3.2. Evaluation Analysis of Cloud Model of Each Company
According to the ESEP evaluation framework constructed above, this study adopts the comprehensive evaluation cloud model to conduct equivalent evaluation of each sample company. The brief evaluation steps are as follows:
(1) AHP-EW method is selected to determine the factor subset of each index weight. (1) Firstly, on the basis of fully combing and referring to the ideas and methods of AHP, the subjective weight is obtained according to the operation steps of AHP; (2) Secondly, on the basis of obtaining relevant index data, the objective weight is obtained according to the operation steps of the entropy weight method (Formulas (1)–(4)). After getting the subjective weight by AHP and the objective weight by entropy weight method, calculating according to Formula (5), the comprehensive weight of SHEE management evaluation of mineral resource-based listed companies can be obtained.
(2) According to the data value range of each index, determine the evaluation grade theory domain. By referring to relevant literature, this paper divides each indicator into five grades, which are used to evaluate the level of a company in a certain index: I-alert level, II-improvement level, III-transition level, IV-acceptable level, V-claimable level. Specific index levels are divided as follows: Taking index X1 (degree of perfection of mechanism system) as an example, the level I interval is [1, 1.5], the level II interval is [1.5, 2.5], the level III interval is [2.5, 3.5], the level IV interval is [3.5, 4.5], and the level V interval is [4.5, 5]. In the same way, all index grades can be obtained according to the Formula (10).
(3) According to Formula (6), the evaluation level corresponding to each indicator is represented by the corresponding cloud parameters . Taking indicator X1 (degree of perfection of mechanism system) as an example, the parameters of level I interval cloud model are . The parameters of the II level interval cloud model are . The parameters of level III interval cloud model are . The parameters of IV level interval cloud model were . The parameters of the V level interval cloud model were ; Similarly, according to Formula (10), cloud parameter matrices of all indicators at all levels can be obtained.
(4) Taking the screened indicator data and acquired cloud digital characteristic values as parameters, and the X-conditional cloud generator in the model is used to input the algorithm program into Matlab2014 software for calculation, so as to obtain the membership degree of an experiment. In order to improve the accuracy and credibility of the data, the number of experiments was set as K = 2000, and the final membership degree could be obtained according to Formula (7). Due to space limitations, the membership calculation results of SINOPEC in 2017 are taken as an example (see
Table 4).
Comprehensive evaluation results vector are obtained by computing Formula (8): {0.0000, 0.0000, 0.4582, 0.4657, 0.0761}, based on the principles of maximum membership degree, corresponding to the maximum membership degree of evaluation grade as a result of comprehensive evaluation, that is, the comprehensive evaluation results for IV SINOPEC in 2017 indicate that its ESEP management level is at an acceptable level.
Similarly, the evaluation cloud level of all sample companies can be obtained, and the company level can be visualized after quantitative processing, as shown in
Figure 5.
Figure 6 shows that the ESEP management level of most listed companies in the energy industry is between level II and III, indicating that the ESEP management level of most companies is between “transition level” and “improvement level”. Further statistics on the number of samples at all levels showed that 1.32% (S = 5) of the samples belonged to class V, indicating that their ESEP management level reached the “claimable level”; 15.87% (S = 60) of the samples belonged to level IV, indicating that the ESEP management level reached the “acceptable level”; 56.611% (S = 214) of the samples belonged to level III, indicating that their ESEP management level reached the “transition level” level; 24.07% (S = 91) of the samples belonged to level II, indicating that their ESEP management level was at the “improvement level”; 2.11% (S = 8) of the samples belong to level I, indicating that their ESEP management level is at the “alert level”. Further research shows that different industries have different ESEP management levels. The ESEP management levels from high to low are the coal mining and washing industry, oil and natural gas extraction industry, gas production and supply industry, water production and supply industry, power and heat production and supply industry. Among them, the coal mining and washing industry, oil and gas industry, electricity, heat production and supply industry, gas production and supply industry, water production and supply industry of 2018 ESEP management benchmarking enterprise respectively for China Shenhua (V), SINOPEC (IV), China Yangtze Power (IV), Shenzhen Gas (IV), Grandblue Environment (IV), etc. Some studies have found that the internationalization of the board of directors would enhance the tendency of listed companies’ green business behavior [
39], and the incentives of championships would also have a positive impact on the CEOs of listed companies to take environmental responsibility [
40]. In the future, it can try to improve the level of energy saving and environmental protection practices of listed companies by guiding the internationalization of their boards of directors and actively carrying out ESEP activities in bidding competitions.
