Epc engineering information management system supporting two-way matching of business strategy and technical path

Abstract

The present application relates to the technical field of EPC engineering information management, in particular to an EPC engineering information management system supporting two-way matching of business strategy and technical path. It comprises a data acquisition and integration unit for acquiring multi-source heterogeneous data from the whole process of project implementation; a resource optimization allocation model is constructed based on the pre-processed business side data, and an actual executable resource allocation scheme is generated; based on the resource allocation scheme and the model constraint conditions, combined with the technical side data and external environment data, a candidate set of technical paths conforming to the business constraint conditions is generated; based on the candidate set of technical paths and the business side data, a two-way coupled optimization model is constructed, and multi-objective optimization is solved to obtain the implementation scheme with the optimal score. By mapping the resource allocation scheme to each construction node, combining the BIM model and external environment data, a candidate set of technical paths conforming to the business constraints is generated, and the precise scheduling of materials, equipment and labor and the construction period prediction are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of EPC engineering information management, in particular to an EPC engineering information management system supporting two-way matching of business strategy and technical path. BACKGROUND

[0002] The EPC general contracting mode is the current international common engineering project organization and implementation method, and its core advantage lies in integrating design, procurement and construction links to realize overall optimization control of project cost, construction period and quality. However, in the actual project management process, the business management dimension and the technical management dimension are often independent of each other, forming a "data island" and "decision fragmentation", which leads to low project management efficiency and difficulty in achieving expected goals.

[0003] The engineering information management systems (such as traditional ERP, project management system or BIM management platform) on the market at present mostly focus on single-dimensional information recording and process approval, for example, or focus on financial management, or focus on construction progress tracking; and in the project execution process, in the face of changes (such as design changes, price fluctuations, weather abnormalities), it is unable to generate and filter out implementation schemes that perform well in cost, construction period, risk, efficiency and other multiple targets based on real-time data, resulting in poor project management flexibility and weak anti-interference ability; therefore, an EPC engineering information management system supporting two-way matching of business strategy and technical path is designed. SUMMARY

[0004] The purpose of the present application is to provide an EPC engineering information management system supporting two-way matching of business strategy and technical path to solve the problem of poor project management flexibility and weak anti-interference ability caused by the inability to generate and filter out implementation schemes that perform well in cost, construction period, risk, efficiency and other multiple targets based on real-time data in the face of changes (such as design changes, price fluctuations, weather abnormalities) in the project execution process as proposed in the background art.

[0005] To achieve the above purpose, the present application provides an EPC engineering information management system supporting two-way matching of business strategy and technical path, comprising:

[0006] A data acquisition and integration unit, which is used to acquire multi-source heterogeneous data from the whole process of project implementation and pre-process the multi-source heterogeneous data;

[0007] Among them, the multi-source heterogeneous data includes business side data, technical side data and external environment data;

[0008] A business strategy unit, which constructs a resource optimization allocation model based on the pre-processed business side data and generates an actually executable resource allocation scheme ;

[0009] The resource optimization allocation model includes a target function and a model constraint condition, and a contract cycle constraint is introduced and a risk coefficient are used to impose constraint filtering on the candidate scheme generated by the resource allocation model;

[0010] The technical path unit determines the required materials, equipment and labor for each stage through a BIM model based on the resource configuration scheme and the model constraint condition, in combination with technical side data and external environment data, to generate a candidate set of technical paths that meet the business constraint condition ;

[0011] The bidirectional matching unit obtains the implementation scheme with the optimal score by constructing a bidirectional coupling optimization model and performing multi-objective optimization solution using a non-dominated sorting genetic algorithm based on the candidate set of technical paths and the business side data .

[0012] As a further improvement of the technical solution, the data acquisition and integration unit includes a multi-source heterogeneous data acquisition module and a data preprocessing module;

[0013] The multi-source heterogeneous data acquisition module is used to acquire multi-source heterogeneous data of the whole process of engineering construction, and the acquired multi-source heterogeneous data is preprocessed by the data preprocessing module to smooth noise and unify dimensions of the collected data;

[0014] The business side data at least includes a project fund budget cost , a financing interest rate , a contract constraint construction period , a risk coefficient ;

[0015] The technical side data at least includes construction progress , material strength parameters , construction procedures , mechanical equipment utilization rate , energy consumption data ;

[0016] The external environment data includes climate temperature , precipitation .

[0017] As a further improvement of the technical solution, the business strategy unit includes a resource allocation module and a cost constraint module;

[0018] The resource allocation module is used to receive preprocessed business-side data and construct an objective function based on the business-side data. The resource optimization allocation model is used to allocate business-side resources and output resource allocation candidate vectors.

[0019] The cost constraint module introduces model constraints based on preprocessed business-side data to the resource optimization allocation model, which is used to apply constraint filtering to the candidate solutions generated by the resource allocation model.

[0020] The resource allocation module and the cost constraint module perform collaborative calculations through a solver to obtain an optimized resource allocation scheme under model constraints. .

[0021] As a further improvement to this technical solution, a resource optimization allocation model is constructed based on the aforementioned business-side data. The specific steps involved are as follows:

[0022] The preprocessed business-side data is normalized to generate an input vector. ;

[0023] The project construction process is divided into: The project is divided into several phases, and various resources in the project are categorized into different types. kind;

[0024] Obtain the planned allocation of each type of resource at each stage. and actual usage Calculate each type of resource Different construction stages utilization efficiency ;

[0025] Based on utilization efficiency The utilization efficiency of each stage is weighted and calculated to obtain the first stage. Reference values ​​for the comprehensive utilization efficiency of similar resources at all stages of the entire project construction process ;

[0026] Based on the Class resources in the The allocation of resources for each construction phase is defined by the resource allocation vector. , resource allocation vector As a decision variable to be optimized;

[0027] Based on resource allocation vector , construct the first Comprehensive utilization efficiency function of class resources ;

[0028] Based on the Comprehensive utilization efficiency function of class resources and the The cost of the funds corresponding to the resource , and introduce a risk coefficient , to maximize resource utilization efficiency to build the objective function.

[0029] As a further improvement of the technical solution, the model constraint includes a capital cost constraint, a contract period constraint and a risk constraint;

[0030] The capital cost constraint is used to limit the total capital expenditure to not more than the project capital budget cost ;

[0031] The contract period constraint is used to limit the total construction period of the entire project to not more than the upper limit of the contract period;

[0032] The risk constraint is used to limit the overall risk index introduced by the resource allocation scheme to not more than the preset risk threshold.

