Risk Evaluation of Hazardous Chemical Road Transportation Accidents Based on a Combined Empowerment-Cloud Model

Abstract

To improve the accuracy of the risk evaluation results of hazardous chemical road transportation systems, an accident risk evaluation model based on the combined weight-cloud model is proposed to address the problems of ambiguity and randomness in the risk factors for hazardous chemical road transportation accidents. First, incorporating the risk factor identification results, the risk evaluation index system for hazardous chemical road transportation accidents is established. On the basis of determining the subjective and objective weights of the indices by using the analytic hierarchy process and entropy weight method, the comprehensive weights of the evaluation indices are calculated by the linear weighting method. Then, the weighting results are combined with the cloud model to calculate the risk level of the evaluation indices at all levels and to output the results in the form of a risk cloud diagram to visually determine the risk level of hazardous chemical road transportation. Finally, the model is applied to evaluate the accident risk in an example. The results demonstrate that the model is both practical and applicable in the evaluation of the risk of hazardous chemical road transportation accidents. The evaluation results can provide a valuable reference for hazardous chemical road transportation enterprises to prevent accidents.

1. Introduction

Hazardous chemicals are basic raw materials that are indispensable for people’s production and life. Road transportation is the best choice for the logistics transportation of hazardous chemicals [1]. As hazardous chemicals are flammable, explosive, and highly toxic, accidents in transit lead to leaks, fires, and explosions [2], resulting in casualties, property damage, environmental pollution, ecological damage, and other serious consequences. For example, on 2 January 2022, a heavy semi-trailer tanker collided with a vehicle moving in the opposite direction in Jiangyou city, Mianyang, Sichuan Province, resulting in eight deaths, 20 injuries, and direct economic losses of approximately 11.35 million yuan. On 11 March 2022, a heavy-duty warehouse truck lost control within the section of Ebao Town, Qilian County, on National Highway 227 and crashed into the temporary checkpoint of Ebao, resulting in seven deaths, two injuries, and direct economic losses of approximately 3.513 million yuan; additionally, the surrounding houses and environment were damaged to different degrees. All these accidents have brought significant losses and exerted negative social impacts on the country and its people. The road transportation system of hazardous chemicals is a complex and huge system. Its safety is easily disturbed by the fuzziness and randomness of many influencing factors, which makes it difficult to determine the risk of road transportation accidents of hazardous chemicals. How to consider the fuzziness and randomness and accurately evaluate the risk of road transportation accidents of hazardous chemicals, so as to take effective measures to prevent and control them, has become a social problem that needs to be solved.

The problem of the risk evaluation of hazardous chemical road transportation accidents has received increasing attention from scholars in China and elsewhere. Fuzzy comprehensive evaluation [3], gray theory [4], and other probabilistic risk assessment methods have been widely used in the safety assessment of the road transportation of hazardous chemicals. The fuzzy comprehensive evaluation method considers the fuzziness of the influencing factors, but does not consider their randomness. The gray prediction model is very simple to implement, but it ignores the correlation between various factors, relies heavily on historical data, and may produce huge errors that cannot reflect the actual situation well [2]. The cloud model theory provides many novel strategies for solving more complex problems in the field of decision-making. Peng et al., 2020 [5], constructed a tunnel water inrush risk assessment model by combining the analytic hierarchy process and the cloud model, and applied and verified the model in practice. Wang and Zhang [6], 2020, proposed a cloud model evaluation method to evaluate the safety status of highways and their supporting facilities. Therefore, the combination of the analytic hierarchy process and cloud models can be applied in the transportation of hazardous chemicals on roads. Compared with traditional fuzzy evaluation, which relies on the affiliation function provided by experts’ a priori knowledge or obtained through statistical methods, a cloud model can effectively reflect the fuzziness and randomness in the data of evaluation indices, making the evaluation results more accurate and reasonable. Therefore, the cloud model is superior to the fuzzy comprehensive evaluation method and the gray theory evaluation method in solving the uncertainty problem.

Due to the uncertainty of the risk evaluation system for hazardous material road transport accidents, further research on the fuzziness and randomness of the evaluation process is needed. Cloud models have been widely used to address uncertainty problems in many fields. However, there are few studies on the application of cloud modeling to the risk evaluation of hazardous chemical road transport accidents. Therefore, a cloud model is introduced to address the ambiguity and randomness in the risk assessment of hazardous chemicals road transport accidents.

