related to the design components, establish a multivariate regression model based on the historical construction logs, and calculate the construction delay days, design parameter deviations, and material arrival delays to output the construction influencing factors. ; S14, introduce the environmental sensitivity coefficient and calculate the dynamic impact coefficient across stages.

[0007] Optionally, the construction impact factor It is obtained by constructing a multiple regression model, which is expressed as: , Indicates the actual number of days of delay on the critical path of the component; is the mean value of parameter deviation caused by procurement, defined as: , is the regression coefficient obtained from model training, For components The construction impact factor reflects the comprehensive impact of construction plan deviation on design objectives. For components factors influencing purchasing.

[0008] Optionally, the dynamic influence coefficient is expressed as: ; in, For components The dynamic impact coefficient in the current cycle, is the influence coefficient of the previous cycle, These are the procurement and construction impact factors calculated during this period.

[0009] Optionally, the collection of operation data at each stage in S2 specifically includes: The design node calls the version management interface of the building information model to extract the parameter change operations in the record. Within the set time window, the number of change events of all design parameters is counted to measure the dynamic change frequency of the design link. The procurement node connects to the material transportation scheduling platform, analyzes the geographic trajectory data of the transport vehicles, and extracts the time difference between the actual arrival time and the scheduled time. At the same time, combined with the acceptance information based on the blockchain, it eliminates the records with failed signature verification or anomalies, and only retains the valid data; The construction node obtains the workload data per unit time through the engineering machinery controller, and combines the rated power information of the equipment to calculate the actual work efficiency under the current operation state as the mechanical efficiency value of the construction node.

[0010] Optionally, the stage weight distribution specifically includes: Based on the extracted procurement influencing factors and environmental sensitivity coefficients, corresponding fusion weights are assigned to design nodes to reflect the sensitivity of the design process to downstream deviations; According to the construction impact factor and the cross-stage dynamic image coefficient, the fusion weight is assigned to the procurement node to quantify the amplification or buffering effect of the intermediate link in the transmission of target deviation; The fusion weight of the construction node is determined by the sum of the weights of the design and procurement nodes, so that the overall weight maintains a normalized distribution among the three stages and ensures the quantitative closure of data fusion.

[0011] Optionally, the design node vector includes its stage identification, weighted parameter change frequency index and corresponding component number; the procurement node vector includes the stage identification, weighted arrival time difference index and material batch number; the construction node vector includes the stage identification, weighted work efficiency index and associated process code, and S2 also includes vectorized coding, which performs structured expression after weight correction on the data of the three nodes, and adds the stage identification and auxiliary information to form a node state vector.

[0012] Optionally, the establishment of the Markov transition chain in S3 includes defining a set of state transition directions according to the stage identifier in the node state vector, , construct a binary transition probability matrix, where: D→P transition probability ,The percentage of procurement node abnormalities after design node abnormalities in historical data is ,calculated; P→C transition probability ,Count the percentage of construction node abnormalities after procurement node abnormalities.

[0013] Optionally, the S3 also includes the setting of warning values: based on the reservoir environment complexity index Setting alert thresholds , where the more complex the terrain The larger the value, the or Activates the generation of reverse compensation instructions.

[0014] Optionally, the generating of the reverse compensation instruction includes: If the probability of D→P exceeds the standard, a parameter optimization instruction is sent to the design node, and the weight of the procurement influencing factor α is simultaneously reduced; If the probability of P→C exceeds the limit, the construction resource reallocation instruction is triggered and the construction resource is reallocated according to the actual number of days of delay. Update construction impact factors; The actual deviation data after compensation is input into the multivariate regression model of S1, and the dynamic impact coefficient is recalculated to form a cross-stage control closed loop.

[0015] Beneficial effects of the present invention: This invention, by constructing a multivariate regression model linking the design, procurement, and construction phases, combined with multi-source data fusion and a dynamic weight allocation strategy, can capture the state changes of key nodes in real time and vectorize them into structured state codes. Compared to traditional approaches that rely on single-stage progress tracking, this invention can identify potential deviation nodes in the cross-stage transmission chain, shifting target control from "delayed response" to "feedforward warning," thereby improving the accuracy of deviation identification and the timeliness of intervention during the EPC process.

