Evaluation of Coal Supply and Demand Security in China and Associated Obstacle Factors

Abstract

In China, coal security issues are strongly linked to national energy security and economic and social stability. Facing environmental protection constraints, research on China’s coal security should analyze both supply and demand security. In this study, 19 criteria were selected for four subsystems (coal supply chain, coal market, economy and demographics, and social ecology) to construct an evaluation system for China‘s coal supply and demand security. Entropy and TOPSIS methods were used to evaluate coal security for 2002–2019. The obstacle factors affecting coal security for each subsystem were determined, and the grey model was used to predict the security and obstacle degree for each subsystem for 2020 and 2021. The results indicate that (1) China’s coal supply and use safety level in the period 2002–2019 was below a relatively safe level, i.e., at a safe early warning level during the period 2010–2014 and at a general safety level in the remaining years. (2) The basic coal reserves, the reserve-production ratio of the basic reserves, the balance of social coal stocks at the beginning and the end of the year, and the proportion of coal imports to consumption, urbanization rate, carbon dioxide emissions, and coal consumption in thermal power generation are the main obstacle factors that affect the safety of coal supply and use in China. (3) The obstacle degree of the coal production and supply evaluation subsystem is higher than that of the other three sub-systems, and the overall change trend during the study period showed a downward trend at first, followed by an upward trend.

1. Introduction

China is currently the world’s largest producer and consumer of energy, and coal is the main energy source, with the highest degree of stability, economy, and independent guarantee in China. For a long time, coal has occupied a dominant position in China’s energy production and consumption structure, which has served as a strong impetus for China‘s rapid economic and social development. According to data from the National Bureau of Statistics of China, from 1949 to the end of 2021, the coal industry generated nearly 96.5 billion tons of coal, with the annual output increasing from 34.32 million tons in 1949 to 680 million tons in 1978 and to 4.07 billion tons in 2021, while China’s gross domestic product (GDP) rose from CNY 364.5 billion in 1978, which was in the early days of reform and the opening of the country to international markets, to CNY 114.4 trillion in 2021. The numerous advantages of coal for China, such as its availability, production capacity, consumption proportion, and low cost, mean that there is unlikely to be a short-term shift in China’s coal-based energy system.

However, given the complex and changeable nature of the international environment, China’s economy has begun to move toward a low-carbon future based on its dual carbon goals (i.e., to reach peak carbon emissions in 2030 and to become carbon neutral by 2060). While China’s energy security strategy is now promoting the development of green and low-carbon technologies, it remains necessary to maintain the role of coal in guaranteeing energy security during this energy transformation. On 22 March 2022, Han Zheng, a member of the Standing Committee of the Political Bureau of the CPC Central Committee and Vice Premier of the State Council, stressed at a symposium on the clean and efficient utilization of coal that “We must persist in promoting clean and efficient utilization of coal from the actual situation of our country, give full play to the role of coal as the bottom guarantee, and ensure the national energy and electricity security”. Thus, it is necessary to clearly understand the dominant position of coal in China’s national energy system, consolidate its role in preserving the energy security of the country, and improve the security of coal supply and demand in order to promote the smooth transition to a more sustainable and environmentally friendly energy system while maintaining a safe and stable supply of energy.

However, China’s coal supply chain and coal market mechanisms are not perfect. In recent years, due to the influence of global energy patterns, international trade friction, coal import policies, spatial changes in coal production and demand, environmental constraints of carbon emissions, and the conflict in supply and demand between coal and electricity, China’s coal supply and demand security has been consistently under threat, particularly with the emergence of coal supply shortages and the imbalance between supply and demand. China’s energy security does not rely only on the energy supply; energy consumption is a major consideration as well. Similarly, coal security concerns both supply and demand. On the one hand, a consistent supply of coal is a basic guarantee for China’s energy security; thus, it is important to identify and address any shortcomings in the coal supply chain and promote the balanced development of the coal market. On the other hand, in response to the requirements for the development of high-quality technologies that can contribute to meeting the dual-carbon target and actively respond to climate change, the impact of coal demand on the economy, society, and the ecological environment also needs to be considered.

In this paper, we study coal security in China by identifying the factors affecting both the supply of and demand for coal and presenting corresponding policy suggestions. This research is important for ensuring the safe and stable supply of coal and coal-generated power in China, promoting the development of the coal industry, maintaining the role of coal in national energy security, and providing solid support for stable and sustainable economic development. Based on the research results, we present policy suggestions that address the weak links in China’s coal supply chain and promote the balance between supply and demand in the coal market. Additionally, this research provides a policy basis for the development of high-quality energy infrastructure, coping with climate change, reducing the adverse impact of coal use on the social ecology, and promoting the security of coal supply and demand in order to support the role of coal in ensuring China’s energy security.

2. Literature Review

The two oil crises in the 1970s threatened the energy security of various countries; thus, energy security has become a focus of many researchers worldwide. Initially, energy security research focused on oil security, with the intention of preventing oil supply interruptions and ensuring oil supply security [1]. However, in accordance with the constant changes in the global energy structure, research on energy security in various countries has shifted from oil security to overall energy security, encompassing oil, coal, natural gas, and electric power, in addition to individual studies on various energy sources.

Energy security is vital to the economic and social development of all countries. Early research on energy security centered on the security of energy supplies [2], defining energy security as obtaining an adequate energy supply at a reasonable price. This definition has received widespread recognition from international energy research institutions such as the International Energy Agency (IEA) [3,4] and the Asia Pacific Energy Research Center [5]. With the development of the economy, society, science, and technology, and the rise in public awareness regarding environmental protection and sustainable development, the definition and connotations of energy security have expanded. For example, Radovanovic et al. [6] pointed out that energy security should include environmental and social perspectives, and the sustainability of energy demand has also become the focus of attention. Consequently, the traditional perspective of supply security has evolved into a comprehensive energy concept aimed at supply and demand security. In particular, Wei et al. [7] defined energy security as an available, affordable, and sustainable energy supply that meets the needs of national economic development while ensuring that the production and use of this energy will not damage the sustainable development of the ecological environment. In recent years, quantitative research on energy security in China and internationally has steadily increased, and the research methods involved in the evaluation of energy security have diversified [8]. A summary of significant representative studies on the comprehensive evaluation of energy security both domestically and internationally is presented in

Table 1. Representative research on evaluation indices for energy security in China and internationally.

