IDI full-project periodic risk digital assessment method and system

Abstract

The invention discloses an IDI full-project cycle risk digital assessment method and system, which have the beneficial effects that potential risk factors are identified from project data by adopting a text mining technology, risk factor importance assessment is carried out by utilizing a mode of weighting word frequency through information entropy, risk influence factors can be comprehensively identified, and the risk assessment accuracy is improved. Exploring the internal influence of project quality and risk factors; a multi-level and multi-dimensional risk evaluation system is constructed by adopting an analytic hierarchy process, so that the quality risk state of a project can be comprehensively reflected, and a quantitative standard of an evaluation index system is given; according to the method, a project risk evaluation model is constructed, a project risk scoring model is constructed from four dimensions of a hierarchical weight matrix, a severity matrix, a conditional probability matrix and an index occurrence probability matrix of risk indexes, various risk factors are considered, and the mutual relation and occurrence possibility between risks can be known more comprehensively; a Monte Carlo simulation method is adopted to grade and divide project risks, and the risk can be quickly traced, monitored and managed according to a result.

Description

[0001] The present invention mainly relates to the field of risk assessment of construction projects, and in particular to an IDI full-project cycle risk digital assessment method and system. Background Art

[0002] Project quality is the foundation of construction projects, but in reality, due to the large scale and wide coverage of construction projects, various quality problems occur frequently. According to the data released by the China Consumers Association, the proportion of quality problems in the data of complaints received for housing and building materials in recent years has been counted. The number of consumer complaints in the field of housing and building materials has always remained high, among which the proportion of complaints due to quality has continued to remain at a level of more than 25%, including roof leakage, wall peeling, structural plate cracks and other quality problems are common. The high number of housing and service complaints not only reflects the current situation of consumer rights being damaged, but also reflects that in the process of construction and delivery, due to the lack of a systematic and strong supervision mechanism, project quality problems frequently occur, which in turn increases the management pressure of government regulatory departments and the society's demand for an effective complaint resolution mechanism. At the same time, due to the characteristics of a posteriori and hidden nature of construction project quality, many quality problems are difficult to find during acceptance. When this happens, due to the large number of construction project entities and the vague boundaries of responsibility division, quality problems are difficult to be effectively resolved, causing damage to the rights and interests of homeowners.

[0003] In 2019, the Ministry of Housing and Urban-Rural Development proposed in the "Guiding Opinions on Improving the Quality Assurance System and Improving the Quality of Construction Projects" that it is necessary to "organize and carry out pilot projects for engineering quality insurance and accelerate the development of engineering quality insurance". Among them, Inherent Defect Insurance (IDI) is the most widely studied and promising insurance. The insured purchases engineering latent defect insurance for the developed buildings. When the buildings are damaged due to potential internal defects, the insurance company bears the compensation liability. The basic coverage of IDI is foundation and main structure engineering (10 years), insulation and waterproofing engineering (5 years), decoration and decoration engineering (2 years). Among them, the probability of major defects in foundation and main structure engineering is low, but the compensation cost is high, which is the focus of risk management. Insulation and waterproofing engineering, decoration and decoration, installation engineering, etc. are common problems in the industry. The probability of occurrence is high, and complaints are frequent. It is the difficulty of risk management.

[0004] At present, most existing risk management institutions only consider from their own perspective and have not yet established a full-process and comprehensive project risk system. The existing engineering quality risk assessment is still at the stage of simple analysis of historical accidents based on experience, and cannot provide insurance companies with a solid basis for insurance. The existing technology mainly uses the following methods to conduct project risk assessment: 1. Quality risk assessment method based on fuzzy theory FSE; Disadvantages: insufficient analysis accuracy, insufficient analysis of the comprehensive effects of various factors, unable to cover all influencing factors, and has certain limitations in application. The evaluation results depend more on the domain knowledge and rich experience of the decision-making team. Affected by the subjectivity and ambiguity of expert knowledge, the evaluation results are prone to deviations; 2. Quality assessment methods based on Bayesian networks, deep learning, evidence theory, and scalable cloud; Disadvantages: The construction engineering environment is complex, the risk influencing factors are intricate, and the uncertainty is high. The existing artificial intelligence methods cannot fully reflect the hierarchy and correlation between project quality and risk factors, nor can they fully reflect the complex internal connections between risk factors; at the same time, this type of method has high requirements on the quality and data of training samples, and requires high-quality, large-scale data sets. However, the data quality in the engineering field is uneven, and the actual accumulated data samples are limited, which leads to challenges in the adaptability and accuracy of the model.

[0005] Therefore, how to design a simple, efficient, full-cycle and comprehensive project risk assessment method and system that can fully reflect the hierarchy and correlation between project quality and risk factors is a technical problem to be solved. Summary of the invention

[0006] Based on this, it is necessary to provide an IDI full-project cycle risk digital assessment method and system to address existing problems.

