Engineering change management and risk control intelligent decision support system

Abstract

The invention discloses an engineering change management and risk control intelligent decision support system, and relates to the technical field of engineering management, the system comprises the following components: a data acquisition module, a quantum computing simulation module, a risk assessment module, a decision support module and a knowledge base module; according to the invention, through integrating a plurality of modules of data acquisition, quantum calculation simulation, risk assessment and decision support, comprehensive intelligent support of engineering change management is realized, especially the quantum calculation simulation module can rapidly traverse massive change possibility combinations by using superposition and entanglement characteristics of quantum bits, and the change probability is improved. And simulation analysis is carried out in different risk scenes for each combination, so that the decision-making efficiency is greatly improved, a more accurate and reliable change strategy is provided for a decision maker through accurate simulation and analysis, and the risk of decision-making errors is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering management, and in particular to an intelligent decision support system for engineering change management and risk control. Background Art

[0002] In the field of engineering management, engineering change is a common phenomenon, which may be caused by a variety of factors, such as design adjustments, changes in construction schedules, changes in cost budgets, or updates to risk assessment results. Traditional engineering change management usually relies on manual judgment and experience-based decision-making, which is not only inefficient, but also easily affected by personal subjective factors, resulting in a higher risk of decision-making errors.

[0003] With the continuous development of information technology, especially the widespread application of big data and artificial intelligence technology, new solutions have been provided for engineering change management. However, most of the existing engineering change management systems are still at the level of data collection and simple analysis, lacking efficient simulation and analysis capabilities for massive possible combinations of changes, and it is difficult to provide decision makers with accurate and reliable change strategies in complex and changing risk scenarios.

[0004] In addition, risk assessment, as an important part of engineering change management, also faces many challenges. Traditional risk assessment methods are often based on qualitative analysis of historical data and expert experience. It is difficult to quantify the risk probability and risk impact of change plans under different risk scenarios, which makes risks more difficult to control and respond to.

[0005] Therefore, it is particularly important to develop an intelligent decision support system for engineering change management and risk control. Summary of the invention

[0006] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide an intelligent decision support system for engineering change management and risk control. It can achieve comprehensive intelligent support for engineering change management by integrating multiple modules such as data acquisition, efficient simulation, precise analysis and intelligent decision-making. It also provides decision makers with more accurate and reliable change strategies through precise simulation and analysis, thereby reducing the risk of decision-making errors.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: an intelligent decision support system for engineering change management and risk control, the system comprising the following components: a data acquisition module, a quantum computing simulation module, a risk assessment module, a decision support module and a knowledge base module;

[0008] The data acquisition module collects multi-source heterogeneous data related to engineering changes, including engineering design documents, construction progress data, cost budget data, and risk assessment data. It connects with the design software, construction management system, and cost accounting system database interfaces, uses data extraction algorithms, obtains accurate data in real time according to preset formats and time intervals, and performs preliminary cleaning and preprocessing operations on the collected data to remove duplicate, erroneous, or incomplete data records.

[0009] The quantum computing simulation module: Based on the superposition and entanglement characteristics of quantum bits, a quantum computing model of the engineering change plan is constructed, and various possible combinations of the engineering change plan are encoded into quantum states. Each quantum bit corresponds to a different value state of a change factor, and a multi-objective optimization algorithm based on quantum Monte Carlo simulation is used to evaluate the change plan. The formula is:

[0010]

[0011] Where E represents the comprehensive evaluation value of the change plan, n is the number of evaluation targets, and w i is the weight of the i-th evaluation target, f i (x) is the function value of the i-th evaluation target of the change plan x, and the weighted disadvantage formula is:

[0012] w i =α×W AHP (i)+(1-α)×W entropy (i)

[0013] Among them, α is the adjustment coefficient, and its value range is [0, 1]. This module uses the parallel processing capability of quantum computing to quickly traverse a large number of possible change combinations and conduct simulation analysis for each combination under different risk scenarios;

[0014] The risk assessment module: Based on the results of the quantum computing simulation module, a risk assessment algorithm combining Bayesian network and fuzzy comprehensive evaluation is used to construct a Bayesian network structure for engineering change risk, and the conditional probability table of each node is determined based on historical data and expert knowledge to determine the evaluation factor set U = {u1, u2, ..., u m} and the comment set V = {v1,v2,…,v n}, establish the fuzzy judgment matrix R, and calculate the comprehensive risk assessment value B by the following formula:

[0015] B=W·R

[0016] Among them, W is the weight vector of the evaluation factors, which is determined by combining the Delphi method with the hierarchical analysis method. This module quantitatively evaluates the risk probability and risk impact degree of various change plans under different risk scenarios;

[0017] The decision support module: Based on the results of the quantum computing simulation module and the risk assessment module, the scheme ranking algorithm based on multi-attribute decision-making is used to normalize different evaluation indicators and risk assessment results to eliminate the dimensionality effect, and then construct a decision matrix D. j (j=1,2,…,k), whose evaluation value under different attributes i is d ij , the comprehensive score S of each solution is calculated by the following formula j :

[0018]

[0019] Among them, w i The weight of each attribute is used to sort the change plans according to the comprehensive score, providing the decision maker with the theoretically optimal change strategy;

[0020] The knowledge base module stores knowledge and experience related to engineering change management and risk control. The knowledge representation adopts a combination of semantic networks and production rules. During the operation of the system, new engineering change cases and risk events are analyzed through knowledge mining algorithms to extract valuable knowledge and update the knowledge base content. The knowledge mining algorithm is based on association rule mining technology and mines potential knowledge rules from a large amount of historical data by setting support and confidence thresholds.

[0021] Furthermore, when the data acquisition module interfaces with various information systems, it adopts specific interface protocols for different types of databases. For relational databases, it adopts the SQL interface protocol. By writing customized SQL query statements, data is extracted according to preset fields and conditions. For non-relational databases, corresponding drivers and API interfaces are used to obtain data. In the process of data cleaning and preprocessing, for numerical data, an outlier detection algorithm based on statistical methods is used. For text data, natural language processing technology is used to remove noise words and perform text standardization. These operations ensure the quality of collected data and provide accurate and reliable data support for subsequent modules. When using the 3σ principle to detect outliers in numerical data, the data sequence is assumed to be x1, x2, …, x N , first calculate the mean of the data and standard deviation If the data point x j satisfy It is determined as an outlier and removed. When processing text data in natural language, the lexical analysis tool is used to segment the text and split the text into words or phrases. The grammatical relationship between words is then determined through syntactic analysis, and noise words that are not related to engineering changes are removed. The remaining vocabulary is standardized to ensure data consistency and availability.

[0022] Furthermore, in the quantum computing simulation module, the quantum state encoding of the change scheme adopts Gray code encoding, which can effectively reduce the energy consumption and error rate in the process of quantum bit state conversion. In the multi-objective optimization algorithm based on quantum Monte Carlo simulation, the number of iterations of each simulation is dynamically adjusted according to the complexity of the change scheme and the number of evaluation targets. When the change scheme involves more change factors and the evaluation targets are complex, the number of iterations is increased to improve the simulation accuracy. Conversely, the number of iterations is reduced to improve the calculation efficiency. The adjustment formula of the number of iterations is:

[0023]

[0024] Where N is the adjusted number of iterations, N0 is the initial number of iterations, β is the adjustment coefficient, m is the number of change factors, n is the number of evaluation targets, M and N are the maximum values ​​of the preset change factors and evaluation targets, respectively. In a more complex engineering change scenario, the change factor m reaches 80% of M, and the number of evaluation targets n reaches 70% of N. At this time, if β = 0.5 and N0 = 100, then according to the formula, By dynamically adjusting the number of iterations, the advantages of quantum computing parallelism can be fully utilized to improve the overall performance of the system.