Reference	Date	Focus	Methods	Index Construction
Cabalu [10]	June 2009	Natural gas supply security	Comprehensive index	Natural gas supply security evaluation indicators, including intensity, net import dependence, the ratio of domestic production to total domestic consumption, and geopolitical risk
Prambudia et al. [11]	May 2012	Energy security	Integrated simulation model	Energy security indicator system comprising 12 specific indicators based on four dimensions: effectiveness, ability to pay, efficiency, and acceptability
Jewell et al. [12]	February 2014	Energy security	Scenario analysis and integrated evaluation model	Energy security indicator system for a long-term energy transformation scenario with multiple specific indicators based on sovereignty, resilience, and robustness
Xue, J.J. et al. [1]	May 2014	Energy supply security	Entropy, set pair analysis, obstacle degree	Comprehensive evaluation index for China’s energy supply security with 13 specific indicators based on energy availability, affordability, efficiency, and technology research and development
Tian, S.Z. et al. [13]	September 2014	Coal security	AHP, entropy, and TOPSIS	Coal security evaluation index with 16 specific indicators based on resources, the market, the economy, and social factors that affect coal supply and demand security
Wang, X.Q. et al. [14]	October 2014	Coal security	Improved integrated evaluation using AGA–EAHP–EM–GRA–TOPSIS	Coal security evaluation system with 16 indicators based on four dimensions: coal availability, accessibility, sustainability, and technical development
Chen, Z.R. et al. [15]	July 2015	Energy security	Entropy weight, matter–element extension model	Energy security evaluation system with 20 specific indicators based on five elements of energy security: driving forces, pressure, the state, influences, and policy responses
Xue, J.J.et al. [16]	October 2015	Regional energy supply security	Entropy, comprehensive scoring, and the grey prediction model	Regional energy supply security evaluation index system with 11 indicators based on energy availability, affordability, efficiency, and technology research and development
Kisel et al. [17]	August 2016	Energy security	Matrix technique	Energy security matrix with related short-term (operation and technical flexibility) and long-term (technical vulnerability, economic dependence, and political influence) indicators for energy security
Meng, C. et al. [18]	August 2016	Coal safety	Back propagation	Chinese coal security evaluation system with nine specific indicators based on resources, supply and demand, transportation, disasters, environment, and market factors that affect coal security
Fan, A.J. et al. [19]	February, 2018	Energy security	Factor analysis	Comprehensive evaluation system for China’s energy security with 14 measurement indicators based on three dimensions: energy supply, consumption, and environmental security
Sun, H. et al. [20]	April 2018	Regional energy security	Entropy–TOPSIS and difference analysis	Evaluation system for China’s regional energy security based on four areas: energy supply security, use security, economic security, and environmental security
Xu, X.M. et al. [21]	September 2018	Vulnerability of coal market supply	Factor analysis	Evaluation system for regional coal market supply vulnerability that includes five indicators: coal production–sales ratio, storage–sales ratio, external dependence, consumption intensity, and coal consumption proportion
Li, P. et al. [22]	January 2019	Regional energy supply security	Vertical and horizontal grading and secondary weighting	Evaluation index system for energy supply security with 11 specific indicators based on the four dimensions of availability, affordability, environmental acceptability, and energy technology and efficiency
Li, Y.H. et al. [23]	April 2020	Natural gas security	4-As radar chart	Evaluation index system for natural gas energy security constructed from four dimensions: resource availability, technology availability, environmental affordability, and national affordability.
Wang, D.Q. et al. [24]	October 2020	Energy security	Entropy weighted aggregation and sensitivity analysis	In the DESI model, based on different energy uses, an evaluation index system for China’s energy security constructed based on three dimensions: energy supply, energy consumption, and environment
Zhang, J. et al. [25]	November 2020	Natural gas security	Entropy–fuzzy comprehensive evaluation	Risk evaluation index for natural gas investment based on six dimensions: political environment, economic environment, commercial environment, resource potential, environmental constraints, and bilateral relations, and 26 indicators. Additional safety index system for the natural gas supply with nine indicators
Shi, D. et al. [8]	January 2021	Primary energy security	Entropy–TOPSIS distance function model and grey relational analysis	China’s energy security index based on energy availability, the economy, cleanliness, and sustainability with 27 indicators

, with the format and style of the table referring to references [9].

International quantitative research on the evaluation of energy security has mainly focused on energy security indicators and multi-index comprehensive evaluation systems for global or national energy security, with most targeting oil and natural gas or the energy system in general. In contrast, there have been few studies on coal security. At present, the most widely used energy security indicators include the Shannon–Wiener index, the Herfindahl–Hirschman index, the energy price index, energy dependence on foreign countries, and the energy consumption intensity. Although these indices focus on the key points of the energy security problem, they do not comprehensively and objectively reflect the complexity of energy security in practical applications, and subsequently, cannot be used to comprehensively evaluate energy security [9]. A multi-index evaluation framework should reflect the multifaceted and complex nature of energy security systems along multiple dimensions and should comprehensively measure and analyze energy security. For example, Cabalu [10] constructed an index system for natural gas supply security and used it to evaluate the natural gas supply security of seven natural gas-importing countries in Asia. In addition, Prambudia et al. [11,12,17] applied an integrated simulation model and the matrix method to construct an energy security evaluation framework based on different dimensions.

Domestic quantitative research on the evaluation of energy security has mainly developed in two directions: the comprehensive evaluation of China’s energy security as a whole and the assessment of primary energy sources such as coal, oil, and natural gas. Xue et al. [1,8,15,19,20,22] included different types of energy, such as coal, oil, natural gas, and clean energy, in their analysis of energy security and created a comprehensive energy security evaluation index based on different dimensions to evaluate, analyze, and predict national or regional energy security. In contrast, Li et al. [24] constructed a quantitative analysis framework for the energy security of natural gas only based on different dimensions, whereas Tian et al. [13,14,18,21] constructed a coal security evaluation index consisting of a number of factors to systematically study the coal supply and demand in China. These assessments have employed a variety of methods for determining weights for their respective indices, including the entropy method, the analytic hierarchy process (AHP), factor analysis, vertical and horizontal grading, set pair analysis, the technique for order of preference by similarity to ideal solution (TOPSIS), and BP neural networks, with one study combining associative group analysis (AGA), the express analytic hierarchy process (EAHP), expectation maximization (EM), grey relational analysis (GRA), and TOPSIS.