[0007] In a first aspect, an embodiment of the present application provides a method for digitally assessing the risks of an IDI full project cycle, comprising the following steps: Obtain project-related information and historical data, and build a project risk database; Based on the project corpus and expert knowledge base, a domain dictionary associated with the project is constructed, and data cleaning, screening, verification and text segmentation are performed on the project risk database to obtain a project risk factor set consisting of multiple project risk factors; The project risk factor set is processed by using an entropy weight method to obtain a weighted word frequency set consisting of weighted word frequencies of each project risk factor in the project risk factor set, and the weighted word frequency set is filtered based on a preset threshold to obtain a key risk factor set; Processing the key risk factor set through subject coding to obtain a standard key risk factor set; Performing risk quantification and matrix conversion on the standard key risk factor set to obtain a key risk factor occurrence matrix and a key risk factor probability matrix; Use multi-level analysis to classify the standard key risk factor set and build a project risk assessment system; Based on the project risk assessment system and the project historical data, construct a severity weight matrix of key risk factors of the project; Using an expert scoring method to perform risk quantification processing on each standard key risk factor in the standard key risk factor set to obtain a key risk factor weight matrix; Inputting the key risk factor occurrence matrix, the key risk factor probability matrix, the key risk factor severity weight matrix and the key risk factor weight matrix into a project risk assessment model to obtain a comprehensive project risk score; The Monte Carlo simulation method is used to classify the comprehensive risk score of the project and output the project risk assessment result.

[0008] Preferably, the obtaining of project-related information and project historical data and building of a project risk database includes: Comprehensively collect project-related information and historical data, and based on a detailed understanding of project-related data, analyze, confirm, summarize, and organize project risks, and finally compile them into a project risk database.

[0009] Preferably, the project risk factor set is processed by using an entropy weight method to obtain a weighted word frequency set consisting of weighted word frequencies of each project risk factor in the project risk factor set, including: Based on the probability distribution of each project risk factor in the project risk factor set, the information entropy of the project risk factor set is obtained using formula (1): ; The information entropy is combined with the word frequency of the project risk factor through formula (2) to obtain the weighted word frequency of the project risk factor: ; in, is the first in the set of project risk factors Project risk factors, , is the number of project risk factors in the project risk factor set; in, For the The average probability of occurrence of project risk factors, It is expressed by formula (3): ; in, For the The project risk factors are The probability of occurrence in a project is obtained by formula (4): ; in, is the first in the set of project risk factors The project risk factors are The number of times it occurred in each project, , is the number of items; For the The word frequency of each project risk factor is expressed by formula (5): .

[0010] Preferably, the key risk factor set is processed by subject coding to obtain a standard key risk factor set, including: Based on the professional knowledge and experience of experts, each key risk factor of the key risk factor set is standardized and coded and sorted according to preset rules to obtain the standard key risk factor set.

[0011] Preferably, risk quantification and matrix conversion are performed on the standard key risk factor set to obtain a key risk factor occurrence matrix and a key risk factor probability matrix, including: Manual combing and NPL algorithm are used to annotate the data of the standard key risk factor set; Based on the standard key risk factor set, the standard key risk factor set after data annotation is mapped and binarized according to the standard key risk factor ID to obtain a key risk factor occurrence matrix and a key risk factor probability matrix; The key risk factor occurrence matrix is ​​expressed by formula (6): ; The key risk factor probability matrix is ​​expressed by formula (7): ; in, The first of the set of standard key risk factors The key risk factors of the standard are in The probability of occurrence in a project, ; in, , , For the The probability of occurrence of the key risk factors of the standard For the Key risk factors and The probability of the three standard key risk factors occurring simultaneously.

[0012] Preferably, the standard key risk factor set is classified using a multi-level analysis method to construct a project risk assessment system, including: Firstly, the standard key risk factor set is classified by multi-level analysis method, and a subjective analysis model with hierarchical structure is constructed, which is divided into four levels, namely, target level, factor level, project level and indicator level; Among them, the target layer represents the overall risk status of the project; the factor layer represents the type of standard key risk factors, which is a first-level indicator; the project layer represents the detection items of standard key risk factors, which is a second-level indicator; the indicator layer represents the specific sub-indicators of standard key risk factors, which is a third-level indicator.

[0013] Preferably, based on the project risk assessment system and the project historical data, a severity weight matrix of key risk factors of the project is constructed, including: Assigning weights to each standard key risk factor in the standard key risk factor set based on project historical data and the Delphi expert method to obtain an initial weight for each standard key risk factor; The Cronbach's Alpha model is used to perform consistency test on the initial weight of each standard key risk factor in the standard key risk factor set to obtain the key risk factor severity weight matrix of the project; The key risk factor severity weight matrix is ​​expressed by formula (8): ; in, ; , For the matrix The maximum value in is the minimum value in the matrix; , For the The key risk factors of the standard are in The probability of occurrence in an item; For the The initial risk weights of the standard key risk factors are obtained by formula (9): ; in, For the The weighted values ​​of the project historical data of the standard key risk factors are For the Expert weightings of standard key risk factors.

[0014] Preferably, an expert scoring method is used to perform risk quantification processing on each standard key risk factor in the standard key risk factor set to obtain a key risk factor weight matrix, including: Based on the Delphi expert method, weights are assigned to the standard key risk factor set to obtain a key risk factor weight matrix; The key risk factor weight matrix is ​​expressed by formula (10): ; in, For the Project No. The weights of key risk factors are determined based on the criteria.

[0015] Preferably, the project risk assessment model is represented by formula (11): ; in, Provide a comprehensive risk score for the project.

