CN115641004A - House building construction risk grading evaluation method based on complex network - Google Patents

Abstract

The invention discloses a house building construction risk grading evaluation method based on a complex network, which comprises the following steps: 986 accident survey reports are collected and analyzed, and extraction and classification of risk factors are realized by combining relevant building standard specifications and engineering management practice; by introducing a complex network theory, a house building construction risk cause network model is established by applying Gephi9.0.2 software, the distribution condition of the values is obtained by using related topological parameters, and the result of the descending order of the total values is classified by using an ABC classification method. In order to overcome the limitation of the degree value in determining the risk factor or node risk, the node degree value category and the consequence severity degree of the node in an abnormal state are comprehensively considered, and the risk grade of the risk factor is judged. The method improves the credibility and objectivity of the evaluation result and effectively improves the construction safety management level of the house building.

Description

House building construction risk grading evaluation method based on complex network

Technical Field

The invention relates to a construction risk evaluation method, in particular to a house building construction risk grading evaluation method based on a complex network.

Background

The house building construction has the characteristics of fluidity, variable construction process and environment, complex construction technology, heavy physical labor, large personnel mobility and the like, and is an important field of the building construction industry. In recent years, the number of accidents and death in the field of house building and municipal engineering in China has been on a continuous rising trend, as shown in FIG. 1. In the house municipal engineering safety accidents, the occupied proportion of house building construction accidents is large. Building construction accidents are a result of the interrelation of numerous risk factors. Therefore, for the safety problems of the current house building construction safety production situation and exposure, the dual prevention mechanism of the safety risk classification management and control and the hidden danger investigation and treatment is particularly urgent, and the risk evaluation is used as an important means of the safety risk management and control of the forward movement of the gateway, so that the method has an important value for improving the house building construction safety management level.

At present, methods such as a safety check chart, an LEC method and an LS method are generally adopted in the building industry, but the method has the problems of strong subjectivity, strong subjective intention of scoring or evaluating personnel, lack of objectivity, low accuracy of evaluation results and the like. In academic circles, scholars at home and abroad also develop highly effective research work around the aspects of accident causes, risk evaluation and the like, and obtain great results, but when the method is applied to the risk evaluation of actual buildings, the problems of high doorsill, high complexity, poor operability and the like of evaluators exist, and the method cannot adapt to the current situation of the risk evaluation of the construction of buildings. In the case of accidents, it can be seen that many major accidents are not caused by a single adverse factor, but are generated under the action of catalysis and amplification of adverse conditions. Therefore, a new risk evaluation method suitable for house building construction production practice needs to be developed urgently, relevance and occurrence frequency among risk factors are considered comprehensively, severity of consequences caused by the risk factors is combined, house building construction risks are evaluated, accuracy and scientificity of risk evaluation are improved, and important practical significance is achieved for improving building construction safety management level.

Disclosure of Invention

The invention mainly aims to provide a building construction risk grading evaluation method based on a complex network, which considers the relevance and occurrence frequency among risk factors and evaluates the building construction risk by combining the severity of consequences caused by the risk factors so as to improve the accuracy and the scientificity of risk evaluation.

The technical scheme adopted by the invention is as follows: a house building construction risk grading evaluation method based on a complex network comprises the following steps:

based on the accident survey report, combining the national standard and the actual engineering management, the extraction and classification of risk factors are realized;

building a house building construction risk cause network model by introducing a complex network theory and applying Gephi9.0.2 software, obtaining a value distribution condition by using related topological parameters, and classifying the descending order arrangement result of the total value by using an ABC classification method to obtain three categories of a high value, a medium value and a low value;

comprehensively considering the node value category and the consequence severity of the node in an abnormal state, and judging the risk level of the risk factor;

and ranking the risk factors with the same risk value again based on the value of the node risk factor to obtain a more detailed and accurate evaluation result.