[0033] As a further improvement of the technical solution, the technical path unit includes a construction resource mapping module and a technical path generation optimization module;

[0034] The construction resource mapping module maps the resource allocation scheme to the construction nodes of each construction stage based on the BIM model, outputs a candidate technical path scheme set , and determines the materials, equipment and labor required for each stage, while calculating the predicted construction efficiency and construction period prediction value of each stage;

[0035] The technical path generation optimization module filters and sorts the candidate technical path scheme set based on the model constraint conditions of the business strategy unit, to generate a technical path candidate set that meets the business constraint conditions .

[0036] As a further improvement of the technical solution, the BIM model maps the resource allocation scheme to the construction nodes of each construction stage, outputs a candidate technical path scheme set, and the specific steps are:

[0037] Divide the project construction process into phases, get the node set of each phase , wherein each node corresponds to a number of components, and the material requirement of each component is ;

[0038] The allocation amount of the th resource in the resource allocation scheme in the stage The required amount of materials for nodes is The actual material usage at the node was calculated. And by analyzing the actual material usage of the node sub-units. Summing yields the total actual material usage for each node. ;

[0039] Material usage at all nodes within the phase Summing gives the total material usage for each stage. ;

[0040] Based on the Class resources in the stage Allocation amount Calculate the number of devices required for each stage. ;

[0041] Based on node construction procedures And introduce the construction phase. workload This yields the labor demand for each type of job. ;

[0042] Phase The manual requirements of all nodes within the process are aggregated to obtain the stage. Total internal human resource utilization demand ;

[0043] Based on construction stage Total internal material usage Quantity of equipment Labor demand Construct a nonlinear regression model to output the construction phase. The internal construction efficiency is expected, and this is achieved by incorporating climate temperature. and precipitation The optimized nonlinear regression model was used to calculate the optimized expected construction efficiency. ;

[0044] Based on the expected construction efficiency and stages Internal workload To obtain the project duration forecast ;

[0045] The material usage, equipment scheduling, labor requirements, expected construction efficiency, and projected construction period at each stage are mapped to generate a set of candidate technical path solutions. .

[0046] As a further improvement to this technical solution, the set of candidate technical path solutions is... Screening and sorting are performed to generate a candidate set of technical paths meeting the business constraint conditions, and the specific steps are as follows:

[0047] Based on the candidate technical path scheme set and the model constraint conditions, the candidate technical path scheme set is subjected to constraint judgment one by one, wherein the model constraint conditions at least include a capital cost constraint, a contract period constraint and a risk constraint;

[0048] The candidate scheme that does not meet any of the above constraint conditions is eliminated;

[0049] For the candidate technical path scheme meeting the constraint conditions, a candidate set of technical paths meeting the business constraint conditions is generated based on a Pareto optimal sorting algorithm .

[0050] As a further improvement of the technical solution, the bidirectional matching unit includes a conflict detection module, a bidirectional coupling optimization module and a multi-sorting decision module;

[0051] The conflict detection module is based on the candidate set of technical paths , and introduces capital cost constraints, contract period constraints and risk constraints to perform contract period conflict detection, capital conflict detection and risk conflict detection on the candidate schemes one by one, and outputs a feasible scheme set meeting all constraint conditions ;

[0052] The bidirectional coupling optimization module is based on the feasible scheme set to construct a multi-objective optimization model of bidirectional coupling, and based on a non-dominated sorting genetic algorithm, the feasible scheme set is solved to generate a Pareto frontier solution set ;

[0053] The business side optimization target includes total cost minimization and risk level minimization, and the technical side optimization target includes construction duration minimization and construction efficiency maximization;

[0054] The multi-sorting decision module is used for weighted sorting of the candidate schemes in the Pareto frontier solution set to determine the optimal implementation scheme .

[0055] Compared with the prior art, the present application has the following advantages:

[0056] 1、The support business strategy and technical path two-way matching EPC engineering information management system, based on the pretreated business data, constructs the resource optimization allocation model containing the objective function and multiple constraints (funds, cycle, risk), outputs the optimal resource allocation scheme, ensures that the resource allocation realizes the maximum utilization efficiency within the controllable range of budget, duration and risk;

[0057] At the same time, by mapping the resource allocation scheme to each construction node, combining the BIM model and external environment data, a candidate set of technical paths conforming to business constraints is generated, and precise scheduling of materials, equipment and labor and duration prediction are realized.

[0058] 2、The support business strategy and technical path two-way matching EPC engineering information management system, based on non-dominated sorting genetic algorithm (NSGA-II) to construct a two-way coupled optimization model, multi-objective collaborative optimization (cost, risk, duration, efficiency) is carried out on the candidate set of technical paths, and finally the Pareto optimal solution is output, and the final implementation scheme is determined through weighted sorting, realizing the balance of business target and technical target. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 The overall flowchart of the present application.

[0060] The meanings of the various labels in the figure are:

[0061] 1、Data acquisition and integration unit;

[0062] 2、Business strategy unit; 21、Resource allocation module; 22、Cost constraint module;

[0063] 3、Technical path unit; 31、Construction resource mapping module; 32、Technical path generation and optimization module;

[0064] 4、Two-way matching unit; 41、Conflict detection module; 42、Two-way coupled optimization module; 43、Multi-sorting decision module. DETAILED DESCRIPTION

[0065] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0066] Please refer to Figure 1As shown, an EPC engineering information management system supporting two-way matching of business strategy and technical path is provided, comprising a data acquisition and integration unit 1 for acquiring multi-source heterogeneous data from the whole process of project implementation and pre-processing the multi-source heterogeneous data;

[0067] The multi-source heterogeneous data includes business side data, technical side data and external environment data.

[0068] Further, the data acquisition and integration unit 1 includes a multi-source heterogeneous data acquisition module and a data preprocessing module.

[0069] The multi-source heterogeneous data acquisition module is used to acquire multi-source heterogeneous data in the whole process of engineering construction, and the acquired multi-source heterogeneous data is pre-processed by the data preprocessing module for noise smoothing and dimension unification of the collected data.

[0070] The business side data at least includes project funding budget cost , financing interest rate , contract constraint construction period , risk coefficient .

[0071] The technical side data at least includes construction progress , material strength parameter , construction process , mechanical equipment utilization rate , energy consumption data .

[0072] The external environment data includes climate temperature , precipitation .