[0016] In a second aspect, the embodiment of the present application provides an IDI full-project cycle risk digital assessment system, including: Data acquisition module, used to obtain project-related information and project historical data, and build a project risk database; A preprocessing module is used to construct a domain dictionary associated with the project based on the project corpus and expert knowledge base and to perform data cleaning, screening, verification and text segmentation on the project risk database to obtain a project risk factor set consisting of multiple project risk factors; A first data processing module is used to process the project risk factor set by using an entropy weight method to obtain a weighted word frequency set composed of weighted word frequencies of each project risk factor in the project risk factor set and filter the weighted word frequency set based on a preset threshold to obtain the key risk factor set; A data coding module, used for processing the key risk factor set through subject coding to obtain a standard key risk factor set; A second data processing module is used to perform risk quantification and matrix conversion on the standard key risk factor set to obtain a key risk factor occurrence matrix and a key risk factor probability matrix; The data analysis module uses a multi-level analysis method to classify the standard key risk factor set and build a project risk assessment system; A first matrix calculation module is used to construct a severity weight matrix of key risk factors of the project based on the project risk assessment system and the project historical data; The second matrix calculation module is used to perform risk quantification processing on each standard key risk factor in the standard key risk factor set by using an expert scoring method to obtain a key risk factor weight matrix; A risk assessment module, used to input the key risk factor occurrence matrix, the key risk factor probability matrix, the key risk factor severity weight matrix and the key risk factor weight matrix into a project risk assessment model to obtain a comprehensive project risk score; The result output module is used to classify the comprehensive risk score of the project using the Monte Carlo simulation method and output the project risk assessment result.

[0017] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention adopts text mining algorithm technology to identify potential risk factors from the project quality risk description, designs a text mining framework, and calculates the information entropy weighted word frequency TF-H to evaluate the importance of risk factors, which can effectively and comprehensively identify risk influencing factors and explore the intrinsic influence of building quality and risk factors; (2) The analytic hierarchy process (AHP) is used to construct a reasonable and scientific multi-level and multi-dimensional risk evaluation system for the quality status of construction projects, which can reasonably and comprehensively reflect the quality risk status of construction projects and provide quantitative standards for the evaluation index system; (3) A new project quality risk evaluation model is constructed, and a project quality risk scoring model is constructed from four dimensions: the hierarchical weight matrix, severity matrix, conditional probability matrix, and indicator occurrence probability matrix of risk indicators. The index system comprehensively considers various risk factors, including but not limited to technical requirements, material quality, process specifications, etc. This indicator system will help project managers to have a more comprehensive understanding of the interrelationships between risks and their potential occurrence; (4) Construct a project risk assessment model, quantify project quality risks, and use Monte Carlo simulation to rate and classify project risks. Through a comprehensive assessment of these factors, the risks of construction projects can be quantitatively scored, guiding project managers to quickly trace and monitor management risks, and to develop targeted risk closure strategies and improvement measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] By referring to the following drawings, the exemplary embodiments of the present invention can be more completely understood. The drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present invention and do not constitute a limitation of the present invention. In the drawings, the same reference numerals generally represent the same parts or steps.

[0019] Figure 1 A flowchart of a digital risk assessment method for the entire IDI project cycle provided according to an exemplary embodiment of the present application; Figure 2A word cloud diagram of key risk factors of an IDI full-project cycle risk digital assessment method provided according to an exemplary embodiment of the present application; Figure 3 A matrix diagram showing the binarization of standard key risk factors of an IDI full-project cycle risk digital assessment method provided by an exemplary embodiment of the present application is shown; Figure 4 A schematic diagram of a project risk analysis model provided by an exemplary embodiment of the present application is shown; Figure 5 A distribution diagram of predicted variables of a Monte Carlo simulation provided by an exemplary embodiment of the present application is shown; Figure 6 A schematic diagram of the structure of an IDI full-project cycle risk digital assessment system provided according to another exemplary embodiment of the present application; Figure 7 A risk matrix diagram of project and factor layers provided by an exemplary embodiment of the present application is shown; Figure 8 A risk matrix diagram of the project and indicator layers provided by an exemplary embodiment of the present application is shown; Fig. 9 A risk quantification diagram of various indicators in a project provided by an exemplary embodiment of the present application is shown; DETAILED DESCRIPTION

[0020] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0021] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0022] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0023] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0024] The embodiment of the present application provides a digital risk assessment method for the entire IDI project cycle, which is described below in conjunction with the accompanying drawings.

[0025] Reference Figure 1 , which shows a digital risk assessment method for the entire IDI project cycle provided by some embodiments of the present application, including the following steps: S101: Obtain project-related information and historical data, and build a project risk database; Specifically, the assessment data of all implemented IDI projects are collected through the information collection platform, including the survey task list, project risk tracking, assessment work traces, project risk collection, engineering entity inspection list, etc. At the same time, comprehensive collection of project-related information and historical data, on the basis of a detailed understanding of the project-related data, the project risks are analyzed, confirmed, summarized, and sorted, and finally summarized to form a project risk database.

[0026] In this embodiment, taking the Guangzhou regional data in South China as an example, Guangzhou has been implementing regional pilot projects and full-area mandatory policies and technical standards since 2020. As of December 2023, the data has accumulated for 57 projects, with a total of 2,165 rounds of assessments and 28,597 standardized risk data, mainly major risk data. In terms of the project data cycle span, currently projects with fewer inspection rounds (less than or equal to 10 rounds) account for 62.5%; projects with a duration that is too short (less than or equal to 5 months) account for 23.75%. Data from projects with more than 5 rounds and a duration greater than 5 months are selected, with data related to major quality risks based on the construction process as the data basis for subsequent project risk assessment models.

