Recommendations for Refined Preventive Maintenance Management of Concrete Bridges in China Based on Environmental Risk Zoning

Abstract

Service environment has an important influence on the durability of concrete bridges. Environmental factors are often considered in the design stage, however, they tend to be ignored in the process of maintenance management. It is of great significance in prolonging the service life of a bridge to pay attention to the influence of the environment on the deterioration of the bridge during the preventive maintenance process. Concrete carbonation, chloride ion corrosion, sulfate corrosion, concrete cracking, concrete abrasion, and freeze–thaw damage are the main causes of decline of bridge durability. Based on these six kinds of distress, indexes of environmental zoning are proposed. For this purpose, variable weight theory is introduced into the matter-element extension model, and the comprehensive environmental risk values of various regions are calculated. The environmental risk zoning map of the concrete bridge’s location is then drawn using data visualization software (Tubiaoxiu). These maps show that the east, south, and southwest of China are the most high-risk areas for the deterioration of concrete bridges, and the validity of the results is proved by the distribution of collected bridge collapse accidents. Through the environmental risk zoning map, managers can intuitively grasp the degree of environmental risk in each region in real time. This approach could help bridge management department to implement differential management of concrete bridges in different regions, adjust bridge maintenance intervals dynamically, predict the state of bridges scientifically, and realize refined preventive maintenance management.

Keywords

infrastructure management and system preservation structures maintenance bridges sustainability and resilience transportation and sustainability transportation and economic development environmental impact

By the end of 2019, the stock of highway bridges in China had exceeded 870,000, of which concrete bridges accounted for more than 90%, and the bridge stock continues to expand ( 1 , 2 ). At the same time, more and more bridges enter the maintenance stage ( 3 ). With the improvement of bridge construction quality in China, collapse events occurring in the construction stage show a downward trend, while the proportion of collapse events occurring in the operation stage has risen to about 60% ( 4 ). At present, the designed service life of Chinese bridges is 100 years, but the service life of most bridges in practice is far less than the designed service life. How to enhance bridge maintenance and management capacity has become a key issue for China in its aim to become a world bridge power ( 5 ). The traditional approach to bridge maintenance is to take remedial measures after bridge damage has developed to a certain extent. Such an approach lacks periodicity, planning, and global awareness, and cannot meet the functional requirements of the bridge. In the 1990s, American scholars put forward the concept of preventive maintenance ( 6 ). Bridge preventive maintenance is a proactive maintenance measure to slow down the occurrence and development of bridge damage, prolong the service life of the bridge, and reduce the maintenance cost in the life cycle of the bridge while maintaining it under the condition of good bridge structure and function ( 7 , 8 ). The preventive maintenance mode has been integrated into the Bridge Management System (BMS) for its high safety and low total investment cost (Figure 1).

Figure 1.

Promotion path of preventive maintenance management.

As shown in Figure 1, the BMS mainly consists of three core parts: database module, simulation analysis module, and decision support module. The evaluation and prediction of the bridge’s technical condition is the core content of BMS, which provides theoretical support for decisions on preventive bridge maintenance ( 9 ). At present, Markov chain models are widely used in BMS prediction models ( 10 – 12 ), which provide a framework for interpreting uncertainty and are compatible with existing bridge evaluation systems. The approach assumes that the structural health state of the bridge is independent and simulates the possibility of the bridge changing from one state to another. In the absence of external intervention, the condition state of the bridge will not be improved, and the bridge will always be in the final state before maintenance ( 13 ). However, the period between different bridge states varies, and the interval time between state transitions is shortened with the deterioration of conditions, which is contrary to the model’s assumption that the state transition probability is constant during the lifetime of the bridge ( 14 ). Therefore, this model cannot accurately reflect the degradation process of an actual bridge ( 15 ). To further describe the inhomogeneity of bridge degradation processes, Manafpour et al. ( 16 ) estimated the transition probability and residence time of concrete bridge deck deterioration based on the semi-Markov time model of accelerated failure time Weibull fitting parameters. Fang and Sun ( 15 ) used Weibull distribution to characterize the degradation behavior of bridges in various states, used semi-Markov process to calculate the transition probability of bridges, and predicted the service life of urban bridges on this basis. Bush et al. ( 13 ) comprehensively applied expert evaluation and data mining to propose a time-based stochastic bridge deterioration prediction model under the condition of limited data based on the degree of bridge use and severity of bridge degradation. These studies suggest that Markov models with duration as a random variable have the potential to simulate more realistic bridge degradation processes ( 14 ). In addition, with the development of science and technology, research on simulating bridge degradation processes based on artificial intelligence technology and realizing intelligent bridge prediction is gradually increasing. Huang ( 17 ) developed a neural network based prediction model for the health state of concrete bridge deck, and identified 11 important factors affecting bridge degradation: bridge age, design load, maintenance history, bridge length, average daily traffic, deck area, deck environment, span number, slope, area, and pre-condition. The model has the ability to accurately predict the condition of the bridge deck and can provide relevant information for maintenance planning and decision-making at the project level and the network level ( 17 ). Although these methods have played an important role in the prediction of the state of bridges, there are still some shortcomings. One is insufficient consideration of environmental factors; environmental factors have a great influence on bridge state transition probability and residence time. Secondly, the above models have a high demand for both quality and quantity of data, typically requiring over 20 years of bridge inspection data. At present, China’s bridge inspection data accumulation is insufficient, so the accuracy of the above prediction model is limited.

Factors leading to bridge degradation can be roughly divided into three categories: design and construction factors, natural aging of the bridge, and environmental factors ( 18 , 19 ). There is some in-depth research on the first two categories, but there is still a lack of research on environmental factors ( 13 – 16 ). Huang ( 17 ) showed that environment and region are important factors for the prediction of bridge condition grade. Especially with the improvement of bridge construction level, and more bridges being constructed across rivers and seas, the impact of environmental factors on bridges needs greater attention ( 20 , 21 ). Tan et al. ( 4 ) concluded that flooding, scouring, earthquake, wind, fire, and other natural factors will damage the bridge structure and accelerate the degradation and failure of the bridge structure. Moomen et al. ( 22 ) incorporated natural environmental factors such as temperature and freeze–thaw cycles into the prediction model. Davis-McDaniel et al. ( 23 ) suggested that the top five critical factors leading to failure of segmental concrete box girder bridges, from most to least critical, are: flood, scour, overloading, corrosion of post-tensioning tendons, and earthquake. Several scholars have developed risk-based approaches that have achieved success in assessing factors affecting the risk of bridge failure, prioritizing appropriate mitigation measures, and thus improving the allocation of scarce monitoring and maintenance resources. Suo et al. ( 24 ) developed a decision support approach for critical infrastructure risk assessment in a complex interdependent scenario using DEMATEL. By using that approach, the overall risk profile is clarified and the risk sources are identified. They confirmed the need to consider complex interdependent scenarios in predicting bridge structural deterioration based on the inherent characteristics of critical infrastructure ( 24 ). Sasidharan et al. ( 25 ) introduced into the decision-making process a conceptual risk information approach to consider the direct and indirect consequences associated with bridge closure or failure caused by scouring, and proposed a risk-based asset management approach to assess the relative effectiveness of different asset management strategies on climate change and budgetary constraints. Karamouz et al. ( 26 ) proposed an optimization program using ideal solution and genetic algorithm to improve the accuracy of infrastructure performance evaluation. Chen et al. ( 27 ) put forward the view of environmental zoning, classified the working environment of bridges based on the characteristics of environmental factors, and applied it to the durability design of bridges. Notwithstanding this body of research, there is no differential regulation on bridge inspection and maintenance under different environmental conditions in the current Chinese bridge maintenance management standard.

To further improve the accuracy of prediction of bridge state, improve bridge maintenance levels, and prolong the service life of bridges, considering the large regional differences in the environment of bridges in China, this paper integrates the idea of environmental zoning into the risk-based bridge management method. Based on the degradation mechanism of environmental forces applied to the durability of concrete bridges, the causes of bridge distress are analyzed. Six kinds of distress environments affecting the durability of concrete bridges are obtained: concrete carbonation , chloride ion corrosion, sulfate corrosion, concrete cracking, concrete abrasion, and freezing–thawing damage. Secondly, environmental risk zoning standards are determined by referring to existing research results and relevant requirements of bridge design specifications. According to the publicly released environmental data, the risk levels of all kinds of environment in each province are determined. At the same time, the risk level zoning maps of various environments are drawn. Finally, the variable weight theory and matter-element extension model are used to calculate the comprehensive environmental risk value. Data visualization technology (Tubiaoxiu) is then used to draw the comprehensive environmental risk zoning map. The map can reflect the environmental risk to concrete bridges in different regions intuitively and concisely, and provides a new idea for bridge management departments to improve the fault detection system in their BMS, improve the prediction accuracy of bridge degradation, and develop a refined preventive maintenance plan.