[0073] In this embodiment, the project funding budget cost is collected from the financial management platform, including material procurement cost, labor cost, equipment rental fee;

[0074] The financing interest rate is collected from the bank or financing platform;

[0075] The construction progress is collected from the on-site construction monitoring system, including the process completion ratio and the planned progress deviation;

[0076] The material strength parameter is collected from the material laboratory and the supplier quality evaluation parameter;

[0077] The energy consumption data includes material price, labor cost, equipment rental fee, material consumption and energy consumption;

[0078] External environment data is collected from a meteorological monitoring system;

[0079] Wherein, the construction progress and energy consumption data , using adaptive Kalman filtering algorithm for noise smoothing;

[0080] Project funding budget cost , financing interest rate , contract constraints construction period , risk coefficient , using Z-score standardization algorithm unified dimension;

[0081] Material strength parameters , construction process sequence , mechanical equipment utilization rate , using Min-Max normalization algorithm for numerical mapping;

[0082] Climate temperature and precipitation , using sliding window average and normalization processing to eliminate short-term fluctuations.

[0083] The EPC engineering information management system supporting the bidirectional matching of business strategy and technical path further includes a business strategy unit 2, which constructs a resource optimization allocation model based on the preprocessed business side data, and generates an actual executable resource configuration scheme ;

[0084] Wherein, the resource optimization allocation model includes an objective function and a model constraint condition, and by introducing contract cycle constraints and risk coefficient , used to constrain and filter the candidate scheme generated by the resource allocation model;

[0085] The optimized resource configuration scheme and the model constraint condition result are output to the bidirectional matching unit and coupled with the technical path unit result;

[0086] In this embodiment, the business strategy unit 2 includes a resource allocation module 21 and a cost constraint module 22;

[0087] The resource allocation module 21 is used to receive the preprocessed business side data, and construct an objective function based on the business side data. The resource optimization allocation model is used to optimally allocate the business side resources, output a resource allocation candidate vector, and ensure the reasonable scheduling of resources between various processes or stages of the project;

[0088] The cost constraint module 22 introduces model constraint conditions for the resource optimization allocation model based on the preprocessed business side data, (project fund budget cost including material cost, labor cost and equipment use cost) for applying constraint filtering to the candidate scheme generated by the resource allocation model to ensure that the resource optimization allocation model does not break the project budget boundary when performing scheduling optimization, thereby realizing economic rationality and cost controllability of resource scheduling while ensuring construction progress and quality targets.

[0089] The resource allocation module 21 and the cost constraint module 22 perform collaborative calculation through a solver, which adopts linear programming, integer programming or heuristic search algorithm to obtain an optimized optimal resource allocation scheme under model constraint conditions

[0090]

[0091] In the formula, represents the optimal allocation amount of the th resource category in the th construction stage in the optimal resource allocation scheme; represents the resource category index, wherein the resource categories include materials, labor, mechanical equipment, etc.; represents the total number of resource categories; represents the construction stage index, wherein the construction stages include civil engineering, equipment installation, material transportation, assembly, etc.; represents the total number of construction stages; represents the optimized optimal resource allocation scheme obtained by the solver under the conditions of fund cost constraint, contract period constraint and risk constraint.

[0092] In this embodiment, the cost constraint module 22 performs feasibility determination on the candidate resource allocation scheme based on the model constraint conditions, that is:

[0093] The resource allocation vector obtained by solving is input to the cost constraint module 22, and the cost constraint module 22 checks the candidate scheme;

[0094] If the fund cost, contract period or risk constraint is not satisfied, an adjustment suggestion is generated and fed back to the resource allocation module 21 for iterative optimization, so that the optimization process forms a closed loop in the fund, period and risk three dimensions.

[0095] The resource optimization allocation model is constructed based on the business side data, and the specific steps involved are:

[0096] The preprocessed business side data is normalized to generate an input vector , wherein,​ denotes the normalized business side data feature, denotes the normalized business side input vector, and the input vector

[0097] denotes the i-th construction stage, including civil engineering, equipment installation, material transportation, assembly and other construction stages;

[0098]

[0099]

[0100]

[0101]

[0102]

[0103] ​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​

[0104] wherein, represents the comprehensive utilization efficiency of the class resource in all stages of the entire project construction process; represents the utilization efficiency of the class resource in the th construction stage; represents the weight coefficient of the construction stage , used to represent the relative importance of the stage in the entire construction process, which is set according to the sensitivity of the construction stage to cost, progress or risk, for example, the civil engineering stage (capital intensive) has a larger weight, and the material handling stage has a smaller weight; ; represents the total number of construction stages, including civil engineering, equipment installation, material handling, assembly, etc.;

[0105] define the resource allocation vector as the decision variable to be optimized, wherein, represents the allocation amount of the class resource in the th construction stage (decision variable before optimization); represents the resource allocation vector set, representing the overall resource allocation scheme to be optimized;

[0106] based on the resource allocation vector , the comprehensive utilization efficiency function of the class resource is constructed ;

[0107]

[0108] wherein, represents the comprehensive utilization efficiency function of the class resource in the entire construction period, used to measure the overall utilization level of the class resource in all stages. Specifically, when the comprehensive utilization efficiency function of the class resource is high, it means that the class resource is reasonably allocated and fully utilized in the entire project cycle, and when the comprehensive utilization efficiency function of the class resource is low, it indicates that there is resource waste or insufficient allocation, which needs to be adjusted ;

[0109] based on the comprehensive utilization efficiency function of the class resource and the corresponding capital cost of the class resource , and introducing the risk coefficient , the objective function is constructed to maximize the resource utilization efficiency;

[0110]

[0111] In the formula, represents the target function value, and represents the comprehensive optimization benefit under the current resource allocation vector ; represents that the target function is maximized in all feasible resource allocation schemes to obtain an optimal scheme ; The optimal scheme contains the optimal allocation results of all resource categories and all construction stages. represents the capital cost of the resource, including material cost, labor cost, equipment use cost, etc. represents the risk penalty coefficient, which is used to adjust the influence weight of risk in the target function. represents the overall risk coefficient, which is used to quantify the construction risk introduced by the current resource allocation scheme , which is obtained by weighted aggregation of multiple key risk factors (at least including schedule risk, cost risk, safety risk, quality risk and supply chain risk).

[0112] In the target function of the resource optimization allocation model in this embodiment, the risk coefficient is introduced as a risk penalty term, which is used to quantitatively evaluate the overall risk level of the resource allocation scheme, and as a risk penalty term to reduce the comprehensive optimization value of the high-risk scheme, so as to balance the resource utilization efficiency, financial economy and controllability of construction risk in the model optimization process.

[0113] Further, the model constraint conditions include capital cost constraint, contract period constraint and risk constraint.