[0027] S102: constructing a domain dictionary associated with the project based on the project corpus and expert knowledge base and performing data cleaning, screening, verification and text segmentation on the project risk database to obtain a project risk factor set consisting of multiple project risk factors; Specifically, by using text mining technology to conduct in-depth analysis of the collected engineering quality risk data, the aim is to systematically identify and extract key elements and find out potential quality risk influencing factors. Specifically, the following steps are included: 1. Clean and normalize the quality report data through text preprocessing, delete duplicate data, remove text noise, and standardize the text, such as unifying industry abbreviations or terminology, and normalizing risk status. For example, different assessors on the front line may have fuzzy deviations in their assessment of the same type of risk status, and business experts are required to normalize the fuzzy risk status. Each quality hazard item has only one status (including: risk retention, minor risk, serious risk, general risk, normal risk), etc.

[0028] 2. Construct a dictionary for the construction engineering field based on a combination of corpus and expert knowledge. Select the dictionary for the construction and civil engineering industry as the basic corpus, such as the synonym dictionary and stop word list of Harbin Institute of Technology. At the same time, business experts simultaneously sort out and expand the domain dictionary and synonym mapping table to reduce the discreteness in the text description to achieve better text segmentation. 3. Through the term-document matrix, the more than 20,000 standardized risk texts in this embodiment are vectorized into a sparse two-dimensional matrix, that is, a set of project risk factors.

[0029] S103: Processing the project risk factor set using an entropy weight method to obtain a weighted word frequency set consisting of weighted word frequencies of each project risk factor in the project risk factor set, and filtering the weighted word frequency set based on a preset threshold to obtain a key risk factor set; Specifically, based on the probability distribution of each project risk factor in the project risk factor set, the information entropy of the project risk factor set is obtained using formula (1): ; The information entropy is combined with the word frequency of project risk factors through formula (2) to obtain the weighted word frequency of project risk factors: ; in, The first in the project risk factor set Project risk factors, , is the number of project risk factors in the project risk factor set; in, For the The average probability of occurrence of project risk factors, It is expressed by formula (3): ; in, For the The project risk factors are The probability of occurrence in a project is obtained by formula (4): ; in, is the first in the set of project risk factors The project risk factors are The number of times it occurred in each project, , is the number of items; For the The word frequency of each project risk factor is expressed by formula (5): .

[0030] Specifically, the information entropy weighted word frequency (TF-H) is calculated. Based on the probability distribution of words, the information entropy of the text is first calculated, and then the information entropy is combined with the traditional word frequency to obtain the information entropy weighted word frequency. The smaller the entropy value, the greater the discreteness of the risk indicator, the greater the amount of information carried, and the greater the weight of the indicator in the risk indicator system.

[0031] Specifically, term frequency (TF) refers to the frequency of a word appearing in a document, and is usually used to measure the importance of a word in a document. It can be obtained by calculating the ratio of the number of times a word appears in a document to the total number of words in the document. Term frequency statistics provide new methods and perspectives for academic research. It is a commonly used weighting technique in intelligence retrieval and text mining to evaluate the degree of repetition of a word for a document or a set of documents in a field in a corpus. The probability of a word appearing in a document usually refers to the TF-IDF value of the word, which comprehensively considers the term frequency (TF) and the inverse document frequency (IDF). Among them, the inverse document frequency (IDF) is a measure that the importance of a word decreases as the frequency of its occurrence in a document increases. It is used to adjust the weights of words that appear frequently in multiple documents. The calculation of IDF is based on the assumption that the importance of a word decreases inversely with the frequency of its occurrence in a document set. The calculation of the TF-IDF value combines term frequency and inverse document frequency, aiming to measure the importance of a word in a document while taking into account the rarity of the word in the entire corpus.

[0032] In this embodiment, the weighted word frequency set is filtered based on a preset threshold to obtain the TOP100 high-frequency feature factors, and redundant information is manually filtered to form knowledge discovery, thereby obtaining a set of key risk factors for the project.

[0033] S104: Process the key risk factor set through subject coding to obtain a standard key risk factor set; Specifically, the extracted risk factors are standardized through expert experience combined with project data and literature data. In this embodiment, taking the above data as an example, 45 key risk indicators are finally formed. For example: 'Wall quality defects': problems exist during the construction process, such as loose masonry, cracks, deformation, etc., resulting in defects in the structural stability, sealing or appearance quality of the wall itself. 'Basement leakage': Water leakage in the basement of a building may be caused by defects in the waterproof layer, too high groundwater level, failure of the basement drainage system, etc.

[0034] 'Roof leakage': Water leakage on the roof of a building may be caused by defects in the roof waterproofing layer, unreasonable structural design, aging of materials, etc. 'Waterproof leakage in toilets and bathrooms': Water leakage in toilets and bathrooms of a building may be caused by defects in the waterproofing layer, loose pipe connections, improper construction technology, etc. 'Leakage in other parts': Other types of leakage accidents that occur during the construction of a building project other than those listed above. These leakage problems may involve different parts of the building, such as bathrooms, toilets, stairwells, and walls. The coding rules of the subject are based on the professional knowledge and experience of experts and are formulated. They can be changed in different embodiments. In this embodiment, 45 key risk indicators are obtained in this way.