Variable weight theory is a comprehensive decision-making method that was first proposed by Wang Peizhuang, a well-known scholar in China, in the 1980s. It has been widely applied in many fields and hasachieved remarkable results to the present ( 28 – 30 ). Variable weight theory holds that the weight of factors should be changed according to the change of corresponding factor state index value, so that the weight can better reflect the application of corresponding factor state change in the evaluation and decision system. Li and Wu ( 28 ) applied variable weight theory to mines, evaluating coal floor water inrush, and proved that, compared with constant weight evaluation, variable weight theory can effectively improve accuracy when dealing with mutation of hydrogeological conditions. Wang et al. ( 29 ) proposed a risk assessment approach based variable weights and matter-element extension, and applied it to subway foundation pit engineering. The results showed that the evaluation method can be more widely used in engineering than other evaluation methods ( 29 ). Wang et al. ( 30 ) proposed an improved risk assessment model for water inrush in karst tunnels based on matter-element theory and ideal point method, and verified the accuracy of the assessment results. The matter-element extension model studies the evaluation object, its features, and its magnitude as a whole. This evaluation method eliminates the interference of human factors analysis and evaluation results, improves the approximation of traditional algorithms, and has been widely used ( 31 , 32 ). Shan et al. ( 31 ) applied the fuzzy matter-element extension model to the evaluation of river health, and proved that the method can evaluate river health systematically, comprehensively, and accurately. Zhou et al. ( 32 ) introduced extension theory into the safety appraisal of existing concrete members. The results showed that the matter-element extension model based on comprehensive weight is more accurate and rational. It can be seen that the variable weight theory and the matter-element extension model can effectively improve the accuracy of risk assessment. The environmental factors affecting bridge degradation are constantly changing, and it is beneficial to improve the accuracy of environmental factor risk assessment by using the variable weight theory and the matter-element extension model.

The remainder of this paper is structured as follows. The next section describes the method of environmental risk zoning, which consists of four steps: Step 1 is analysis of the causes of concrete bridge deterioration; Step 2 determines the risk level criteria; Step 3 determines the weight; Step 4 calculates the comprehensive environmental risk value. By this method, the results of environmental risk zoning of concrete bridges in China are obtained. Based on the results of environmental risk zoning, four suggestions on preventive maintenance management of concrete bridges are put forward. Finally, the content and results of this research are summarized, and future works are recommended.

Methods

In this section, the method of environmental risk zoning for concrete bridge operation is expounded. The technical roadmap for this section is shown in Figure 2. Firstly, the causes of concrete bridge deterioration are analyzed. Based on research results in the existing literature, the mechanism of environmental impact on bridge degradation is analyzed, six kinds of bridge environments are summarized, and the influencing factors of accelerating bridge degradation under different environments are clarified. Secondly, the risk level criteria are determined. By referring to the existing relevant research and design standards, the assessment standards of various environmental risk levels are determined, and the indicators are comprehensively considered by using the accumulation method. Public environmental data was collected from the China Statistical Yearbook, the official website of China Meteorological Administration, and other government departments. The risk levels of the environment in each province are determined, and the environmental zoning map of six types of environments drawn. Thirdly, the variable weight theory is used to determine the weight. A total of 156 cases of concrete bridge collapse accidents caused by environmental factors were collected, and the initial weight of each type of environment was calculated by statistical method. Considering the imbalance of various environmental risks, variable weight theory is used to calculate the variable weight of various environments. Finally, the matter-element extension model is used to calculate the comprehensive environmental risk value. At the same time, the calculated results are translated into the concrete bridge environmental zoning map by using the visualization tool, and the environmental zoning results are analyzed.

Figure 2.

Research process of concrete bridge environmental risk zoning.

Step 1: Analysis of Causes of Concrete Bridge Deterioration

The deterioration of bridges includes both chemical and physical actions. Through the literature review, it was found that alkali aggregate reaction, concrete carbonation, sulfate corrosion, reinforcement corrosion, freeze–thaw failure of concrete, concrete cracking, and concrete abrasion are the main causes of the deterioration of concrete bridges ( 33 – 35 ). Among them, corrosion of steel reinforcement is the key to the decline in durability of bridges and damage to concrete structures ( 8 , 36 ). The factors that affect the deterioration of bridges are complex, including internal material, structure, and external environment. This paper only focuses on the influence of external environmental factors on the deterioration of bridges.

Chemical Actions

Alkali Aggregate Reaction

Alkaline oxide in concrete reacts with silicon dioxide in aggregate, resulting in silicate gel attaching to the surface of the aggregate. After absorbing water, the volume expands, eventually leading to concrete cracking. The reaction will continue to occur inside the concrete member until the structure is destroyed. This reaction is inevitable and belongs to the natural aging phenomenon of bridges. Natural factors such as humidity and precipitation will accelerate the occurrence of this reaction ( 27 ).

Concrete Carbonation

Alkaline substances in concrete react with carbon dioxide to produce carbonate and water, which in turn lowers the PH and damages the alkalinity of concrete (Equation 1). Concrete carbonation has little effect on the performance of concrete, but it destroys the passive film formed on the surface of reinforcement under the action of sodium hydroxide and accelerates corrosion of the reinforcement. The first factor affecting concrete carbonation is the contact area between concrete and carbon dioxide and the diffusion rate of carbon dioxide, which is related to the porosity and compactness of the concrete material itself, carbon dioxide volume fraction, and wind. The second is the chemical reaction conditions, which are related to the temperature, humidity, rainfall, and mineral admixtures in the concrete ( 27 ). carbonation is the most serious when the humidity is moderate. In addition, higher temperatures can accelerate the air flow and increase the reaction rate. Mineral admixtures can react with Ca(OH)₂ to reduce concrete basicity and accelerate carbonation.

Ca {(OH)}_{2} + C O_{2} = CaC O_{3} ↓ + H_{2} O

(1)

Sulfate Corrosion

There are many types of sulfate corrosion, such as sulfate crystalline type, ettringite crystalline type, gypsum crystalline type, magnesium sulfate dissolution crystalline type, and carbon sulfur calcium silicate crystalline type. Among these, ettringite crystalline type is the most common ( 37 ). Taking sodium sulfate as an example, sulfate reacts with calcium hydroxide in cement to produce calcium sulfate (Equation 2), and then reacts with solid calcium aluminate hydrate in cement to produce ettringite (Equation 3). At first, ettringite strengthens the concrete, but this effect is temporary—as the acicular crystals of ettringite form, destructive internal stresses are created in the concrete, leading to cracks in the concrete surface. The factors affecting sulfate corrosion include sulfate ion concentration and temperature. Sulfate ions not only come from the internal components of the bridge, but also exist in the external environment such as air, water, and soil, and their concentration can be expressed by PH value ( 22 ).

\begin{matrix} N a_{2} S O_{4} \cdot 10 H_{2} O + Ca {(OH)}_{2} \to CaS O_{4} \cdot 2 H_{2} O \\ + 2 NaOH + 8 H_{2} O \end{matrix}

(2)

\begin{matrix} 3 (CaS O_{4} \cdot 2 H_{2} O) + 4 CaO \cdot A l_{2} O_{3} \cdot 12 H_{2} O \\ + 14 H_{2} O \to 3 CaO \cdot A l_{2} O_{3} \cdot 3 CaS O_{4} \cdot 32 H_{2} O \\ + Ca {(OH)}_{2} \end{matrix}

(3)

Reinforcement Corrosion

Corrosion of reinforcement is a complex process, including chemical corrosion, electrochemical corrosion, and biological corrosion. The corrosion state of reinforcement depends on the surrounding environment. When the passive film on the surface of the reinforcement is destroyed and in an active state, electrochemical corrosion occurs in contact with water and rust forms on the surface of the reinforcement. The volume of rust causes expansion, leading to concrete cracking and falling off (Equations 4 and 5). In addition, chloride ions contact the reinforcement through concrete voids, producing red rust under the combined action of oxygen and water (Equations 6 and 7). The corrosion degree of reinforcement is a progressive process, from the appearance of rust spots on the surface of reinforcement to the formation of red rust (Fe(OH)₃) and black rust (Fe₃O₄). Finally, the steel sectional area becomes smaller and separates from the concrete protective layer, the concrete structure is destroyed, and the mechanical properties of the structure deteriorate. The first factor that accelerates corrosion of steel reinforcement is the contact area between steel reinforcement and corrosion gas, which is related to the compacting property of the concrete, thickness of the concrete protective layer, carbonation degree of the concrete, and chloride ion content. Chloride ion comes from a wide range of sources, such as the concrete material, snowmelt agent, and seawater ( 21 ). The second is the conditions under which chemical reactions take place, which are related to humidity, temperature, precipitation, and wind ( 21 ).