Further, the house building construction risk grading evaluation method based on the complex network further comprises the following steps:

the method specifically comprises the steps of analyzing risk factors related to accident occurrence processes, accident causes, occurrence time and places, responsibility units and personnel handling suggestions from different levels of relevant units, operation environments, operators, external supervision and the like according to an accident development process under the context of house construction accidents based on a house construction safety accident investigation report, comprehensively considering production elements such as personnel, equipment, methods, environments, materials and the like related to house construction in combination with national standards, classifying terms with similar semantics into one class, and obtaining a semantic merging table, wherein the terms are classified into safety technology bottoms with poor pertinence, safety technology bottoms missing, safety technology bottoms with incomplete contents and the like, so that the risk factors are extracted and classified.

Furthermore, the method for evaluating the risk classification of the house building construction based on the complex network further comprises the following steps:

the ranking principle of the risk factors is as follows:

the time sequence of the development and evolution process of the building construction accident;

the logic dependence relationship among risk factors in the development process of the building construction accident;

triggering relation among risk factors in the development process of the building construction accident;

a house building construction operation method, an operation flow and a construction program.

Furthermore, the method for evaluating the risk classification of the house building construction based on the complex network further comprises the following steps:

analyzing and identifying risk factors of the whole process of evolution and development of each building construction accident based on an accident investigation report, after coding each risk factor, sequencing the risk factors according to the sequence and logic association relationship according to the idea and principle of the establishment of a risk chain to form a complete accident chain containing human-object-ring-pipe factors, taking management factors in accident evolution as front-end factors of the accident risk chain, taking the accident as the terminal point of the accident risk chain, and forming the complete accident risk chain according to the risk factors 1 → the risk factors 2 → DEG- → the accident;

each risk factor is used as a node of the network, and if a plurality of risk factors occur in the same accident, the risk factors are connected in pairs by a plurality of directed lines;

the network nodes on the risk chain are connected by directed edges representing incidence relations; integrating and superposing the accident risk chains, wherein the sum of the times that the same directed edge appears in different accident risk chains is the weight of the edge;

the in-degree of a node represents a direct risk and the out-degree represents an indirect risk.

Furthermore, the method for evaluating the risk classification of the house building construction based on the complex network further comprises the following steps:

introducing a complex network, establishing a house building construction accident risk network by applying Gephi9.0.2 software, wherein the network consists of a plurality of risk chains, network nodes represent accident risk factors, and connecting edges among the network nodes represent cause-effect relationships among the risk factors, namely the network is formed by connecting a plurality of chain accident risk chains, and the building accidents are explained by using the topological characteristics of the network;

the risk factor judgment level adopts the node value and the severity of the consequences thereof to carry out comprehensive judgment, and the calculation formula is as follows:

in the formula (I), the compound is shown in the specification,Ris a node risk value;Dassigning a node value class;Sthe severity of the consequence caused by the abnormal state of the node;

and performing descending order on all the possible risk results to obtain the overall accumulated value of all the risk results.

According to the accident grade division rule, dividing the serious injury quantity division basis in different grades of the accident as the basis of the risk grade value range division, and acquiring the value range of the risk grade according to the principle of taking the near value for the risk value.

The invention has the advantages that:

the method comprehensively considers the classification assignment of the node value and the severity of the abnormal state of the node, so that the evaluation result is more accurate;

the risk factor extraction system is comprehensive. According to the whole process of accident development, from different levels of relevant units, operation environments, operating personnel, external supervision and the like, the method combines 'classification and codes of dangerous and harmful factors in the production process' (GB/T13861-2009) and 'classification standards of casualty accidents of employees of enterprises' (GB 6441-1986) to realize extraction and classification of risk factors of the house building construction accidents;

when a plurality of risk factors are at the same risk value, the importance degree of the risk factors is different; therefore, the risk factors with the same risk value are sorted again based on the value of the node risk factor to obtain a more detailed and accurate evaluation result;

the method is simple and convenient to calculate, strong in operability, strong in data support, easy to judge by evaluation personnel according to the scoring standard, capable of reflecting the actual construction site and capable of improving the accuracy of the risk evaluation result.

In addition to the objects, features and advantages described above, other objects, features and advantages of the present invention are also provided. The present invention will be described in further detail below with reference to the drawings.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention.