[0025] Furthermore, in the process of constructing the Bayesian network of the risk assessment module, a structural learning algorithm is used to determine the network structure. The algorithm adopts a scoring search-based method, uses the Bayesian information criterion as a scoring function, scores different network structures, and selects the structure with the highest score as the optimal Bayesian network structure. When determining the conditional probability table of each node, for nodes with sufficient historical data, the maximum likelihood estimation method is used to perform parameter estimation. For nodes with insufficient data, subjective estimation is performed in combination with expert knowledge. In the fuzzy comprehensive evaluation part, the Delphi method is used to invite experts to determine the evaluation set V. Invite multiple experts in the engineering field to evaluate, and determine reasonable comment levels and corresponding descriptions based on expert feedback and statistical analysis results. Invite 10 experts to evaluate the comment set of engineering change risks. The experts vote and evaluate the possible comment levels based on their own experience and professional knowledge. After statistical analysis, if the majority of experts believe that the comment set of engineering change risks should include five levels: "high risk", "relatively high risk", "medium risk", "relatively low risk" and "low risk", then the comment set V will be determined based on this. Through this scientific construction and evaluation method, the accuracy and reliability of risk assessment can be improved.

[0026] Furthermore, in the normalization process of the multi-attribute decision-making scheme ranking algorithm in the decision support module, different normalization methods are used for different types of attributes. For benefit-type attributes, a linear proportional transformation method is used, and the formula is:

[0027]

[0028] For cost-type attributes, the inverse transformation method is used, and the formula is:

[0029]

[0030] in, is the normalized evaluation value, d ij is the original evaluation value, min(d i ) and max(d i ) are the minimum and maximum values ​​of attribute i, respectively. When determining the attribute weight by combining the information gain method with the coefficient of variation method, the information gain method is first used to calculate the information gain of each attribute to measure the influence of the attribute on the decision result. Then, the coefficient of variation method is used to calculate the coefficient of variation of each attribute to reflect the discrete degree of the attribute. Finally, the weights obtained by the two methods are combined by weighted average to obtain the final attribute weight. For example, for attribute i in a decision matrix D, it contains the evaluation value d 1i ,d 2i ,…,d ki , first calculate the information gain IG i and coefficient of variation CV i , let the weight obtained by the information gain method be w il , the weight is w i2 , through the weighted average formula w i =γ×w i1 +91-γ)×w i2 , γ is the fusion coefficient, and the final attribute weight w is obtained i ,This normalization and weight determination method for different attributes can more accurately reflect the ,importance of each attribute in decision making and improve the ,scientific nature of decision making.

[0031] Furthermore, in terms of knowledge representation, the nodes of the semantic network of the knowledge base module represent concepts, entities or attributes, the edges represent the relationship between them, and the form of the production rules is IF condition THEN conclusion. By combining the semantic network with the production rules, the knowledge in the field of engineering change management and risk control can be more comprehensively and accurately represented. In the process of knowledge mining, the association rule mining technology adopts the Apriori algorithm. On the basis of the traditional Apriori algorithm, a pruning strategy is introduced. According to the support and confidence thresholds, the item sets that cannot generate strong association rules are deleted in advance to reduce the amount of calculation. At the same time, in order to improve the mining efficiency, a distributed computing framework is adopted to divide the data into multiple subsets, and the mining tasks are performed in parallel on different computing nodes. Finally, the results are merged. For a data set containing a large amount of engineering change case data, it is divided into 5 subsets, and association rule mining is performed simultaneously on 5 computing nodes. Each node processes its own subset according to the preset support and confidence thresholds to mine local association rules. Finally, these local rules are merged to obtain global association rules, thereby improving the knowledge update efficiency and quality of the knowledge base and providing the system with richer and more accurate knowledge support.

[0032] Furthermore, the system also includes a user interaction module, which provides a friendly graphical interface to facilitate decision makers to interact with the system. The interface design uses intuitive charts and visualization elements to display information related to the engineering change plan. Users can input specific requirements and constraints through the interface, and the system screens and re-evaluates the change plan based on these conditions. At the same time, users can provide feedback on the change plan recommended by the system. The system adjusts the corresponding algorithm parameters and models based on user feedback to achieve human-computer collaborative optimization function. After viewing the risk level distribution map of the change plan, the user believes that the weight setting of a certain risk factor in the current risk assessment model is unreasonable. The user enters the adjustment suggestion through the interface. After receiving the suggestion, the system recalculates the weight of the risk factor in the risk assessment model, updates the risk assessment results of the change plan, and provides the user with the adjusted plan information again, thereby improving user experience and decision satisfaction.