2 Fe + O_{2} + 2 H_{2} O \to 2 F {e_{2}}^{+} + 4 O H^{-} \to 2 Fe {(OH)}_{2}

(4)

4 Fe {(OH)}_{2} + O_{2} + 2 H_{2} O \to 4 Fe {(OH)}_{3}

(5)

F {e_{2}}^{+} + 2 C l^{-} + 4 H_{2} O \to FeC l_{2} \cdot 4 H_{2} O

(6)

\begin{matrix} FeC l_{2} \cdot 4 H_{2} O \to 2 Fe {(OH)}_{2} ↓ + 2 C l^{-} + 2 H^{+} \\ + 2 H_{2} O, 4 Fe {(OH)}_{2} + O_{2} + 2 H_{2} O \to 4 Fe {(OH)}_{3} \end{matrix}

(7)

Physical Actions

Freeze–Thaw Destruction

Freeze–thaw failure is a kind of physical failure which occurs under the conditions of concrete saturation and freezing–thawing cycles. The volume of water expands after freezing, squeezing the concrete structure, increasing the porosity and reducing the mechanical properties of concrete. The factors affecting freeze–thaw damage to concrete include temperature, precipitation, duration of freezing–thawing cycles, humidity, wind and chloride ion concentration ( 22 ).

Concrete Abrasion

Concrete abrasion is mainly caused by water erosion and wind-sand erosion. Long-term abrasion will cause surface damage, cracking, and peeling of concrete bridges, and at the same time, aggravate other distresses ( 18 , 38 ). The factors influencing concrete abrasion include wind speed, wind direction, and velocity of water currents. In particular, excessive sand mining will damage the riverbed, change the direction of water flow, and intensify river scour during bridge service ( 4 , 21 , 23 ).

Concrete Cracking

The forms and causes of cracks are diverse, and their degree of influence on the bridge structure is also different. On the one hand, the structural form and material properties of concrete bridges determine that concrete can work with cracks. On the other hand, external forces on the bridge may also lead to concrete cracks, such as earthquakes, landslides, and other natural disasters, as well as overloaded vehicles and other bad conditions of use ( 21 , 23 ). “Overload” means that the weight of a single freight vehicle exceeds its carrying capacity, and the total load of the vehicle exceeds the bridge bearing limit. Cracking of concrete increases the contact area between concrete and external gas and liquid, which creates conditions for other chemical reactions and accelerates the deterioration of the structure.

According to the above mechanism analysis, alkali aggregate reaction belongs to the natural aging reaction of bridges, which will not be discussed much in this paper. In the aspect of steel corrosion, the corrosion caused by chloride ion is mainly considered. The other environmental factors that affect bridge degradation and their influencing factors are shown in Table 1.

Table 1.

The Main Environmental Factors Affecting Bridge Distresses

	Distress environment	Environmental factors
	Concrete carbonation	Chloride ion corrosion	Sulfate corrosion	Freeze–thaw destruction	Concrete abrasion	Concrete cracking
Humidity	•	•	na	•	na	na
Temperature	•	•	○	•	na	○
Precipitation	○	○	na	•	na	na
Wind speed	○	○	na	○	•	○
Carbon dioxide concentration	•	na	na	na	na	na
Chloride ion concentration	na	•	na	•	na	na
Sulfate ion concentration	na	na	•	na	na	na
Earthquake	na	na	na	na	na	•
Flow rate	na	na	na	na	•	na
Overloading	na	na	na	na	na	•

Note: • = Major factor; ○ = Secondary cause; na = not applicable.

Step 2: Determine the Risk Level Criteria

In different areas of China, the durability of concrete bridges decreases at different rates. Understanding the service environment of bridges in different regions and the risk degree of the environment can effectively improve the durability and prolong the service life of bridges from the aspects of design, construction, maintenance, and management. Through the analysis of the causes of bridge degradation, environmental factors such as temperature, humidity, precipitation, wind speed, carbon dioxide concentration, chloride ion concentration, sulfate ion concentration, natural disaster frequency, overload and so on all have an impact on the degradation rate of concrete structure (4, 21 –23, 27, 39, 40). According to the existing research results and the requirements of relevant specifications, the risk level criteria are determined. The following is a detailed description of the risk level criteria for each type of environment.

Concrete Carbonation

Carbon dioxide concentration, relative humidity, and temperature are the main environmental factors that lead to concrete carbonation. According to the Code for durability design of concrete structures (GB/T 50476-2008), the corrosive carbon dioxide concentration in water (unit: mg/L) is 15–30 in a low-risk environment, 30–60 in a medium-risk environment, and 60–100 in a high-risk environment. However, the difference in the volume fraction of carbon dioxide in various regions is small ( 27 ). Existing studies show that the influence of relative humidity (H, unit: %) on the degree of carbonation of concrete presents a normal distribution: 55 ≤ H < 70 represents a high-risk carbonation environment, 50 ≤ H < 55 and 70 ≤ H < 80 represent a medium-risk carbonation environment, while lower relative humidity levels represent low-risk carbonation environments ( 41 ). In addition, temperature (T, unit: °C) is directly proportional to the carbonation rate ( 42 , 43 ). Comprehensively considering the influence of temperature and humidity on concrete carbonation, these risk levels were accumulated to obtain the risk levels of concrete carbonation environment (as shown in Table 2).

Table 2.

Classification of Concrete carbonation Environmental Risks

Influence factors	Classification criteria for risk levels
Temperature (R_t)	R_t = 1 (T < 10°C)	R_t = 2 (10°C ≤ T < 20°C)	R_t = 3 (20°C ≤ T)
Humidity (R_h1)	R_h1 = 1 (H < 50%; 80% ≤ H)	R_h1 = 2 (50% ≤ H < 55%; 70% ≤ H < 80%)	R_h1 = 3 (55% ≤ H < 70%)
Concrete carbonation environmental risk level (R_ca)	R_ca = R_t + R_h1

Note: R_ca < 3 is a low-risk carbonation environment; 3 ≤ R_ca < 5 is a medium-risk carbonation environment; 5 ≤ R_ca is a high-risk carbonation environment.

Chloride Ion Corrosion

The chloride ion concentration on the surface of concrete structures and the relative humidity and temperature of the environment are the main environmental factors affecting chloride ion corrosion ( 44 ). According to the Design Specification for Durability of Concrete, chloride ion concentration in water (mg/L) of 100–500, 500–5,000, and above 5,000 represents low-risk environment, medium-risk environment, and high-risk environment, respectively. However, there is no statistical data on the mass fraction of chlorine ions. Generally speaking, the mass fraction of chlorine ions is relatively large in coastal areas (within 1.5 km of the coastline) ( 45 ), in areas where de-icing salt is used in winter, and in areas with severe chemical pollution ( 27 ). Increase of ambient temperature will increase the diffusion coefficient of chloride ions ( 46 ). When other conditions are the same, higher relative humidity is related to more serious chloride corrosion ( 47 ). Comprehensively considering the influence of chloride ion concentration, temperature (T, unit: °C), and humidity (H, unit: %) on chloride ion corrosion, the risk levels of environmental chloride ion corrosion were accumulated to obtain the risk levels (as shown in Table 3).

Table 3.

Classification of Chloride Ion Corrosion Environmental Risks

Influence factors	Classification criteria for risk levels
Temperature (R_t)	R_t = 1 (T < 10°C)	R_t = 2 (10°C ≤ T < 20°C)	R_t = 3 (20°C ≤ T)
Humidity (R_h2)	R_h2 = 1 (H < 60%)	R_h2 = 2 (60% ≤ H < 80%)	R_h2 = 3 (80% ≤ H)
Chloride ion concentration (R_cl)	R_cl = 1 (snowmelt agent in winter)	R_cl = 2 (areas with heavy industrial pollution)	R_cl = 3 (coastal areas)
Chloride ion corrosion environmental risk level (R_ch)	R_ch = R_t + R_h2 + R_cl

Note: R_ch < 4 is a low-risk chloride ion corrosion environment; 4 ≤ R_ch < 7 is a medium-risk chloride ion corrosion environment; 7 ≤ R_ch is a high-risk chloride ion corrosion environment.

Sulfate Corrosive Environment

The concentration of sulfate ions in the environment is the main factor leading to sulfate corrosion. According to the Design Specification for Durability of Concrete, the concentration of sulfate ion in water (mg/L) of 200–1,000, 1,000–4,000 and above 4,000 represent low-risk environment, medium-risk environment, and high-risk environment, respectively. There are no relevant statistical data, however. Sulfate ions account for 74% of the total anion content in China’s acid rain, and sulfur dioxide is one of the main components of industrial pollution. Therefore, sulfur dioxide emissions (E_s, unit: 10,000 tons) and precipitation PH value data in industrial pollution statistics were used to replace the indicators of sulfate ion concentration, and their risk levels were accumulated to obtain the environmental risk levels of sulfate corrosion (as shown in Table 4).

Table 4.

Classification of Sulfate Corrosion Environmental Risks

Influence factors	Classification criteria for risk levels
Industrial pollution (R_p)	R_p = 1 (E_s < 30)	R_p = 2 (30 ≤ E_s < 60)	R_p = 3 (60 ≤ E_s)
Precipitation PH value (R_PH)	R_PH = 1 (5.5 < PH ≤ 6.5)	R_PH = 2 (4.5 < PH ≤ 5.5)	R_PH = 3 (PH ≤ 4.5)
Sulfate corrosion environmental risk level (R_s)	R_s = R_p + R_PH

Note: R_s < 3 is a low-risk sulfate corrosion environment; 3 ≤ R_s < 5 is a medium-risk sulfate corrosion environment; 5 ≤ R_s is a high-risk sulfate corrosion environment.