3.3. Evaluation and Analysis of Each Indicator Cloud Model
Based on the screening index data, this study uses cloud generator in the cloud model, inputs the algorithm program operations into Matlab2014 software, sets all samples of each target cloud characteristic parameters (see
Table 4), and sets cloud characteristic parameters of the criterion layer and target layer in turn by fuzzy arithmetic according to the Formula (8) (see
Table 5).
After calculating, the cloud characteristic parameters of ESEP management are (2.7598, 0.0019, 0.1199). Based on the cloud characteristic parameters obtained above, combine with the cloud evaluation scale (Formula (6)), and use the forward cloud generator in the model to input the algorithm program into Matlab2014 software for calculation, so as to get the evaluation cloud map of target layer and criterion layer (see
Figure 6).
As can be seen from
Figure 6, the expected value of the comprehensive cloud of energy saving and environmental protection evaluation of listed companies in the energy industry
falls between the “improvement level” and the “transition level”, and it is more inclined to the “transition level”. It can be seen that the energy conservation and environmental protection management of the energy industry is at the level between the “improvement level” and the “transition level”. In addition, the entropy
of the evaluation result cloud is much smaller than that of the evaluation cloud, so it can be concluded that the evaluation result has a small range and good stability, reflecting that there is little difference between listed companies in energy conservation and environmental protection management, which may be caused by the fact that most companies are weak in energy conservation and environmental protection management. The result shows that
is relatively large, reflecting that cloud thickness is larger than the evaluation cloud, indicating that the energy conservation and environmental protection management of each company needs to be improved.
Similarly, cloud model graphs of B1-ESEP governance framework, B2-ESEP management implementation process, B3-ESEP governance efficiency, B4-ESEP public welfare and other criteria can be obtained, as shown in
Figure 7.
This study further visualized
and its standard deviation in cloud model parameters for each indicator. It can be seen from
Figure 8 that the cloud expectation value of most indicators fluctuated up and down the dividing line of level II~III, among which C2- management culture had the highest expectation value. This is followed by C3-management system, C4-clauses and policies, C1-institutional system, C8-energy efficiency management, and C9-tackling climate change management, indicating that most listed companies perform better in these aspects. It is worth noting that the C18-ESEP influence, C17-ecological environment construction, C16-greenhouse gas emissions, C14-energy consumption, and C13-waste emissions are weak. This indicates that the C18-ESEP influence, C17-ecological environment construction, C16-greenhouse gas emissions, C14-energy consumption, and C13-waste emissions are the key to further improving energy conservation and environmental protection management.
This study further analyzes the original data of companies of all levels to clarify the focus of improvement of companies at all levels. The specific results are shown in
Table 6.
3.4. Limitations
In the construction of the ESEP index system and quantitative research, this study strives to be scientific and rigorous, but there are still some deficiencies due to the limitations of many factors, and the specific limitations are as follows.
(1) The evaluation framework system and its indicators need to be further supplemented and modified. Due to the restriction of data availability, the index system itself cannot fully guarantee that it covers all the evaluation indicators reflecting ESEP management level, especially the evaluation of ESEP management performance. With the deepening of people’s understanding of ESEP management, related evaluation indicators would be further expanded.
(2) The method of data acquisition needs to be further expanded. The ESEP management evaluation information in this paper mainly comes from the social responsibility report, sustainable development report, CSMAR database and company website issued by listed companies, which may lead to incomplete ESEP management information.
(3) The rationality of the evaluation results needs to be further verified. As some companies have adopted non-disclosure or selective disclosure in ESEP management, the evaluation results of this study may not fully represent the ESEP management level of these companies, and more comprehensive information can be collected by further combining questionnaire survey and other methods in subsequent research.