[0033] Furthermore, the system has a data security and backup mechanism. During the data collection and transmission process, encryption technology is used to encrypt sensitive data to prevent data leakage and tampering. The data stored in the system database is backed up regularly. The backup strategy adopts a combination of full backup and incremental backup. The full backup completely backs up the entire database at a set time interval, and the incremental backup records the changed data between two full backups. The backup data is stored in an off-site data center to prevent data loss in the event of a local data center failure or disaster. At the same time, the system sets up an access permission management module to assign different system access permissions according to user roles and responsibilities to ensure that only authorized users can access and operate relevant data, thereby ensuring system data security and stable operation. For highly sensitive data in engineering design files, access is only granted to engineers and managers with senior permissions, and the AES encryption algorithm is used for encryption during data transmission. The encryption key is updated regularly to enhance data security. In terms of backup, a full backup is performed at 2 a.m. every Sunday, and an incremental backup is performed at 3 a.m. every day. The backup data is transmitted to an off-site data center for storage via a dedicated network to ensure that the data can be effectively protected under any circumstances.

[0034] Compared with the existing technology, this engineering change management and risk control intelligent decision support system has the following beneficial effects:

[0035] 1. This system realizes comprehensive intelligent support for engineering change management by integrating multiple modules including data acquisition, quantum computing simulation, risk assessment and decision support. In particular, the quantum computing simulation module uses the superposition and entanglement characteristics of quantum bits to quickly traverse a large number of possible change combinations and conduct simulation analysis for each combination under different risk scenarios. This not only greatly improves the efficiency of decision-making, but also provides decision makers with more accurate and reliable change strategies through precise simulation and analysis, thereby reducing the risk of decision-making errors.

[0036] 2. This system uses a risk assessment algorithm that combines Bayesian networks with fuzzy comprehensive evaluation to quantitatively assess the risk probability and risk impact of various change plans under different risk scenarios. This quantitative assessment method makes risks more intuitive and controllable, helping decision makers to better identify and respond to potential risks. At the same time, the continuous updating and improvement of the knowledge base module also provides the system with richer risk management experience and response strategies, further enhancing the system's risk management and control capabilities.

[0037] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1 It is a process operation diagram of an engineering change management and risk control intelligent decision support system. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0041] Embodiment 1

[0042] This embodiment describes a large-scale bridge construction project. Due to complex geological conditions, the pier foundation design needs to be changed during the construction process. At this time, the engineering change management and risk control intelligent decision support system begins to play a role.

[0043] This module interfaces with the bridge design software, construction progress management system, and cost accounting system. For the relational database in the design software, the SQL interface protocol is used. By writing SQL query statements, the engineering design file data related to the pier foundation design, such as the pier size and foundation type, is extracted. For the non-relational database in the construction progress management system, the corresponding driver and API interface are used to obtain the construction progress data, such as the completed pier construction progress and the remaining construction volume. The cost budget data is obtained from the cost accounting system, including the cost of the original design plan and the estimated cost after the change. At the same time, risk assessment data of similar bridge engineering changes in the past are also collected. In the data cleaning and preprocessing stage, for numerical data, such as the bearing capacity of the pier foundation, the 3σ principle is used for outlier detection. Assume that the acquired pier foundation bearing capacity data sequence is x1, x2,…, x N , first calculate the mean and standard deviation If the data point x j satisfy It is determined as an outlier and removed. For text data, such as records in construction logs, lexical analysis tools are used to perform word segmentation, remove noise words that are not related to the changes in the pier foundation, such as weather descriptions, and standardize the remaining words.

[0044] The various possible combinations of the pier foundation change scheme are encoded into quantum states using Gray code encoding. For example, the change factors include the foundation form (such as expanded foundation, pile foundation), foundation size (different lengths, widths, depths), and each quantum bit corresponds to a different value state of a change factor. The change scheme is evaluated using a multi-objective optimization algorithm based on quantum Monte Carlo simulation. The evaluation targets include project cost, construction period, and structural stability. The number of evaluation targets n = 3, and the formula is E = ∑ i =1 3 w i ×f i (x), where E represents the comprehensive evaluation value of the change plan, w i is the weight of the i-th evaluation target, f i (x) is the function value of the i-th evaluation target of the change plan x, and the weight calculation formula is w i =α×W AHP (i)+(1-α)×W entropy (i), α is 0.5, assuming that the number of change factors m = 4, the preset maximum value of change factors M = 5, the maximum number of evaluation targets N = 5, the initial number of iterations N0 = 100, and the adjustment coefficient β = 0.5, then the number of iterations Through the parallel processing capabilities of quantum computing, we can quickly traverse a large number of possible change combinations and conduct simulation analysis for each combination under different risk scenarios.