[0114] Among them, the capital cost constraint is used to limit that the total capital expenditure cannot exceed the project capital budget cost .

[0115] In this embodiment, the capital cost constraint is specifically:

[0116]

[0117] In the formula, represents the project capital budget cost, including material cost, labor cost, equipment rental cost, etc. represents the resource category index, wherein the resource category includes material, labor, mechanical equipment, etc. represents the total number of resource categories. represents the construction stage index, wherein the construction stage includes civil engineering, equipment installation, material transportation, assembly, etc. represents the total number of construction stages. represents the total amount of fund expenditure on all resource categories and all construction stages, and considers the construction stages corresponding financing interest rate after the actual fund demand, represents the resource allocation scheme the allocation amount of the category resource in the stage ;

[0118] The contract cycle constraint is used to limit the total construction cycle of the entire project to not exceed the upper limit of the contract cycle;

[0119] In this embodiment, the contract cycle constraint is specifically:

[0120]

[0121] In the formula, represents the duration (construction length) of the construction stage under the resource allocation scheme ; represents the upper limit of the contract cycle, that is, the contract constraint duration;

[0122] The risk constraint is used to limit the overall risk index introduced by the resource allocation scheme to not exceed the preset risk threshold;

[0123] In this embodiment, the risk constraint is specifically:

[0124]

[0125] In the formula, represents the overall risk index value under the resource allocation scheme ; represents the preset risk tolerance threshold, that is, the maximum tolerable risk level specified by the business side or the contract.

[0126] The EPC engineering information management system supporting the two-way matching of business strategies and technical paths further includes a technical path unit 3, which determines the materials, equipment and labor required for each stage based on the resource allocation scheme and the model constraint condition, in combination with technical side data and external environment data, through a BIM model, to generate a candidate set of technical paths that meet the business constraint conditions ;

[0127] In this embodiment, the technical path unit 3 includes a construction resource mapping module 31 and a technical path generation optimization module 32;

[0128] The construction resource mapping module 31 maps the resource allocation scheme The BIM model maps the resource allocation scheme to each construction node of each construction stage, and outputs a candidate technical path scheme set , to determine the required materials, equipment and labor for each stage, and to calculate the predicted construction efficiency and duration prediction value of each stage;

[0129] In this embodiment, the BIM model contains the construction nodes of the project, node component information (length, area, volume, coordinate position), material type and quantity, construction procedure sequence, equipment demand and labor input information (work type and work quantity), and the like;

[0130] The BIM model is used to map the resource allocation scheme to each construction node, and to calculate the stage material use plan, equipment scheduling and labor input quantity in combination with the node attributes, so as to generate the stage construction efficiency and duration prediction;

[0131] The technical path generation optimization module 32 screens and sorts the candidate technical path scheme set based on the candidate technical path scheme set and the model constraint conditions of the business strategy unit 2, to generate a technical path candidate set that meets the business constraint conditions . .

[0132] The BIM model maps the resource allocation scheme to each construction node of each construction stage, and outputs a candidate technical path scheme set, and the specific steps involved are as follows:

[0133] The project construction process is divided into stages, to obtain a node set for each stage , wherein represents the total number of nodes included in the th construction stage, represents the th node in the th construction stage , wherein each node corresponds to a plurality of components, and the material demand of each component is ;

[0134] The allocation amount of the th type of resource in the resource allocation scheme to the stage is , and the node material demand is , to calculate the actual material use amount of the node , and to obtain the actual material use amount of the node summary by summing up the actual material use amount of the node subunit of the node ;

[0135] ​

[0136] In the formula, Represents a node The actual amount of materials used; Represents a node The material requirements;

[0137] The material usage of all nodes within the stage for:

[0138]

[0139] Material usage at all nodes within the phase Summing gives the total material usage for each stage. ;

[0140]

[0141] In the formula, Indicates traversing nodes All components within are used to calculate the total material quantity of the nodes; Represents a node The total amount of materials used within the node The amount of material used for all internal components is summed up. Indicates the first Total material usage for each construction phase; Indicates from the first Node set of each construction stage In the middle, each node is extracted in turn. ; This indicates the node index, representing the construction phase. A construction node within;

[0142] Based on the Class resources in the stage Allocation amount Calculate the number of devices required for each stage. This is used to implement device scheduling mapping;

[0143] Based on node construction procedures And introduce the construction phase. workload This yields the labor demand for each type of job. , used to implement artificial resource mapping;

[0144] Phase The manual requirements of all nodes within the process are aggregated to obtain the stage. Total internal human resource utilization demand ;

[0145] In the present embodiment, the number of devices required per stage wherein, represents rounding up, ensuring the number of devices is an integer; represents the efficiency coefficient of the type of device (i.e. the amount of tasks that can be completed per hour per device); represents the number of type of devices required for the construction stage (rounded to the nearest integer);

[0146] wherein, the amount of labor required per type of work is:

[0147]

[0148] wherein, represents the amount of labor required within the node ; represents the set of construction processes corresponding to the node ; represents the proportion of tasks within the construction process ; represents the unit task labor demand coefficient of the process ; represents the node construction process index; represents the amount of tasks in the construction stage ;

[0149] the total demand for human resources usage within the stage is:

[0150]

[0151] wherein, represents the total labor demand within the construction stage;

[0152] Based on the total material usage , the number of devices , and the labor demand within the construction stage, a non-linear regression model is constructed to output the predicted construction efficiency within the construction stage , and by introducing the climate temperature and precipitation , the non-linear regression model is optimized to calculate the optimized predicted construction efficiency ;

[0153] Based on the predicted construction efficiency and the stage ​​​Task volume inside , get the duration prediction value ;

[0154] In this embodiment, the specific steps involved in building a nonlinear regression model based on a machine learning algorithm are as follows:

[0155] The total material usage , the number of equipment , and the labor demand are taken as inputs;

[0156] At the same time, the climate temperature and precipitation are taken as additional influencing quantities;

[0157] The construction efficiency influencing factor vector is constructed, and all resource inputs and environmental conditions are mapped into a unified feature space (the above features are mapped into a unified dimension feature space, which can be processed by normalization or standardization);

[0158] Based on historical sample data, a machine learning algorithm (i.e., a nonlinear regression model) is used to fit the construction efficiency influencing factor vector to obtain the construction efficiency :

[0159]