[0016] The present invention integrates the terrain complexity index K calculated by the geographic information system, adaptively adjusts the warning threshold of deviation transfer, and cooperates with the sparse Markov chain model that only retains the D→P→C path to reduce the false alarm rate and computational complexity. In scenarios such as reservoir projects with large terrain undulations and complex conduction paths, the present invention has stronger environmental adaptability and real-time performance, and improves the stability and robustness of deviation chain diagnosis.

[0017] The present invention activates the reverse compensation instruction through deviation propagation analysis, links the influencing factors in the S1 stage with the regression model to perform incremental coefficient updates, and forms an adaptive closed-loop path of "anomaly identification-compensation adjustment-model reconstruction". Compared with the static target assessment mechanism in the traditional EPC system, the present invention has the dynamic regulation capability under the scenario of continuous evolution of target deviation, can effectively suppress system oscillation and maintain the consistency and controllability of the whole process target. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 Schematic diagram of a method flow in an embodiment of the present invention; Figure 2 Schematic diagram of multi-stage data fusion and state quantization according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Those skilled in the art may also implement some known technologies in other alternative ways. The accompanying drawings are only for describing the embodiments in more detail and are not intended to limit the present invention in any specific way.

[0021] like Figure 1-Figure 2 As shown, a method for dynamically identifying target deviations at different stages of a reservoir EPC project includes the following steps: S1, stage target association modeling: Based on the BIM parameters of the design phase, the material list of the procurement phase, and the process plan of the construction phase, a three-stage correlation multivariate regression model is constructed. A dynamic influence coefficient is configured for each EPC phase node. The dynamic influence coefficient is calculated and generated based on the cross-stage deviation transmission law in historical data. S2, multi-stage data fusion and state quantization: Real-time collection of operational data for each stage, including material arrival time differences at procurement nodes, mechanical efficiency values ​​at construction nodes, and parameter change frequencies at design nodes. Dynamic influence coefficients are used to assign stage weights to the data and generate node state vectors that include stage identifiers. S3: Cross-stage deviation propagation identification and closed-loop control: A Markov transfer chain is established based on the stage identifier of the node state vector, and the deviation transfer probability between adjacent stages is calculated. When the transfer probability of design→procurement or procurement→construction exceeds the warning threshold, the reverse stage compensation instruction is triggered and the dynamic influence coefficient of S1 is updated to form a closed-loop control.

[0022] S1 specifically includes: S11, extract the physical parameter set of the components in the BIM model in the design phase, including concrete strength grade, steel bar specifications and anti-seepage coefficient, and generate a design feature matrix with a unique component ID. The build is as follows:

[0023] in, For the A unique identifier for a component, Indicates the concrete strength grade (C30, C35, etc.), For steel bar specifications, is the anti-seepage grade coefficient, The total number of components.

[0024] S12, match the material technical parameters in the purchase material list with the design feature matrix and establish the component ID Material code association mapping table: , and match the design parameters of each component with the procurement parameters to construct the deviation matrix: ; Calculate the weight of the impact of material parameter deviation on design parameters based on historical supply data, and record it as the procurement impact factor , further calculate the purchasing impact factor : ;in, The actual supply value of the corresponding component parameter in the purchase list, that is, The first component that matches the purchase list The delivery value of each technical parameter, is the design parameter value, is the relative deviation, For the The influence weight of the item parameter satisfies ,determined by simulation or expert scoring method, For components factors influencing purchasing.

[0025] i represents a component (uniquely identified by the component ID in the BIM model); j represents the jth design technical parameter related to the component, for example: j=1: concrete strength grade (for example, the design value is C30, and the corresponding purchase and supply value is the compressive strength MPa value, such as 32.5 MPa); j=2: steel bar diameter and yield strength (for example, if the design is HRB400-16 and the supply is marked as HRB400, the actual yield strength is 420MPa); j=3: Anti-seepage performance (e.g., if the design anti-seepage grade is P6 and a certain brand of water-stop material is purchased, the corresponding permeability coefficient is obtained); S13: Analyze the process constraints in the construction process plan, identify the key construction paths that are strongly related to the design components, establish a multivariate linear regression model based on the historical construction logs, and calculate the construction delay days, design parameter deviations, and material arrival delays to output the construction influencing factors. First, extract the critical path extension information and historical records associated with the component, including the actual extension days on the critical path where the component is located ; Mean value of parameter deviation caused by procurement , defined as: , using multiple linear regression model to fit construction influencing factors : ,in, is the regression coefficient obtained by model training, which meets the normalization conditions ,Training based on historical project log dataset, For components The construction influencing factor reflects the comprehensive impact of construction plan deviation on design objectives.