[0045] Based on the results of the quantum computing simulation module, a risk assessment algorithm combining Bayesian network and fuzzy comprehensive evaluation is adopted. The Bayesian network structure is determined by using the structural learning algorithm. The Bayesian information criterion is used as the scoring function to score different network structures. The structure with the highest score is selected to determine the evaluation factor set U = {u1,u2,…,u m} (such as geological conditions, construction technology difficulty, material supply) and the comment set V = {v1,v2,…,v n The evaluation criteria are as follows: 1. The evaluation criteria are as follows: 1. The evaluation criteria are as follows:} (such as low risk, medium risk, and high risk). The comment set is evaluated by inviting multiple bridge engineering experts through the Delphi method, and a fuzzy judgment matrix R is established. The weight vector W is determined by combining the Delphi method with the hierarchical analysis method. The comprehensive risk evaluation value B = W·R is used to quantitatively evaluate the risk probability and risk impact degree of various pier foundation change plans under different risk scenarios.

[0046] According to the results of the quantum computing simulation module and the risk assessment module, the scheme ranking algorithm based on multi-attribute decision-making is used to normalize different evaluation indicators (such as cost, construction period, stability, risk) and risk assessment results. For cost attributes (such as engineering cost), the inverse transformation method is used. The formula is: For benefit-type attributes (such as structural stability), the linear proportional transformation method is used, and the formula is: Construct the decision matrix D, determine the attribute weights by combining the information gain method with the coefficient of variation method, set the fusion coefficient γ = 0.6, and calculate the comprehensive score of each solution The change plans are ranked according to the comprehensive scores to provide decision makers with the theoretically optimal pier foundation change strategy.

[0047] The knowledge and experience related to bridge engineering change management and risk control are stored. The knowledge representation adopts the combination of semantic network and production rules. For example, the nodes in the semantic network represent bridge piers, foundations, and geological concepts, and the edges represent the relationship between them, such as "bridge piers are located on the foundation" and "the foundation is affected by geological conditions". Production rules include "IF the geological conditions are complex AND the foundation form changes THEN the risk increases". In the knowledge mining process, the Apriori algorithm is used, and the pruning strategy is introduced. According to the support and confidence thresholds, the item sets that cannot generate strong association rules are deleted in advance. The distributed computing framework is used to mine knowledge in parallel, and potential knowledge rules are mined, such as "the association rule between specific foundation changes and extension of construction period under certain geological conditions", and the knowledge base content is updated.

[0048] Embodiment 2

[0049] This embodiment describes that during the construction of a high-rise commercial building, due to commercial planning adjustments, the internal space layout of the building needs to be changed, including floor function zoning and store area adjustments.

[0050] Data is connected with architectural design software (such as AutoCAD), construction management system (including progress, personnel, and equipment information), cost accounting system (involving construction materials, labor cost data) and risk assessment database. For the relational database in the design software, the SQL interface protocol is used to extract the design file data related to the internal space layout of the building, such as the floor plan and space dimensions of each floor. The construction progress data is obtained through the API interface of the construction management system, such as the construction progress percentage of each floor, the current number of construction personnel and equipment usage. The cost budget data is obtained from the cost accounting system, such as the cost of different decoration materials and the construction cost of different functional areas. At the same time, risk assessment data on past commercial building space layout changes are collected. In the data cleaning and preprocessing stage, for numerical data, such as the store area value, the 3σ principle is used to detect outliers. Assuming that the store area data sequence is x1, x2,…, x N , calculate the mean and standard deviation If the data point x j satisfy It is considered as an outlier and removed. For text data, such as change application records during the construction process, lexical analysis tools are used to segment words, remove noise words such as application date and applicant name, and perform standardization.