FIG. 1 is a graph of the number of accidents and the number of deaths from municipal works and buildings;

FIG. 2 is a network diagram of the building construction accident risk of the present invention;

FIG. 3 is a hierarchy distribution diagram of risk factors of the present invention;

FIG. 4 is an accident risk node distribution plot for a summary value of greater than 10 of the present invention;

fig. 5 is a frame diagram of the hierarchical evaluation method for the risk of building construction based on the complex network.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Data collection and data processing analysis

Data collection

According to 986 house building construction accident survey reports during 2015-2019, the accident survey reports are summarized on the basis of site survey, data review, accident occurrence related units and personnel interview and accident site reduction by an expert survey group constructed by an emergency management department. The accident investigation report is used as an important data source for production safety accident analysis, and the content comprises the following components: general accident unit, accident occurrence and rescue conditions, accident causes and properties, responsibility unit and personnel handling suggestions, accident prevention measures and suggestions and the like.

Data processing analysis

(1) According to the time scale of accidents, the number of accidents and the number of dead people in the house construction production during 2015-2020 are shown in Table 1. Statistically, these 986 accidents caused 1214 deaths, and on average 1.23 deaths per accident. According to incomplete statistics, the number of house construction accidents and the number of dead people in the year 2015 to 2020 show a rising trend overall, and fall back in the year 2020 as shown in fig. 1. The statistical data show that the safe production situation in the field of house building construction in China is severe.

TABLE 1 2015-2020 Accident number and death number (case)

(2) According to the spatial scale of the accident, namely the geographical position of each province city and autonomous region, china is divided into seven geographical regions of northeast, north China, east China, south China, northwest China and the like, and the construction and production safety accident condition of buildings in each region is shown in table 2. The number of accidents in the areas of east and south China is relatively large, and the number of accidents in the northeast and northwest China is small. In addition, the number of accidents is concentrated in provincial cities or economically developed cities. The method mainly has two reasons, on one hand, the safe production level of the construction industry in different areas is uneven due to the fact that the economic development conditions of all areas are different and the productivity development level of the construction industry is not uniform; on the other hand, the timely degree and the integrity of the accident survey reports issued by the official websites of the provinces and cities are greatly different.

TABLE 2 safety accident situation of building construction and production in each area

(3) According to casualties caused by production safety accidents and direct economic loss classification specified in production safety accident report and investigation and treatment regulations, accidents are classified into general accidents, major accidents and particularly major accidents according to the severity. Through preliminary analysis, 986 accidents in the case library occur, and among the accidents, the general accident 902 is 91.48%, the major accident 79 is 8.01%, the major accident 5 is 0.51%, and no particularly major accident occurs, as shown in table 3.

TABLE 3 percentage of accidents by severity

Data analysis method

Complex network

The complex network is a mode for abstract description of a complex system based on graph theory, and is a network structure formed by a huge number of nodes and complex and intricate relations among the nodes, the topological structure characteristics are emphasized, each unit structure of the complex system is abstracted into the network, each unit is abstracted into the nodes in the complex network, and the relation among the units is abstracted into the edges of the complex network. At present, the method is widely applied to aspects of social relations, transportation, environmental protection, resource utilization and the like. Mutual dependency and cascade effect among causative factors are analyzed through topology, gephi is used for network analysis software, and modeling is conducted on the overall topological structure of the system based on dynamic and hierarchical diagram interactive visualization and detection open source tools of a complex system.

The accident occurrence process is quite complex, and the accident is caused by failure of the multi-stage barriers only when a plurality of causative factors are arranged in a certain sequence and are triggered simultaneously or sequentially. The building construction accident risk network is composed of a plurality of risk chains, network nodes represent accident risk factors, and connecting edges among the network nodes represent cause-and-effect relationships among the risk factors, namely, the network formed by connecting the plurality of chain accident risk chains utilizes the topological characteristics of the network to further explain building accidents. The node degree in the risk chain is defined as the number of risk nodes directly connected with a certain risk node. The degree of the node degree is a first-order neighbor number, represents the centrality of a node in a topological network and the influence degree of the node on adjacent nodes, and can reflect the importance degree of risk nodes in the network to a certain degree. In the directed topology network, the number of edges pointing to other nodes from a node is called the out degree of the node, the number of edges pointing to a node from other nodes is called the in degree of the node, and the total value of the node is the sum of the in degree and the out degree.