Freeze–Thaw Damage Environment

The coldest average monthly temperature is the main factor leading to damage from freezing and thawing. Factors such as humidity before freezing also affect the degree of freeze–thaw damage. According to the Design Specification for Durability of Concrete, average temperature of the coldest month of −3°C to 2.5°C, −8°C to −3°C, and below −8°C represent low-risk environment, medium-risk environment, and high-risk environment, respectively. Considering the influence of the coldest average monthly temperature (T_c, unit: °C) and the humidity before freezing (H_c, unit: %) on freeze–thaw damage, the risk levels were accumulated to obtain the environmental risk levels of freeze–thaw damage (as shown in Table 5).

Table 5.

Classification of Freeze–Thaw Damage Environmental Risks

Influence factors	Classification criteria for risk levels
The coldest average monthly temperature (T_c)	R_tc = 1 (–3°C < T_c ≤ 2.5°C)	R_tc = 2 (–8°C < T_c ≤ –3°C)	R_tc = 3 (T_c ≤ –8°C)
Humidity before freezing (H_c)	R_hc = 1 (H_c < 60%)	R_hc = 2 (60% ≤ H_c < 80%)	R_hc = 3 (80% ≤ H_c)
Freeze–thaw damage environmental risk level (R_ft)	R_ft = R_tc + R_hc

Note: R_ft < 3 is a low-risk freeze–thaw damage environment; 3 ≤ R_ft < 5 is a medium–risk freeze–thaw damage environment; 5 ≤ R_ft is a high–risk freeze–thaw damage environment.

Concrete Abrasion Environment

Wind erosion and water erosion are the main factors causing concrete abrasion ( 48 , 49 ). Average wind speed (V_w, unit: m/s) and waterlogging frequency (F_r, unit: %) were taken as indicators to measure the degree of wind erosion and water erosion, and their risk levels were accumulated to obtain the environmental risk levels of concrete abrasion (as shown in Table 6).

Table 6.

Classification of Concrete Abrasion Environmental Risks

Influence factors	Classification criteria for risk levels
Average wind speed (V_w)	R_w = 1 (10 < V_w ≤ 14)	R_w = 2 (14 < V_w ≤ 17)	R_w = 3 (17 < V_w)
Waterlogging frequency (F_r)	R_r = 1 (20 < F_r ≤ 50)	R_r = 2 (50 < F_r ≤ 70)	R_r = 3 (70 < F_r)
Concrete abrasion environmental risk level (R_ab)	R_ab = R_w + R_r

Note: R_ab < 3 is a low-risk concrete abrasion environment; 3 ≤ R_ab < 5 is a medium-risk concrete abrasion environment; 5 ≤ R_ab is a high-risk concrete abrasion environment.

Concrete Cracking Environment

Earthquakes and overloading are the main causes of concrete cracking ( 50 , 51 ). The degree of earthquake risk and the degree of overloading risk in each region are divided according to the frequency of catastrophic earthquakes (E, unit: times) and freight volume (G, unit: tons). Taking the influence of earthquake and overloading on concrete cracking into comprehensive consideration, the risk levels were accumulated to obtain the environmental risk levels of concrete cracking (as shown in Table 7).

Table 7.

Classification of Concrete Cracking Environmental Risks

Influence factors	Classification criteria for risk levels
Frequency of catastrophic earthquakes (E)	R_e = 1 (low frequency)	R_e = 2 (medium frequency)	R_e = 3 (high frequency)
Freight volume (G)	R_g = 1 (100,000 < G ≤ 200,000)	R_g = 2 (200,000 < G ≤ 300,000)	R_g = 3 (300,000 < G)
Concrete cracking environmental risk level (R_ck)	R_ck = R_e + R_g

Note: R_ck < 3 is a low-risk concrete cracking environment; 3 ≤ R_ck < 5 is a medium-risk concrete cracking environment; 5 ≤ R_ck is a high-risk concrete cracking environment.

According to the above risk level criteria, the service environmental risks of concrete bridges in all provinces of China need to be further determined.

Step 3: Determine the Weight

The weight value of each evaluation index plays a decisive role in the evaluation result. The traditional matter-element extension model usually uses the constant weight method to determine the weight, and then takes the maximum membership criterion to evaluate the grade. However, the constant weight method often produces deviation in the evaluation result because of the subjective factors existing in the evaluation process. Variable weight theory has been widely used in the field of risk assessment ( 52 – 54 ). When some indicators in the index system seriously deviate from the normal value, the risk level of the top event can be directly affected according to the actual situation ( 55 ). If the constant weight is used, it may lead to the neutralization of the influence degree of the indicators and make the evaluation results seriously inconsistent with the actual situation ( 56 ). To reflect accurately the impact of each index, the variable weight method is adopted to adjust the weight of each index dynamically.

Determine the Initial Weight

Risk assessment requires comprehensive consideration of the probability of risk occurrence and the impact of risk occurrence ( 25 ). Based on the statistical data of bridge collapse accidents, this study roughly determines the probability of the occurrence of risk in each region. A total of 156 cases of concrete bridge collapse accidents caused by environmental factors (including both service conditions and natural environment) in China from 1995 to 2019 were collected. Data sources on major bridge collapses include reports on official government websites and local news, and literature reviews. A distribution map of the collected concrete bridge collapse accidents is drawn (as shown in Figure 3). Black dots represent bridge collapse accidents, and the larger the black dots are, the more frequent the bridge collapse accidents. According to the limited data statistics, the concrete bridge collapse accidents are mainly distributed in the southeast coastal area and the southwest area of China. In addition, according to the 2019 China Statistical Yearbook, the number of bridges in various regions was marked, and it was found that the eastern and southern coastal areas had the larger bridge stocks. In addition to environmental factors, bridge collapse frequency may be related to bridge stock to some extent.

Figure 3.

Distribution of urban bridge and concrete bridge collapse accidents.

The collected bridge collapse information was counted by province to describe the possibility of bridge collapse caused by environmental factors. The risk level data obtained in Step 1 were used to describe the influence degree of environmental factors on bridge collapse. According to Step 1, the risk level data calculated in this paper were scored on a five-point scale, from a risk level of 1, meaning very low risk, to a risk level of 5, meaning very high risk. The probability of occurrence of risk and the influence degree of risk are considered comprehensively, and the initial weight is calculated. The initial weight of each environment is calculated as $W_{j}^{0}$ according to Equation 8.

\begin{matrix} W_{j}^{0} = (\sum_{P = 1}^{P = n} N_{pj} R_{pj}) / T \\ T = \sum_{P = 1}^{P = n} N_{pj} R_{j} \end{matrix}

(8)

where j = 1 refers to the concrete carbonation environment, j = 2 refers to the chloride ion corrosion environment, j = 3 refers to the sulfate corrosion environment, j = 4 refers to the freeze–thaw damage environment, j = 5 refers to the concrete abrasion environment, and j = 6 refers to the concrete cracking environment. P refers to province; R_pj refers to the risk level of each province’s j environment. N_pj refers to the number of collapse accidents caused by j environmental factors in “P” province.

Determine the Variable Weight

The balance function $B (X) = \sum_{j = 1}^{m} x_{j}^{α} (α \in [0, 1])$ is introduced to calculate the variable weight according to Equation 9.

W_{j} (x_{1}, x_{2} \dots x_{n}, W_{1}^{0}, W_{2}^{0} \dots W_{n}^{0}) = \frac{W_{j}^{0} x_{j}^{α - 1}}{\sum_{j = 1}^{n} W_{j}^{0} x_{j}^{α - 1}}

(9)

In Equation 9, α is the balance coefficient; the size of its value represents the requirement for the equilibrium of the index. α ≤ 0.5 means that decision-makers have high requirements for the balance of various environmental risks. α > 0.5 means that decision-makers have low requirements for the balance of various environmental risks. α = 1 is a constant weight model. x_j refers to the risk level of each type of environment.

Step 4: Calculate the Comprehensive Environmental Risk Value

Matter-element extension theory is a research method which can effectively solve the problem of incompatibility and uncertainty of evaluation objects and is widely used in the comprehensive evaluation of various engineering fields. The general idea is to transform all evaluation indexes into quantitative indexes by establishing a multi-index matter-element evaluation model, and then determine the final result by quantitative value.

This study comprehensively considers the impact of various environmental factors on deterioration of concrete bridges. Based on matter-element extension theory, the environmental risks to concrete bridges in various provinces were evaluated comprehensively. The comprehensive analysis of the risk assessment level is influenced by many factors ( 57 , 58 ). As a quantitative tool to describe the transformation process of attributes of things, matter-element extension can transform complex problems into the correlation between matter-element networks, thus making mathematical models closer to reality ( 59 ). In recent years, the theory of matter-element extension has been studied in the safety assessment of infrastructure construction, and the effect is remarkable ( 60 , 61 ). Therefore, this paper tries to use this theory to make the environmental risk assessment of concrete bridges more objective and reasonable.

