Construction risk of highway bridges in plateau and mountainous areas based on safety risk assessment

Abstract

The study designed a risk assessment scheme to reduce the risk of highway bridge construction in highland mountainous areas, and optimised the existing hierarchical analysis method used for risk weight calculation, using entropy weight and fuzzy numbers for improvement, and designed an optimised fuzzy hierarchical entropy weight comprehensive risk assessment model. The results found that the maximum affiliation degree of site safety management risk is 0.39, which is a low-level risk; the maximum affiliation degree of personnel safety and operation quality category is 0.42, which is an intermediate risk; the maximum affiliation degree of machinery and equipment is 0.40, which is a high-level risk; the maximum affiliation degree of construction materials is 0.69, which is a low-level risk; and the maximum affiliation degree of environment category is 0.51, which is an intermediate risk. The maximum affiliation of the overall construction risk is 0.369, which indicates that the fuzzy comprehensive evaluation of the project is an intermediate risk. The results of the study show that the proposed construction risk assessment scheme for highway bridges in highland mountainous areas can provide certain reference for the construction of highland mountainous areas and avoid the corresponding safety risks.

Keywords

Plateau bridge construction risk assessment triangular fuzzy numbers AHP

1. Introduction

With the increase of social capital and economic volume, road traffic safety facilities, as economic infrastructure, are developing towards the direction of high construction difficulty and large engineering quantity. The construction of highway bridges, an important branch of transportation infrastructure, has also received widespread attention from scholars from all walks of life. In the history of world bridges, from 2010 to present, there have been several technological breakthroughs in highway bridge construction, resulting in unprecedented wonders. The world’s first all steel tube concrete frame beam bridge, the world’s tallest bridge with a vertical height of 565 m, and a 55 km long sea crossing bridge that runs through underwater, bridge, and island areas [1, 2]. Throughout the distribution of human bridges, they have extended from plains to underwater, from rivers to seas, but bridge construction in plateau mountainous areas (PMA) still accounts for a very small portion to this day. With the construction of the Global Village, the world has gradually shifted its focus to remote areas such as Southeast Asia, where plateaus are the main landforms and a harmonious coexistence with plateaus has emerged. The forest, mineral, hydropower and tourism resources in plateau and mountainous areas are relatively abundant, with high strategic value and development potential. However, due to the complex and variable terrain in plateau areas and the high difficulty of transportation construction, the development of resources in plateau areas has always been very limited. With the breakthroughs in various technologies in bridge construction, the number of highway and bridge construction projects in plateau areas is also increasing. However, the construction of highway bridges has always been a high-risk project, especially in plateau and mountainous areas with complex environments and frequent extreme climates [3]. Not only may accidents occur during the construction process due to construction reasons, but they are also susceptible to component aging and frequent weather disasters, leading to collapse accidents after completion. Based on this research, we will study the construction risk of highway bridges in plateau areas, optimise the existing hierarchical analysis method used for risk weight calculation, improve the entropy weighting method and fuzzy number, and design a safety risk assessment scheme that is designed to meet the plateau environment and is easy to implement. The research is mainly divided into four parts. The first part introduces the construction of highway bridges in PMA, detailing its strategic value and the need for risk and safety assessment; The second part analyzes the factors of highway bridge construction accidents in PMA and decomposes the construction plan, constructs a SRA based on risk identification of various influencing factors, and optimizes the existing calculation methods for risk weight calculation; The third part takes a project in a plateau mountainous area as an example, and uses the proposed SRA for highway and bridge construction in PMA to conduct risk factor assessment and overall risk fuzzy evaluation calculation; The fourth part summarizes and discusses the above content.