[0051] The quantum computing model is constructed with various combinations of the change plans for the internal space layout of the building. Gray code encoding is used. The change factors include the floor functional zoning type (such as public area, commercial area, leisure area), the size range of the store area, and the public area setting. Each quantum bit corresponds to a different value state of a change factor. The multi-objective optimization algorithm based on quantum Monte Carlo simulation is used to evaluate the change plan. The evaluation targets include construction cost, commercial operation benefits, and construction period. n = 3, and the formula is E = ∑ i =1 3 w i ×f i (x), the weight calculation formula is w i =α×W AHP (i)+(1-α)×W entropy (i), α is 0.6, assuming that the number of change factors m = 5, the preset maximum value of change factors M = 6, the maximum number of evaluation targets N = 4, the initial number of iterations N0 = 100, and the adjustment coefficient β = 0.4, then the number of iterations With the parallel processing capabilities of quantum computing, various possible combinations of changes can be simulated and analyzed under different risk scenarios.

[0052] Based on the results of quantum computing simulation, a risk assessment algorithm combining Bayesian network and fuzzy comprehensive evaluation is used to determine the Bayesian network structure through a structural learning algorithm. The optimal structure is selected by scoring with the Bayesian information criterion to determine the evaluation factor set U = {u1,u2,…,u m} (such as changes in market demand, construction technology difficulty, fire safety requirements) and the comment set V = {v1,v2,…,v n The evaluation criteria are as follows: 1. The evaluation criteria are as follows: 1. The evaluation criteria are as follows: (1) low risk, (2) medium risk, (3) high risk, (4) high risk, (5) low risk, (6) medium risk, (7) high risk, (8) high risk, (9) low risk, (10) medium risk, (11) high risk, (12) high risk, (13) high risk, (14) low risk, (15) medium risk, (16) high risk, (17) high risk, (18) high risk, (20) high risk, (21) high risk, (22) high risk, (23) high risk, (24) high risk, (25) high risk, (26) high risk, (27) high risk, (28) high risk, (29) high risk, (30) high risk, (31) high risk, (32) high risk, (33) high risk, (34) high risk, (35) high risk

[0053] According to the results of quantum computing simulation and risk assessment, the scheme ranking algorithm of multi-attribute decision-making is used to normalize the evaluation indicators (such as cost, benefit, cycle, risk) and risk assessment results. The cost-type attributes are transformed by the inverse transformation method, and the benefit-type attributes are transformed by the linear proportional transformation method. The decision matrix D is constructed, and the attribute weights are determined by combining the information gain method with the coefficient of variation method. The fusion coefficient γ is set to 0.7, and the comprehensive score of each scheme is calculated. The change plans are ranked according to the comprehensive scores to provide decision makers with the optimal strategy for changing the internal space layout of the building.

[0054] It stores the knowledge and experience of change management and risk control of commercial construction projects, and adopts a knowledge representation method that combines semantic networks with production rules. Semantic network nodes represent floors, functional areas, and shop concepts, and edges represent relationships, such as "the shop is located in the commercial area of ​​a certain floor". Production rules include "IF the market demand changes greatly AND the functional zoning is adjusted THEN the business operation efficiency is affected". When mining knowledge, the Apriori algorithm and pruning strategy are used to delete irrelevant item sets based on support and confidence thresholds, and use distributed computing frameworks for parallel mining, such as mining "the association rules between changes in specific functional zoning and changes in construction cycles" to update the knowledge base.