[0035] S105: Perform risk quantification and matrix conversion on the standard key risk factor set to obtain a key risk factor occurrence matrix and a key risk factor probability matrix; Specifically, we use manual combing and NPL algorithm to annotate the data of the standard key risk factor set, and use the 45 key risk factors after thematic coding to annotate and map the current 20,000 risk data. At the same time, we generate a word cloud display based on the word frequency statistics of the risk factors, such as Figure 2 shown.

[0036] Specifically, the standard key risk factor set after data annotation is mapped based on the standard key risk factor set and binarized from 0 to 1 according to the standard key risk factor ID, and the occurrence is recorded as "1" and the non-occurrence is recorded as "0". If it occurs multiple times in an inspection, the corresponding number of times "n" is recorded. A sparse matrix with the vertical axis as the time period and the horizontal axis as the risk indicator is formed, such as Figure 3 (Only some examples are shown). Figure 3 The formed sparse matrix is ​​processed to obtain the key risk factor occurrence matrix and the key risk factor probability matrix; Among them, the key risk factor occurrence matrix is ​​expressed by formula (6): ; Among them, the probability matrix of key risk factors is expressed by formula (7): ; in, The first in the standard key risk factor set The key risk factors of the standard are in The probability of occurrence in a project, ; in, , , For the The probability of occurrence of the key risk factors of the standard For the Key risk factors and The probability of the three standard key risk factors occurring simultaneously.

[0037] S106: Use multi-level analysis to classify the standard key risk factor set and build a project risk assessment system; First, the standard key risk factor set is classified by multi-level analysis method to construct a project risk analysis model with a hierarchical structure, such as Figure 4 , is divided into four levels, namely target level, factor level, project level and indicator level; among them, the target level represents the overall risk status of the project; the factor level represents the type of standard key risk factors, which is a first-level indicator; the project level represents the detection items of standard key risk factors, which is a second-level indicator; the indicator level represents the specific sub-indicators of standard key risk factors, which is a third-level indicator.

[0038] S107: Based on the project risk assessment system and historical project risk data, construct a severity weight matrix of key risk factors for the project; Specifically, step S107 includes: (1) Assigning weights to each standard key risk factor in the standard key risk factor set based on project historical data and the Delphi expert method to obtain the initial weight of each standard key risk factor; (2) Using the Cronbach's Alpha model, the initial weight of each standard key risk factor in the standard key risk factor set is checked for consistency, and the key risk factor severity weight matrix of the project is obtained; Among them, the key risk factor severity weight matrix is ​​expressed by formula (8): ; in, ; , For the matrix The maximum value in is the minimum value in the matrix; , For the The key risk factors of the standard are in The probability of occurrence in an item; For the The initial risk weights of the standard key risk factors are obtained by formula (9): ; in, For the The weighted values ​​of the project historical data of the standard key risk factors are For the Expert weightings of standard key risk factors.

[0039] In this embodiment, the Cronbach's Alpha model is expressed by formula (12): ; in, For the The key risk factors of the standard are in The variance of the calculation, It is the total score variance of all standard key risk factors in the standard key risk factor set.

[0040] S108: Using the expert scoring method to perform risk quantification on each standard key risk factor in the standard key risk factor set, and obtaining a key risk factor weight matrix; Based on the Delphi expert method, weights are assigned to the standard key risk factor set to obtain a key risk factor weight matrix; Among them, the key risk factor weight matrix is ​​expressed by formula (10): ; in, For the Project No. The weights of key risk factors are determined based on the criteria.

[0041] In this embodiment, as needed, in the method, according to the results of the chi-square test, the probability values ​​with insignificant correlation are eliminated. When the confidence level of the significant correlation between two indicators is less than 50%, we define it as no significant correlation, and eliminate the probability value (adjust it to 0 to indicate that the appearance of one indicator will not cause the appearance of another indicator) to retain the probability of the occurrence of hidden dangers that may be associated as much as possible to ensure quality and safety. Then the final conditional probability matrix is ​​obtained. (Generally speaking, there are several ways to eliminate confidence levels below 50%, 75%, 80%, 90%, etc. The higher the confidence standard, the more probability values ​​are eliminated, and the risk assessment score will be lower and lower, resulting in a decline in the quality control level).

[0042] S109: Input the key risk factor occurrence matrix, key risk factor probability matrix, key risk factor severity weight matrix and key risk factor weight matrix into the project risk assessment model to obtain a comprehensive project risk score; Among them, the project risk assessment model is expressed by formula (11): ; Specifically, Provide a comprehensive risk score for the project.

[0043] S110: Use the Monte Carlo simulation method to classify the comprehensive risk score of the project and output the project risk assessment results.

[0044] Specifically, step S110 includes: The normal distribution model is used to perform curve fitting on the standard key risk factor set and the key risk factor occurrence matrix to obtain a fitting curve; A preset number of samples are extracted from the fitting curve to perform a model Carlo simulation to divide the project into multiple risk levels and risk level score intervals; Match the comprehensive project risk score with the risk level score range to obtain and output the project risk assessment results.