2. Construction of the Evaluation Index System

Detailed data on 1288 hazardous chemical road transport accidents from 2018 to 2022 were collected from browser websites such as the Chemical Accident Information Network (CAIN), which were collated and analyzed to ultimately produce Figure 1, Accident Data Statistics, and Figure 2, Accident Cause Analysis.

Combining the accident data and the relevant literature on accident risk identification in this field by many scholars in the past, most studies have considered five major factors: personnel, vehicles, hazardous chemicals, environment, and management. However, few have considered the impact of different hazardous chemicals on different equipment on vehicles and the damage to vehicle GPS monitoring equipment. The main risk factors of road transport accidents of hazardous chemicals were finally classified into six aspects, namely, personnel, vehicles, equipment, hazardous chemicals, environment, and management, based on the system safety theory. Risk factors for the road transport of hazardous chemicals and the sources in the literature are summarized in

Table 1. Summary of risk factors and literature sources for road transport of hazardous chemicals.

Main Influencing Factors	Literatures	Risk Influencing Factors
Personnel factors	(Wang et al. [7], Liu et al. [8])	Psychological quality, level of driver training, driver awareness of safety responsibilities, professional and technical competence
Vehicle factors	(Chen et al. [9], Mohammadfam et al. [10])	Vehicle performance status, maintenance status
Equipment factors	(Donaldson [11], Wu and Fan [12])	Equipped with safety accessories, monitoring equipment, emergency supplies
Hazardous chemical factors	(Chen et al. [9], Wu et al. [13])	Hazardous chemical loads, hazardous chemical packaging, physical and chemical properties of hazardous chemicals
Environmental factors	(Wu et al. [13], Fabiano et al. [14])	Road conditions, weather conditions, traffic conditions, density of people along the route
Management factors	(Liu et al. [2], Wu and Fan [12])	Enterprise safety management system, safety education and training, government safety supervision, emergency rescue management

.

2.1. Personnel Factor

Transportation personnel play an important role in the road transportation of hazardous chemicals and are the key influencing factor of accidents. Therefore, transportation personnel will have a greater impact on the stability of the whole transportation system [15].

Hazardous chemical transportation requires a high level of physical and professional skills of the personnel involved, who also need to be licensed to work with the corresponding qualifications [8]. The quality of the personnel involved in the road transportation of hazardous chemicals is relatively low, and their awareness of safety responsibilities is poor, which can lead to unsafe behaviors such as fatigued driving, speeding, and running red lights during transportation, which greatly increases the probability of transportation accidents [7]. Some transport personnel participate in safety training and education content for fewer hours; this mostly involves theoretical knowledge, and less practical training, moral education, and psychological guidance. Thus, they are prone to accidents. Transportation personnel must fully grasp the hazardous characteristics of the hazardous chemicals that they carry before undertaking transportation tasks, and if they lack the ability to recognize hazardous chemicals, they cannot take emergency measures in time when accidents occur, which will increase the losses caused by accidents [16].

2.2. Vehicle Factor

The vehicle is an important factor that ensures the safety of transportation in the road transportation of dangerous chemicals. Because of the dangerous nature of hazardous chemicals, the performance of vehicles in transporting hazardous chemicals needs to meet the corresponding transport standards [9]. The worse the performance of the vehicle is, the greater the risk of hidden dangers in the transport process. When a vehicle is used for a long time, its aging is serious, and there may be large breakages, etc. [10]. Which will lead to a reduction in vehicle safety, making it easy to produce accidents in transit.

2.3. Equipment Factor

When transporting dangerous chemicals, the vehicle needs to be equipped with some equipment to ensure the safety of personnel, and the performance of these devices will have an impact on the stability of the whole transport system [17].

Vehicles transporting hazardous chemicals must be equipped with safety and protective equipment for water, fire, and sun protection as well as firefighting and anti-toxic equipment based on the properties of hazardous chemicals [11]. According to the regulations set by the state, electronic equipment, such as GPS monitoring equipment, needs to be installed in the vehicle to be able to locate and develop a reasonable route in real time. However, the electronic equipment produced at present has inconsistent specifications and uneven quality, which can lead to offline equipment, inaccurate positioning, inaccurate data transmission, and drifting trajectories during transportation, causing interference with the driver’s operating technique and increasing the probability of accident occurrence. Due to the special nature of transporting hazardous chemicals, emergency supplies in the vehicle need to be considered. Such supplies can be used to reduce casualties in case of emergencies.