[0026] S14, introduction of environmental sensitivity coefficient When the storage area temperature hour, ,otherwise , defined as follows: , in In each cycle, the components The cross-stage dynamic influence coefficient To make dynamic updates: ; in, For components The dynamic impact coefficient in the current cycle, is the influence coefficient of the previous cycle, the initial value Determined by looking up the table of project types, such as "dam type = 0.6, diversion tunnel type = 0.4", These are the procurement and construction impact factors calculated during this period.

[0027] S2 specifically includes: S21, real-time collection of multi-source data: S211, Design Node Data Collection: Capture parameter change events through the version control interface of the BIM platform and The frequency of changes in the internal statistical design parameters is defined as: ;in, is the frequency of parameter changes of the design node, is the total number of design parameter change events within the time window, The duration of the statistical period; S212, procurement node data collection: Analyze GPS trajectory data of the material transportation system to obtain the actual arrival time and planned time , and combined with the blockchain inspection records to filter out abnormal data and calculate the net arrival delay: ; S213, construction node data collection: read the workload per unit working hour from the construction machinery controller , and based on rated power Calculating mechanical efficiency : ; S22, dynamic weight allocation: According to the key factors obtained in S1, dynamic weights are assigned to the three types of nodes: Design node weights: ; Procurement node weight: ; Construction node weight: ; in, is the purchasing influencing factor, is the construction influencing factor, is the environmental sensitivity coefficient, is the cross-stage dynamic influence coefficient, Represent the fusion weights of design, procurement, and construction nodes respectively. If , then the automatic normalization process ensures that the total weight is 1.

[0028] S23, vectorized encoding: Encode the data of the three stages according to the weights and add identification information to form a node state vector in a unified format.

[0029] Design node vector: ; Sourcing Node Vector: ; Construction node vector: ; in, is the phase identifier, is the component ID, The batch number of the material. It is the process code, and the second item in the vector is the fused weighted feature value, which is used for subsequent state sequence processing or deviation identification modeling.

[0030] Design node weights middle is the purchasing influence factor, which indicates the feedback degree of purchasing error on the design goal. is the environmental sensitivity coefficient, reflecting the amplifying effect of external factors such as high temperature on design changes. The combination of these two factors represents the indirect impact pressure brought by "actual deviations in the procurement process + environmental risks" on design objectives. Therefore, the weight of the design node is set to the product of these two factors to reflect its upstream sensitivity to deviation transmission.

[0031] Procurement node weight middle, It is the construction impact factor, which indicates the impact of construction delays or inefficiencies on procurement rhythm and material availability. is the cross-stage dynamic impact coefficient, reflecting the accumulation and transmission trend of deviations within the historical cycle. The product of the two represents "the degree of construction's dependence on procurement × the coupling strength of the historical stages." Therefore, the weight of the procurement node is set to this value to reflect the core regulatory role of the downstream coordination function on the overall deviation.

[0032] Construction node weight In the design, all remaining weights are automatically attributed to the construction node. This is because the construction stage is the final executor, and its deviation is more reflected in the results and implementation. When the design and procurement risks are high, the interference weight of the construction-end data should be reduced accordingly to maintain the balance of the overall integration. This design can ensure that the total weight of the three nodes is 1, and the construction stage plays the role of "dynamic self-balancing".

[0033] The weight design logic is to dynamically allocate according to the risk transmission path and the actual error direction, so that each stage bears the data fusion weight corresponding to its "degree of influence" in deviation identification.