Despite the extensive quantitative research that has been conducted on domestic energy security evaluation, this research has a number of shortcomings. First, most of this research, both in China and worldwide, has focused on energy security as a whole, and only a few studies have addressed the security of the coal supply and demand. Second, domestic research on coal safety has mainly focused on the evaluation of coal resources, the market, the economy, society, the environment, and other factors from a macro perspective, or have based their evaluation on the dimensions of availability, sustainability, and technological development. Thus, comprehensive evaluation research that combines the macro and meso perspectives is lacking. Third, research on coal security has primarily investigated the coal supply and production, and few studies have considered current economic and social ecological requirements or comprehensively evaluated the security of the coal supply and demand based on the comprehensive energy view. Finally, there has been little research on the in-depth identification and analysis of factors affecting security levels, especially in the study of coal security.

Due to differences in economic and social development, energy reserves, energy consumption structure and scale, energy policies and systems, ecological and environmental constraints, and the ability to access international resources, individual countries differ in their strategic positions with regard to the security of fossil-based energy sources such as oil, coal, and natural gas. In China, coal remains an irreplaceable source of energy, and it is important to analyze the security of the coal supply and demand. The present study is based on domestic and international research on security evaluation indices for international and domestic energy systems and on primary energy security, such as oil, coal, and natural gas. Given China’s national energy situation and energy reform policies, it is of great theoretical and practical importance to construct a comprehensive evaluation index for coal supply and demand security that encompasses economic, social, environmental, coal market, and coal supply chain dimensions and to identify significant obstacles to coal supply and demand security.

The main goals of the current research are as follows. First, from traditional overall systematic research on energy security to independent research on coal supply and demand security, the current study seeks to enrich energy security research. Second, to extend traditional coal supply security research, this study comprehensively examines the coal supply and demand security and its obstacle factors. Third, in order to understand the changes in the security of China’s coal supply and demand in recent years, the main influencing factors and weak links in the coal supply system, and the adverse effects of coal demand, it is necessary to develop a criteria system and identify obstacles to conduct a quantitative analysis to take effective targeted measures to enhance risk prevention and control coal security. Fourth, because coal is the main energy source in China, it is difficult to change the domestic energy situation in the short term. Therefore, in-depth research on the security of the coal supply and demand will support coal’s role as an important element in the energy transition process.

In the construction of the coal supply and demand security evaluation index system, this study considers coal resource endowment, production, transportation, inventory, and other supply chain links to form a comprehensive evaluation network chain structure. Based on Xue et al.’s [1,13,16,20] research methods, this study uses the entropy method, which is objectively weighted, and TOPSIS, which is suitable for the comprehensive comparison of a limited number of evaluation objects, multiple evaluation criteria, and multi-objective decisions, to conduct a comprehensive evaluation of the security of the coal supply and demand. Furthermore, to examine the coal security situation in China for the period 2002–2019, this study refers to relevant methods from past studies [1,8,26,27] and utilizes the obstacle degree method to quantitatively analyze the factors that affect the safety of coal supply and use, and identify and analyze the main influencing factors and weak links. In addition, the GM(1,1) model is used to predict the degree of security and the obstacle degree of the selected criteria for 2020–2021. Past study on the comprehensive evaluation of China’s coal security over the past five years is summarized in

Table 2. Comparison of previous research on the evaluation of coal security in China in the past five years and the present study.

Reference	Focus	Level	Scope	Dimensions	Time Frame	Number of Indicators	Weight and Evaluation Model	Consideration of Each Link in the Coal Supply Chain	Predictive Analysis	Analysis of Influencing Factors
Meng, C. et al. (2016) [18]	Coal security evaluation framework	National	Coal supply and demand security	Resources, supply and demand, transportation, disasters, environment, market	1995–2013	9	Principal component analysis (PCA), BP neural network	/	√	/
Mo, Y. C. (2017) [29]	Risk evaluation model for the coal supply chain	Industrial	Coal supply risk	Coal mining, transportation, sales, processing, cooperation, overall operation	Random sample	20	Rough set	√	/	√
Xu, X.M.et al. (2018) [21]	Evaluation system for regional coal market supply vulnerability	Regional	Coal supply security	Security response capability, security of supply sensitivity	2015	5	Factor analysis	/	/	/
Wang, Q. (2020) [30]	Security evaluation system for sustainable livelihoods in coal resource-based areas	Region	Sustainable security of coal livelihoods	Economic benefits, ecological security, social equity	2008–2017	27	Entropy comprehensive evaluation	/	/	√
Yang, Z. Q. et al. (2021) [26]	Evaluation index for the ecological security of coal-based cities	Region	Coal ecological security	Eco-environment, social economy	2008–2017	33	Entropy–TOPSIS	/	/	√
Zhou, L. J. (2021) [31]	Index system for the supply and demand risk assessment of coal supply chain participants	Industry	Coal supply and demand risk	Coal companies, transport, storage and distribution centres, consumers, external environment	2010–2019	34	Entropy and an analytic hierarchy process	√	/	√
Zhao, H; Li, J.Y. (2021) [32]	Analysis of safety risks in the coal industry	Industry	Coal supply risk	Resources, macro-economy, coal price, coal supply and demand, coal transportation, safe production, ecological environment	1999–2019	7	Qualitative analysis	/	/	/
This study	Comprehensive evaluation index for coal supply and demand security	National and industrial	Coal supply and demand security	Coal supply chain, coal market factors, economic and demographic factors, social ecological factors	2002–2019	19	Entropy–TOPSIS	√	√	√

, with the format and style of the table referring to references [9,28].

3. Method

3.1. Index System Construction

3.1.1. Study Overview

This study examines the supply and demand security of coal, which is a complex system influenced by multiple factors. In the present study, we refer to the results of domestic and foreign research on energy supply and demand security. Coal supply security has been a core topic of traditional coal security research. At the macro level, coal supply security is affected by macro-environmental factors such as the economy and society. At the meso level, coal mining, production, capital investment, transportation, storage, and terminal consumption in the coal supply chain all have differing degrees of impact on coal supply security. In addition, combined with the governmental dual-carbon target, the present study investigates coal demand security, with a primary focus on the impact of coal consumption on the economy, society, and the ecological environment. Overall, this study analyzes the main factors affecting coal supply and demand security, constructs a quantifiable index system, and comprehensively analyses the historical evolution of coal security.

3.1.2. Selection of criteria

Based on China’s energy situation, the characteristics of coal supply and demand, and previous research on the evaluation of energy and coal security, we selected a preliminary list of 39 criteria related to the economy, society, and ecology (i.e., the macro level) and to the coal industry chain. Following the principles of scientificity, comprehensiveness, systematicness, typicality, dynamics, consistency, and quantification, and according to energy security theory, this paper referred to (a) the IEA’s definition of energy security, (b) the definition of energy security supply and demand developed by Wei et al. [7], (c) results from Xue et al. [1,8,17,19,20] and other domestic and international research regarding the selection of energy security indicators, and (d) domestic research results regarding the selection of coal security evaluation criteria, e.g., Tian et al. [13,14,18,21,26]. We then consulted experts and scholars in the field of energy and coal to streamline the list further. The resulting index included the two target layers of the coal supply and demand security, four subsystems, and 19 specific criteria (

Table 3. Comprehensive evaluation index for coal supply and demand security in China used in the present study.