2. Related works

Highway and bridge construction is a supporting facility for basic economic construction, and a large number of scholars have conducted relevant research on it. To reduce the harm of extreme weather on community large-span bridges and avoid traffic safety issues, Reymert et al. [4] conducted research on system literature search, mapping, and data analysis, and proposed a new traffic safety calculation model and simulation program. Experiments have shown that this model could effectively predict the operating environment of vehicles on bridges, and estimate the wind conditions and vehicle aerodynamic loads. Liu et al. [5] examined the effects of road construction and operation on the behaviour of bridge piers and foundation piles, using a three-dimensional finite element (FE) model to deal with the interactions between the piers, the pile foundations, the surrounding soil and heavy truck loads. And the effectiveness of retrofit countermeasures to improve bridge stability using steel plate walls. The results show that steel plate walls are effective in improving the safety of existing bridge piers. Siwowski et al. [6] proposed an innovative hybrid concept of FRP composite-lightweight concrete structural system for highway bridge construction. Through a series of static and dynamic loading tests on a prototype hybrid structure with a span of 21.0 m, it was demonstrated that in terms of static and dynamic performances, the new hybrid bridge system was fully compliant with the design requirements of the relevant standards, but there were FRP strength Safety Hazards. To calculate the multiple existence coefficient of comprehensive road traffic load in modern cities, Dai et al. [7] proposed an extreme value analysis method to predict the extreme load effect and multiple existence coefficient of highway railway bridges. This method was an unsupervised clustering algorithm based on the generalized extreme value Mixture model, which predicted the extreme value by fitting the mixed distribution characteristic data. The experimental results showed that this method was more reliable than the traditional generalized extreme value method. Cosenza and Losanno [8] put forward a safety assessment and performance grade assessment model suitable for reinforced concrete bridges based on the newly proposed Italian Guidelines for Risk Classification, Safety Assessment and Structural health monitoring of Existing Bridges, assessed the risk elimination and safety assessment of existing reinforced concrete bridges, and proved the integrity and effectiveness of the assessment model.

SRA is a common system analysis method that has been extensively studied in various fields. Wang et al. [9] proposed a network SRA based on Bayesian network attack graph model to avoid the harm caused by network attacks on computers. The test results showed that this model could accurately highlight the degree of computer network vulnerability in chaotic relationships. Chen et al. [10] evaluated and warned the security risks of large-scale group activities based on deep learning, calculated the importance of variables and the weight of security risk index by using the random forest algorithm, and carried out optimization experiments in combination with the model parameters of the random forest algorithm. The results showed that the classification accuracy of the risk analysis method was as high as 0.86, and had good prediction ability. Gunes et al. [11] in order to cope with cybersecurity vulnerabilities and threats brought about by the transformation of seaports, proposed a methodology for a comprehensive cyber risk assessment of container ports from a cyber-physical perspective by analysing four typical cyber-attack scenarios. The results show that the risk is assessed as unacceptable for specific cyber threats. Priscilla et al. [12] proposed a dynamic safety assessment method based on Bayesian networks to address the safety risks associated with the processing and production of large quantities of hazardous materials in the processing industry. This method was based on the fact that Bayesian networks could update prior failure probabilities and perform dynamic risk assessment, which had certain application value. In summary, although many previous scholars and scientists have studied the materials and safety of highway bridge construction and designed a large number of improvement plans, there was still a lack of research on the SRA of highway bridge construction in PMA. The development of PMA had high economic benefits, and the SRA of bridge construction in this area had high application value.

3. Design of SRA and optimization of risk algorithm for highway and bridge construction in PMA

This study aims to design a SRA method for highway bridge construction that is suitable for the environment of plateau areas. Firstly, a complete construction SRA is proposed based on the analysis of the causes and types of safety accidents, as well as the problem prone links in engineering construction. Then, the risk assessment decision algorithm used in the SRA is optimized to make it suitable for SRA in PMA.

3.1 Analysis, identification, and evaluation plan design of risk factors for highway and bridge construction in plateau mountainous areas

To complete the analysis of risk factors in highway bridge construction, the attribution of construction accidents and types of safety accidents from 2011 to 2020 are first summarized and analyzed based on existing research and statistical data. The statistical results are shown in Fig. 1.

Figure 1.

Attribution analysis of types of safety incidents.

The attribution analysis results of 207 bridge accidents that occurred from 2011 to 2020 are shown in Fig. 1(a). From the figure, construction reasons are the main cause of bridge accidents, with 139 accidents caused by construction reasons, accounting for 67.15% of all accidents, most of which occurred during the construction phase. The analysis results of the types of bridge accidents are shown in Fig. 1(b), with high-altitude falls accounting for the largest proportion, reaching 52%. This is the main problem in engineering projects that require high-altitude operations. In bridge construction, it generally occurs in stages such as scaffolding installation and bridge deck pavement. The proportion of object strike and collapse accidents has also reached 17% and 15%, with the former mainly occurring in bridge structure construction and foundation pit excavation, while the latter may cause huge economic losses and casualties, and also have a high social impact. Due to construction reasons being the main cause of bridge accidents, this paper analyzes the risk influencing factors during the construction process of highway bridges in PMA. Considering the wide distribution and complex interrelationships of risk factors affecting bridge construction safety, they can be generally divided into four types of influencing factors, namely personnel, construction technology, management, and environment. Their interrelationships are shown in Fig. 2.