2.4. Chemical Factor

Different hazardous chemicals have different physical and chemical properties, and they are flammable, explosive, corrosive, and toxic, which can cause serious accidents once leaked. The road transportation of hazardous chemicals requires a high level of reliability and sealing of the packaging [18]. If the selected material does not meet the requirements, it will easily lead to a leakage of hazardous chemicals during transportation, which will cause serious safety accidents. The amount of dangerous chemicals carried will also have an impact on the safety of the road transportation of dangerous chemicals. If the amount of dangerous chemicals carried exceeds the load capacity of the vehicle itself, it will produce a great load on the vehicle’s tires, causing tire deformation, tire blowouts, and other accidents, which greatly increases the risk of transportation [13].

2.5. Environmental Factor

Road transport usually occurs under natural conditions, and the safety and stability of the whole transportation system will be greatly affected by a good or bad transportation environment [14].

Hazardous chemical transport vehicles passing through road sections with large flows of people and traffic will increase the risk potential in the transport process. Transport vehicles driving in bad weather, such as heavy rain, snow, fog, and high and low temperatures, will increase the difficulty of driving for the driver and easily cause driving vehicle collisions or rollover accidents, thus causing dangerous chemical transport accidents [19]. Road conditions are mainly reflected in the road width, road design and road conditions. When the road state is poor, it will increase the psychological pressure of driving for drivers. In a stressful state, there is a higher likelihood of wrong operation, which will ultimately cause safety accidents.

2.6. Management Factor

As a trigger factor in the road transport of dangerous chemicals, the management factor will not directly lead to accidents. It will only be one of the causal factors; the root cause of accidents is poor management.

Hazardous chemical road transport management risk is mainly analyzed by two subjects: government regulatory departments and hazardous chemical transport enterprises. The qualification review of hazardous chemical transport enterprises by governmental supervisory departments, insufficient supervision, and inadequate emergency rescue management after accidents increase the risk of hazardous chemical road transport. Many transport enterprises neglect relevant training for employees, which leads to relevant transport personnel speeding, not operating in accordance with the relevant rules and regulations, and other phenomena, which in turn increases the probability of transport accidents. In addition, the safety management system is not perfect and not in place, resulting in poor awareness of the safety responsibilities of transport personnel, low awareness of safety prevention, and other factors that can cause transport accidents [12].

Based on the risk factor identification results above, a hierarchical index system model for evaluating the safety risk of hazardous chemical road transportation is constructed, with six primary indicators and 20 secondary indicators, as shown in Figure 3.

3. Construction of a Risk Evaluation Model for Hazardous Chemical Road Transportation Accidents

3.1. Determination of the Accident Risk Evaluation Methodology

3.1.1. Determination of Subjective Weights Based on the Analytic Hierarchy Process

The analytic hierarch process calculates the weights through the fuzzy quantification method of qualitative indicators. It has the advantages of being systematic, flexible, and concise [20]. As shown in Figure 3, the risk evaluation index system for hazardous chemical road transport accidents is a three-level hierarchical structure model, and the subjective weights can be calculated by using the analytic hierarchy process, with the following steps:

(1) Construct the judgment matrix.

Using the 1~9-level scale method, the judgment matrix is constructed to determine the relative importance of each factor to the upper level.

(2) Calculate the weights of each factor. The arithmetic mean of all column vectors is used to estimate the weight vector.

$$W_i={\displaystyle \frac{1}{n}}{\displaystyle \sum _j^n\frac{a_{ij}}{{\displaystyle \sum _{k=1}^n{a}_{kj}}}}$$

(1)

where, $i=1,2,3,\dots ,n$.

(3) Test the consistency of the judgment matrix.

$$CI={\displaystyle \frac{{\lambda }_{\max }-n}{n-1}}$$

(2)

where ${\lambda }_{\max }$ is the maximum eigenvalue of the judgment matrix $A={\left(a_{ij}\right)}_{m\times n}$.

(4) Calculate the consistency ratio ($CR$).

$$CR={\displaystyle \frac{CI}{RI}}$$

(3)

where CI is an indicator for measuring the consistency of the judgment matrix; RI is an indicator for comparing the consistency of judgment matrices of different orders; and CR is an indicator to determine whether the matrix has satisfactory consistency.

When CR < 0.10, the consistency of this judgment matrix is considered acceptable; otherwise, it must be corrected.

3.1.2. Determination of Objective Weights Based on the Entropy Weight Method

The entropy weight method determines objective weights based on the magnitude of the variability of indicators. It can reflect, in depth, the differentiation ability of indicators, determine the best weights, and make the assignment more objective. Additionally, calculation is easy. In this paper, the entropy weight method is applied to determine the objective weights of risk factors for hazardous chemical road transport accidents, and the calculation steps are as follows [21,22].