Target	Subsystem	Index Layer	Criteria	Unit	Code	Attribute
Coal supply security	Coal supply chain (A)	Coal resource endowment	Basic coal reserves	10,000 tons	A1	Positive
		Remaining exploitable life	Basic reserve–production ratio	Year	A2	Positive
		Coal production and supply	Raw coal output	10,000 tons	A3	Positive
		Production safety situation	Million-ton mortality rate	%	A4	Negative
		Capital profitability	Total profit of coal wholesale products	CNY 100 million	A5	Positive
		Coal transportation capacity	Coal transportation shipment volume (cumulative completion)	10,000 tons	A6	Positive
		Coal stock	Social stock balance	10,000 tons	A7	Positive
	Coal market (B)	Coal price change	Average annual coal price	USD	B1	Negative
		External dependence of coal	Coal import dependence on foreign countries	%	B2	Negative
		Coal consumption	Coal consumption intensity	%	B3	Negative
		Coal supply and demand relationship	Coal supply–demand ratio	%	B4	Positive
Coal demand security	Economy and demographics (C)	Economic growth	Gross domestic product (GDP)	CNY 100 million	C1	Negative
		Population growth	Natural population growth rate	%	C2	Negative
		Urbanization level	Urbanization rate	%	C3	Negative
		Industrial structure	Contribution rate of secondary industry to GDP	%	C4	Negative
		Energy structure	Proportion of non-coal consumption in energy consumption	%	C5	Positive
	Social ecology (D)	Ecological environment	Carbon dioxide emissions	Megaton	D1	Negative
		Coal consumption	Coal consumption during thermal power generation	10,000 tons	D2	Negative
		Power supply structure	Proportion of coal power generation hours in the total power generation hours	%	D3	Negative

).

The evaluation of coal supply security included two subsystems: the coal supply chain and the coal market. The selection of criteria thus reflected the influence of the economic environment on the coal industry. Considering the importance of the coal supply chain on coal supply security, previous research by Mo Yingcong and other scholars [33,34,35,36,37] on the key elements, risk identification, and security evaluation of the coal supply chain was consulted. The selected criteria included coal resource endowment, availability, production, transportation, and inventory. The coal market criteria included market price, import, consumption, and the supply–demand relationship.

The evaluation of coal demand security included two dimensions: the economy/demographics and the social ecology. The selection of the criteria focused on the influence of macro factors, such as the population and the economy, on coal consumption, as well as the influence of coal demand on the sustainable development of the economy, society, and the ecological environment. Of these factors, the influence of coal use on the economy and society mainly considers the demand for a continuous and stable power supply, whereas the influence of coal use on the ecological environment mainly considers carbon emissions. Data for the criteria used in this study were obtained from the China National Bureau of Statistics, the BP World Energy Statistics Yearbook, the Annual Report on Coal Market Development of China (2021), the China Coal Industry Association, and the China Coal Economic Research Association [38,39,40,41]. Because some data for 2020 and 2021 had not yet been released, we used data up to 2019.

3.2. Study Methods

3.2.1. Data Processing

To eliminate inconsistencies between the subsystems and the orders of magnitude and units of each criterion, it was necessary to standardize the original data and to establish the data matrix i = 1, 2, …, n (where n is the number of evaluation years), j = 1, 2, …, m (where m is the number of evaluation criteria). In this paper, we chose the maximum-minimum normalization method because this method performs better than classical vector normalization and sum normalization. However, any other normalization method can also be selected [42,43]. The direction of positive criteria is consistent with the coal supply and demand security in China evaluated in this study. That is, the larger the value, the better, whereas the direction of negative criteria is the opposite, and the smaller the value, the better. Equations (1) and (2) were used for data standardization processing $A={(X_{ij})}_{\mathrm{nm}}$ of the positive and negative criteria, respectively [42]:

$$X_{ij}^{\prime }=\frac{X_{ij}-\min (X_{1j},X_{2j},\cdots ,X_{nj})}{\max (X_{1j},X_{2j},\cdots ,X_{nj})-\min (X_{1j},X_{2j},\cdots ,X_{nj})}$$

(1)

$$X_{ij}^{\prime }=\frac{\max (X_{1j},X_{2j},\cdots ,X_{nj})-X_{ij}}{\max (X_{1j},X_{2j},\cdots ,X_{nj})-\min (X_{1j},X_{2j},\cdots ,X_{nj})}$$

(2)

where $X_{ij}^{\prime }$ is the normalized value of $X_{ij}$, and $X_{ij}$ is the original value of criteria j in year i.

3.2.2. Entropy Weighting

There are two types of methods for determining indicator weights: subjective assignment methods and objective assignment methods, and in view of the strong subjectivity of subjective assignment methods, this paper uses the entropy weighting method to determine the criteria weights to objectively reflect the implicit information in the data. In information theory, information entropy is a measure of uncertainty. The entropy weighting method uses entropy to evaluate the dispersion of criteria according to the information provided by its observed values. The greater the dispersion, the greater the influence of the criteria on the overall evaluation index, and the higher the weight assigned. For evaluating the security of supply and demand of coal in China, the greater the gap between observations of a particular evaluation indicator, the less stable that criterion is, and the greater the impact of uncertainty on security issues; the entropy value method is useful for identifying such criteria and assigning a higher weight to them. Furthermore, with different normalization methods, the correlation of rankings had a smaller variance when the entropy weighting method was used [43], and it is beneficial to guarantee the robustness of the research results.