ABC classification method

The ABC Classification (Activity Based Classification) is a mathematical analysis method proposed by the italian economist pareto, and its basic principle is to analyze and manage a plurality of factors, sort and classify objects to be managed according to object attributes or frequency, usually into three levels, i.e., a level a, a level B and a level C, which correspond to three levels, i.e., a large level, a medium level and a small level, and the Classification standard is shown in table 4.

TABLE 4 ABC level Classification criteria

The building construction has many working procedures, complex equipment, large personnel flow and long construction period, and the occurrence of accidents can be prevented and controlled by taking targeted measures only after finding the primary and secondary relations of accident risk factors. The ABC classification method has the advantages of simple and convenient calculation, visual and clear data expression and the like. The ABC classification method is applied to risk analysis of house construction accidents, and is mainly used for classifying and arranging indexes to be analyzed according to relevant statistical data to find key factors or weak links, so that risk factors of the building construction accidents can be objectively analyzed, and safe production of house construction projects is promoted.

House building construction accident risk factor extraction

Based on 986 survey reports of construction safety accidents of building construction, under the context of construction accidents of building construction, according to the development process of accidents and the application of system thinking, risk factors related to contents such as accident occurrence processes, accident causes, occurrence time and places, responsibility units, personnel handling suggestions and the like are analyzed from different levels such as related units, operation environments, operators, external supervision and the like, production process danger and harmful factor classification and codes (GB/T13861-2022) and enterprise worker casualty accident classification standards (GB 6441-86) are combined, production elements such as personnel, equipment, methods, environments, materials and the like related to building construction are comprehensively considered, semantically similar terms are classified into one class, semantically merged tables (parts) are obtained, and if the safety technology bottom of hand is not strong, the safety technology bottom of hand is lacked, the safety technology bottom of hand is not completely inconsistent in classification is not unified as the safety technology bottom of hand is not in place, and extraction and classification of the risk factors are realized. For convenience of representation, the types of house construction accidents and risk factors are coded, see table 6.

TABLE 5 semanteme merge table (part)

TABLE 6 list of risk factors and coding table for accident type

Idea and principle for risk chain establishment

The building idea of the building construction accident risk chain is based on risk transmission, the core idea is that risk factors of the risk accident are interacted, and the risk factors are transmitted along with the accident evolution development process to cause the accident. The occurrence, evolution, development of an accident is the result of the interaction of various risk factors along a chain with a certain regularity, called the risk chain. In order to enhance the rationality of building construction accident risk cause chain establishment and comprehensively analyze an accident investigation report, the following 4 accident risk chain establishment principles, namely the risk factor sequencing principle, are provided: (1) the time sequence of the development and evolution process of the building construction accident; (2) the logic dependence relationship among risk factors in the development process of the building construction accident; (3) triggering relation among risk factors in the development process of the building construction accident; (4) a house building construction operation method, an operation flow, a construction program and the like.

Take the eminence accident that falls together that certain building site of china eastern portion took place as an example, according to thinking and the principle that accident risk chain established, then the risk chain of this accident is: the construction unit sends the package not in place, leads to the illegal contract under the condition that the construction unit does not possess corresponding qualifications, and the construction unit does not strictly implement the safety production responsibility system, and the safety education training is not in place, and the construction operation personnel have thin safety protection consciousness, do not correctly post the safety belt and carry out the rendering operation on the side without setting the protection railing, and then lead to the emergence of eminence falling accident.

Construction accident risk network construction of house building

Based on 986 accident survey reports, risk factors in the whole process of evolution and development of each building construction accident are analyzed and identified, after the risk factors are coded, the 986 risk factors are sequenced according to the sequence and the logical association relation according to the concept and the principle of the establishment of the risk chain, a complete accident chain containing the human-object-ring-pipe factors is formed, the management factors in the accident evolution are used as the front end factors of the accident risk chain, the accident is used as the terminal point of the accident risk chain, and the complete accident risk chain is formed according to the risk factors 1 → the risk factors 2 → the accident, as shown in table 7. A risk chain of 5 of accidents is enumerated. Based on this, 986 accident risk chains were established. In order to visualize the accident chain, find out the key risk factors and the key risk chain, represent the sorting mode of the risk factors in the risk chain in the csv file, and draw a house building construction accident risk network model composed of 118 nodes and 478 connecting edges as shown in fig. 2 after introducing gephi0.9.2 software.