(1) Construct the numerical matrix. Let the model have n secondary evaluation indicators and m primary evaluation indicators. $x_{ij}\left(i=1,2,3,\dots ,m;j=1,2,3,\dots ,n\right)$ is the value of the jth evaluation indicator under the ith primary evaluation indicator, forming a numerical matrix $X_{ij}$ of evaluation indicators of order $m\times n$.

$$X_{ij}=\left(\begin{array}{ccc}{X}_{11}& \dots & X_{1n}\\ ⋮& \ddots & ⋮\\ X_{m1}& \dots & X_{mn}\end{array}⋮\right)$$

(4)

(2) Standardize the indicators. Equation (5) is used to standardize positive indicators, and Equation (6) is used to standardize negative indicators.

$$L_{ij}={\displaystyle \frac{X_{ij}-X_{\min }}{X_{\max }-X_{\min }}}$$

(5)

$$L_{ij}={\displaystyle \frac{X_{\max }-X_{ij}}{X_{\max }-X_{\min }}}$$

(6)

where (5) is the normalization of positive indicators, and (6) is the normalization of negative indicators, $X_{\min },X_{\max }$ are the minimum and maximum values of the same criterion element index, respectively, and $L_{ij}$ is the standardized $X_{ij}$.

(3) Calculate the weight value of the evaluation indicators. Calculate the weight value $T_{ij}$ of the jth secondary evaluation indicator of the evaluation unit relative to the ith primary evaluation indicator.

$$T_{ij}={\displaystyle \frac{L_{ij}}{{\displaystyle {\sum }_{i=1}^m{L}_{ij}}}}$$

(7)

(4) Calculate the entropy value of the index. Calculate the entropy $E_j$ of the output of the jth evaluation index.

$$E_j=-{\displaystyle \frac{1}{\ln m}}{\displaystyle \sum _{i=1}^m{T}_{ij}}\ln T_{ij}$$

(8)

(5) Calculate the objective weights of the indicators.

$$W_j={\displaystyle \frac{1-E_j}{{\displaystyle {\sum }_{i=1}^{n}1-E_j}}}$$

(9)

3.1.3. Comprehensive Weight Determination Based on the Linear Weighting Method

Since the subjective weights cannot fully reflect the actual data information and the objective weights cannot highlight the importance attached to different indicators by the participating decision-makers, the subjective weights calculated by the analytic hierarchy process and the objective weights calculated by the entropy weight method are combined and assigned by the linear weighting method to obtain the comprehensive weights ${\omega }_j$. The calculation formula is as follows.

$${\omega }_j=\alpha {\omega }_{jz}+\beta {\omega }_{jk}$$

(10)

where $j=1,2,3\dots ,n$ and j is the number of indicators, ${\omega }_{jz}$ is the subjective weight, and ${\omega }_{jk}$ is the objective weight. $\alpha$ is the weighting coefficient of the subjective weight, $\beta$ is the weighting coefficient of the objective weight, and the value range of the weighting coefficient is (0, 1).

The calculation using the coefficient of variation method leads to the specific value of $\alpha$ [23]. The arithmetic formula is as follows.

$$\left\{\begin{array}{l}\alpha ={\displaystyle \frac{n}{n-1}}\left[{\displaystyle \frac{2}{n}}\left(P_1+2P_2+\dots +n{P}_n\right)-{\displaystyle \frac{n-1}{n}}\right]\\ \beta =1-\alpha \end{array}\right.$$

(11)

where $P_n$ is the value of each evaluation index weight in the subjective weight vector ${\omega }_{jz}$ ranked from smallest to largest.

3.2. Cloud Model Evaluation Method

The cloud model evaluation method can complete the direct conversion of qualitative concepts and quantitative information, taking into account the ambiguity and randomness of the process of the risk evaluation of hazardous chemical road transportation accidents [24].

3.2.1. Basic Theory of Cloud Models

Let $U$ be a quantitative thesis domain expressed in exact numerical terms and $C$ be a qualitative concept on $U$ if the quantitative value $x$ is satisfied:

• $x\in U$;
• $x$ is a random realization of the qualitative concept $C$; and
• The certainty $\mu (x)\in \left[0,1\right]$ of $x$ over $C$ is a random number with a stable tendency.

Then, the distribution of $x$ over the quantization domain $U$ is called a cloud, and each $x$ is called a cloud droplet.