First, the characteristic proportion of the evaluation object ${\mathrm{P}}_{ij}$ and the entropy of each criterion in evaluation index $e_j$ were calculated using Equations (3) and (4), respectively [8,15,43]:

$${\mathrm{P}}_{ij}=\frac{X_{ij}\prime }{{\displaystyle \sum _{i=1}^n{X}_{ij}\prime }}$$

(3)

$$e_j=-\frac{1}{\ln n}{\displaystyle \sum _{j=1}^m({\mathrm{P}}_{ij}\times \ln {\mathrm{P}}_{ij})},(0\le e_j\le 1)$$

(4)

The difference coefficient $g_j$ and weight for each criterion $w_j$ were then calculated as follows:

$$g_j=1-e_j$$

(5)

$$w_j=\frac{g_j}{{\displaystyle \sum _{j=1}^m{g}_j}}$$

(6)

3.2.3. Comprehensive Evaluation Index Using TOPSIS

Multi-criteria decision-making (MCDM) techniques are important and popular mathematical methods used in a variety of human activities. One of the most frequently used MCDM methods is the technique for order of preference by similarity to ideal solution (TOPSIS) [42]. The TOPSIS evaluation method makes full use of raw data information, has no special requirements for sample information, is not disturbed by the choice of reference series, and has the advantages of intuitive geometric meaning, low distortion of data information, reliable reflection of results, a wide range of application, flexible and convenient operation, etc., and is widely used in various research fields. TOPSIS is a ranking method that is based on the distance from an ideal value. By calculating the weighted Euclidean distance between each evaluation object and the positive and negative ideal solutions, we can judge how close each evaluation object is to the optimal scheme, which can be used as the standard to rank the advantages and disadvantages. For the TOPSIS method, rankings obtained using maximum–minimum, vector, and other normalization methods could be quite similar, especially for the big number of alternatives, and in this case, the entropy method performed worse than equal weights [43]. In conclusion, this paper selected the combination of the maximum–minimum normalization method, entropy method, and TOPSIS method to conduct a comprehensive evaluation of coal supply and demand security level, which can ensure the stability and reliability of the results. The TOPSIS evaluation method has the following steps [8,20,44].

First, we calculated the weighted normalized weight matrix $V_{ij}$:

$$V_{ij}=w_j·{\mathrm{P}}_{ij},i=\left(1,2,\dots ,n\right) j=\left(1,2,\dots ,m\right)$$

(7)

Second, we calculated the positive and negative ideal solutions ($V^{+}$ and $V^{-}$, respectively) using Equations (8) and (9), respectively:

$$V^{+}=(V_1^{+},V_2^{+},\cdot \cdot \cdot ,V_m^{+})=\left\{\max V_{ij}|j=1,2,\cdot \cdot \cdot ,m\right\}$$

(8)

$$V^{-}=(V_1^{-},V_2^{-},\cdot \cdot \cdot ,V_m^{-})=\left\{\min V_{ij}|j=1,2,\cdot \cdot \cdot ,m\right\}$$

(9)

Third, we calculated the Euclidean distance for each criterion from the positive and negative ideal solutions ($S_i^{+}$ and $S_i^{-}$, respectively):

$$S_i^{+}=\sqrt{{\displaystyle \sum _{j=1}^m{(V_j^{+}-V_{ij})}^2}}$$

(10)

$$S_i^{-}=\sqrt{{\displaystyle \sum _{j=1}^m{(V_j^{-}-V_{ij})}^2}}$$

(11)

Finally, we calculated the relative closeness of each evaluation object to the ideal solution $C_i$:

$$C_i=\frac{S_i}{S_i^{+}+S_i^{-}}$$

(12)

Relative closeness has a range of [0, 1], with a higher value indicating a higher level of security in this study [8,45]. Using this relative closeness score, we divided the security level for China’s coal supply and demand into five grades with a step size of 0.2 (

Table 4. Grading the security level of coal supply and demand in China according to relative closeness.

Security Level	Unsafe/Insecure	Warning	Generally Secure	Relatively Secure	Secure
Relative closeness range	[0, 0.2)	[0.2, 0.4)	[0.4, 0.6)	[0.6, 0.8)	[0.8, 1]

).

3.2.4. Obstacle Degree Model

Using the entropy and TOPSIS methods, the status and trend in China’s coal supply and demand security could be measured for the research period, but the important factors that most strongly affect coal supply and demand security were not identified. Therefore, we utilized the obstacle degree model to analyze the obstacle degree for each subsystem and individual criteria, identify the primary obstacle factors, and present targeted suggestions based on this. The specific steps in the obstacle degree model are described below [46,47,48].

First, factor contribution $F_{ij}$ and deviation $I_{ij}$ were calculated as follows:

$$F_{ij}=W_j{U}_k,$$

(13)

where $W_j$ is the weight of a single criterion, and $U_k$ is the weight of each subsystem.

$$I_{ij}=1-X_{ij}^{\prime }$$

(14)

Next, we calculated the obstacle degree of single criterion $p_{ij}$ as follows:

$$p_{ij}=\frac{F_i{I}_{ij}}{{{\displaystyle \sum }}_{i=1}^m\left(F_i{I}_{ij}\right)}\times 100\%$$

(15)

Finally, based on the analysis of the obstacle degree of individual criteria, the obstacle degree for each subsystem on the security of the coal supply and demand $B_j$ was further calculated as follows [26,49]:

$$B_j=\sum p_{ij}$$

(16)

3.2.5. Grey Prediction Model

This study also utilized a grey prediction model to forecast China’s basic coal reserves and the security of the coal supply and demand for the 2020–2021 period. In this study, a GM(1,1) model was constructed using cumulative generation using the following process [50,51,52].

First, the original data formed the sequence $x^{\left(0\right)}=\left(x^{\left(0\right)}\left(1\right),x^{\left(0\right)}\left(2\right),\cdots ,x^{\left(0\right)}\left(n\right)\right)$, and these data were accumulated once to generate the sequence

$$\begin{gathered} x^{\left(1\right)}=\left(x^{\left(1\right)}\left(1\right),x^{\left(1\right)}\left(2\right),\cdots ,x^{\left(1\right)}\left(n\right)\right)=\left(x^{\left(1\right)}\left(1\right),x^{\left(1\right)}\left(1\right)+x^{\left(0\right)}\left(2\right),\cdots ,x^{\left(1\right)}\left(n-1\right)+x^{\left(0\right)}\left(n\right)\right), \\ {x}^{\left(1\right)}\left(k\right)={{\displaystyle \sum }}_{i=1}^k{x}^{\left(0\right)}\left(i\right),\left(k=1,2,\cdots ,n\right). \end{gathered}$$

(17)

The mean value sequence with $z^{\left(1\right)}$ is given by $x^{\left(1\right) } z^{\left(1\right)}=\left(z^{\left(1\right)}\left(2\right),z^{\left(1\right)}\left(3\right),\cdots ,z^{\left(1\right)}\left(n\right)\right)$, where $z^{\left(1\right)}=0.5x^{\left(1\right)}\left(k\right)+0.5x^{\left(1\right)}\left(k-1\right),\left(k=2,3,\cdots ,n\right)$.