Classical Domain and Joint Domain

The concrete bridge environment is divided into three levels of risk, 1 = low-risk environment, 2 = medium-risk environment, 3 = high-risk environment. Assume that N_j is the risk level (j = 1,2,3) and C_i is the risk assessment factor (i = 1,2...6), V_j is the classical domain, a_jn and b_jn are the upper and lower limits of the classical domain (Equation 10).

R_{j} = (N_{j}, C_{j}, V_{j}) = [\begin{matrix} N_{j} & c_{1} & (a_{j 1}, b_{j 1}) \\ c_{2} & (a_{j 2}, b_{j 2}) \\ ⋮ & ⋮ \\ c_{n} & (a_{jn}, b_{jn}) \end{matrix}]

(10)

Assume that N_o is the risk assessment criteria, V_p is the joint domain, a_pn and b_pn are the upper and lower limits of the joint domain (Equation 11).

R_{p} = (N_{p}, C_{i}, V_{p}) = [\begin{matrix} N_{o} & c_{1} & (a_{p 1}, b_{p 1}) \\ c_{2} & (a_{p 2}, b_{p 2}) \\ ⋮ & ⋮ \\ c_{n} & (a_{pn}, b_{pn}) \end{matrix}]

(11)

Evaluation of the Matter-Element

P_k is the object to be evaluated, V_k is the value of C_i (Equation 12).

R_{pk} = (P_{k}, C_{i}, V_{k}) = [\begin{matrix} P_{k} & c_{1} & v_{k 1} \\ c_{2} & v_{k 2} \\ ⋮ & ⋮ \\ c_{n} & v_{kn} \end{matrix}]

(12)

Correlation Function

The concept of distance is introduced to reflect the process of matter element changing from quantity to quality (Equations 13 and 14).

\begin{matrix} K_{j} (v_{i}) = {\begin{matrix} \frac{- ρ (v_{i}, V_{ji})}{| V_{ji} |} v_{i} \in V_{ji} \\ \frac{ρ (v_{i}, V_{ji})}{ρ (v_{i}, V_{pi}) - ρ (v_{i}, V_{ji})} v_{i} \notin V_{ji} \end{matrix} \\ | V_{ji} | = (b_{ji} - a_{ji}), (i = 1, 2 \dots n; j = 1, 2 \dots m) \end{matrix}

(13)

{\begin{matrix} ρ (v_{i}, V_{xi}) = a - v_{i} v_{i} \leq \frac{a_{i} + b_{i}}{2} \\ ρ (v_{i}, V_{xi}) = v_{i} - b v_{i} > \frac{a_{i} + b_{i}}{2} \end{matrix}

(14)

where V_xi refers to V_pi or V_ji, and $ρ (v_{i}, V_{pi})$ and $ρ (v_{i}, V_{ji})$ represent the distance between evaluation index and joint domain and classical domain respectively.

Composite Rating

K_j(p_k) is the weighted correlation value, K_j(v_i) is the correlation value, w_j is the weight of each environment (Equations 15 and 16).

K_{j} (p_{k}) = \sum_{i = 1}^{n} w_{j} K_{j} (v_{i})

(15)

\bar{K_{j}} (p_{k}) = \frac{K_{j} (p_{k}) - \min [K_{j} (p_{k})]}{\max [K_{j} (p_{k})] - \min [K_{j} (p_{k})]}

(16)

The characteristic variable j*, was further calculated, and the comprehensive risk level was obtained according to the bias of j* (Equation 17).

j^{*} = \frac{\sum_{j = 1}^{m} j {\bar{K}}_{j} (p_{k})}{\sum_{j = 1}^{m} {\bar{K}}_{j} (p_{k})}

(17)

Results

According to relevant environmental data released by China Statistical Yearbook 2019, the official website of China Meteorological Administration, and China Earthquake Information Network, risk levels of concrete bridges in various provinces were classified. According to the requirements of Step 2, all kinds of environmental risk levels in each province can be calculated, as shown in Table 8. The data in Table 8 are transformed into an environmental risk level zoning map by means of data visualization (Tubiaoxiu), as shown in Figure 4.

Table 8.

Grade of Deteriorating Environment for Concrete Bridges in All Provinces

	Provinces	Environment
	Concrete carbonization environment R_ca	Chloride ion corrosion environment R_ch	Sulfate corrosion environment R_s	Freeze–thaw damage environment R_ft	Concrete abrasion environment R_ab	Concrete cracking environment R_ck
Beijing	2	2	0	2	0	0
Tianjin	3	2	0	2	0	0
Hebei	2	2	2	1	1	2
Shanxi	2	2	2	2	0	2
Neimenggu	1	2	2	3	2	2
Liaoning	2	2	1	3	2	1
Jilin	2	1	1	3	1	0
Heilongjiang	2	2	1	3	1	0
Shanghai	2	3	1	0	1	1
Jiangsu	2	3	2	0	1	1
Zhejiang	2	3	2	0	1	1
Anhui	2	0	2	0	1	2
Fujian	3	3	2	0	1	1
Jiangxi	2	0	2	0	1	1
Shandong	3	0	2	1	1	2
Henan	3	0	1	1	0	1
Hubei	2	0	2	0	0	1
Hunan	2	0	2	0	0	1
Guangdong	2	3	2	0	2	2
Guangxi	3	3	2	0	1	1
Hainan	2	3	0	0	2	0
Chongqing	2	0	2	0	0	1
Sichuan	2	0	3	0	2	2
Guizhou	2	3	3	0	0	1
Yunan	2	0	2	0	0	2
Xizang	1	0	0	0	1	1
Shanxi	3	0	1	1	0	1
Gansu	2	0	1	3	1	1
Qinghai	2	1	0	2	2	1
Ningxia	2	2	1	2	0	0
Xinjiang	2	1	1	2	2	2

Note: The high-risk area value is 3 (dark gray cells), the medium-risk area value is 2 (gray cells), and the low-risk area value is 1.

Figure 4.

Grade distribution (low, medium and high risk) of environmental factors causing deterioration of concrete bridges: (a) concrete carbonation, (b) chloride ion penetration, (c) sulfate corrosion, (d) freeze–thaw damage, (e) concrete abrasion, (f) concrete cracking.

As can be seen from Figure 4, the environmental risk level zoning of different provinces varies greatly. Concrete carbonation is widespread; most areas in China are concrete carbonation environments, and the risk level is mostly intermediate. The areas with severe chloride ion corrosion are mainly concentrated in the southeast coastal areas which have high humidity and high chloride ion concentration. The areas with severe sulfur ion corrosion are mainly in the south, where industry is most developed and acid rain is relatively serious. Freezing–thawing damage is mainly concentrated in the cold northern regions (especially the northeast region). The high-risk areas for concrete abrasion environment are mainly the northwest region, which has both rich wind energy resources and high sand content ( 48 ), and the southwest region with frequent mountain torrents. Concrete cracking environments are also widespread; Yunnan, Xinjiang, Sichuan, and other places with high frequency of catastrophic earthquakes, as well as Beijing-Tianjin-Hebei and the surrounding areas with large truck flow, are high-risk areas ( 50 ).

Following the content of Step 3, the weight of each type of environment is calculated. Because of the lack of environmental zoning data from Taiwan, the three bridge collapses collected from Taiwan were excluded, and a total of 153 collapses were analyzed. Although the data statistics have some limitations, they can reflect the effect of each environmental factor on the deterioration of the bridge to some extent. The calculation results of initial weight and variable weight are shown in Table 9.

Table 9.

Weight of the Various Environmental Factors

	Environmental factor	Weight
	Average value	Initial weight	Fixed weight	Variable weight
	Average value	Initial weight	Fixed weight	α = 0.2	α = 0.5	α = 0.8
Concrete carbonation environment R_ca	2.13	0.27	0.27	0.18	0.21	0.25
Chloride ion corrosion environment R_ch	1.39	0.16	0.16	0.15	0.16	0.16
Sulfate corrosion environment R_s	1.45	0.22	0.22	0.20	0.21	0.22
Freeze–thaw damage environment R_ft	1.00	0.07	0.07	0.09	0.08	0.07
Concrete abrasion environment R_ab	0.87	0.13	0.13	0.18	0.16	0.14
Concrete cracking environment R_ck	1.10	0.16	0.16	0.19	0.18	0.17

Note: Different α values represent different risk preferences: α < 0.5 indicates that it pays more attention to the balance of various environmental risk levels, α > 0.5 represents a certain degree of tolerance for various environmental risks.

The results of risk values in each region, calculated based on matter-element extension theory, are shown in Table 10. Using the data in Table 10, the environmental zoning map of concrete bridges in China was drawn (as shown in Figure 5).

Table 10.