The cloud model uses three numerical features, i.e., expected value $E_x$, entropy $E_n$ and hyper entropy $H_e$, to characterize a concept as a whole, which is denoted as $C(E_x,E_n,H_e)$. The specific meanings and formulas for calculation are shown in

Table 2. Numerical Characterization of Cloud Models.

Digital Characteristic	Meaning	Formula
$E_x$	The point in the space of the argument domain that reflects the qualitative concept is the cloud center of gravity of the cloud droplet group. The greater the expectation, the higher the risk level.	$E_x=\overline{X}={\displaystyle \frac{1}{n}}{\displaystyle \sum _{i=1}^n{x}_i}$
$E_n$	Uncertainty measures of qualitative concepts, determined by the fuzziness and probability of the metric qualitative concepts, can reflect the degree of discrete cloud droplets. The greater the entropy value, the greater the ambiguity, and the lower the credibility of the evaluation results.	$E_n=\sqrt{{\displaystyle \frac{\pi }{2}}}\times {\displaystyle \frac{1}{n}}{\displaystyle \sum _{i=1}^n\left|x_i-\overline{X}\right|}$
$H_e$	An uncertain measure of entropy that reflects the degree of cohesion of the cloud droplets. The greater the hyperentropy, the thicker the cloud layer, the greater the randomness, and the lower the stability value.	$H_e=\sqrt{\left|S^2-E_n^2\right|}$

. In Table 2, the variance calculation formula is as follows:

$$S^2={\displaystyle \frac{1}{n-1}}{{\displaystyle \sum _{i=1}^n\left(x_i-\overline{X}\right)}}^2$$

(12)

3.2.2. Construction of Cloud Models

A comprehensive risk cloud for hazardous chemical road transport accidents is constructed with the following steps [25].

(1) Determine the risk level standard cloud.

The evaluation of the indicators of the cloud model is expressed as the concept of expressing the fuzzy good and bad of things, and the comment of the evaluation indicators in each comprehensive evaluation system has different characteristics according to the evaluator, so the determination of the standard cloud of the cloud model is an important prerequisite for the subsequent risk cloud and comprehensive cloud of the cloud model.

Considering the reality of road transport of dangerous chemicals and the suggestions of experts from various parties, combined with the habit of the language expression of road transport accidents of dangerous chemicals, the risk evaluation level of hazardous chemical road transport accidents is divided into 5 levels: “very low risk”, “low risk”, “medium risk”, “high risk”, and “very high risk”. The numerical features of each level are transformed into standard clouds by using the forward cloud generator of the cloud model. The cloud parameters for each level interval $\left[x_{\min },x_{\max }\right]$ are calculated as follows [26].

$$\left\{\begin{array}{l}{E}_{x_0}={\displaystyle \frac{x_{\max }+x_{\min }}{2}}\\ E_{n_0}={\displaystyle \frac{x_{\max }-x_{\min }}{6}}\\ H_{e_0}=k\end{array}\right.$$

(13)

where k is the super entropy value, reflecting the cohesiveness of the cloud droplets. The smaller its value is, the smaller the dispersion of the cloud droplets and the thinner the normal cloud [27]. k is the constant, which can be adjusted based on the uncertainty and randomness of the specific index. It is generally taken as 0.1.

The numerical characteristics of the standard cloud can be obtained based on the equation above, as shown in

Table 3. Digital eigenvalue of standard cloud.

Security Level	Grade Range	Cloud Characteristic Values
Very High Risk	(7–10]	(8.5, 0.5, 0.1)
High Risk	(5–7]	(6, 0.333, 0.1)
Medium Risk	(3–5]	(4, 0.333, 0.1)
Low Risk	(1–3]	(2, 0.333, 0.1)
Very low risk	(0–1]	(0.5, 0.167, 0.1)

where: In the table, (0–1] includes 1 but does not include 0, and so on.

.

(2) Calculate the risk cloud of the evaluation indicators.