Next, we established the grey differential equation as follows:

$$x^{\left(1\right)}\left(k\right)+a{z}^{\left(1\right)}\left(k\right)=b,\left(k=2,3,\cdots ,n\right)$$

(18)

The corresponding whitening differential equation for the GM(1,1) model was as follows:

$$\frac{d{x}^{\left(1\right)}}{dt}+a{x}^{\left(1\right)}\left(k\right)=b,$$

(19)

where $a$ is the development ash number, and $b$ is the endogenous control ash number.

Parameter vector $\widehat{u}$ was then solved using the least squares method:

$$\widehat{u}={\left(\widehat{a},\widehat{b}\right)}^T={\left(B^{T}B\right)}^{-1}{B}^{T}Y,$$

(20)

where $u={\left(a,b\right)}^{T}B=\left[\begin{array}{c}\begin{array}{cc}-z^{\left(1\right)}\left(2\right) & 1\\ -z^{\left(1\right)}\left(3\right)& 1\\ \cdots & \\ -z^{\left(1\right)}\left(n\right)& 1\end{array}\end{array}\right]Y=\left[\begin{array}{c}\begin{array}{c}{x}^{\left(0\right)}\left(2\right) \\ x^{\left(0\right)}\left(3\right)\end{array}\\ \begin{array}{c}\cdots \\ x^{\left(0\right)}\left(n\right)\end{array}\end{array}\right]$.

Next, substituting the parameter vector, the whitening response equation for GM(1,1) was obtained as shown in Equation (21):

$$\left\{\begin{array}{c} {\widehat{x}}^{\left(1\right)}\left(k+1\right)=\left(x^{\left(0\right)}\left(1\right)-\frac{b}{a}\right)e^{-ak}+\frac{b}{a},\left(k=1,2,\cdots ,n\right)\\ {\widehat{x}}^{\left(0\right)}\left(k+1\right)={\widehat{x}}^{\left(1\right)}\left(k+1\right)-{\widehat{x}}^{\left(1\right)}\left(k\right),\left(k=1,2,\cdots ,n\right)\end{array}\right.$$

(21)

To ensure the reliability of the model, this study utilized the commonly used posterior difference test method to test the reliability. For this purpose, the variance of the original sequence and the residual sequence $S_1^2$ and the deviation of the residual $S_2^2$ were calculated [27,52]; the residual error $e\left(k\right)=x^{\left(0\right)}\left(k\right)-{\widehat{x}}^{\left(0\right)}\left(k\right)\left(k=1,2,\cdots ,n\right)$.

$$S_1^2=\frac{1}{n}{{\displaystyle \sum }}_{k=1}^n{\left[x^{\left(0\right)}\left(k\right)-\overline{x}\right]}^2$$

(22)

$$S_2^2=\frac{1}{n-1}{\displaystyle \sum }_{k=2}^n{\left[e^{\left(0\right)}\left(k\right)-\overline{e}\right]}^2$$

(23)

Subsequently, we calculated the posterior error ratio:

$$C=\frac{S_2}{S_1}$$

(24)

The goal is to achieve a lower posterior error ratio C. The smaller the C, the larger the S₁, and the smaller the S₂, i.e., the larger the original dataset, the smaller the residual error, indicating that the difference between the calculated and the actual values obtained using the model was not too discrete. The accuracy of the grey model was divided into four grades, as shown in

Table 5. Grading of the inspection accuracy of the grey prediction model.

Model Accuracy Level	Value Range for Posterior Error Ratio C
Grade Ⅰ (good)	C ≤ 0.35
Grade II (qualified)	0.35 < C ≤ 0.50
Grade III (barely qualified)	0.50 < C ≤ 0.65
Grade IV (unqualified)	0.65 < C

[53,54].

4. Results and Discussion

4.1. Weight Calculation Results

Using Equations (1)–(6), the entropy weighting method was used to calculate the weights of the individual criteria and each subsystem. It is evident that the subsystems and criteria within these subsystems had different effects on the overall evaluation system. The top five criteria in terms of their weights were the reserve–production ratio of the basic reserves, the proportion of non-coal energy consumption in energy consumption, carbon dioxide emissions, the proportion of coal imports in consumption, and the proportion of total power generation hours from coal (

Table 6. Weight for China’s coal supply and demand security evaluation criteria.

Subsystem	Reference Number	Criteria	Hierarchical Weight	Weight
A. Coal supply chain (0.3368)	A1	Basic coal reserves	0.1375	0.0463
	A2	Basic reserve–production ratio	0.3028	0.1020
	A3	Raw coal output	0.0885	0.0298
	A4	Million-ton mortality rate	0.0675	0.0227
	A5	Total profit of wholesale coal products	0.1376	0.0463
	A6	Coal transportation shipment volume (cumulative completion)	0.1428	0.0481
	A7	Balance of the social inventory of coal at the beginning of the year	0.1233	0.0415
B. Coal market (0.2072)	B1	Average annual coal price	0.1725	0.0357
	B2	Proportion of coal imports	0.3409	0.0706
	B3	Coal consumption intensity	0.2582	0.0535
	B4	Coal supply–demand ratio	0.2285	0.0473
C. Economy and demographics (0.2479)	C1	Gross domestic product (GDP)	0.1417	0.0351
	C2	Natural population growth rate	0.1210	0.0300
	C3	Urbanization rate	0.1944	0.0482
	C4	Contribution rate of secondary industry to GDP	0.1966	0.0487
	C5	Proportion of non-coal energy consumption consumption	0.3461	0.0858
D. Social ecology (0.2081)	D1	Carbon dioxide emissions	0.3861	0.0803
	D2	Coal consumption for thermal power generation	0.2880	0.0599
	D3	Proportion of coal power generation hours to total power generation hours	0.3259	0.0678

).

4.2. Analysis of the Comprehensive Evaluation Results

TOPSIS was used to calculate the Euclidean distance between each evaluation criteria and the positive and negative ideal solutions using Equations (7)–(11). Equation (12) was used to calculate the strength and ranking of China’s coal supply and demand security for the 2002–2019 period (

Table 7. Safety degree, ranking, and security score for coal supply and demand in China.