TABLE 7 Accident Risk chain summary sheet

Each risk factor in the topological network graph is used as a node of the network, and if multiple risk factors occur in the same accident, the risk factors are connected in pairs by using multiple directed lines. The network nodes on the risk chain are connected by directed edges representing incidence relations; and integrating and superposing the accident risk chains, wherein the sum of the times of the same directed edge appearing in different accident risk chains is the weight of the edge. The in-degree of a node represents a direct risk and the out-degree represents an indirect risk. The association degree of the nodes is represented by the thickness of the lines, and the thicker the directed lines are, the deeper the association degree between the nodes is.

House construction accident risk network analysis

Calculating a topological parameter analysis result of the house building construction accident risk network diagram to display: the network diameter is 6, which means that the maximum value of the shortest distance between any two nodes in the network is 6; the average degree is 4.051, which means that each node is connected with 5 edges on average; the average path length is 2.248, which means that the connection among the network node factors is close, and the accident can be triggered by passing 3 nodes on average. As can be seen from Table 6, the overall process of various accidents runs through the complex risk system formed by the 4 major factors of "human-object-ring-tube" to different extents, the management factor is an indirect factor of the accidents, and the "human-object-ring" factor is a direct factor of the accidents and conforms to the accident cause theory. In addition, research shows that factors close to the accident node mainly originate from operators, and most of factors far from the accident node are management factors and mainly originate from managers. Because of the numerous risk factors for building construction, it is particularly necessary to find out the key risk factors. According to the related research result of the complex network, the network state can be realized by the influence on the internal important nodes, and the control on the whole network is realized by controlling the operation state of a few nodes, wherein the important nodes are called key nodes of the network and are also called key risk factors.

And analyzing each accident risk path in the building construction accident risk network based on comprehensive discussion, finding out node pairs and high-frequency risk chains with strong relevance in the risk network, and representing the node strength by taking the number of edge connecting times among risk factors as the weight of the risk edge connecting. There are 478 connection relationships between nodes inside the network. According to the ABC classification method, the nodes are sorted according to the weight of the connecting edge, and 120 nodes with strong relevance exist, namely the weight of the connecting edge is not less than 2. Because of limited space, the risk chain is formed by a source node, a target node and a connecting edge between the two nodes, and the ABC classification method is used to calculate and select the connecting edge with the weight coefficient larger than 10 as a node pair with stronger relevance, such as 1 → 2, 22 → 72, 22 → 23, 80 → A, 8 → 22, the connecting edge times between the risk factor pairs are larger than 10, and the more the connecting edge times, the more the weight of the path, the more important the node pair is, as shown in Table 8. Further, there are 1 → 2 → 10 → 17 → 21, 1 → 2 → 8 → 21 → 23 → 80, 1 → 2 → 8 → 22 → 79, 1 → 2 → 8 → 21 → 79, 1 → 2 → 8 → 21 → 80, and the like, which find out a high-frequency causative chain affecting the occurrence of an accident. Therefore, these node pairs and high frequency risk chains are the focus of risk management.

TABLE 8 highly correlated node pairs

The evaluation method proposes

A complex network is introduced, gephi9.0.2 software is used for establishing a house building construction accident risk network, the network consists of a plurality of risk chains, network nodes represent accident risk factors, edges among the network nodes represent cause-effect relationships among the risk factors, namely the network is formed by connecting a plurality of chain accident risk chains, and the building accidents are explained by using the topological characteristics of the network. The node degree in the risk chain is defined as the number of risk nodes directly connected with a certain risk node. The degree of the node degree is a first-order neighbor number, represents the centrality of a node in a topological network and the influence degree of the node on adjacent nodes, and can reflect the importance degree of risk nodes in the network to a certain degree. In enterprise safety management, the risk factors with high occurrence probability need to be managed intensively, and the risk factors with low occurrence probability but high severity need to be managed intensively. However, the importance of the risk factors is determined by the application value, and although the relevance and the occurrence frequency of the risk factors are considered, the severity of the consequences in the abnormal state of the nodes is not considered. Therefore, in order to make up for the defect that the degree value determines the risk factor classification and enhance the scientificity of the classification result of the importance of the risk factors, the judgment level of the risk factors adopts the node degree value and the severity of the consequences thereof to carry out comprehensive judgment, and the calculation formula is as follows:

in the formula (I), the compound is shown in the specification,Ris a node risk value;Dassigning values to the node value classes;Sthe severity of the consequences of the abnormal state of the node is obtained.