The raw data are obtained by distributing questionnaires to experts and scoring the risk level. The raw data were input into the backward cloud generator and calculated using the following formula to determine the risk cloud of hazardous chemical road transportation accident evaluation indicators.

$$\left\{\begin{array}{l}{E}_{xj}=\overline{X}={\displaystyle \frac{1}{n}}{\displaystyle \sum _{i=1}^n{x}_{ij}}\\ E_{nj}=\sqrt{{\displaystyle \frac{\pi }{2}}}\times {\displaystyle \frac{1}{n}}{\displaystyle \sum _{i=1}^n\left|x_{ij}-\overline{X}\right|}\\ H_{ej}=\sqrt{\left|S_j^2-E_{nj}^2\right|}\\ S_j^2={\displaystyle \frac{1}{n-1}}{{\displaystyle \sum _{i=1}^n\left(x_{ij}-\overline{X}\right)}}^2\end{array}\right.$$

(14)

where $E_{xj},E_{nj},H_{ej},S_j^2$ are the sample expected value, entropy, hyper entropy, and variance of the jth indicator, respectively; $x_{ij}$ is the ith value of the jth indicator.

(3) Calculate the comprehensive risk cloud.

Based on the combination weighting results, the comprehensive cloud algorithm can calculate the digital features of the comprehensive cloud model. The formula is as follows [28].

$$\left\{\begin{array}{l}{E}_x^{\ast }={\displaystyle \sum _{j=1}^n{\omega }_j{E}_{xj}}\\ E_n^{\ast }={\displaystyle \sum _{j=1}^n\sqrt{{\omega }_j^2{E}_{nj}^2}}\\ H_e^{\ast }={\displaystyle \sum _{j=1}^n\sqrt{{\omega }_j^2{H}_{ej}^2}}\end{array}\right.$$

(15)

Based on the formula above, the risk cloud of indicators and comprehensive weights can be combined to realize a comprehensive evaluation of the risk level of hazardous chemical road transportation accidents.

3.2.3. Risk Assessment of Road Transport Accidents Involving Hazardous Chemicals

On the basis of the determined risk assessment index system for hazardous chemical road transportation accidents, each index is processed, the subjective and objective weights are determined separately, and weight combinations are carried out. Cloud model theory is introduced, and a risk assessment model for hazardous chemical road transportation accidents based on cloud models is established. Finally, with the help of MATLAB2020a software, a comprehensive cloud map is output through a backward cloud generator, and this map is compared with the standard cloud map to ultimately determine the risk level of hazardous chemical road transportation accidents. The specific risk assessment process for road transportation accidents involving hazardous chemicals is shown in Figure 4.

As shown in Figure 4, based on the data collection of expert ratings, cloud digital feature values are obtained through a forward cloud generator, and then combined with a backward cloud generator to determine the combination weighting of weights to obtain the standard cloud and comprehensive cloud, ultimately determining the accident risk level.

4. Example Application

4.1. Overview of an Enterprise

A road transport enterprise for hazardous chemicals in Hunan Province was selected as the object of accident risk assessment. The company is a professional logistics enterprise engaged in dangerous goods warehousing, transportation, distribution, logistics, and financial integration services. The company owns more than 60 sets of controllable tank vehicles, including 15 new 45 m stainless steel tank cars purchased in 2011. The company has a professional hazardous chemical transportation management team. All the vehicles use information management methods and are equipped with a GPS/GIS positioning system and video surveillance system to realize the whole process of visual safety supervision, ensuring the safe and efficient driving of the vehicles.

The combined weighting-cloud model was used to evaluate the risk of road transport accidents involving hazardous chemicals in this enterprise.

4.2. Calculation of Comprehensive Weights

The subjective weights were determined by the analytic hierarchy process, and then the objective weights were determined by the entropy weight method. Finally, the linear weighting method was applied to find the comprehensive weights. The specific values of the calculated weighting coefficient results are shown in

Table 4. Calculated value of weighting factor.

Subjective Weighting Factor	Numerical Value	Objective Weighting Factor	Numerical Value
${\alpha }_0$	0.34	${\beta }_0$	0.66
${\alpha }_A$	0.32	${\beta }_A$	0.68
${\alpha }_B$	0.33	${\beta }_B$	0.67
${\alpha }_C$	0.30	${\beta }_C$	0.70
${\alpha }_D$	0.38	${\beta }_D$	0.62
${\alpha }_E$	0.20	${\beta }_E$	0.80
${\alpha }_F$	0.30	${\beta }_F$	0.70

, and the specific results of the comprehensive weights are shown in

Table 5. Comprehensive weights of road transport safety risk assessment indexes of hazardous chemicals.