[0045] In this embodiment, a variety of distribution models can be used, not limited to normal distribution, such as lognormal, Weibull, beta, Pareto, Poisson distribution, etc. The distribution of risk data is first fitted, and the data distribution law and trend are determined by the Anderson-Darling test P value and related fit parameters. The lognormal distribution AD test is 1.1496, Weibull is 1.0268, Beta is 1.5566, and normal distribution is 2.3204. Finally, the lognormal distribution is selected as the most suitable fitting distribution to construct the fitting curve.

[0046] In this embodiment, Monte Carlo simulation is used to randomly extract 5000 samples from the fitting curve for Monte Carlo simulation, and finally the logarithmic normal distribution of WN accident type data is determined, where the mean value of WN2 accident type is 42.99 and the standard deviation is 22.25; the mean value of WN3 accident type is 47.22 and the standard deviation is 26.66; the mean value of WN4 accident type is 45.50 and the standard deviation is 26.76. The predicted value is determined on the logarithmic normal distribution fitting curve determined above, and the distribution diagram of the predicted variable is drawn. Figure 5 Only results for WN4 in the project layer are given.

[0047] Finally, the simulation results were divided into four equal parts to determine the four-level rating of this project, namely "1-prone to defects", "2-needs to be improved", "3-controllable quality", and "4-excellent". The specific rating standards are as follows: Excellent Quality control Room for improvement Defect prone Level meaning There are very few hidden dangers in the project and their consequences are minimal The project occasionally has hidden dangers and the consequences of the accidents are worth noting. Hidden dangers often occur in projects and the consequences of accidents cannot be ignored Hidden dangers frequently occur in the project and the consequences of the accidents are very serious. The present invention has the following beneficial effects: (1) The present invention adopts text mining technology to identify potential risk factors from the project quality risk description, designs a text mining framework, and evaluates the importance of risk factors by calculating the information entropy weighted word frequency TF-H, which can effectively and comprehensively identify risk influencing factors and explore the intrinsic influence of building quality and risk factors; (2) A reasonable and scientific multi-level and multi-dimensional risk evaluation system for the quality status of construction projects is constructed by using the analytic hierarchy process (AHP), which can reasonably and comprehensively reflect the quality risk status of construction projects and provide quantitative standards for the evaluation index system; (3) A new project quality risk evaluation model is constructed, and a project quality risk scoring model is constructed from four dimensions: the hierarchical weight matrix of risk indicators, the severity matrix, the conditional probability matrix, and the indicator occurrence probability matrix. The index system comprehensively considers various risk factors, including but not limited to technical requirements, material quality, process specifications, etc. This indicator system will help project managers to have a more comprehensive understanding of the interrelationships between risks and their potential occurrence possibilities; (4) Construct a project quality risk assessment model, quantify project quality risks, and use Monte Carlo simulation to rate and classify project risks. Through a comprehensive assessment of these factors, the risks of construction projects can be quantitatively scored, guiding project managers to quickly trace and monitor management risks, and to develop targeted risk closure strategies and improvement measures.

[0048] In the above embodiment, a method is provided, and correspondingly, the present application also provides a system. The system provided in the embodiment of the present application can implement the above method, and the system can be implemented by software, hardware, or a combination of software and hardware. For example, the system may include integrated or separate functional modules or units to perform the corresponding steps in the above methods.

[0049] In some implementations of the embodiments of the present application, the system 20 provided in the embodiments of the present application is based on the same inventive concept as the method provided in the aforementioned embodiments of the present application and has the same beneficial effects.

[0050] like Figure 6 As shown, the system 20 may include: Data acquisition module 201, used to acquire project-related information and project historical data, and build a project risk database; A preprocessing module 202 is used to construct a domain dictionary associated with the project based on the project corpus and the expert knowledge base and to perform data cleaning, screening, verification and text segmentation on the project risk database to obtain a project risk factor set consisting of multiple project risk factors; The first data processing module 203 is used to process the project risk factor set by using the entropy weight method, obtain a weighted word frequency set composed of weighted word frequencies of each project risk factor in the project risk factor set, and filter the weighted word frequency set based on a preset threshold to obtain a key risk factor set; A data coding module 204 is used to process the key risk factor set through thematic coding to obtain a standard key risk factor set; The second data processing module 205 is used to perform risk quantification and matrix conversion on the standard key risk factor set to obtain a key risk factor occurrence matrix and a key risk factor probability matrix; The data analysis module 206 uses a multi-level analysis method to classify the standard key risk factor set and construct a project risk assessment system; The first matrix calculation module 207 is used to construct a severity weight matrix of key risk factors of the project based on the project risk assessment system and the historical risk data of the project; The second matrix calculation module 208 is used to perform risk quantification processing on each standard key risk factor in the standard key risk factor set using an expert scoring method to obtain a key risk factor weight matrix; Risk assessment module 209, used to input the key risk factor occurrence matrix, the key risk factor probability matrix, the key risk factor severity weight matrix and the key risk factor weight matrix into the project risk assessment model to obtain a comprehensive project risk score; The result output module 210 is used to classify the comprehensive risk scores of the project using the Monte Carlo simulation method and output the project risk assessment results.

[0051] The IDI full project cycle risk digital assessment system provided in this embodiment is used to execute the IDI full project cycle risk digital assessment method disclosed in the previous embodiment, and has the same technical concept as the previous embodiment, so this embodiment will not be repeated here.