[0055] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. An intelligent decision support system for engineering change management and risk control, characterized in that: The system includes the following components: data acquisition module, quantum computing simulation module, risk assessment module, decision support module and knowledge base module; The data acquisition module collects multi-source heterogeneous data related to engineering changes, including engineering design documents, construction progress data, cost budget data, and risk assessment data. It connects with the design software, construction management system, and cost accounting system database interfaces, uses data extraction algorithms, obtains accurate data in real time according to preset formats and time intervals, and performs preliminary cleaning and preprocessing operations on the collected data to remove duplicate, erroneous, or incomplete data records. The quantum computing simulation module: Based on the superposition and entanglement characteristics of quantum bits, a quantum computing model of the engineering change plan is constructed, and various possible combinations of the engineering change plan are encoded into quantum states. Each quantum bit corresponds to a different value state of a change factor, and a multi-objective optimization algorithm based on quantum Monte Carlo simulation is used to evaluate the change plan. The formula is: Where E represents the comprehensive evaluation value of the change plan, n is the number of evaluation targets, and w i is the weight of the i-th evaluation target, f i (x) is the function value of the i-th evaluation target of the change plan x, and the weighted disadvantage formula is: In i =α×W AHP (i)+(1-α)×W entropy (and) Among them, α is the adjustment coefficient, and its value range is [0, 1]. This module uses the parallel processing capability of quantum computing to quickly traverse a large number of possible change combinations and conduct simulation analysis for each combination under different risk scenarios; The risk assessment module: Based on the results of the quantum computing simulation module, a risk assessment algorithm combining Bayesian network and fuzzy comprehensive evaluation is used to construct a Bayesian network structure for engineering change risk, and the conditional probability table of each node is determined based on historical data and expert knowledge to determine the evaluation factor set U = {u1, u2, ..., u m } and the comment set V = {v1,v2,…,v n }, establish the fuzzy judgment matrix R, and calculate the comprehensive risk assessment value B by the following formula: B=W·R Among them, W is the weight vector of the evaluation factors, which is determined by combining the Delphi method with the hierarchical analysis method. This module quantitatively evaluates the risk probability and risk impact degree of various change plans under different risk scenarios; The decision support module: Based on the results of the quantum computing simulation module and the risk assessment module, the scheme ranking algorithm based on multi-attribute decision-making is used to normalize different evaluation indicators and risk assessment results to eliminate the dimensionality effect, and then construct a decision matrix D. j (j=1,2,…,k), whose evaluation value under different attributes i is d ij , the comprehensive score S of each solution is calculated by the following formula j : Among them, w i The weight of each attribute is used to sort the change plans according to the comprehensive score, providing the decision maker with the theoretically optimal change strategy; The knowledge base module stores knowledge and experience related to engineering change management and risk control. The knowledge representation adopts a combination of semantic networks and production rules. During the operation of the system, new engineering change cases and risk events are analyzed through knowledge mining algorithms to extract valuable knowledge and update the knowledge base content. The knowledge mining algorithm is based on association rule mining technology and mines potential knowledge rules from a large amount of historical data by setting support and confidence thresholds.

2. The intelligent decision support system for engineering change management and risk control according to claim 1 is characterized in that: When the data acquisition module interfaces with various information systems, it adopts specific interface protocols for different types of databases. For relational databases, it adopts SQL interface protocol. By writing customized SQL query statements, data is extracted according to preset fields and conditions. For non-relational databases, corresponding drivers and API interfaces are used to obtain data. In the process of data cleaning and preprocessing, for numerical data, an outlier detection algorithm based on statistical methods is used. For text data, natural language processing technology is used to remove noise words and perform text standardization. Through these operations, the quality of collected data is ensured, and accurate and reliable data support is provided for subsequent modules. When using the 3σ principle to detect outliers in numerical data, the data sequence is assumed to be x1, x2, …, x N , first calculate the mean of the data and standard deviation If the data point x j satisfy It is determined as an outlier and removed. When processing text data in natural language, the lexical analysis tool is used to segment the text, split the text into words or phrases, and then the grammatical relationship between words is determined through syntactic analysis, and then the noise words that are not related to the engineering change are removed, and the remaining words are standardized.

3. The intelligent decision support system for engineering change management and risk control according to claim 1 is characterized in that: In the quantum computing simulation module, the quantum state encoding of the change scheme adopts Gray code encoding, which can effectively reduce the energy consumption and error rate in the process of quantum bit state conversion. In the multi-objective optimization algorithm based on quantum Monte Carlo simulation, the number of iterations of each simulation is dynamically adjusted according to the complexity of the change scheme and the number of evaluation targets. When the change scheme involves many change factors and the evaluation targets are complex, the number of iterations is increased to improve the simulation accuracy. The adjustment formula of the number of iterations is: Where N is the adjusted number of iterations, N0 is the initial number of iterations, β is the adjustment coefficient, m is the number of change factors, n is the number of evaluation targets, M and N are the maximum values ​​of the preset change factors and evaluation targets, respectively. In a more complex engineering change scenario, the change factor m reaches 80% of M, and the number of evaluation targets n reaches 70% of N. At this time, if β = 0.5 and N0 = 100, then according to the formula, By dynamically adjusting the number of iterations, the parallel advantages of quantum computing can be fully utilized.