DThe node value class is assigned, see table 9;Sthe severity of the consequences of a node in an abnormal state is shown in table 10. And performing descending order on all the possible risk results to obtain the overall accumulated value of all the risk results. According to the accident grade division rule of the national production safety accident report and investigation handling regulations, dividing the basis of the number of the serious injuries in different grades of the accident as the basis of the risk grade value range division, obtaining the value range of the risk grade according to the principle of taking the near value for the risk value, and showing the quantitative risk grade judgment in the construction process in a table 9. The risk level distribution of all nodes is calculated by the method in combination with the actual engineering and is shown in the table 10.

TABLE 9 Risk level determination

TABLE 10 Risk factors importance (examples)

Node degree value category analysis

In fig. 2, nodes with significantly larger values are as follows: 21 (the potential hazards are not dealt with in time), nodes 22 (safety education is not trained in place), 79 (the work method, work procedure or operation posture is unsafe), 80 (personal protective articles, appliances are not worn, used or correctly worn), and the like. A large node area indicates that the node is more important. The node gradually spreads outwards from the center of the circle, and the number of the connecting edges of the node is less and the importance degree is smaller along with the reduction of the node value; on the contrary, the more nodes are connected with edges near the periphery of the original circle center, the greater the importance degree is. Because of more risk factors, 27 nodes with the total value greater than 10 are regarded as key risk factors, and the hierarchical distribution of the node values is shown in fig. 3.

The effective safety management means is adopted to guarantee that people, objects and the environment are kept in good states, and among key risk factors, the factor of management errors is up to 51.85 percent. The node state in the network can influence the state of the whole network, and further can influence the accident evolution result. Therefore, the state of the key nodes in the network is adjusted to enable the key node factors (key risk factors) to be in a controllable state, and the occurrence of the construction accidents of the house building can be fundamentally restrained.

The application of the ABC classification method to the risk factor classification management is beneficial to the building construction units to centralize capital and personnel investment into important risk points, so that the classification management is realized, and the safety management efficiency is improved. Based on this, to the difference of node importance degree, adopt the control degree and the safety inspection frequency of different grades, in time eliminate accident hidden danger to optimize traditional accident management and control measure, promote housing construction safety control efficiency. Therefore, according to the ABC classification method, the height values of the 109 node factors are 27 height value factors, 24 middle value factors, and 58 low value factors, and the assignment of the values is based on the risk probability in the reference LEC, and the distribution and assignment of the values are shown in table 11.

TABLE 11 Risk factual value classification for building construction

The node state in the network can influence the state of the whole network, and further can influence the accident evolution result. Therefore, the state of the key nodes in the network is adjusted, so that the key node factors are in a controllable state, and the occurrence of the construction accidents of the house buildings can be fundamentally restrained. Corresponding management measures are taken aiming at risk factors of different levels, the safety management strategy is favorably optimized, a new thought is provided for accident analysis and prevention, and the corresponding relation between main accidents and the risk factors is shown in a table 12.

TABLE 12 correspondence of major accidents to risk factors

Severity analysis of consequences caused by abnormal node state

The abnormal state of a node may cause the Severity of the consequences of the accident (Severity,S) The node abnormal state is the size of casualties or direct property loss caused by accidents possibly caused by abnormal states of the nodes, the personal injury or the property loss caused by the accidents can change within a large range, the possible result of slight injury needing rescue is regulated, and the score is specified as 1, which is taken as a datum point; which would be a possible outcome of many deaths, a score of 15 was assigned as another point of reference. Inserting corresponding intermediate values between 1 and 15: the method comprises four evaluation grades of light injury, general accident, major accident and serious accident. Describing accident grade in the comprehensive accident investigation report of the severity of the consequences caused by the abnormal state of the node and consulting professional technicians to determine and applyThe evaluation criteria of the severity of the consequences of an accident which may be caused by abnormal conditions are shown in Table 13.