Figure 2.

Interrelationship of the four elements.

Figure 3.

Construction risk mapping for highland mountain road bridges.

Personnel, construction technology, management, and environment interact with each other and jointly affect the construction of highway bridges. The personnel element is at the core, and the impact of technology, management, and environment is reflected through people. They are also the main victims of safety accidents, and attention should be paid to ensuring personnel safety throughout the entire construction [13]. People mainly include personnel quality and status, which mainly refer to physical fitness, professional skills, safety awareness, management level of management personnel, and psychological stability; Status refers to the mental state and physical fitness of the person who updates in real-time. The environment can be mainly divided into the external and internal environment. The internal environment is the construction site environment where workers live and produce, including the lighting environment, air humidity, noise, pollution and other factors that may have a direct impact on construction operations and personnel status. The external environment is the natural environment and social environment. The natural environment that should be paid attention to when carrying out highway and bridge construction operations in plateau and mountainous areas mainly includes complex and changeable terrain, harsh air conditions in high-altitude areas, geological disasters due to complex geological conditions, severe weather disasters such as snowstorm, rainstorm, severe cold, and extreme heat, as well as the consequent disasters such as landslides and debris flows. The social environment not only refers to the socio-economic conditions and conventional cultural, legal, moral and other common knowledge conditions that the construction project relies on, but also needs to pay attention to the local cultural customs, dietary habits, customs and beliefs, cultural level, etc. when carrying out construction in highland mountainous areas. The terrain in PMA is complex and frequently changes. The construction of highway bridges in this area requires extremely high levels of construction technology [14, 15, 16]. In order to facilitate the extraction of construction risk according to the hierarchical analysis method, its dimensions are divided as shown in Fig. 3.

Four categories of construction safety risks were classified based on the four risk influencing factors proposed in the study, namely personnel risk, mechanical equipment and construction material risk, management risk and environmental risk.

3.2 Fuzzy hierarchical entropy weight integrated risk assessment method design

In the calculation of risk assessment, it is necessary to determine the weight of each risk factor, usually using the hierarchical analysis method (Analytic Hierarchy Process, AHP) to build a weighted decision-making analysis model [17, 18, 19], the hierarchical analysis method can be used for multi-objective decision-making problems in accordance with the established principles of layer-by-layer decomposition, and then in accordance with the composition of the elements of the internal interrelationships and the degree of subordination to classify the construction of the progressive hierarchical model The hierarchical structure is shown in Fig. 6, which makes the target problem hierarchical and finally obtains the weight ranking.

Figure 4.

AHP model.

Hierarchical analysis model from top to bottom in order for the target layer, criterion layer, element layer, the target layer that the target problem to be solved, the criterion layer for the existence of the structure of each risk problem, the element layer for the risk problem corresponding to the influence factors. Hierarchical analysis method can qualitatively or quantitatively analyze the influencing factors of the target problem, generally through the qualitative indicators of the target problem influencing factors quantitative sorting calculation of the composition and order of each layer, in order to achieve the decision-making problem of multi-scenario optimization. When using the hierarchical analysis method, first consider the interaction of the influencing factors, then the factor analysis is compared to divide the different levels, and finally compare the differences of the elements between and within the levels in order to assess the relative importance of each element. After constructing the target problem want AHP system, the judgment matrix is constructed by comparing the importance two by two

$\displaystyle A=\left[{a_{ij}}\right]_{n\times m}=\left[{{\begin{array}[]{cccc% }{a_{11}}&{a_{12}}&\ldots&{a_{1n}}\\ {a_{21}}&{a{}_{22}}&\ldots&{a_{2n}}\\ \vdots&\vdots&\vdots&\vdots\\ {a_{m1}}&{a_{m2}}&\ldots&{a_{mn}}\\ \end{array}}}\right]$ (1)

As shown in Eq. (1), the $a_{ij}$ denotes the factor $i$ relative to the $j$ importance degree, hierarchical single sorting according to the judgment matrix, the calculation formula is shown in Eq. (2).

$\displaystyle\textit{AW}=\lambda_{\max}W$ (2)

As shown in Figure Eq. (2), the maximum eigenvalue is solved for the matrix $\lambda_{\max}$ , and its corresponding eigenvector $W$ , after which W in order to avoid the existence of self-contradictory items in the experts’ scores, a consistency test is performed, see Eq. (3).