First-Level Indicators	Subjective Weights	Objective Weights	Comprehensive Weights	Secondary Indicators	Subjective Weights	Objective Weights	Comprehensive Weights	Sorting
A	0.2950	0.3437	0.3271	A1	0.0354	0.0628	0.0538	5
				A2	0.0969	0.1533	0.1349	1
				A3	0.1131	0.0927	0.0989	3
				A4	0.0496	0.0349	0.0395	12
B	0.1291	0.0789	0.0960	B1	0.0861	0.0246	0.0452	8
				B2	0.0430	0.0543	0.0508	6
C	0.0964	0.1164	0.1096	C1	0.0299	0.0496	0.0437	9
				C2	0.0189	0.0317	0.0274	17
				C3	0.0476	0.0351	0.0385	13
D	0.2468	0.1539	0.1855	D1	0.0733	0.0275	0.0435	10
				D2	0.0403	0.0337	0.0351	14
				D3	0.1332	0.0927	0.1069	2
E	0.0591	0.1417	0.1136	E1	0.0118	0.0263	0.0206	20
				E2	0.0236	0.0545	0.0455	7
				E3	0.0119	0.0306	0.0237	19
				E4	0.0118	0.0303	0.0238	18
F	0.1736	0.1654	0.1682	F1	0.0212	0.0872	0.0676	4
				F2	0.0735	0.0264	0.0406	11
				F3	0.0394	0.0275	0.0312	15
				F4	0.0395	0.0243	0.0288	16

.

Where α₀ is the subjective weight of the first-level indicator, and β₀ is the objective weight of the first-level indicator. α_A is the subjective weight of the secondary indicator, and β_A is the objective weight of the secondary indicator, and so on.

4.3. Comprehensive Risk Cloud Model

(1) Standard cloud

The risk standard cloud of hazardous chemical road transportation accidents is constructed, and the cloud diagram is drawn using MATLAB to obtain the risk level standard cloud, as shown in Figure 5.

(2) Indicator risk cloud

The numerical characteristic values of each indicator are calculated and substituted into Equation (14) to determine the weighted cloud numerical characteristics of each level of indicators, as shown in

Table 6. Cloud digital characteristic values of first-level indicators.

Tier 1 Indicators	Cloud Characteristic Values
A	(5.987, 0.537, 0.183)
B	(4.675, 0.305, 0.090)
C	(3.484, 0.616, 0.149)
D	(5.041, 0.602, 0.312)
E	(2.921, 0.480, 0.148)
F	(5.164, 0.440, 0.140)

. The cloud diagram is generated using MATLAB, as shown in Figure 6.

As shown in Figure 6A–F are the first-level evaluation indicator cloud maps of factors such as personnel, vehicle, equipment, chemical, environmental, and management. The risk level is obtained by comparing with the standard cloud map.

(3) Comprehensive risk cloud

The numerical eigenvalues and comprehensive weights of the first-level indicators calculated in the previous step are substituted into Equation (15) to obtain the numerical eigenvalues of the comprehensive cloud (4.924, 0.512, 0.183), and the comprehensive risk cloud is plotted, as shown in Figure 7.

As shown in Figure 7, the distribution of cloud drops in the comprehensive risk cloud is between “medium risk” and “high risk”, with a slight bias toward the “medium risk” level. Therefore, it can be judged that the risk level of the enterprise’s hazardous chemical road transportation accidents is “medium risk”. During the road transport of hazardous chemicals, the risk of transport accidents still needs to be controlled in order to be effectively prevented and controlled.

5. Analysis of the Evaluation Results and Countermeasures

5.1. Analysis of the Indicator Weights

As shown in Table 5, the order of the ratio of the comprehensive weights of the first-level indicators is personnel factor > hazardous chemical factor > management factor > environmental factor > equipment factor > vehicle factor. The personnel factor in the road transport of hazardous chemicals accounted for 32.71% of the weight, the largest proportion, as the primary consideration in the transport process. The hazardous chemical factor is second only to the personnel factor, accounting for approximately 18.55%. The evaluation index weight of the vehicle factor is the smallest, accounting for approximately 9.6%, but the index remains a valuable component of the evaluation process, and is an influencing risk factor in transportation that cannot be ignored.

The top five evaluation indicators with the greatest indicator weights in the secondary indicators are ranked as follows: psychological quality > carrying capacity of dangerous chemicals > professional and technical ability > safety education and training > safety responsibility awareness. Among them, the indicator weight of psychological quality is the largest, and it is the most critical risk factor in the road transportation of hazardous chemicals. Hazardous chemical transport enterprises need to focus on the psychological quality of operators. Among the secondary indicators, the density of people and property along the route, traffic conditions and road conditions accounted for the smallest weight. However, in the overall system of risk indicators of the road transport of dangerous chemicals, this weight is not negligible. The density of people and property along the road in the process of the road transport of dangerous chemicals accounted for the smallest proportion, but the density of people and property along the road is one of the risk factors that drivers must pay attention to in real time during the driving process. The driver can plan a suitable time or route for transport tasks to avoid areas with a dense population and dense property, thereby reducing the risk of transport accidents.