[0034] S3 specifically includes: S31, Markov transition chain modeling: Based on the stage identifier in the node state vector, construct the transition state set: ; The transition probability matrix of abnormal conduction in the construction phase is defined as follows: design Probability of purchase transfer: ; purchase Construction transfer probability: ; in, The probability that the abnormality of the design node will lead to the abnormality of the procurement node, is the probability that a construction node abnormality will be caused by a procurement node abnormality, Indicates the number of times a design node anomaly is followed by a procurement node anomaly in the history. 、 Indicates the total number of exceptions that occurred at design nodes or procurement nodes in history.

[0035] S32, dynamic warning value setting: according to the reservoir area terrain complexity index Adaptively adjust the warning threshold: ,in, is the warning trigger threshold of the current cycle, It is the terrain complexity index, which is calculated by the geographic information system based on slope, fault density, surface undulation, etc. The larger the value, the more complex the environment. The trigger judgment conditions are: ; The reverse compensation mechanism is activated when any condition is met.

[0036] S33, reverse compensation instruction generation: S331, if the transition probability from design to procurement exceeds the threshold, then: Issue parameter optimization instructions to design nodes; The procurement impact factor is updated simultaneously, and its weight coefficient is reduced as follows: ,in is the reduction coefficient, set to 0.8.

[0037] S332: If the probability of transition from procurement to construction exceeds the threshold, then: Triggering the reallocation of construction resources; And the construction impact factor is corrected according to the following incremental model: ,in, is the actual number of days of delay for the current construction node, is the adjusted construction impact factor.

[0038] S34, re-enter the feedback data after compensation adjustment into the multivariate regression model established in S1, and recalculate the cross-stage dynamic impact coefficient based on the latest design deviation, procurement difference and construction delay , thereby realizing deviation conduction identification Compensation feedback Closed-loop control mechanism for model recalculation.

[0039] By limiting state transitions to design → procurement and procurement → construction, a sparse Markov transition model is constructed to reduce invalid state traversal paths and improve the real-time and efficiency of anomaly identification. The terrain complexity index is introduced as a tuning parameter to dynamically lower the alarm threshold in reservoir construction environments with complex geological conditions, alleviating false alarms and missed alarms and enhancing the model's adaptability in extreme environments. A smoothing correction factor that increases with delay time is introduced into the construction deviation compensation mechanism to prevent parameter mutations from causing oscillations in the scheduling system, thereby maintaining the stability of the target control chain at each stage.

[0040] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0041] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for dynamic identification of target deviations at different stages of a reservoir EPC project, characterized in that: The following steps are involved: S1, stage target association modeling: Based on the BIM parameters of the design phase, the material list of the procurement phase, and the process plan of the construction phase, a three-phase correlation multivariate regression model is constructed, and a dynamic influence coefficient is configured for each EPC phase node. The dynamic influence coefficient is calculated and generated based on the cross-phase deviation transmission law in historical data; S2, multi-stage data fusion and state quantization: Real-time collection of operational data for each stage, including material arrival time differences at procurement nodes, mechanical efficiency values ​​at construction nodes, and parameter change frequencies at design nodes, is performed. The data is weighted according to the dynamic influence coefficients, generating a node state vector including a stage identifier. S3: Cross-stage deviation propagation identification and closed-loop control: A Markov transfer chain is established based on the stage identifier of the node state vector, and the deviation transfer probability between adjacent stages is calculated. When the transfer probability of design→procurement or procurement→construction exceeds the warning threshold, the reverse stage compensation instruction is triggered and the dynamic influence coefficient of S1 is updated to form a closed-loop control.

2. A method for dynamic identification of target deviations at different stages of a reservoir EPC project according to claim 1, characterized in that: Said S1 specifically includes: S11, extracting a set of physical parameters of components in the BIM model in the design phase, including concrete strength grade, steel bar specifications, and anti-seepage coefficient, and generating a design feature matrix with a unique component ID; S12, matching the material technical parameters in the purchase material list with the design feature matrix, establishing a component ID → material code association mapping table, and calculating the impact weight of material parameter deviation on design parameters based on historical supply data, recorded as the purchase impact factor ; S13: Analyze the process constraints in the construction process plan, identify the key construction paths that are strongly related to the design components, establish a multivariate regression model based on the historical construction logs, and calculate the construction delay days, design parameter deviations, and material arrival delays to output the construction influencing factors. ; S14, introduce the environmental sensitivity coefficient and calculate the dynamic impact coefficient across stages.