[0052] In this embodiment, the system uses project risk rating and displays it in a visual way. The system also includes the following features: 1. Regional overall project analysis The project's location information is matched with the basic geographic data in the GIS platform, and a regional project quality distribution heat map is generated based on the concentration and level differences of project quality risks.

[0053] 2. Detailed analysis of project-level risk matrix Conduct a layered and detailed analysis of project risks, through project and factor level analysis (such as Figure 7 ), project and indicator level analysis (such as Figure 8 ), intuitively find out which areas have higher risk concentration, and quickly identify which factor layers and their indicators are the main sources of quality risks. At the same time, analyze the indicator scores of specific projects (such as Fig. 9 ), showing that this project has serious problems such as defects in cantilever components and floating foundation slabs. By tracing back and analyzing the severity of risk issues of each indicator, we can quickly trace and monitor management risks and formulate targeted risk closure strategies and improvement measures.

[0054] 3. Dynamic evaluation process traceability analysis Based on the quality status level of each project in each inspection round, a time trend chart is used to display the quality dynamics within each project evaluation cycle, revealing the distribution characteristics and abnormal points of the evaluation data, ensuring the continuity of quality management and the targeted improvement.

[0055] like Fig. 9 , showing the quality performance level results corresponding to each inspection and its changing trend between each inspection. The data of the above embodiment show that this project has been evaluated and inspected 20 times. There are fewer hidden dangers in the early stage of the project, and the quality level of leakage accidents in the mid-term has dropped significantly, indicating that quality hidden dangers related to leakage accidents occur frequently at this stage and the consequences of accidents are relatively serious. Starting from October 2022, the quality level of leakage accidents has dropped significantly and basically remained at the "lowest" level, indicating that quality hidden dangers related to leakage accidents occur very frequently at this stage and the consequences of accidents are very serious. In the following two months, new hidden dangers gradually decreased, and the quality of the project gradually improved. The indicator layers at different stages can be further analyzed to strengthen the prevention and investigation of potential hidden dangers with high risk impact.

[0056] In this embodiment, in building a comprehensive risk and quality management system, a hierarchical and multi-dimensional visual analysis framework is adopted to better carry out refined project management and decision support.

[0057] It should be noted that the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0058] Technicians in the relevant field can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0059] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0060] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0061] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0062] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should all be included in the scope of the claims and specification of the present application.

Claims

1. A digital risk assessment method for the entire IDI project cycle, characterized in that: The steps include: Obtain project-related information and historical data, and build a project risk database; Based on the project corpus and expert knowledge base, a domain dictionary associated with the project is constructed, and data cleaning, screening, verification and text segmentation are performed on the project risk database to obtain a project risk factor set consisting of multiple project risk factors; The project risk factor set is processed by using an entropy weight method to obtain a weighted word frequency set consisting of weighted word frequencies of each project risk factor in the project risk factor set, and the weighted word frequency set is filtered based on a preset threshold to obtain a key risk factor set; Processing the key risk factor set through subject coding to obtain a standard key risk factor set; Performing risk quantification and matrix conversion on the standard key risk factor set to obtain a key risk factor occurrence matrix and a key risk factor probability matrix; Use multi-level analysis to classify the standard key risk factor set and build a project risk assessment system; Based on the project risk assessment system and the project historical data, construct a severity weight matrix of key risk factors of the project; Using an expert scoring method to perform risk quantification processing on each standard key risk factor in the standard key risk factor set to obtain a key risk factor weight matrix; Inputting the key risk factor occurrence matrix, the key risk factor probability matrix, the key risk factor severity weight matrix and the key risk factor weight matrix into a project risk assessment model to obtain a comprehensive project risk score; The Monte Carlo simulation method is used to classify the comprehensive risk score of the project and output the project risk assessment result.

2. According to claim 1, a digital risk assessment method for the entire IDI project cycle is characterized in that: The obtaining of project-related information and project historical data and building of a project risk database includes: Comprehensively collect project-related information and historical data, and based on a detailed understanding of project-related data, analyze, confirm, summarize, and organize project risks, and finally compile them into a project risk database.

3. According to claim 2, a digital risk assessment method for the entire IDI project cycle is characterized in that: The project risk factor set is processed by using the entropy weight method to obtain a weighted word frequency set consisting of the weighted word frequencies of each project risk factor in the project risk factor set, including: Based on the probability distribution of each project risk factor in the project risk factor set, the information entropy of the project risk factor set is obtained using formula (1): ； The information entropy is combined with the word frequency of the project risk factor through formula (2) to obtain the weighted word frequency of the project risk factor: ； in, is the first in the set of project risk factors Project risk factors, , is the number of project risk factors in the project risk factor set; in, For the The average probability of occurrence of project risk factors, It is expressed by formula (3): ； in, For the The project risk factors are The probability of occurrence in a project is obtained by formula (4): ； in, is the first in the set of project risk factors The project risk factors are The number of times it occurred in each project, , is the number of items; For the The word frequency of each project risk factor is expressed by formula (5): 。 4. The IDI full-project cycle risk digital assessment method according to claim 3 is characterized in that: The key risk factor set is processed by subject coding to obtain a standard key risk factor set, including: Based on the professional knowledge and experience of experts, each key risk factor of the key risk factor set is standardized and coded and sorted according to preset rules to obtain the standard key risk factor set.