Year	S+	S−	Security Score	Ranking	Security Level
2002	0.1404	0.1825	0.5652	1	Generally secure [0.4, 0.6]
2003	0.1522	0.1635	0.5179	2
2004	0.1417	0.1506	0.5153	3
2005	0.1571	0.1361	0.4642	6
2006	0.1573	0.1274	0.4474	8
2007	0.1643	0.1160	0.4139	12
2008	0.1532	0.1213	0.4418	10
2009	0.1603	0.1122	0.4117	13
2015	0.1606	0.1202	0.4282	11
2016	0.1634	0.1295	0.4421	9
2017	0.1666	0.1350	0.4475	7
2018	0.1694	0.1470	0.4647	5
2019	0.1751	0.1558	0.4709	4
2010	0.1657	0.1071	0.3926	14	Warning [0.2, 0.4]
2011	0.1898	0.0953	0.3343	18
2012	0.1863	0.1012	0.3519	17
2013	0.1866	0.1026	0.3549	16
2014	0.1774	0.1066	0.3754	15

), with the overall security level classified according to the evaluation criteria presented in Table 3. Overall, China’s coal supply and demand security was below the relatively secure level during the study period; it was at a warning level during the 2010–2014 period and at a generally secure level for the remaining years. During the warning period, the reserve–production ratio for China’s coal resources (A2) and the difference in the coal social stock between the beginning and the end of the period were low. In addition, the proportion of coal imports (B2), the natural population growth rate (C2), the contribution of secondary industry to GDP (C4), carbon dioxide emissions (D1), and coal consumption for thermal power generation (D2) were relatively high. Coal supply and demand security for 2002–2019 is shown in Figure 1.

Overall, the coal supply and demand security level in China initially declined, followed by an increase, with the inflection point occurring in 2011. From 2002 to 2011, China’s economy developed rapidly, and the GDP growth rate was above 9%. With the continuous and rapid increase in coal consumption, the reserve–production ratio for basic coal reserves dropped rapidly from 214 in 2002 to 82 in 2010. This extensive development also led to significantly higher carbon emissions; thus, the security level of China’s coal supply and demand gradually declined during this period. After 2011, as China’s economic development entered a new stage of high-quality development, economic growth slowed, the industrial structure was improved, and the energy structure was optimized. The government also introduced a series of policies such as coal capacity reduction, output control, long-term guarantees, coal price stabilization, and the development of new energy sources. Consequently, China’s energy sustainability improved, the basic reserves and output ratio for coal resources increased, the proportion of non-coal consumption increased significantly, and the number of coal power generation hours increased.

Constructing the GM(1,1) model, the security score of the coal supply and demand for the 2011–2019 period was used in Equations (17)–(24) to estimate the coal security for 2020 and 2021. The predicted relative closeness for coal supply and demand security in China was 0.5058 in 2020 and 0.5320 in 2021, whereas the mean square deviation ratio of the grey prediction was 0.2709. The accuracy level was classified as Grade I, and the fit to the data was good. Combined with the forecast results, the security level of coal supply and demand in China is expected to further improve in the next few years but remain at the generally secure level.

4.3. Obstacle Factor Analysis

Using the obstacle degree model and Equations (13)–(16), the safety obstacle degree for China’s coal supply and demand from 2002 to 2019 was calculated. The seven evaluation criteria with the highest obstacle degrees for each year were selected as the main obstacle factors (

Table 8. Obstacle factors for China’s coal supply and demand security (2002–2019).

Year	Obstacles	Ranking
		1	2	3	4	5	6	7
2002	Factor	B4	A5	A7	A6	C2	C5	D3
	Degree	17.1513	13.8813	11.8326	10.5356	10.0048	9.9214	7.7003
2003	Factor	B4	A5	C5	A7	A6	C4	D3
	Degree	13.6083	11.9292	9.8863	9.7280	8.7207	8.1028	7.6628
2004	Factor	A2	A5	C5	A7	A6	B4	C2
	Degree	10.6749	10.1543	9.8065	9.3999	8.0487	7.8016	6.9608
2005	Factor	A2	B4	C5	A5	D1	A6	D3
	Degree	12.1533	11.5068	9.8922	8.3545	8.0930	6.5284	6.3635
2006	Factor	A2	D1	C5	A5	B4	A7	D3
	Degree	13.4762	9.8018	9.5326	8.2265	7.2504	7.0723	6.6146
2007	Factor	A2	D1	B4	A7	C5	D2	A5
	Degree	13.6310	10.3969	8.6782	8.6023	8.4942	6.8780	6.4435
2008	Factor	A2	D1	A7	C5	B4	D2	B1
	Degree	14.7922	11.0519	9.4525	8.0894	7.8963	7.5695	6.8230
2009	Factor	A2	D1	A7	D2	C5	B2	C3
	Degree	15.5679	11.5988	9.9619	8.1915	7.8097	7.3990	5.6555
2010	Factor	A2	D1	B2	D2	A1	A1	B1
	Degree	16.5943	11.6609	8.5155	8.3695	7.7384	7.6131	6.3271
2011	Factor	A2	A1	D1	D2	B2	C3	A7
	Degree	15.9628	12.9148	10.9245	8.5050	7.5880	6.7183	6.4460
2012	Factor	A2	A1	D1	B2	D2	C3	A7
	Degree	15.8548	11.4121	11.3081	9.5000	9.1336	7.4324	6.2976
2013	Factor	A2	D1	A1	B2	D2	C3	A7
	Degree	15.7508	11.8392	10.7308	10.5392	10.0576	8.0542	6.5237
2014	Factor	A2	D1	A1	D2	B2	C3	A7
	Degree	15.5621	12.0292	10.4156	9.6808	9.6234	8.7091	6.4721
2015	Factor	A2	D1	A1	C3	D2	B2	C1
	Degree	16.8843	13.2811	11.0521	10.4569	9.8325	7.4407	7.1916
2016	Factor	A2	D1	C3	D2	D2	B2	C1
	Degree	15.6538	13.0100	10.9977	10.2223	9.9020	9.6751	7.7606
2017	Factor	A2	D1	C3	D2	B2	A1	C1
	Degree	15.2790	12.9525	11.2624	10.4704	9.8577	9.7421	8.4928
2018	Factor	A2	D1	C3	D2	B2	C1	A1
	Degree	15.6491	13.6107	12.2392	11.6074	10.3057	9.6991	8.9257
2019	Factor	A2	D1	C3	D2	B2	C1	A7
	Degree	15.3641	13.7258	12.4505	11.8101	10.6781	10.3252	8.4902

). Overall, the main obstacles that affected the security of the coal supply and demand in China during the study period were the basic coal reserves (A1), the reserve–production ratio for basic reserves (A2), the social stock balance of coal at the beginning and end of the year (A7), the ratio of coal imports to consumption (B2), the urbanization rate (C3), carbon dioxide emissions (D1), and coal consumption during thermal power generation (D2).