Comprehensive Environmental Risk Value for All Provinces

Province	Fixed weight risk value	Variable weight risk value			Province	Fixed weight risk value	Variable weight risk value
Province	Fixed weight risk value	α = 0.2	α = 0.5	α = 0.8	Province	Fixed weight risk value	α = 0.2	α = 0.5	α = 0.8
Beijing	2.05	2.01	2.02	2.04	Hubei	1.61	1.56	1.56	1.58
Tianjin	2.17	2.10	2.12	2.15	Hunan	1.61	1.56	1.56	1.58
Hebei	2.43	2.39	2.41	2.42	Guangdong	2.60	2.60	2.60	2.60
Shanxi	2.59	2.50	2.62	2.60	Guangxi	2.89	2.92	2.90	2.89
Neimenggu	2.45	2.50	2.48	2.46	Hainan	2.12	2.07	2.09	2.10
Liaoning	2.36	2.37	2.37	2.36	Chongqing	1.61	1.56	1.56	1.58
Jilin	1.33	1.07	1.08	1.23	Sichuan	2.84	2.87	2.87	2.86
Heilongjiang	2.67	1.35	2.90	2.73	Guizhou	2.74	2.43	2.53	2.65
Shanghai	1.76	1.54	1.56	1.65	Yunnan	2.58	2.14	2.24	2.41
Jiangsu	2.52	2.42	2.48	2.51	Xizang	1.28	1.28	1.28	1.28
Zhejiang	2.52	2.42	2.48	2.51	Shanxi	1.75	1.55	1.61	1.69
Anhui	2.24	2.50	1.82	2.12	Gansu	1.47	1.43	1.44	1.46
Fujian	2.89	2.92	2.90	2.89	Qinghai	1.52	1.54	1.53	1.52
Jiangxi	1.53	1.44	1.46	1.50	Ningxia	2.64	2.26	2.35	2.49
Shandong	2.87	2.69	2.78	2.84	Xinjiang	2.28	2.31	2.29	2.29
Henan	1.75	1.55	1.61	1.69

Note: The higher the value, the greater the risk of concrete bridge damage.

Figure 5.

Environmental zoning of concrete bridges: (a) fixed weight value (α = 1), (b) variable weight value (α = 0.2), (c) variable weight value (α = 0.5), (d) variable weight value (α = 0.8).

The comprehensive environmental risk zoning map of concrete bridges (as shown in Figure 5) can intuitively show the risk level of bridges in various regions. First, it can be observed that east and south coastal areas, and southwest and north areas are the main areas with high environmental risk of concrete bridge damage in China. Secondly, the risk value of the environmental zoning map is consistent with the collected distribution of bridge collapse accidents, so the zoning map has certain credibility. Thirdly, comparing the zoning map with the different weights, the variation of most regions has changed little, while the variation of Yunnan, Heilongjiang, Anhui, and a few other provinces have changed by a large degree. This is because some indicators of risk in these regions are uneven.

Discussion: Improvement of Preventive Maintenance Based on Environmental Zoning

Based on the research results of this study, the importance and feasibility of environmental risk zoning for concrete bridges are discussed in this section.

Necessity of Environmental Risk Analysis

Through the theoretical analysis of the degradation of concrete bridges, it is not difficult to find that environmental factors can cause fatal damage to concrete bridges. Environmental factors not only reduce the durability of concrete and steel, but also accelerate the degradation process of bridges. The influence of environmental factors on bridge structure is complex, but the existing research on environmental factors is not sufficient ( 57 , 58 , 62 ). Environmental factors such as flood, scour, earthquake, and wind, as well as temperature and freeze–thaw cycles, are usually considered in bridge degradation analysis (4, 21 –23). However, further degradation mechanisms are rarely mentioned. In this paper, the reasons for the degradation of bridges caused by environmental factors are analyzed from the micro level, and the environments that are easily observed to cause bridge damage are divided into six types. The factors influencing the various environments are further analyzed, as shown in Table 1. This lays a theoretical foundation for determining the environmental risk assessment index. According to the results of environmental zoning (Figure 5), environmental risk levels in different regions of China vary greatly. Therefore, it is necessary to carry out environmental zoning to realize the differential management of concrete bridges and improve the refinement of bridge management.

The Role of Environmental Risk Zoning

On the one hand, environmental risk zoning can guide relevant departments to improve the bridge environment in a targeted manner, and on the other hand, it can guide relevant departments to formulate differentiated bridge maintenance standards. According to the results of environmental risk zoning (Table 8), the risk levels of various environments in different regions can be obtained. By conducting an in-depth study on the high-risk environment and analyzing the causes of high risk, relevant departments can take corresponding measures to intervene for non-force majeure reasons. This approach can fundamentally improve the environment of concrete bridges in the region. For example, it can be seen from Table 8 that in Hebei province “concrete cracking environment” is a medium-level risk. After in-depth investigation, it may be found that the main cause of concrete cracking in Hebei province is long-term overloading of vehicles. Relevant departments can take corresponding measures to avoid the occurrence of vehicle overloading to reduce the risk level to bridges. In addition, specific maintenance standards could be developed for the high-risk areas identified in Table 8. The durability of concrete bridge structure varies greatly in different environments. The same bridge inspection intervals will not only cause the waste of maintenance resources, but also increase the potential safety hazard of the bridge.

Recommendations on Preventive Maintenance Refinement Management

Based on the perspective of environmental risk, some suggestions are put forward to improve the BMS (Figure 6). These measures could help to ensure the safety of the bridge structure, to prolong the service life of the bridge, and to promote the reasonable arrangement of maintenance resources.

Bridge inspection should be managed differently according to the environmental risk values of different regions. According to the requirements of the Specification for Maintenance of Highway Bridges and Culverts (JTG 5120-2021), the interval of bridge inspection cannot be more than three years, but within that interval, bridge management departments in each region may adjust inspection intervals according to local conditions. At present, the bridge inspection interval is only set according to the bridge’s technical condition. It can be seen from Figure 5 that environmental risks vary greatly across the country, and it is unscientific to apply the same inspection interval in different risk areas. It is suggested to realize the differential management of inspection interval by referring to the environmental risk value in Figure 5 while considering the technical condition of the bridge. Scientifically set inspection intervals can ensure that effective protective measures are identified before the bridge reaches a dangerous condition and thus extend the bridge’s life.

At present, the data collection mainly focuses on the technical condition of the bridge, ignoring environmental data. According to the analysis results in the second section of this paper, the environment has an important influence on the deterioration of bridges, to which more attention should be paid. The environmental database should contain information about temperature, humidity, precipitation, wind speed, carbon concentration, ion concentration, sulfate ion concentration, natural disaster frequency, freight traffic, and other factors. The database is the foundation of the BMS and the pre-condition for other modules to run. Therefore, attention should be paid to environmental factors in data collection.

Environmental factors should be added into the bridge state prediction model to improve the prediction accuracy. The core of the simulation analysis module is to construct the bridge degradation model, which can predict the bridge service state by reading the database data, and provide the basis for bridge maintenance decisions. The question of how to improve the accuracy of the model has attracted the attention of scholars. The existing prediction models pay little attention to environmental factors ( 13 – 16 ). For example, Moomen et al. ( 22 ) only incorporated temperature and freeze–thaw cycles into the bridge prediction model. Compared with the environmental factors analyzed in this paper, this is far from enough. After adding environmental data collection into the database module, environmental factors can be taken into account in the prediction, to make the simulation data closer to reality and improve the accuracy of prediction of the bridge state.

Targeted preventive measures should be taken in accordance with the specific environmental conditions of each locality. As can be seen from Figure 4, the service environment of bridges is different in each region. For example, in northeast China, freezing–thawing damage and chloride ion corrosion caused by the use of snowmelt agent are the main environmental factors that degrade the durability of concrete bridges. However, in southwest China, frequent earthquakes and severe sulfur ion corrosion caused by acid rain are the main reasons for the deterioration of concrete bridges. All kinds of environmental zoning maps can quickly and intuitively show the environmental conditions of a specific area. And taking timely, targeted preventive measures can slow down the deterioration of bridges.

Figure 6.

Recommendations for refinement of preventive maintenance management based on environmental risk perspective.

Conclusions

Many concrete bridges in China are entering the stage of maintenance. The increasing demand for bridge maintenance conflicts with the incomplete bridge maintenance level. How to use modern technical means to improve the efficiency of bridge maintenance and prolong the service life of bridges is a major problem facing bridge management departments. BMS has been widely used in China, but it still faces the problem of inaccuracy of bridge condition prediction. This paper puts forward the idea of environmental risk zoning from the perspective of risk management. At the same time, a method for implementation of environmental risk zoning is proposed and proved to be feasible. According to the calculation results, some suggestions for improving the BMS are put forward.

Environmental factors have a great impact on the deterioration of bridges ( 57 ). Most of the existing studies focus on specific regions and environments, without comprehensive considerations ( 58 , 62 ). In this paper, on the basis of the analysis of the causes of the deterioration of concrete bridges, six types of main environments are determined, and the main impact indexes and risk grade classification standards of all kinds of environments are determined based on the relevant codes. Secondly, the risk levels of all kinds of environments in each province are classified according to the established risk classification standards by using public data such as the China statistical yearbooks, and the risk distribution diagrams of all kinds of environments are drawn. Thirdly, bridge collapse accidents caused by environmental factors are collected, and initial weight calculated based on the data. The variable weight theory is introduced, the matter-element extension model is constructed, the comprehensive risk values of different regions under different weights are calculated, and the comprehensive environmental risk zoning map is drawn. Finally, based on the above analysis, the policy recommendations for sustainable bridge management are put forward. The environmental risk zoning map can intuitively show the environmental risk situation of different regions. For the bridge management department, bridge inspection intervals can be planned according to regional differences to carry out targeted prevention and to take account of the high-risk environmental factors of each province. This approach can improve the effectiveness of preventive maintenance and extend the life of the bridge.