4. The intelligent decision support system for engineering change management and risk control according to claim 1 is characterized in that: In the process of constructing the Bayesian network of the risk assessment module, a structural learning algorithm is used to determine the network structure. The algorithm adopts a scoring search-based method, with the Bayesian information criterion as the scoring function, to score different network structures, and select the structure with the highest score as the optimal Bayesian network structure. When determining the conditional probability table of each node, for nodes with sufficient historical data, the maximum likelihood estimation method is used to perform parameter estimation. For nodes with insufficient data, subjective estimation is performed in combination with expert knowledge. In the fuzzy comprehensive evaluation part, the Delphi method is used to determine the comment set V, and multiple engineering experts are invited to evaluate. According to the expert feedback and statistical analysis results, reasonable comment levels and corresponding descriptions are determined.

5. The intelligent decision support system for engineering change management and risk control according to claim 1 is characterized in that: In the normalization process of the multi-attribute decision-making scheme ranking algorithm in the decision support module, different normalization methods are used for different types of attributes. For benefit-type attributes, a linear proportional transformation method is used, and the formula is: For cost-type attributes, the inverse transformation method is used, and the formula is: in, is the normalized evaluation value, d ij is the original evaluation value, min(d i ) and max(d i ) are the minimum and maximum values ​​of attribute i, respectively. When determining the attribute weight by combining the information gain method with the coefficient of variation method, the information gain method is first used to calculate the information gain of each attribute to measure the degree of influence of the attribute on the decision result. Then, the coefficient of variation method is used to calculate the coefficient of variation of each attribute to reflect the discrete degree of the attribute. Finally, the weights obtained by the two methods are combined by weighted average to obtain the final attribute weight. For attribute i in a decision matrix D, it contains the evaluation value d 1i ,d 2i ,…,d ki , first calculate the information gain IG i and coefficient of variation CV i , let the weight obtained by the information gain method be w il , the weight is w i2 , through the weighted average formula w i =γ×w i1 +(1-γ)×w i2 , γ is the fusion coefficient, and the final attribute weight w is obtained i .

6. The intelligent decision support system for engineering change management and risk control according to claim 1 is characterized in that: In terms of knowledge representation, the nodes of the semantic network of the knowledge base module represent concepts, entities or attributes, the edges represent the relationship between them, and the production rules are in the form of IF conditions and THEN conclusions. By combining the semantic network with the production rules, the knowledge in the field of engineering change management and risk control can be more comprehensively and accurately represented. In the process of knowledge mining, the association rule mining technology adopts the Apriori algorithm. On the basis of the traditional Apriori algorithm, a pruning strategy is introduced. According to the support and confidence thresholds, the item sets that cannot generate strong association rules are deleted in advance to reduce the amount of calculation. A distributed computing framework is used to divide the data into multiple subsets, and the mining tasks are executed in parallel on different computing nodes, and the results are finally merged.

7. The intelligent decision support system for engineering change management and risk control according to claim 1 is characterized in that: The system also includes a user interaction module, which provides a friendly graphical interface. The interface design uses intuitive charts and visualization elements to display information related to the engineering change plan. Users can input specific requirements and constraints through the interface. The system screens and re-evaluates the change plan based on these conditions. At the same time, users can provide feedback on the change plan recommended by the system. The system adjusts the corresponding algorithm parameters and models based on user feedback to achieve human-computer collaborative optimization function.

8. The intelligent decision support system for engineering change management and risk control according to claim 1 is characterized in that: The system has a data security and backup mechanism. During the data collection and transmission process, encryption technology is used to encrypt sensitive data to prevent data leakage and tampering. The data stored in the system database is backed up regularly, and the backup strategy uses a combination of full backup and incremental backup. The full backup completely backs up the entire database at a set time interval, and the incremental backup records the data changes between two full backups. The backup data is stored in an off-site data center to prevent data loss in the event of a local data center failure or disaster. At the same time, the system sets up an access permission management module to assign different system access permissions according to user roles and responsibilities.

Images