5. According to claim 4, a digital risk assessment method for the entire IDI project cycle is characterized in that: Perform risk quantification and matrix conversion on the standard key risk factor set to obtain a key risk factor occurrence matrix and a key risk factor probability matrix, including: Manual combing and NPL algorithm are used to annotate the data of the standard key risk factor set; Based on the standard key risk factor set, the standard key risk factor set after data annotation is mapped and binarized according to the standard key risk factor ID to obtain a key risk factor occurrence matrix and a key risk factor probability matrix; The key risk factor occurrence matrix is ​​expressed by formula (6): ； The key risk factor probability matrix is ​​expressed by formula (7): ； in, The first of the set of standard key risk factors The key risk factors of the standard are in The probability of occurrence in an item, ; in, , , For the The probability of occurrence of the key risk factors of the standard For the Key risk factors and The probability of the three standard key risk factors occurring simultaneously.

6. According to claim 5, a digital risk assessment method for the entire IDI project cycle is characterized in that: The multi-level analysis method is used to classify the standard key risk factor set and build a project risk assessment system, including: Firstly, the standard key risk factor set is classified by multi-level analysis method, and a subjective analysis model with hierarchical structure is constructed, which is divided into four levels, namely, target level, factor level, project level and indicator level; Among them, the target layer represents the overall risk status of the project; the factor layer represents the type of standard key risk factors, which is a first-level indicator; the project layer represents the detection items of standard key risk factors, which is a second-level indicator; the indicator layer represents the specific sub-indicators of standard key risk factors, which is a third-level indicator.

7. The IDI full-project cycle risk digital assessment method according to claim 6 is characterized in that: Based on the project risk assessment system and the project historical data, a severity weight matrix of key risk factors of the project is constructed, including: Assigning weights to each standard key risk factor in the standard key risk factor set based on project historical data and the Delphi expert method to obtain an initial weight for each standard key risk factor; The Cronbach's Alpha model is used to perform consistency test on the initial weight of each standard key risk factor in the standard key risk factor set to obtain the key risk factor severity weight matrix of the project; The key risk factor severity weight matrix is ​​expressed by formula (8): ； in, ; , For the matrix The maximum value in is the minimum value in the matrix; , For the The key risk factors of the standard are in The probability of occurrence in an item; For the The initial risk weights of the standard key risk factors are obtained by formula (9): ； in, For the The weighted values ​​of the project historical data of the standard key risk factors are For the Expert weightings of standard key risk factors.

8. The IDI full-project-cycle risk digital assessment method according to claim 7 is characterized in that: The expert scoring method is used to perform risk quantification processing on each standard key risk factor in the standard key risk factor set to obtain a key risk factor weight matrix, including: Based on the Delphi expert method, weights are assigned to the standard key risk factor set to obtain a key risk factor weight matrix; The key risk factor weight matrix is ​​expressed by formula (10): ； in, For the Project No. The weights of key risk factors are determined based on the criteria.

9. The IDI full-project cycle risk digital assessment method according to claim 8 is characterized in that: The project risk assessment model is expressed by formula (11): ； in, Provide a comprehensive risk score for the project.

10. The IDI full-project cycle risk digital assessment method according to claim 9 is characterized in that: The Monte Carlo simulation method is used to classify the comprehensive risk score of the project and output the project risk assessment results, including: Using a normal distribution model to perform curve fitting on the standard key risk factor set and the key risk factor occurrence matrix to obtain a fitting curve; Extracting a preset number of samples from the fitting curve to perform a model Carlo simulation to divide the project into a plurality of risk levels and risk level score intervals; The comprehensive project risk score is matched with the risk level score range to obtain and output the project risk assessment result.

11. An IDI full-project cycle risk digital assessment system, characterized in that: include: Data acquisition module, used to obtain project-related information and project historical data, and build a project risk database; A preprocessing module is used to construct a domain dictionary associated with the project based on the project corpus and expert knowledge base and to perform data cleaning, screening, verification and text segmentation on the project risk database to obtain a project risk factor set consisting of multiple project risk factors; A first data processing module is used to process the project risk factor set by using an entropy weight method to obtain a weighted word frequency set composed of weighted word frequencies of each project risk factor in the project risk factor set and filter the weighted word frequency set based on a preset threshold to obtain a key risk factor set; A data coding module, used for processing the key risk factor set through subject coding to obtain a standard key risk factor set; A second data processing module is used to perform risk quantification and matrix conversion on the standard key risk factor set to obtain a key risk factor occurrence matrix and a key risk factor probability matrix; The data analysis module uses a multi-level analysis method to classify the standard key risk factor set and build a project risk assessment system; A first matrix calculation module is used to construct a severity weight matrix of key risk factors of the project based on the project risk assessment system and the project historical data; The second matrix calculation module is used to perform risk quantification processing on each standard key risk factor in the standard key risk factor set by using an expert scoring method to obtain a key risk factor weight matrix; A risk assessment module, used to input the key risk factor occurrence matrix, the key risk factor probability matrix, the key risk factor severity weight matrix and the key risk factor weight matrix into a project risk assessment model to obtain a comprehensive project risk score; The result output module is used to classify the comprehensive risk score of the project using the Monte Carlo simulation method and output the project risk assessment result.

Images