[0160] In the formula, represents a nonlinear regression model based on a machine learning algorithm, which is used to map the construction efficiency influencing factor to the predicted construction efficiency, wherein the machine learning algorithm includes but is not limited to support vector regression (SVR), random forest regression (RF), gradient boosting decision tree (GBDT), multilayer perception (MLP), or long short-term memory network (LSTM); is a model parameter (such as weight, bias, tree node splitting parameter, etc.), which is obtained by minimizing the loss function ;

[0161] In the formula, is the loss function value, which is used to measure the error between the predicted construction efficiency and the historical actual construction efficiency, and is used to train the nonlinear regression model;

[0162]

[0163] In the formula, represents the loss function value, which is used to measure the error between the predicted construction efficiency and the historical actual construction efficiency, and is used to train the nonlinear regression model; represents the predicted construction efficiency of the th construction stage based on the nonlinear regression model; represents the Historical construction efficiency for each construction phase (historical sample data provided, calculated based on the actual amount of work completed and the actual time spent in each construction phase);

[0164] Specifically, historical samples and historical construction efficiency Used for training nonlinear regression models By minimizing the loss function To update model parameters ;

[0165] For tree-based models (such as RF and GBDT), model parameters are trained using split gain or residual fitting methods.

[0166] Specifically, in this embodiment, the machine learning algorithm preferably employs Support Vector Regression (SVR):

[0167] Support Vector Regression (SVR) Model Parameters This mainly includes the weights and bias terms of the support vectors. In addition, the present invention preferably uses a Gaussian kernel as the kernel function, and the regularization parameter is set between 5 and 15, preferably 10; the insensitive loss parameter is set between 0.01 and 0.2, preferably 0.1.

[0168] Using the preprocessed training dataset (construction efficiency influencing factor vector) ) and corresponding tags (historical construction efficiency) ), and historical data (historical construction efficiency influencing factor vector) Corresponding historical construction efficiency The support vector regression (SVR) model was trained by randomly dividing the dataset into training and test sets in a 7:3 ratio and then using a combination of grid search and cross-validation to find the optimal parameter combination.

[0169] In this embodiment, the material usage, equipment scheduling and labor requirements, expected construction efficiency and projected construction period at each stage are mapped and a set of candidate technical path solutions is generated. ;

[0170]

[0171] In the formula, Represents a set of candidate technology path solutions; Indicates the first Each of the candidate technology paths is 100. It represents a complete sequence of construction phase plans, consisting of resource usage and construction efficiency information for each phase, including material usage, equipment demand, labor demand, construction efficiency, and schedule prediction, which is used for subsequent construction plan optimization and scheduling decisions. Total number of project construction stages, Index of project construction stage; Total number of candidate technical path schemes, Index of candidate technical path scheme.

[0172] In this embodiment, the candidate technical path scheme set is filtered and sorted to generate a technical path candidate set that meets the business constraint conditions, involving the following specific steps:

[0173] Based on the candidate technical path scheme set and the model constraint conditions, the candidate technical path scheme set is subjected to constraint determination one by one, wherein the model constraint conditions at least include the fund cost constraint, the contract cycle constraint and the risk constraint;

[0174] Candidate schemes that do not meet any of the above constraint conditions are eliminated;

[0175] For candidate technical path schemes that meet the constraint conditions, a technical path candidate set that meets the business constraint conditions is generated based on the Pareto optimal sorting algorithm .

[0176] In this embodiment, a technical path candidate set that meets the business constraint conditions is generated based on the Pareto optimal sorting algorithm , involving the following specific steps:

[0177] For each scheme in the candidate technical path scheme set , its multi-objective performance vector is defined:

[0178]

[0179] Wherein:

[0180] , represents the total duration target (unit: day or hour) of the candidate technical path scheme ;

[0181] , represents the total cost of the candidate technical path scheme ;

[0182] , represents the overall risk indicator (dimensionless and normalized to ) of the candidate technical path scheme ; in the formula, represents the total number of risk factors (for example , represents the schedule risk, the cost risk, the safety risk, the quality risk, the supply chain risk, etc.); represents the index of risk factor; represents the weight of the th risk factor (determined based on historical project data and combined with expert experience method), normalized to make ; represents the risk metric of the candidate technical path scheme on the th risk factor, normalized to map to ;

[0183] , represents the construction efficiency of each stage of the candidate technical path scheme divided by the comprehensive cost of the stage, representing the construction efficiency per unit cost input (dimensionless);

[0184] wherein, represents the multi-objective performance vector of the candidate technical path scheme , used for subsequent Pareto optimal sorting as a multi-dimensional measurement basis for distinguishing the pros and cons of different candidate technical path schemes; represents the index of project construction stage; represents the total number of project construction stages; represents the capital cost corresponding to the th type of resource; represents the capital cost rate or financing interest rate (dimensionless) corresponding to the construction stage ; represents the comprehensive cost of the construction stage (reflecting the total input cost of the construction stage , unit: yuan), which is the sum of material cost, labor cost, equipment rental cost, material consumption value and energy consumption value; represents the expected construction efficiency of the construction stage under the candidate technical path scheme ; represents the risk value calculated for the candidate technical path scheme .

[0185] For any two schemes , if the following conditions are met:

[0186]

[0187] and there is at least one target such that ;

[0188] and , the scheme Dominance scheme , denoted as ;

[0189] In the set , if there exists a scheme that is not dominated by any other scheme, i.e.:

[0190]

[0191] then this scheme is determined as a Pareto optimal solution and is included in the Pareto front set:

[0192]

[0193] where denotes the objective function index, denotes the objective to be minimized, denotes the condition that holds for all objective functions (here, duration, cost, and risk), denotes the unit cost construction efficiency of scheme (the larger the better, with the objective to maximize), denotes the unit cost construction efficiency of scheme ; denotes any other candidate scheme relative to scheme ; denotes the dominance relation symbol, if , then it means that scheme is not inferior to in all objectives to be minimized, and is superior to in at least one objective, and is not lower than in the efficiency objective; denotes the Pareto front set, which contains all candidate schemes that are not dominated by other schemes, and is the final optimal solution set available for decision makers to choose from;

[0194] The non-dominated sorting method is adopted to divide all schemes into several levels:

[0195] The first level ;

[0196] After deleting the schemes in , the dominance relation is recalculated in the remaining set to obtain the second level ;

[0197] This process is repeated until all schemes are stratified;

[0198] Within the same non-dominated level, to avoid the concentration and degeneration of solutions, the crowding distance of each scheme is calculated:

[0199]

[0200] wherein, denotes the crowding distance of the scheme , which is used to measure the relative distribution sparsity of the solution in the same non-dominated layer. The greater the crowding distance, the fewer the solutions around the scheme, and the scheme is preferentially selected; denotes the index of the objective function, ; denotes the value of the scheme on the first objective function; and , denote the maximum value and the minimum value of the objective function , respectively (for normalization, to ensure that the objective functions have comparability under different dimensions); , are adjacent schemes before and after sorting on the objective function , respectively, which are the previous and the next schemes, respectively, used to calculate the field distance and measure the distribution density of the solution.