$\displaystyle\textit{CI}=\frac{\lambda_{\max}-n}{n-1}$ (3)

In Eq. (3), the CI denotes the consistency index, in case it is not equal to 0, the consistency ratio needs to be calculated, and according to the average stochastic consistency RI table to solve for $\textit{CR}=\frac{\textit{CI}}{\textit{RI}}$ . The judgment matrix satisfies the consistency requirement only when it satisfies $\textit{CR}<0.1$ when, the judgment matrix satisfies the consistency requirement.

When risk analysis is conducted, because the risk factor hierarchical comparison ranking mainly relies on specialized knowledge and a priori knowledge, the weight assignment is affected by subjective factors, so the objective assignment through the entropy weighting method is carried out in order to carry out the weight assignment with both subjective and objective considerations, and to meet the requirements of scientificity on the basis of retaining the influence of experience. Entropy is initially a thermodynamic function In information theory, entropy is mainly expressed as information entropy, which is used to measure the uncertainty of information. An increase in information entropy represents an increase in the uncertainty or randomness of information, while a decrease represents information becoming more certain or organized. In security risk analysis, the entropy value can be used to assess the randomness and disorder of events, and also reflects the degree of dispersion of the indicator. A high degree of dispersion of an indicator means that it has a higher impact in a comprehensive assessment. Therefore, the entropy weighting method can help to determine the objective weights of indicators. Its calculation steps are as follows, the judgment matrix that meets the consistency requirements is normalized to obtain the standard matrix $r_{ij}=\frac{a_{ij}}{\sum_{i=1}^{n}{a_{ij}}}$ as in Eq. (4).

$\displaystyle R=\left\{{r_{ij}}\right\}=\left({{\begin{array}[]{cccc}{r_{11}}&% {r_{12}}&\ldots&{r_{1n}}\\ {r_{21}}&{r_{22}}&\ldots&{r_{2n}}\\ \vdots&\vdots&\vdots&\vdots\\ {r_{m1}}&{r_{m2}}&\ldots&{r_{mn}}\\ \end{array}}}\right)$ (4)

Equation (4) has that $r_{ij}=\frac{a_{ij}}{\sum_{i=1}^{n}{a_{ij}}}$ , the entropy of each evaluation index can be solved for the criterion matrix, as shown in Eq. (5).

$\displaystyle E_{j}=-\frac{\sum\nolimits_{i=1}^{n}{r_{ij}\ln r_{ij}}}{\ln n}$ (5)

In Eq. (5), the $E_{j}$ denotes the entropy, there is $n$ denotes the matrix dimension, and the deviation degree of the indicator can be solved from the evaluation indicator is entropy value, as shown in Eq. (6).

$\displaystyle d_{j}=1-E_{j}$ (6)

In Eq. (6), the degree of deviation corresponding to the evaluation indicator $j$ is $d_{j}$ . From the degree of deviation, the correction coefficient of the indicator is further calculated, and the calculation formula is shown in Eq. (7).

$\displaystyle\mu_{j}=\frac{d_{j}}{\sum\nolimits_{j=1}^{n}{d_{j}}}$ (7)

By correcting the AHP by the correction coefficients, its initial weights can be solved $W_{j}$ , after which the weight coefficients optimized by entropy weighting method can be solved. The weight coefficient calculation formula is shown in Eq. (8).

$\displaystyle\theta_{j}=\frac{\mu_{j}W_{j}}{\sum\nolimits_{j=1}^{n}{\mu_{j}W_{% j}}}$ (8)

The use of AHP and entropy weighting method can be both subjective and objective benefits of the consideration, the use of its evaluation indicators can be calculated to obtain a combination of indicators weight, there is a comprehensive weight coefficient calculation formula is shown in Eq. (9).

$\displaystyle w_{j}=\rho W_{j}+(1-\rho)\theta_{j}$ (9)

In Eq. (9), the $W_{j}$ denotes the AHP-derived weight value, and $w_{j}$ denotes the integrated weight coefficient, and $\rho$ denotes the constant parameter, 0.5. The AHP method is based on two-by-two comparison scoring by experts, using nine-point scoring, but the level setting lacks rationality and the logical relationship is not clear, so the evaluation indexes in the results obtained by entropy weighting method are all fuzzy, and quantitative scores can not be directly described to the results, so the fuzzy number method is introduced for the analysis of the results [20, 21]. Fuzzy number can be used to analyze the objective existence of fuzzy phenomena, can be subject to a variety of factors interacting with the overall evaluation of the goal, suitable for solving complex factors interacting with multi-level problems. The calculation steps are as follows, first, in order to expert judgment can form a consistent range of evaluation results to take the value of the establishment of the evaluation of the choice set, the evaluation of the set is expressed in the Eq. (10).