3. A method for dynamic identification of target deviations at different stages of a reservoir EPC project according to claim 2, characterized in that: The construction influencing factors It is obtained by constructing a multiple regression model, which is expressed as: , Indicates the actual number of days of delay on the critical path of the component; is the mean value of parameter deviation caused by procurement, defined as: , is the regression coefficient obtained from model training, For components The construction impact factor reflects the comprehensive impact of construction plan deviation on design objectives. For components factors influencing purchasing.

4. A method for dynamic identification of target deviations at different stages of a reservoir EPC project according to claim 3, characterized in that: The dynamic influence coefficient is expressed as: ； in, For components The dynamic impact coefficient in the current cycle, is the influence coefficient of the previous cycle, These are the procurement and construction impact factors calculated during this period.

5. The method for dynamic identification of target deviations at different stages of a reservoir EPC project according to claim 1 is characterized in that: The collection of operation data at each stage in S2 specifically includes: The design node calls the version management interface of the building information model to extract the parameter change operations in the record. Within the set time window, the number of change events of all design parameters is counted to measure the dynamic change frequency of the design link. The procurement node connects to the material transportation scheduling platform, analyzes the geographic trajectory data of the transport vehicles, and extracts the time difference between the actual arrival time and the scheduled time. At the same time, combined with the acceptance information based on the blockchain, it eliminates the records with failed signature verification or anomalies, and only retains the valid data; The construction node obtains the workload data per unit time through the engineering machinery controller, and combines the rated power information of the equipment to calculate the actual work efficiency under the current operation state as the mechanical efficiency value of the construction node.

6. A method for dynamic identification of target deviations at different stages of a reservoir EPC project according to claim 5, characterized in that: The stage weight distribution specifically includes: Based on the extracted procurement influencing factors and environmental sensitivity coefficients, corresponding fusion weights are assigned to design nodes to reflect the sensitivity of the design process to downstream deviations; According to the construction impact factor and the cross-stage dynamic image coefficient, the fusion weight is assigned to the procurement node to quantify the amplification or buffering effect of the intermediate link in the transmission of target deviation; The fusion weight of the construction node is determined by the sum of the weights of the design and procurement nodes, so that the overall weight maintains a normalized distribution among the three stages and ensures the quantitative closure of data fusion.

7. A method for dynamically identifying target deviations at different stages of a reservoir EPC project according to claim 6, characterized in that: The design node vector includes its stage identifier, the weighted parameter change frequency index and the corresponding component number; the procurement node vector includes the stage identifier, the weighted arrival time difference index and the material batch number; the construction node vector includes the stage identifier, the weighted work efficiency index and the associated process code. The S2 also includes vectorized encoding, which performs structured expression after weight correction on the data of the three nodes, and adds the stage identifier and auxiliary information to form a node state vector.

8. The method for dynamic identification of target deviations at different stages of a reservoir EPC project according to claim 1 is characterized in that: The establishment of the Markov transition chain in S3 includes defining a set of state transition directions according to the stage identifier in the node state vector, , construct a binary transition probability matrix, where: D→P transition probability ,The percentage of procurement node abnormalities after design node abnormalities in historical data is ,calculated; P→C transition probability ,Count the percentage of construction node abnormalities after procurement node abnormalities.

9. A method for dynamically identifying target deviations at different stages of a reservoir EPC project according to claim 8, characterized in that: The S3 also includes the setting of warning values: based on the reservoir environment complexity index Setting alert thresholds , where the more complex the terrain The larger the value, the or Activates the generation of reverse compensation instructions.

10. A method for dynamic identification of target deviations at different stages of a reservoir EPC project according to claim 9, characterized in that: The generation of the reverse compensation instruction includes: If the probability of D→P exceeds the standard, a parameter optimization instruction is sent to the design node, and the weight of the procurement influencing factor α is simultaneously reduced; If the probability of P→C exceeds the limit, the construction resource reallocation instruction is triggered and the construction resource is reallocated according to the actual number of days of delay. Update construction impact factors; The actual deviation data after compensation is input into the multivariate regression model of S1, and the dynamic impact coefficient is recalculated to form a cross-stage control closed loop.

Images