[0201] In the same layer, the solution with a greater crowding distance is preferentially selected to ensure the balanced distribution of the Pareto solution set;

[0202] Finally, the Pareto optimal solution set that meets the business constraint condition is obtained:

[0203]

[0204] wherein, denotes the optimal solution set of the candidate technical path after non-dominated sorting and crowding selection, which meets the business constraint condition; denotes the first Pareto optimal scheme, denotes the number of Pareto optimal solutions, and satisfies , wherein is the total number of candidate technical paths;

[0205] and is output as the candidate set of technical paths that meets the business constraint condition.

[0206] The EPC engineering information management system supporting the bidirectional matching of business strategies and technical paths further comprises a bidirectional matching unit 4, which, based on the candidate set of technical paths and the business side data, constructs a bidirectional coupling optimization model, and uses the non-dominated sorting genetic algorithm NSGA-II for multi-objective optimization solution, and finally obtains the implementation scheme with the optimal score .

[0207] The bidirectional matching unit 4 includes a conflict detection module 41, a bidirectional coupling optimization module 42, and a multi-ranking decision module 43.

[0208] Among them, the conflict detection module 41 is based on the technology path candidate set. Furthermore, it introduces constraints on capital costs, contract periods, and risks, and performs contract period conflict detection, capital conflict detection, and risk conflict detection on each candidate solution, outputting a set of feasible solutions that meet all constraints. ;

[0209] In this embodiment, the conflict detection module 41 further includes a periodic conflict detection submodule, a funding conflict detection submodule, and a risk conflict detection submodule;

[0210] Among them, the periodic conflict detection submodule is used to analyze the candidate set of technical paths. Construction period of each implementation plan Contractual constraints on project duration To make a comparison, if Then it is determined to be a periodic conflict and is eliminated, where, Representation scheme During the construction phase Construction period (unit: days or hours). Indicates the contractually binding construction period (unit: days or hours);

[0211] The funding conflict detection submodule is based on the budget funding requirements of each candidate solution. Project funding budget cost provided by the business side To make a comparison, when Mark as a funding conflict and remove. Representation scheme During the construction phase Budgetary funding requirements (unit: yuan). Project funding budget cost (unit: yuan);

[0212] The risk conflict detection submodule is used to calculate the risk level of each candidate solution. and the risk tolerance threshold set by the business side. To make a comparison, when The event is marked as a risk conflict and removed, among which and All were normalized and mapped to ;

[0213] By detecting conflicts, a set of feasible solutions that meet both constraints is obtained. :

[0214]

[0215] wherein, denotes the set of feasible solutions, denotes all candidate technical path solutions after screening in the conflict detection module 41, satisfying the capital cost constraint, contract period constraint and risk constraint, and the set of feasible solutions Each solution in the set of feasible solutions includes at least resource allocation parameters (material usage, equipment scheduling scheme and labor demand), time and efficiency parameters (stage duration prediction, stage construction efficiency and total duration), and economic and risk parameters (total cost, financing cost and comprehensive risk index); denotes a candidate technical path solution, which is an element in the original candidate set of technical path solutions ; denotes the candidate set of technical path solutions after non-dominated sorting and crowding distance screening;

[0216] The bidirectional coupling optimization module 42 constructs a bidirectional coupling multi-objective optimization model based on the set of feasible solutions , and solves the set of feasible solutions based on the non-dominated sorting genetic algorithm (NSGA-II) to generate a Pareto frontier solution set ;

[0217] In this embodiment, before starting the bidirectional coupling optimization module 42 for multi-objective optimization solution, a solution space preprocessing mechanism is introduced to preliminarily screen the set of feasible solutions by heuristic rules based on prior knowledge to eliminate obviously inferior solutions, thereby reducing the initial search space of the non-dominated sorting genetic algorithm (NSGA-II) and reducing the computational complexity, wherein the heuristic rules based on prior knowledge include but are not limited to:

[0218] If the construction efficiency of a certain solution is lower than the lower quartile of the historical efficiency of similar projects, it is eliminated;

[0219] If the total cost of a certain solution exceeds a certain safety threshold (such as 90%) of the budget, it is considered a high-risk solution and is eliminated;

[0220] If the total duration of a certain solution exceeds a certain buffer percentage (such as 110%) of the contract duration, it is considered an unfeasible solution;

[0221] If the construction efficiency prediction value of a certain solution under extreme environmental conditions (such as heavy rain, high temperature) is lower than the historical minimum value, it is eliminated.

[0222] Further, the bidirectional coupling optimization module 42 adopts a distributed parallel computing architecture (the distributed parallel computing architecture is a parallel computing environment constructed based on a cloud computing platform, realized based on a Master-Worker model, and deployed on a containerized cloud computing platform (such as Kubernetes) for accelerating the execution process of the non-dominated sorting genetic algorithm (NSGA-II); the initial population generation task is decomposed into multiple subtasks, and the initial solutions are generated in parallel by different computing nodes; the crossover and mutation operation of each individual or each pair of parents is independently performed, and is allocated to different computing nodes for parallel execution; a parallel non-dominated sorting algorithm (such as a parallel implementation based on fast non-dominated sorting, the master node randomly and uniformly divides the merged population (parents and offspring) into N sub-populations, and distributes them to N Worker nodes, and each Worker node performs fast non-dominated sorting on its local sub-population in parallel to obtain a local non-dominated level; the master node collects all local frontiers, and through a global merge sorting process, compares and merges the frontiers from different nodes, and finally obtains the global non-dominated sorting result) is used to allocate individuals to different nodes for dominance relationship judgment and stratification; the crowding degree is calculated for each objective function in parallel on different nodes, and the results are aggregated (after determining the global non-dominated level, the individuals in the same frontier layer are allocated to different Worker nodes in parallel to perform the crowding degree calculation task on each objective function, and after each node calculates the crowding degree on the specified objective function, the result is returned to the master node, which sums and aggregates the results to obtain the total crowding degree of each individual)), which is used to parallelize the population initialization, crossover and mutation, non-dominated sorting and crowding degree calculation tasks in the non-dominated sorting genetic algorithm (NSGA-II), and deploy them on a cloud computing platform for collaborative computing, so as to compress the optimization process that originally takes several hours or even several days into an acceptable time limit for engineering decision-making (such as minutes or hours), and meet the needs of real-time decision-making on site.