$\displaystyle V=\left\{{V_{j}}\right\}=\left\{{v_{1},v_{2},v_{3}\ldots,v_{n}}\right\}$ (10)

In Eq. (10), $V_{j}$ denotes the evaluation grade, and $n$ denotes the number of evaluation levels. From the evaluation set, the value range of the evaluation result can be limited. The expert scores based on the evaluation set and calculates the weights to derive the single-factor affiliation matrix. As shown in Eq. (11)

$\displaystyle R=\left[{{\begin{array}[]{cccc}{r_{11}}&{r_{12}}&\ldots&{r_{1n}}% \\ {r_{21}}&{r{}_{22}}&\ldots&{r_{2n}}\\ \vdots&\vdots&\vdots&\vdots\\ {r_{m1}}&{r_{m2}}&\ldots&{r_{mn}}\\ \end{array}}}\right]$ (11)

In Eq. (11), the $r_{ij}$ denotes the effect of the factor layer’s first $i$ element of the factor layer to the first element of the alternative set $j$ element of the factor layer to the first element of the choice set $v_{j}$ degree of affiliation. By considering the shadow of all factors in the subordinate synthesis, the fuzzy comprehensive evaluation set can be obtained, as in Eq. (12).

$\displaystyle B=W\cdot R=\left[{b_{1}\;\;b2\;\;\ldots\;\;b_{m}\;}\right]$ (12)

In Eq. (12), the $b_{j}$ denotes the fuzzy comprehensive evaluation index, according to the calculated value of the fuzzy comprehensive evaluation index, the solution of the target problem can be determined based on the original maximum affiliation degree. The calculation formula is shown in Eq. (13).

$\displaystyle V=\left\{{v|v_{t}\to\max(b_{j})}\right\}$ (13)

4. Application test of safety risk assessment for highway and bridge construction projects in highland mountainous areas

After completing the design of the safety risk assessment scheme for the construction of highway bridges in highland mountainous areas and optimizing the hierarchical analysis method with the entropy weight method and triangular fuzzy numbers to derive the design of the fuzzy hierarchical entropy weight risk assessment method, in order to validate its application effect in the analysis of the actual risk problems, the risk prediction is carried out with the safety risk assessment scheme proposed by the study in the construction of a highland mountainous highway bridge in Tibet, as an example.

4.1 Risk identification for highway and bridge construction projects in highland mountainous areas

This highway and bridge construction project is located in the Linjiang area of a city in Yunnan, with a maximum altitude of 6740 meters and an average altitude of 2000 meters, with a wide range of altitude changes. The average annual temperature reaches 15 ${{}^{\circ}}$ C, and there are two frost-free seasons throughout the year, the climate is characterized by plateau cold temperate monsoon climate, the average annual precipitation distribution is uneven, and the dry and wet seasons are distinct. The distribution of earthquakes is characterized by high frequency and low intensity, which is conducive to bridge construction. According to the actual local construction process, considering the geological environment factors, and the mutual influence between risk factors, through the expert assessment, data query, analogical analysis and other methods, risk identification and risk analysis to build a highland mountainous highway bridge construction risk factors in line with the AHP structure, the results are shown in Table 1.

Table 1
Risk factors for highway and bridge construction in highland mountainous areas

	Level 1 indicators	Underlying indicators
Risk mapping of highway and bridge construction in highland mountainous areas	Site security management (U1) Personnel Safety and Operational Literacy (U2)	Whether safety inspections are carried out on a regular basis (U11) Due diligence of full-time safety managers (U12) Specification of incoming test methods (U13) Short or inexperienced construction personnel (U21)Educational level of construction personnel (U22) Lack of safety awareness among construction workers (U23) Level of professionalism of construction personnel (U24) Training on security measures (U25) Compliance of safety operations (U26)
	Machinery and equipment (U3)	Mechanical strength compliance (U31) Regularity of maintenance and repair (U32)
	Construction materials (U4)	Material incoming inspection pass rate (U41) Regularity of storage and transportation of materials (U42) Whether material is used as designed (U43) Availability of safety regulations (U44)
	Environment (U5)	Whether the highland environment affects the health of personnel (U51)
		Whether the plateau environment affects the quality of building materials (U52)
		Natural disaster occurrence (U53)
		Reasonableness of environmental layout (U54)
		Completeness of on-site protective measures (U55)

In accordance with the methodology proposed in the study, the evaluation set was established following the design of the questionnaire and the experts were asked to evaluate the evaluation indicators, which had four evaluation levels based on the level of risk, which were low, medium, high and very high, as shown in Table 2.