This paper puts forward the research idea and research path for environmental risk zoning, but it is still in the preliminary exploration stage, there are still many limitations. One is that this study ignores the interaction between environmental factors. For example, when other conditions are the same, the corrosion life of concrete structures is significantly reduced after experiencing freeze–thaw cycles, and the risk value in northern coastal areas that are subject to freezing and thawing should be appropriately increased ( 45 ). The carbonation environment will accelerate the rate of chloride ion corrosion ( 43 ). Wind erosion can accelerate the formation of cracks on concrete surface under freezing and thawing action ( 48 ). When concrete in cold areas is subjected to river scouring and freeze–thaw cycles for a long time, its mechanical properties and durability will deteriorate to a greater extent; tests show that the durability degradation of concrete is two to three times more serious than for a single factor under the action of both freezing–thawing and washing abrasion ( 49 ). Secondly, the initial weight setting of this paper is based on accident data. However, the occurrence of bridge collapse accidents is often caused by a variety of factors—we can only choose from numerous cases mainly caused by environmental factors for analysis. This method can only reflect the occurrence probability of environmental risks roughly and has limitations in accuracy. In addition, the limited amount of data affects the accuracy of the risk value to some extent. This method needs the support of more measured data.

This study can also be further deepened in the following ways. (i) More measured data can be obtained by using advanced technical means to improve the accuracy of zoning. (ii) One could explore further how to analyze quantitatively the correlation between environmental impact factors. (iii) Improvement of the environmental factors in the bridge condition prediction model could improve the prediction accuracy.

Footnotes

Author Contributions

The authors confirm contribution to the paper as follows: study conception and design: D. Su, Y. Liu; data collection: Q. Duan, Z. Cao; analysis and interpretation of results: D. Su, Y. Liu; draft manuscript preparation: D. Su. All authors reviewed the results and approved the final version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was supported by the National Natural Science Foundation of China (71871014), the China National Key R&D Program during the 13th Five-Year Plan Period (2018YFC0704402-02).

Data Accessibility Statement

Data are available on request to the authors.

References

Ministry of Transport of the People’s Republic of China. Statistical Bulletin on the Development of the Transport Industry in 2019. Ministry of Transport of the People’s Republic of China, Beijing, 2020.

W. H.

Thoughts and Suggestions on Bridge Maintenance Management. Proc., 6th Annual Conference of the Maintenance and Management Branch of China Highway Society, Chengdu, Beijing, 2016. (In Chinese)

Guo

D. J.

Song

Zhou

W. H.

Comprehensive Assessment of Maintenance Status of Inservice Beam Bridges Based on AHP Theory. Journal of Highway and Transportation Research and Development, Vol. 28, No. 3, 2011, pp. 108–112. (In Chinese)

Tan

J. S.

Elbaz

Wang

Z. F.

Shen

J. S.

Chen

Lessons Learnt from Bridge Collapse: A View of Sustainable Management. Sustainability, Vol. 12, No. 3. 2020, p. 1205.

Yan

Q. J.

Wei

J. C.

Shangguan

Zhu

W. D.

Huang

X. Y.

Analysis on Optimal Maintenance Strategy for Highway Concrete Girder Bridges During Service Life. Journal of Highway and Transportation Research and Development, Vol. 36, No. 2, 2019, pp. 95–102. (In Chinese)

FHWA. Insight Into Pavement Preservation: A Compendium. Federal Highway Administration, Washington, D.C., 2000.

Ren

C. F.

Development and Big Data Analysis of Bridge Inspection and Maintenance Management System. Master’s thesis. Chang’an University, Xi’an, China, 2017. (In Chinese)

Liu

M. X.

Theoretical Research on the Preventive Maintenance Technology of Concrete Bridge. Master’s thesis. Chang’an University, Xi’an, China, 2015. (In Chinese)

Srikanth

Arockiasamy

Deterioration Models for Prediction of Remaining Useful Life of Timber and Concrete Bridges: A Review. Journal of Traffic and Transportation Engineering (English Edition), Vol. 7, No. 2, 2020, pp. 152–173.

10.

Barone

Frangopol

D. M.

Life-Cycle Maintenance of Deteriorating Structures by Multi-Objective Optimization Involving Reliability, Risk, Availability, Hazard and Cost. Structural Safety, Vol. 48, 2014, pp. 40–50.

11.

G. P.

Lee

J. H.

Guan

Loo

Y. C.

Blumenstein

Prediction of Long-Term Bridge Performance: Integrated Deterioration Approach with Case Studies. Journal of Performance of Constructed Facilities, Vol. 29, No. 3, 2014, p. 4014089.

12.

Yusuf

Hamid

Developing a Bridge Condition Rating Model Based on Limited Number of Data Sets. Bridge Engineering, 2018, pp. 27–42. https://doi.org/10.5772/intechopen.71556.

13.

Bush

S. J. W.

Henning

T. F. P.

Raith

Ingham

J. M.

Development of a Bridge Deterioration Model in a Data-Constrained Environment. Journal of Performance of Constructed Facilities, Vol. 31, No. 5. 2017, p. 04017080. https://doi.org/10.1061/(ASCE)CF.1943-5509.0001074.

14.

Ishwarya

Arockiasamy

15.

Fang

Sun

L. J.

Developing a Semi-Markov Process Model for Bridge Deterioration Prediction in Shanghai. Sustainability, Vol. 11, No. 19, 2019, p. 5524.

16.

Manafpour

Guler

Radlinska

Rajabipour

Warn

Stochastic Analysis and Time-Based Modeling of Concrete Bridge Deck Deterioration. Journal of Bridge Engineering, Vol. 23, No. 9, 2018, pp. 04018066.1–04018066.9.

17.

Huang

Y. H.

Artificial Neural Network Model of Bridge Deterioration. Journal of Performance of Constructed Facilities, Vol. 24, No. 6, 2010, pp. 597–602.

18.

Deng

Wang

State-of-the-Art Review on the Causes and Mechanisms of Bridge Collapse. Journal of Performance of Constructed Facilities, Vol. 30, No. 2, 2016, p. 04015005.

19.

Zhu

L. M.

Qian

S. Q.

Chen

Q. Y.

Xing

S. L.

Risk Assessment and Prevention Strategy of Bridge Collapse in Service. Journal of Nanjing University of Technology (Natural Science Edition), Vol. 42, 2020, pp. 284–290. (In Chinese)

20.

Minami

Fujiu

Nakayama

Takayama

Chikata

Analysis of Relationship Between Soundness of Bridges and Natural Environments. Journal of Japan Society of Civil Engineers Ser D3, Vol. 72, No. 5, 2016, pp. 251–260.

21.

Liu

M. Y.

Liang

Lifecycle Management and Maintenance of Marine Bridge Engineering. Strategic Study of CAE, Vol. 21, No. 3, 2019, pp. 25–30.

22.

Moomen

Agbelie

B. R.

Labi

Sinha

K. C.

Bridge Deterioration Models to Support Indiana’s Bridge Management System. FHWA Joint Transportation Research Program, Publication No. FHWA/IN/JTRP-2016/03. Purdue University, West Lafayette, IN, 2016.

23.

Davis-McDaniel

Chowdhury

Peng

Dey

A Fault-Tree Model for Risk Assessment of Bridge Failure—A Case Study for Segmental Box Girder Bridges. Journal of Infrastructure Systems, Vol. 19, No. 3, 2012, pp. 326–334. https://doi.org/10.1061/(ASCE)IS.1943-555X.0000129.

24.

Suo

W. L.

Zhang

Sun

X. L.

Risk Assessment of Critical Infrastructures in a Complex Interdependent Scenario: A Four-Stage Hybrid Decision Support Approach. Safety Science, Vol. 120, 2019, pp. 692–705.

25.

Sasidharan

Parlikad

A. K.

Schooling

Risk-Informed Asset Management to Tackle Scouring on Bridges Across Transport Networks. Structure and Infrastructure Engineering, 2021, pp. 1–17. https://doi.org/10.1080/15732479.2021.1899249.

26.

Karamouz

Rasoulnia

Olyaei

M. A.

Zahmatkesh

Prioritizing Investments in Improving Flood Resilience and Reliability of Wastewater Treatment Infrastructure. Journal of Infrastructure Systems, Vol. 24, No. 4, 2018, p. 04018021.

27.

Chen

A. R.

Feng

S. R.

R. J.

Environmental Zoning for Durability Performance of Concrete Bridges Based on Deterioration Mechanism. Journal of Tongji University (Natural Science), Vol. 42, 2014, pp. 331–337. (In Chinese)

28.

Risk Evaluation of Coal Floor Water Inrush Based on Variable Weight Theory and its Application. Journal of Basic Science and Engineering, Vol. 25, No. 3, 2017, pp. 500–508.

29.