Table 13 evaluation standard for severity of possible accident consequence caused by abnormal state of node

Note: in the table, "above" includes the present numbers, and "below" does not include the present numbers.

Multiple risk factors are at the same risk level, but at this point the risk factors are of different importance. Therefore, it is important to rank based on the importance of the node risk factors to further determine the key risk. Under the condition that the risk values are the same, the risk factors are sequenced according to the magnitude of the risk factor values, the key risk factors can be divided again according to the proportion of 10%, 10% and 80% in the ABC classification method, the maximum value is marked as 1, then the key risk factors are sequentially accumulated by taking 1 as a unit, the risk factor serial numbers of the same value are the same, and a more reasonable risk evaluation result is obtained.

Example applications

The applicability and feasibility of the evaluation method are verified by developing the trial application on the construction project site of a certain medical industry base in east China. The project relates to template engineering, scaffold engineering, steel structure installation, temporary electricity utilization, hoisting and the like, the total floor area is about 13183.03 square meters, the total building area is about 77848.83 square meters, the overground building area is 67297.80 square meters, the underground building area is 10551.031 square meters, and the project has construction tasks of various buildings such as 3 production workshops, 1 quality inspection workshop, 1 basement, 1 transformer substation and the like.

Evaluation method comparison and application

A house building construction risk evaluation method based on a complex network is applied to the construction project of the quasi-medical industry base, risk evaluation is carried out aiming at specific risk events existing in the construction project, and the evaluation results obtained by applying a traditional risk evaluation method (LEC method) and the evaluation method of the invention are shown in table 14.

TABLE 14 comparison of evaluation results

Compared with the calculation results of the traditional risk evaluation method, the evaluation on the grades of various risk events is improved to different degrees, and the calculation evaluation results of the risk evaluation method based on the accident cause are identical to the construction safety production condition of the building construction, so that the construction safety management status of the building construction is better met. In the actual project safety management work, it is necessary to strengthen the monitoring and supervising force for major risks, implement standardized operation procedures and improve the risk management and control effect.

The evaluation method disclosed by the invention has the advantages that the data support is relatively firm, the foundation is stronger, the subjective uncertainty of an evaluator is reduced to a certain extent, the risk evaluation result under the multi-factor combination is comprehensively considered, the accuracy of the evaluation result is improved, the evaluation result is more in line with the cognition of the risk in the building construction, and the method has stronger applicability and practicability. In the use of the specific method, the risk evaluation result is closer to the actual engineering project by referring to laws and regulations, department regulations and technical standard specifications of building construction safety production, such as 'project scope of partial projects with larger risk', 'project scope of partial projects with larger risk exceeding a certain scale', 'construction project safety production management regulations' and the like, and distinguishing the risk judgment modes adopted under different conditions.

The method is based on 986 accident survey reports, and extraction and classification of risk factors are realized by combining classification and codes of dangerous and harmful factors in the production process (GB/T13861-2009) and classification standards of casualty accidents of enterprise workers (GB 6441-1986) with engineering management practice; a complex network theory is introduced, gephi9.0.2 software is applied to establish a house building construction risk cause network model, degree value distribution conditions are obtained through relevant topological parameters, an ABC classification method is applied to classify the descending order arrangement results of the total degree values, three categories of high degree values, medium degree values and low degree values are obtained, in order to overcome the limitation that the degree values are used for determining risk factors or node risks, the degree values reflect the node occurrence frequency and the relevance between the nodes and adjacent nodes, but the severity of consequences caused in the abnormal state of the nodes is not considered. In order to enable settlement results to be more accurate, the method comprehensively considers the node value categories and the consequence severity of the nodes in an abnormal state, and judges the risk level of the risk factors; and finally, ranking the risk factors with the same risk value again based on the value of the node risk factor to obtain a more detailed and accurate evaluation result. Example verification shows that the evaluation result of the method accords with evaluation for risk cognition, more accurate risk evaluation for house building construction can be realized, and a frame diagram of the house building construction risk grading evaluation method based on the complex network is shown in FIG. 5.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the present invention, which is intended to cover any modifications, equivalents, improvements, etc. within the spirit and scope of the present invention.