[0223] On the basis of the above-mentioned distributed parallel computing architecture and solution space preprocessing mechanism, the feasible scheme set is solved by the non-dominated sorting genetic algorithm (NSGA-II) to perform multi-objective optimization:

[0224]

[0225] wherein, represents the Pareto frontier solution set, which is calculated by the bidirectional coupling optimization module 42 based on the feasible scheme set If the scheme is not dominated by any scheme , the scheme is a Pareto optimal solution and is included in ;

[0226] wherein the business-side optimization objectives include total cost minimization and risk level minimization, and the technology-side optimization objectives include construction duration minimization and construction efficiency maximization;

[0227] In this embodiment, a bi-directionally coupled multi-objective optimization model is constructed to minimize (construction duration, total cost, risk level) and maximize (construction efficiency) simultaneously:

[0228] Total cost minimization:

[0229]

[0230] Risk level minimization:

[0231]

[0232] Construction duration minimization:

[0233]

[0234] Construction efficiency maximization:

[0235]

[0236] wherein, is the total cost objective function, representing the minimization of total fund expenditure over all stages under the feasible solution set ; is the overall risk objective function, representing the minimization of the sum of risk levels over all stages under the feasible solution set; is the risk level (dimensionless and normalized to ) of construction stage , which is calculated by weighting various risk factors (construction schedule, construction cost, construction safety, construction quality, construction supply chain, etc.); is the total duration objective function, representing the minimization of the sum of construction periods of all construction stages in the feasible solution (unit: day or hour); is the construction efficiency objective function, representing the maximization of the sum of construction efficiencies over all stages under the feasible solution; is the construction efficiency of construction stage (typically obtained through machine learning prediction models);

[0237] The bi-directionally coupled multi-objective optimization model is then:

[0238]

[0239] wherein, represents the feasible solution set One of the candidate schemes in the Pareto front solution set is selected as the final solution; the bi-coupled multi-objective optimization model is used to consider the business side (total cost minimization, risk minimization) and the technical side (construction period minimization, construction efficiency maximization) at the same time.

[0240] The multi-ranking decision module 43 is used to weight and rank the candidate schemes in the Pareto front solution set to determine the optimal implementation scheme .

[0241] In this embodiment, the multi-ranking decision module 43 introduces a business weight vector and a technical weight vector based on the weighted linear aggregation method in the Pareto front solution set to weight and rank the solution set:

[0242]

[0243] wherein , and and are valued based on the expert experience method;

[0244] Finally, the multi-ranking decision module 43 outputs a set of optimal ranked implementation schemes:

[0245]

[0246] wherein represents the weighted total score of the candidate scheme , which is a value obtained by weighting and aggregating the multi-objective performance indicators based on the business weight and the technical weight, and is used for ranking and decision-making; represents the business weight vector, which represents the relative importance of the business objectives (total cost, risk level) in the total score, and the vector dimension is consistent with the number of business objectives; represents the technical weight vector, which represents the relative importance of the technical objectives (construction period, construction efficiency) in the total score, and the vector dimension is consistent with the number of technical objectives; represents the total construction period target value of the candidate scheme , which represents the construction time required to complete the scheme; represents the total cost target value of the candidate scheme , which usually includes comprehensive costs such as materials, labor, equipment, and energy consumption; represents the risk level target value of the candidate scheme , which comprehensively considers progress, cost, safety, quality, supply chain, and other risk factors and performs weighted normalization processing; represents the candidate scheme a construction efficiency target value, usually representing the construction efficiency per unit cost or per unit time; denotes the vector transpose symbol; denotes the optimal solution output by the multi-order decision module 43, i.e. the candidate solution with the minimum weighted total score in the Pareto frontier solution set; denotes the selection of the solution with the minimum weighted total score from the Pareto frontier solution set as the final implementation solution;

[0247] Specifically, the final implementation solution specifically includes resource allocation parameters (material usage, equipment scheduling solution and labor demand), time and efficiency parameters (stage duration prediction, stage construction efficiency and total duration), and economic and risk parameters (total cost, financing cost and comprehensive risk index).

[0248] The above shows and describes the basic principles, main features and advantages of the present application. It should be understood by those skilled in the art that the present application is not limited by the above examples, and the above examples and descriptions in the specification are only preferred examples of the present application and are not intended to limit the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. An EPC engineering information management system supporting two-way matching of business strategy and technical path, characterized in that, The method comprises the following steps: a data acquisition and integration unit (1) is used to acquire multi-source heterogeneous data from the whole process of project implementation, and to pre-process the multi-source heterogeneous data; wherein the multi-source heterogeneous data comprises business side data, technical side data and external environment data; The business strategy unit (2) constructs a resource optimization allocation model based on the preprocessed business side data, and generates an actually executable resource configuration scheme ; The resource optimization allocation model includes a target function and model constraint conditions, and a contract cycle constraint is introduced and a risk coefficient for imposing constraint filtering on candidate schemes generated by the resource allocation model. A technical path unit (3) determines required materials, equipment and labor for each stage through a BIM model based on a resource configuration scheme and model constraints, in combination with technical side data and external environment data, to generate a candidate set of technical paths that meet business constraints ; the technical path unit (3) comprises a construction resource mapping module (31) and a technical path generation and optimization module (32); The construction resource mapping module (31) maps the resource configuration scheme to construction nodes of each construction stage based on the BIM model to output a candidate technical path scheme set , which is used to determine materials, equipment and labor required for each stage, and calculate the predicted construction efficiency and the work period prediction value of each stage. The specific steps involved are: Divide the project construction process into phases, obtain the node set of each phase , wherein each node corresponds to a number of components, and the material requirement of each component is ; Then the resource allocation scheme The first in Class resources in the stage Allocation amount The required amount of materials for nodes is The actual material usage at the node was calculated. And by analyzing the actual material usage of the node sub-units. Summing yields the total actual material usage for each node. ; Total material usage for all nodes in phase Sum to get total material usage for phase ; Based on the first class resource allocation amount in each stage , the required equipment quantity in each stage is calculated; Based on the node construction process , and introduce the construction phase of the task volume , get the amount of labor demand of each type of work ; Phase The manual requirements of all nodes within the process are aggregated to obtain the stage. Total internal human resource utilization demand ; Based on construction phase Internal total material usage , equipment quantity , labor demand Construct a nonlinear regression model for outputting the predicted construction efficiency in the construction phase by introducing climate temperature and precipitation Optimize the nonlinear regression model to calculate the optimized predicted construction efficiency ; Based on predicted construction efficiency and phase volume of tasks within , resulting in a duration prediction value ; The material usage of each stage, equipment scheduling and labor demand, predicted construction efficiency and construction period prediction value are mapped and a candidate technical path scheme set is generated ; A bidirectional matching unit (4) based on a set of technical path candidates and business side data, by constructing a bidirectional coupling optimization model, and using a non-dominated sorting genetic algorithm for multi-objective optimization solution, ultimately obtaining the optimal implementation scheme ; before starting the bidirectional coupling optimization module (42) for multi-objective optimization solution, a solution space preprocessing mechanism is introduced to filter the feasible scheme set by using heuristic rules based on prior knowledge; the business side optimization objectives include total cost minimization and risk level minimization, and the technical side optimization objectives include construction duration minimization and construction efficiency maximization.