Table 2

Composition of the rubric set

Risk level	Acceptance criteria	Descriptive
Level 1 risk Very high)	Unacceptable	The risks in this category are high and should be given high priority, and there is a need for enhanced monitoring, targeted risk treatment measures and the commitment of all necessary resources to reduce the risks to at least an undesirable level.
Level 2 risk Advanced)	Not expecting	The risk in this category is high, and monitoring is strengthened and measures are taken to reduce the risk.
Level 3 risk (intermediate)	Acceptable level	Risks in this category are average and do not require treatment measures, and monitoring is maintained
Level 4 risk (low level)	Negligible	The risk in this category is low and no measures are required.

A total of nine evaluators, including on-site construction workers, managers and experts, were allowed to score the importance of each type of risk based on the questionnaire, and the results of the partial judgment for the environmental secondary indicators are shown in Fig. 5.

Figure 5.

Judgement matrix.

As shown in Fig. 5, Fig. 5(a), 5(b) and 5(c) show the results of construction personnel scoring, management personnel scoring and expert scoring, respectively. After obtaining the judgment results for consistency judgment, less than 0.1 is satisfied, judging the group scoring available, the above secondary indicators judgment results are consistent with the consistency conditions. According to this process, all the scoring tables are calculated, and each participant is given the same weight, so that the environmental risk judgment matrix of the whole group can be derived, and ranked according to the comprehensive degree of importance. After that, the standard matrix calculation and indicator vector normalization were carried out, and the results of the standard matrix and normalization of environmental risk are shown in Fig. 9.

Figure 6.

Natural environmental risk likelihood matrix and normalization results.

As shown in Fig. 6, Fig. 6(a) shows the environmental risk judgment matrix, and Fig. 6(b) shows the results of natural environmental risk normalization. From Fig. 6, we can see the comprehensive weight of each indicator of natural disaster environmental risk, in which the lowest weight of environmental layout is 0.069, the highest weight of whether natural disaster occurs is 0.329, which indicates that it has the greatest impact on safety, whether plateau environment affects personnel health, whether plateau environment affects the quality of building materials, and whether on-site protective measures are complete are roughly equivalent, respectively, to 0.247 and 0.210. Among them, whether or not a natural disaster occurs may be related to the special geographic location of the plateau environment at the site. the highest natural disaster may be related to the special geographic location of the plateau environment of the site, due to the construction of engineering and construction of artificial piles of fill as well as the trench is narrower, the slope is larger related to the easy occurrence of mudslides and other disasters. After completing the comprehensive weight calculation of each indicator of natural disaster risk, according to this method, the calculation of subjective weight vector of personnel safety and operation quality risk, site management risk, machinery and equipment risk, material and tool risk is completed, and the results of each part are summarized to complete the total ranking, and the results are shown in Fig. 7.

Figure 7.

Overall ranking of weights.

As shown in Fig. 7, Fig. 7(a) shows the results of the weight of the bottom element within the layer and the weight ordering among all elements, and Fig. 7(b) shows the results of the weight ordering of the middle layer. Among all the index layers, the construction materials have the highest weight of 0.455, while the personnel safety and operation literacy is the lowest of only 0.086, which may be influenced by the special requirements of construction technology for highway bridge construction in highland mountainous areas. Utilizing the entropy weight method for objective weight calculation, some of the objective weight vectors calculation table and arithmetic mean are shown in Table 2.

Table 3

Results of objective weight calculation by entropy weight method

	U1	U2	U3	U4	U5	Entropy (physics)	Coefficient of variation	Entropy weight
U1	1	1/3	1/3	1/5	1/7	0.7989	0.2011	0.3305
U2	3	1	1	1/3	1/6	0.8100	0.1900	0.1171
U3	3	1	1	1/4	1/5	0.8429	0.1571	0.1011
U4	5	3	4	1	1/3	0.8932	0.1068	0.0399
U5	7	6	5	3	1	0.8510	0.1490	04114
Average of objective weights	0.1355	0.2036	0.0614	0.2437	0.3558	/	/	/

As shown in Table 3, the results of the objective weight vector and the arithmetic mean of the construction personnel, from the table can be seen that the entropy weights of the environment and personnel safety and operational literacy are 0.4114, 0.3305, respectively, much higher than the other level of indicators, indicating that the environment and personnel safety and operational literacy in the objective level of importance of the higher degree of impact on safety.