Wang

Y. H.

Zhai

H. Z.

Zeng

T. M.

Zhang

L. M.

Risk Evaluation of the Subway Foundation Pit Based on Variable-Weight and Matter-Element Theory. Advanced Materials Research, Vol. 941, 2014, pp. 842–848.

30.

Wang

L. P.

Cheng

Liu

Ding

You

Model on Improved Variable Weight-Matter Element Theory for Risk Assessment of Water Inrush in Karst Tunnels. Geotechnical and Geological Engineering, Vol. 39, 2021, pp. 3533–3548.

31.

Shan

C. J.

Dong

Z. C.

D. B.

Wang

Ling

Liu

Study on River Health Assessment Based on a Fuzzy Matter-Element Extension Model. Ecological Indicators, Vol. 127, No. 10, 2021, p. 107742.

32.

Zhou

H. Y.

Yuan

Y. B.

Liu

C. L.

Zhang

Extension Model for Safety Appraisal of Existing Concrete Members Based on an Improved Comprehensive Weighting Method. Advances in Civil Engineering, Vol. 2018, 2018, pp. 1–15.

33.

Yuan

Y. G.

Chen

Han

W. S.

Xie

Wang

Tang

L. L.

Establishment and Updating of Nonstationary Random Resistance Deterioration Model for Existing Concrete Bridges. China Journal of Highway and Transport, Vol. 32, No. 12, 2019, pp. 145–155. (In Chinese)

34.

Long

G. X.

Liu

X. P.

Tang

L. L.

Wang

Huang

P. M.

Study on Concrete Beam Bridge Bearing Capacity Comprehensive Evaluation System. China Safety Science Journal, Vol. 28, No. 10, 2018, pp. 162–168. (In Chinese)

35.

F. H.

Jin

H. S.

D. H.

Jiang

M. C.

Z. W.

G. Y.

Hong

Y. C.

Liu

On the Durability Evaluation of the Reinforced Concrete Bridges. Journal of Safety and Environment, Vol. 18, No. 5, 2018, pp. 1653–1662. (In Chinese)

36.

Caitlyn

D. M.

Fault-Tree Model for Bridge Collapse Risk Analysis. Master’s thesis. Clemson University, SC, 2011.

37.

Basista

Weglewski

Micromechanical Modelling of Sulphate Corrosion in Concrete: Influence of Ettringite Forming Reaction. Theoretical and Applied Mechanics, Vol. 35, 2008, pp. 29–52.

38.

Hao

Y. H.

Liu

Y. C.

Guo

X. Y.

Feng

Y. J.

Jiang

Erosion Morphology, Damage Mechanism and Prediction of Concrete Pavement Under Sandstorm Environment. China Journal of Highway and Transport, Vol. 30, No. 29, 2017, pp. 27–33. (In Chinese)

39.

Gao

Use of Data Analytics and Machine Learning to Improve Culverts Asset Management Systems. PhD thesis. University of Cincinnati, OH, 2019.

40.

Elvik

Sagberg

Langeland

P. A.

An Analysis of Factors Influencing Accidents on Road Bridges in Norway. Accident Analysis & Prevention, Vol. 129, 2019, pp. 1–6.

41.

Cheng

Yang

L. F.

Quantification of Environmental Effect for Carbonation and Durability Analysis of Concrete Structures. China Civil Engineering Journal, Vol. 48, No. 9, 2015, pp. 51–59. (In Chinese)

42.

Tongaria

Mandal

D. S.

Mohan

A Review on Carbonation of Concrete and its Prediction Modelling. Journal of Environmental Nanotechnology, Vol. 7, No. 4, 2019, pp. 75–90.

43.

Gehlot

Sankhla

S. S.

Saini

K. K.

Critical Review on Effect of Chloride Penetration on Concrete and Various Rapid Chloride Penetration Tests Methods. International Journal of Research and Analytical Reviews, Vol. 6, No. 2, 2019, pp. 648–654.

44.

Zhang

H. F.

Zhang

W. P.

X. L.

Jin

X. Y.

Jin

N. G.

Chloride Penetration in Concrete Under Marine Atmospheric Environment – Analysis of the Influencing Factors. Structure and Infrastructure Engineering, Vol. 12, No. 11, 2016, pp. 1–11.

45.

Peng

Y. Z.

Gao

Prediction for Corrosion Life of Reinforced Concrete Structures Under Salt Freeze-Thaw Environment. Advances in Science and Technology of Water Resources, Vol. 39, No. 3, 2019, pp. 44–49. (In Chinese)

46.

Ding

S. J.

J. P.

Chloride Ion Erosion Experiment of Concrete Members Under Different Environmental Conditions. Journal of Harbin Institute of Technology, Vol. 48, No. 12, 2016, pp. 28–33. (In Chinese)

47.

Jiao

J. T.

Y. H.

Yang

R. H.

The Initiation Time for Reinforced Concrete Members Under Chlorine Salt Erosion and Climate. Acta Scientiarum Naturalium Universitatis Sunyatseni, Vol. 55, No. 1, 2016, pp. 85–90. (In Chinese)

48.

Xue

H. J.

Research on Durability Damage and Deterioration Mechanism of Aeolian Sand Concrete in Freeze-Thaw and Salt Corrosion Environment of Wind Erosion Area. PhD thesis. Inner Mongolia Agricultural University, Hohhot, China, 2018.

49.

Liu

M. H.

Han

J. Q.

Research on the Anti-Abrasion Capacity of Concrete Under Freeze-Thaw Cycle. China Civil Engineering Journal, Vol. 52, No. 7, 2019, pp. 100–109. (In Chinese)

50.

Chen

W. K.

Zhou

Z. H.

Gao

Chen

Analysis of Temporal-Spatial Distribution of Life Losses Caused by Earthquake Hazards in Chinese Mainland. Journal of Catastrophology, Vol. 34, No. 1, 2019, pp. 222–228.

51.

Montalvo

Cook

Keeney

Retrospective Analysis of Hydraulic Bridge Collapse. Journal of Performance of Constructed Facilities, Vol. 34, No. 1, 2020, p. 04019111. https://doi.org/10.1061/(ASCE)CF.1943-5509.0001378.

52.

Jin

C. H.

Shen

G. T.

Wang

Liu

Roller Coaster Safety Evaluation Method Based on Matter-Element Extension Theory. Machine Design & Research, Vol. 36, No. 4, 2020, pp. 190–194. (In Chinese)

53.

Luo

Zhu

Zhang

X. Q.

Yao

L. K.

Scheme Evaluation Model of Railway Location Designs Based on Variable Weight Theory. Journal of Railway Engineering Society, Vol. 35, No. 8, 2018, pp. 16–20. (In Chinese)

54.

Liang

Sun

Safety Assessment of Bridges Based on Variable Weight and D-S Evidence Theory. Journal of Northeastern University, Vol. 40, No. 1, 2019, pp. 99–103. (In Chinese)

55.

Ren

Huang

Sun

H. B.

Condition Assessment Method of Suspension Bridges Using Time-Dependent Variable Weight Model. Journal of South China University of Technology (Natural Science), Vol. 46, No. 6, pp. 48–53. (In Chinese)

56.

Yan

Q. Y.

Zhang

M. J.

Qin

G. Y.

Risk Assessment of New Energy Vehicle Supply Chain Based on Variable Weight Theory and Cloud Model: A Case Study in China. Sustainability, Vol. 12, No. 8, 2020, p. 3150. https://doi.org/10.3390/su12083150.

57.

Liu

Y. S.

X. T.

Chen

X. Y.

D. H.

Management Path of Concrete Beam Bridge in China from the Perspective of Sustainable Development. Sustainability, Vol. 12, No. 17, 2020, p. 7145. https://doi.org/10.3390/su12177145.

58.

Luo

D. M.

Wang

Zhang

S. H.

Niu

D. T.

Application of Fuzzy and Rough Sets to Environmental Zonation for Concrete Durability: A Case Study of Shaanxi Province, China. Sustainability, Vol. 12, No. 8, 2020, p. 3128. https://doi.org/10.3390/su12083128.

59.

Wang

W. Q.

Lyu

S. R.

Zhang

Y. D.

S. Q.

A Risk Assessment Model of Coalbed Methane Development Based on the Matter-Element Extension Method. Energies, Vol. 12, No. 20, 2019, p. 3931. https://doi.org/10.3390/en12203931.

60.

Zhao

Y. F.

Wan

J. Q.

An Operational Risk Assessment Model for Air Traffic Control Based on the Set Pair Extension Coupling Method. Journal of Safety and Environment, Vol. 18, No. 6, 2018, pp. 2059–2063. (In Chinese)

61.

Yang

Rong

X. L.

Dong

Risk Assessment on the Tunnel Collapse Probability by the Theory of Extenics in Combination with the Entropy Weight and Matter-Element Model. Journal of Safety and Environment, Vol. 16, No. 2, 2016, pp. 15–19. (In Chinese)

62.

Melchers

R. E.

Chaves

I. A.

Durability of Reinforced Concrete Bridges in Marine Environments. Structure and Infrastructure Engineering, Vol. 16, No. 3, 2019, pp. 1–12.