2. The EPC engineering information management system supporting bidirectional matching of business strategy and technical path according to claim 1, characterized in that: the data acquisition and integration unit (1) comprises a multi-source heterogeneous data acquisition module and a data preprocessing module; wherein the multi-source heterogeneous data acquisition module is used to acquire multi-source heterogeneous data in the whole process of engineering construction, and the acquired multi-source heterogeneous data is pre-processed by the data preprocessing module, for noise smoothing and dimension unification of the collected data; The business side data at least includes project fund budget cost , financing interest rate , contract constraint construction period , risk coefficient ; The technical side data at least includes construction progress , material strength parameter , construction process , mechanical equipment utilization rate , energy consumption data ; The external environment data includes climate temperature , precipitation amount .

3. The EPC project information management system supporting two-way matching of business strategy and technical path according to claim 1, characterized in that: the business strategy unit (2) comprises a resource allocation module (21) and a cost constraint module (22); the resource allocation module (21) is used to receive the pre-processed business side data, and to construct a target function based on the business side data, the resource optimization allocation model is used to allocate the business side resources, and outputs a resource allocation candidate vector; the cost constraint module (22) introduces a model constraint condition for the resource optimization allocation model based on the pre-processed business side data, for applying constraint filtering to the candidate scheme generated by the resource allocation model; The resource allocation module (21) and the cost constraint module (22) are cooperatively calculated by a solver, and an optimized optimal resource allocation scheme is obtained under the model constraint condition .

4. The EPC project information management system of claim 3, wherein: the specific steps for constructing the resource optimization allocation model based on the business side data are as follows: The pre-processed business side data is normalized to generate an input vector ; The project construction process is divided into phases, and various resources in the project are divided into categories; Obtain the planned allocation of each type of resource at each stage. and actual usage Calculate each type of resource Different construction stages utilization efficiency ; Based on utilization efficiency The utilization efficiency of each stage is weighted to calculate the comprehensive utilization efficiency of the whole project The comprehensive utilization efficiency reference value of the resource type in all stages of the whole project construction process ; based on the first class resource allocation amount of the first construction phase defines a resource allocation vector the resource allocation vector as a decision variable to be optimized; Resource allocation vector based on resource allocation vector , constructing a first class resource utilization efficiency function ; Based on the first Comprehensive utilization efficiency function of resources of the same type And the Corresponding capital cost of resources of the same type And introduce risk coefficient To maximize resource utilization efficiency to build objective function.

5. The EPC project information management system of claim 4, wherein: the model constraint condition comprises a fund cost constraint, a contract period constraint and a risk constraint; wherein the capital cost constraint is used to limit the total capital expenditure to not exceed the project capital budget cost ; the contract period constraint is used to limit the total construction period of the whole project to be less than the upper limit of the contract period; the risk constraint is used to limit the overall risk index introduced by the resource allocation scheme to be less than a preset risk threshold.

6. The EPC project information management system supporting two-way matching of business strategy and technical path according to claim 1, characterized in that: For the set of candidate technology paths The specific steps involved in filtering and sorting to generate a candidate set of technology paths that meet business constraints are as follows: Based on the candidate technology path scheme set and model constraints, the candidate technology path scheme set The constraint judgment is performed one by one, wherein the model constraints at least include a capital cost constraint, a contract cycle constraint and a risk constraint; candidate schemes that do not meet any of the above constraint conditions are eliminated; For the candidate technology path scheme meeting the constraint condition, based on a Pareto optimal sorting algorithm, a candidate set of technology paths meeting the business constraint condition is generated .

7. The EPC project information management system of claim 1, wherein: the system is configured to support a two-way matching of business strategy and technical path. The technical path generation optimization module (32) generates a candidate set of technical path schemes based on the candidate set of technical path schemes and the model constraints of the business strategy unit (2) The candidate set of technical path schemes is screened and sorted to generate a candidate set of technical paths that meet the business constraints .

8. The EPC project information management system of claim 1, wherein: the system is configured to support a two-way matching of business strategy and technical path. the bidirectional matching unit (4) comprises a conflict detection module (41), a bidirectional coupling optimization module (42) and a multi-order decision module (43); The conflict detection module (41) detects the conflicts based on a candidate set of technical paths , and introduces fund cost constraints, contract cycle constraints and risk constraints, and performs contract cycle conflict detection, fund conflict detection and risk conflict detection on the candidate schemes one by one, and outputs a feasible scheme set meeting all constraint conditions ; The bidirectional coupling optimization module (42) is based on a set of feasible solutions A multi-objective optimization model of bidirectional coupling is constructed, and the set of feasible solutions is solved based on a non-dominated sorting genetic algorithm to generate a Pareto front solution set ; wherein the heuristic rules based on prior knowledge include: if the construction efficiency of a certain scheme is lower than the lower quartile of the efficiency of historical similar projects, it is eliminated; if the total cost of a certain scheme exceeds a certain safety threshold of the budget, it is considered as a high-risk scheme and is eliminated; if the total duration of a certain scheme exceeds a certain buffer ratio of the contract duration, it is considered as an unfeasible scheme; if the predicted construction efficiency of a certain scheme under extreme environmental conditions is lower than the historical minimum value, it is eliminated; The multi-ranking decision module (43) is configured to rank the candidate solutions in the Pareto front solution set to determine the optimal implementation solution ; wherein the multi-ranking decision module (43) ranks the Pareto frontier solution set based on a weighted linear aggregation method incorporating a business weight vector and a technology weight vector to the solution set. ​

Images