4.2 Fuzzy evaluation test of the risk of highway and bridge construction projects in highland mountainous areas

According to the evaluation set of evaluation indicators and calculate the evaluation set corresponding to each expert, then according to the weight of the indicators to obtain the evaluation set of indicators, and then according to the weight of each expert to obtain the final comprehensive evaluation set of each intermediate layer. The results of all calculations are summarized to produce the affiliation matrix of all risk factors, and the visualization results are shown in Fig. 8.

Figure 8.

Risk factors Affiliation Matrix Visualisation results.

Figure 8 shows the results of the visualization of the affiliation matrix of the risk factors of each underlying element, from Fig. 8, we can see that Defects in the material whether the layout of the environment is reasonable (U54) has the highest level4 affiliation of 0.75, the risk level of the low level, there is no one factor affiliated with level1 very high level of security risk, based on the matrix to calculate the intermediate layer of the matrix of risk factors, the results are shown in Fig. 9.

Figure 9.

Fuzzy integrated judgement.

As shown in Fig. 9, Fig. 9(a) shows the average affiliation matrix of the expert ratings of the layer elements, and Fig. 9(b) shows the average affiliation results of the expert ratings of the final assessment. It can be seen that the affiliation matrix of each intermediate layer risk factors, and in accordance with the formula for each risk layer affiliation calculation, the final results indicate that the maximum affiliation of the site safety management risk is 0.39, for level 4 risk belongs to the low-level risk; personnel safety and operational literacy category maximum affiliation of 0.42, for level 3, belongs to the intermediate risk; machinery and equipment maximum affiliation of 0.40, for level 2, belonging to high-level risk; construction materials maximum affiliation degree is 0.69, for level 4 belongs to low-level risk; environmental category maximum affiliation degree is 0.51, for level 3, belonging to intermediate risk. Based on the above first-level fuzzy evaluation comprehensive judgment matrix to calculate the second-level fuzzy evaluation comprehensive judgment matrix, get the comprehensive judgment vector, because of its overall construction risk maximum degree of affiliation 0.369 $\in v_{3}$ , so the fuzzy comprehensive evaluation of the project for intermediate risk.

5. Conclusion

In order to reduce the incidence of accidents and avoid the risk of highway bridge construction in highland mountainous areas, the study designed a safety risk assessment program for the characteristics of highland mountainous area highway bridge construction, and proposed a fuzzy hierarchical entropy weight comprehensive risk assessment method. The results of risk identification analysis show that the comprehensive weight of each indicator of environmental risk is 0.069 for the lowest environmental layout weight, and 0.329 for the highest weight of whether natural disasters occur, which has the greatest impact on the safety; among all the indicator layers, the weight of construction materials reaches 0.455 at the highest level, and the lowest weight of personnel safety and operation quality is only 0.086. The fuzzy comprehensive evaluation of the construction of highway bridges in highland mountainous areas is carried out, and there are four evaluation sets according to the risk level. According to the risk level setting evaluation set there are four evaluation levels are very high, high level, intermediate, low level. The maximum affiliation degree of site safety management risk is 0.39, which is level 4 risk belonging to low-level risk; the maximum affiliation degree of personnel safety and operation quality category is 0.42, which is level 3, belonging to intermediate risk; the maximum affiliation degree of machinery and equipment is 0.40, which is level 2, belonging to high-level risk; the maximum affiliation degree of construction materials is 0.69, which is level 4 belonging to low-level risk; The maximum affiliation of the environmental category is 0.51, level 3, which is an intermediate risk. The overall construction risk has a maximum affiliation of 0.369, so the final result is that the fuzzy comprehensive evaluation of the project is intermediate risk. Therefore, the FCE of this project was ultimately determined to be a lower risk. The experimental results indicated that the proposed SRA method could provide a certain reference for highway and bridge construction in PMA. However, it still has problems such as computational complexity, high requirements for data timeliness, and limited timeliness of calculation results. This is also an area for further research to improve.