Claims (5)

1. A house building construction risk grading evaluation method based on a complex network is characterized by comprising the following steps:

based on the accident survey report, combining the national standard and the actual engineering management, the extraction and classification of risk factors are realized;

building a house building construction risk cause network model by introducing a complex network theory and applying Gephi9.0.2 software, obtaining a value distribution condition by using related topological parameters, and classifying the descending order arrangement result of the total value by using an ABC classification method to obtain three categories of a high value, a medium value and a low value;

comprehensively considering the node value category and the consequence severity of the node in an abnormal state, and judging the risk level of the risk factor;

and ranking the risk factors with the same risk value again based on the value of the node risk factor to obtain a more detailed and accurate evaluation result.

2. The hierarchical evaluation method for building construction risk based on complex network as claimed in claim 1, characterized in that the hierarchical evaluation method for building construction risk based on complex network further comprises:

the method specifically comprises the steps of based on a building construction accident investigation report, monitoring different levels from related units, operation environments, operating personnel and the outside according to an accident development process under the context of a building construction accident by using system thinking, analyzing risk factors related to the accident occurrence process, the accident reason, the occurrence time and place, responsibility units and personnel processing suggestions, comprehensively considering production elements such as personnel, equipment, methods, environments and materials related to the building construction by combining national standards, classifying terms with similar semantics into one class, and obtaining a semantic merging table, wherein the terms are classified into safety technology with poor pertinence, safety technology with missing and safety technology with incomplete contents, so that the risk factors are extracted and classified.

3. The hierarchical evaluation method for risk of building construction based on complex network as claimed in claim 1, characterized in that the hierarchical evaluation method for risk of building construction based on complex network further comprises:

the ranking principle of the risk factors is as follows:

the time sequence of the development and evolution process of the building construction accident;

the logic dependence relationship among risk factors in the development process of the building construction accident;

triggering relation among risk factors in the development process of the building construction accident;

a house building construction operation method, an operation flow and a construction program.

4. The hierarchical evaluation method for building construction risk based on complex network as claimed in claim 1, characterized in that the hierarchical evaluation method for building construction risk based on complex network further comprises:

analyzing and identifying risk factors of the whole process of evolution and development of the construction accident of each house based on the accident investigation report, coding each risk factor, sequencing the risk factors according to the order and the logical association relationship according to the idea and the principle of establishing a risk chain, forming a complete accident chain containing human-object-ring-pipe factors, taking management factors in accident evolution as front end factors of the accident risk chain, taking an accident as a terminal point of the accident risk chain, and forming the complete accident risk chain according to the risk factors 1 → the risk factors 2 → · · · · · · · · · → the accident;

each risk factor is used as a node of the network, and if a plurality of risk factors occur in the same accident, the risk factors are connected in pairs by a plurality of directed lines;

the network nodes on the risk chain are connected by directed edges representing the incidence relation; integrating and superposing the accident risk chains, wherein the sum of the times that the same directed edge appears in different accident risk chains is the weight of the edge;

the in-degree of a node represents a direct risk and the out-degree represents an indirect risk.

5. The hierarchical evaluation method for building construction risk based on complex network as claimed in claim 1, characterized in that the hierarchical evaluation method for building construction risk based on complex network further comprises:

introducing a complex network, establishing a house building construction accident risk network by applying Gephi9.0.2 software, wherein the network consists of a plurality of risk chains, network nodes represent accident risk factors, and connecting edges among the network nodes represent cause-effect relationships among the risk factors, namely the network is formed by connecting a plurality of chain accident risk chains, and the building accidents are explained by using the topological characteristics of the network;

the risk factor judgment level adopts the node value and the severity of the consequences thereof to carry out comprehensive judgment, and the calculation formula is as follows:

in the formula (I), the compound is shown in the specification,Ris a node risk value;Dassigning values to the node value classes;Sthe severity of the consequences caused in the abnormal state of the node;

performing descending order on all risk results which possibly occur to obtain the overall accumulated values of all risk results;

according to the accident grade division rule, dividing the serious injury quantity division basis in different grades of the accident as the basis of the risk grade value range division, and acquiring the value range of the risk grade according to the principle of taking the near value for the risk value.

Images