5.2. Analysis of the Evaluation Results

(1) Based on the numerical characteristics of the primary index cloud and the risk cloud diagram, we know that the environmental factor is a “low risk”, the equipment factor and vehicle factor are a “medium risk”, and the hazardous chemical factor, management factor, and personnel factor are a “high risk”. Therefore, for low-risk environmental factors, weather conditions and transportation routes should be considered. For medium-risk equipment and vehicle factors, equipment and vehicle inspections need to be strengthened. However, due to the significant threat posed by high-risk factors, targeted measures can be taken for personnel, management, and hazardous chemical factors to improve the management of hazardous chemical road transportation.

(2) According to the comprehensive cloud diagram, compared with the distribution of cloud drops of the standard cloud, the comprehensive cloud drops are more dispersed and discrete, and the thickness of the cloud is larger; that is, the entropy value and super entropy of the comprehensive cloud are larger, which indicates that the risk uncertainty of the enterprise’s hazardous chemical road transportation system is larger and the stability is poorer. The comprehensive cloud risk level is influenced by the risk factors for evaluation indices at all levels together. Through the numerical characteristics of each index evaluation cloud, it can be found that there are still higher risks in personnel, management, and dangerous chemicals, which need to be controlled in a focused manner.

(3) The obtained evaluation results are consistent with the assessment results of the production safety risk classification and grading of transportation enterprises by local transportation authorities, indicating that the evaluation model has good credibility and practicality.

5.3. Countermeasures and Suggestions

According to the results of the analysis, to reduce the possibility of accidents in the road transport of dangerous chemicals, the first level of indicators in the “high-risk” factors can be targeted to take targeted prevention and control measures to improve road transport safety levels.

(1) For transport personnel

Focus on transport personnel, encourage transport personnel to actively participate in training, enhance safety awareness, learn the basic knowledge of road transport safety of hazardous chemicals and the handling methods of emergencies, and improve the level of relevant business knowledge. Relevant responsible persons should pay attention to the physical health and emotional state of drivers, escorts, and other transport personnel to reduce the possibility of road transport accidents involving hazardous chemical factors and to improve the level of transport safety.

(2) Enterprise management

Enterprises should strengthen the management of safety training education and training, safety management systems, safety supervision, and emergency rescue management, increase the content of driver safety training, improve the internal safety management system, strengthen the safety supervision of the road transportation of hazardous chemicals, take effective measures in time when problems are encountered, and pay attention to learning emergency rescue-related skills, with a view to fundamentally curbing the occurrence of transport accidents and ensuring safe transportation.

(3) Chemicals

The hazardous chemical factor exhibits the highest super entropy, indicating significant instability; that is, the risk is unstable. Therefore, enterprises should pay attention to the characteristics of dangerous chemicals, choose packaging materials that meet the standards, and not overload vehicles to transport dangerous chemicals. When carrying out loading, the packaging status of dangerous chemicals should be checked one by one to ensure that the packaging is intact, safe, and effective to reduce the possibility of accidents in the transport of dangerous chemicals and to improve transport safety.

6. Conclusions

Aiming to address the ambiguity and randomness of the risk of hazardous chemical road transport accidents, a risk evaluation model for hazardous material road transport accidents based on the combined empowerment-cloud model is proposed, and the study obtains the following conclusions:

(1) Through risk identification to obtain the risk impact factors of hazardous chemical road transport accidents and to establish an evaluation index system, the combination of the analytic hierarchy process and the entropy weight method is used to realize a combination of subjective and objective evaluation index weights so that the evaluation index weights are more reasonable.

(2) A risk evaluation model for hazardous chemical road transportation accidents based on the cloud model is built. It takes into account the ambiguity and randomness of risk factors and makes the evaluation results more intuitive through cloud diagrams.

(3) Through example application, the reasonableness and feasibility of the evaluation model are verified, and the obtained evaluation results and countermeasures can provide a reference for hazardous chemical road transportation enterprises to prevent accidents and improve their safety management level.

(4) The risk assessment model of hazardous chemicals road transportation accidents based on the combined empowerment-cloud model is still insufficient. This method does not consider the coupling between uncertain risk factors in hazardous chemicals road transportation accidents, and further research is needed.