References

Xiong

Guo

. Establishment and repetition survey of primary GNSS control network of Hong Kong-Zhuhai-Macao Bridge. J Surv Eng. 2022; 148(1): 210-216.

Jiang

Chen

. Understand space-time accessibility using a visual metaphor: a case study in Hong Kong-Zhuhai-Macao Belt. Environ Earth Sci. 2022; 81(4): 130-137.

Morgan

. A biocultural examination of home food production and child growth in highland Peru. Am J Hum Biol. 2020; 32(4): e23438.

Reymert

Ronnquist

Oiseth

. Systematic metadata analysis of wind-exposed long-span bridges for road vehicle safety assessments. J Bridge Eng. 2022; 27(2): 04021104.

Liu

Jiang

. Safety analysis for bridge pier under nearby road construction and operation. Measurement. 2020; 15: 107169.

Siwowski

Rajchel

Kulpa

. Design and field evaluation of a hybrid FRP composite - Lightweight concrete road bridge. Compos Struct. 2019; 230: 111504.

Dai

. Investigation of multiple-presence factor for traffic loads on road-rail bridges based on a novel extreme value analysis approach. Struct Saf. 2022; 96: 102199.

Cosenza

Losanno

. Assessment of existing reinforced-concrete bridges under road-traffic loads according to the new Italian guidelines. Struct Concr. 2021; 22(5): 2868-2881.

Wang

Zhu

Shen

Lin

Liu

Zhao

. A network security risk assessment method based on a B_NAG model. Comput Syst Sci Eng. 2021; 38(1): 103-117.

10.

Chen

Zheng

Huang

. Large group activity security risk assessment and risk early warning based on random forest algorithm. Pattern Recognit. Lett, 2021; 144: 1-5.

11.

Gunes

Kayisoglu

Bolat

. Cyber security risk assessment for seaports: A case study of a container port. Comput Secur. 2021; 103(10): 102-196.

12.

Priscilla

Renjith

. Evolution of safety and security risk assessment methodologies towards the use of Bayesian networks in process industries. Process Saf Environ. 2021; 149: 758-775.

13.

Kaneko

Yuzui

. Novel method of dynamic event tree keeping the number of simulations in risk analysis small. Reliab Eng Syst Saf. 2023; 231: 109-129.

14.

Saemi

Komijan

Tavakkoli-Moghaddam

Fallah

. Solving an integrated mathematical model for crew pairing and rostering problems by an ant colony optimisation algorithm. Eur J Ind Eng. 2022; 16(2): 215-240.

15.

Chen

Deng

Niu

. Identification of core risk factors and risk diffusion effect of urban underground engineering in China: A social network analysis. Safety Sci. 2022; 147: 105591.

16.

Beutel

Dullemond

. An improved representative particle monte carlo method for the simulation of particle growth. Astron Astrophys. 2023; 670: A134.

17.

Zhu

Xiang

Luo

. Evaluation of failure risk for prestressed anchor cables based on the AHP-ideal point method: An engineering application. Eng Fail Anal. 2022; 138: 106293.

18.

Eynizadeh

Dehghani

. Selecting elite tomato genotypes through a modified analytical hierarchy process. Crop Pasture Sci. 2020; 71(9): 822-830.

19.

Cebi

Gundogdu

Kahraman

. Consideration of reciprocal judgments through decomposed fuzzy analytical hierarchy process: A case study in the pharmaceutical industry. Appl Soft Compu. 2023; 134: 110000.

20.

Workman

Darch

Dooley

Lomax

Maltby

Pollitt

. Climate policy decision making in contexts of deep uncertainty-from optimisation to robustness. Environ Sci Policy. 2021; 120: 127-137.

21.

Wei

Wang

Zhu

. Fuzzy subspace clustering noisy image segmentation algorithm with adaptive local variance & non-local information and mean membership linking. Eng Appl Artif Intell. 2022; 110: 104672.

Construction risk of highway bridges in plateau and mountainous areas based on safety risk assessment

Abstract

Keywords

1. Introduction

2. Related works

3. Design of SRA and optimization of risk algorithm for highway and bridge construction in PMA

3.1 Analysis, identification, and evaluation plan design of risk factors for highway and bridge construction in plateau mountainous areas

4.1 Risk identification for highway and bridge construction projects in highland mountainous areas

Table 1 Risk factors for highway and bridge construction in highland mountainous areas

References

Table 1
Risk factors for highway and bridge construction in highland mountainous areas