The reserve–production ratio (A2) for China’s basic coal reserves ranked first among the evaluation criteria for coal supply and demand security from 2004 to 2019. This is an important criterion because it reflects the rate of use of China’s basic coal reserves, and it acts as the strongest constraint on coal supply security. According to the analysis, the reserve–production ratio dropped from 159 in 2004 to its lowest point of 57 in 2011. This can be attributed to the rapid decline in basic reserves due to the rapid increase in coal consumption to meet the rapid economic development. After 2011, with the slowdown in economic growth, the increase in basic coal reserves, and the slowdown in coal consumption, the reserve–production ratio for basic coal reserves increased and has remained above 70 since 2016; however, it remained at a low level during the study period.

The difference in the social stock of coal at the beginning of the year (A7) appeared 13 times in the top seven criteria for the 2002–2019 period, and this was the main obstacle factor affecting the security of the coal supply in China. In particular, the difference in coal’s social stock was positive for only five years (2005, 2006, 2015, 2016, and 2017), and negative for the remaining years, indicating that the coal inventory level in China was generally insufficient during the study period, which had an adverse impact on coal supply security.

The ratio of coal imports to consumption (B2) appeared 11 times among the top seven obstacle criteria for the 2002–2019 period, particularly after 2009, indicating that it was one of the main obstacles affecting China’s coal supply security. In 2009, China became a coal importer for the first time. During that year, the coal import volume reached 125.84 million tons, more than three times that in 2002. Since then, the volume of coal imports has continued to expand, with China becoming the largest importer of coal in the world. The ratio of imports to consumption increased from 0.74% in 2002 to 7.46% in 2019 and has been high in recent years, indicating a significant dependence on foreign countries, thus increasing the obstacles to coal supply security.

The urbanization rate (C3) appeared 10 times among the top seven obstacles for 2002–2019, particularly after 2009. The ranking of this obstacle rose consistently during the study period from its lowest ranking of 17th in 2003 to third in 2016, thus increasing its impact on coal demand security in China. After 2009, China’s urbanization rate rose from 46.59% in 2009 to 60.60% in 2019. Urbanization directly leads to a substantial increase in China’s energy demand and coal consumption. Urbanization is inevitable for economic and social development; however, from the perspective of coal supply and demand security, its impact on coal demand security is increasing.

Carbon dioxide emissions (D1) were ranked second for the 2006–2010 and 2013–2019 periods and appeared 15 times in total during the study period. This is the most important factor affecting the security of coal demand in China. From 2002 to 2019, China’s carbon dioxide emissions rose from 3.843 billion tons to 9.811 billion tons, with coal carbon emissions accounting for more than 70% of the total energy carbon emissions. These emissions were primarily from coal consumption, which was the most important obstacle factor affecting the ecological environment and the security of coal use and needs to be a focus for carbon reduction to achieve carbon neutrality.

Coal consumption in thermal power generation (D2) occurred 13 times among the top seven obstacle factors during the 2002–2019 period, starting in 2007. This criterion was ranked between fourth and sixth from 2007 to 2019. From 2002 to 2019, coal consumption for China’s thermal power generation, which is the foundation for industrial electricity demand and ensures the stable operation of the economic and social power supply, increased from 597.98 million tons in 2002 to 2101.59 million tons in 2019. The rapid growth in China’s economy led to a continuous increase in electricity demand, and the power supply structure dominated by thermal power generation subsequently led to a significant increase in coal consumption. This was another main obstacle affecting coal demand security, and the pressure to meet the economic and social electricity demand is increasing under the goal of carbon neutrality.

Of the top seven criteria with the highest obstacle degree during the 2002–2019 period, other criteria with a high frequency included the total wholesale profit of coal and products (A5), the coal transportation volume (A6), the coal supply and demand ratio (B4), the GDP (C1), the proportion of non-coal energy consumption (C5), and the proportion of power generation hours for coal (D3). Of these, the GDP (C1) became a factor after 2015, mainly due to the continuous expansion of China’s economy and the increasing size of the GDP. The other obstacle factors were mostly observed before 2010, and their influence has weakened in recent years. In the future, it will be necessary to monitor the change in the obstacle degree of each individual criteria and combine the overall management of the security of the coal supply and demand with the management of individual criteria to further improve coal security.

Based on the results for the obstacle degree of individual criteria, we further calculated the obstacle degree for the four subsystems associated with China’s coal supply and demand security for the 2002–2019 period and predicted the obstacle degree for each subsystem in 2020 and 2021 using the grey model based on Equations (17)–(24) (Figure 2). Of the four evaluation subsystems, the obstacle degree of the coal supply chain was higher than that of the other three subsystems. For this subsystem, the obstacle degree decreased and then increased, with an inflection point occurring in 2016. The obstacle degree was predicted to continue to rise in 2020 and 2021, gradually increasing the impact on coal supply security. In contrast, the obstacle degree of the coal market subsystem was the lowest of the four subsystems. During the study period, the obstacle degree fluctuated from 2004 to 2017, but has reduced since then, weakening its impact on coal supply security. Finally, the obstacle degree for the social ecology subsystem and the population economy subsystem remained relatively consistent, though the obstacle degree of the former has increased since 2016, whereas it has decreased for the latter since the same year.

5. Conclusions and Suggestions

5.1. Conclusions

During its economic development and the transitional period for energy transformation, coal remains the main energy source for China due to its economic reliability and independent security. Establishing a comprehensive evaluation index to study the security of the coal supply and demand in recent years, and identifying and analyzing the main obstacle factors, are crucial to improving the ability to compensate for the weak links in the coal supply chain, balancing supply and demand in the coal market, enhancing coal security, promoting high-quality coal development, and guaranteeing the smooth operation of the economy and society. For this reason, this paper presents conclusions and recommendations that provide a theoretical basis for ensuring coal supply and demand security and overall energy security in China. By constructing an evaluation index for coal supply and demand security based on the entropy and TOPSIS methods and using obstacle degree analysis to identify of obstacle factors for the period 2002–2019, a number of significant findings can be reported.

(1)

The comprehensive evaluation results based on entropy–TOPSIS show that the level of China’s coal supply and use during the period 2002–2019 was generally below the safety level, i.e., it was at the safe warning level during the period 2010–2014 and the critical safety level in the other years. The change in China’s coal supply and demand security level initially decreased until 2011 before increasing after that year. Using the GM(1,1) model, this study predicted the security for 2020 and 2021. The results show that the security level and use of the coal supply in China will continue to rise; however, there is still room to improve this security.

(2)

Barrier factors affecting China’s coal supply and use safety during the period 2002–2019 were calculated and identified, and the evaluation index with the top seven obstacle degrees in each year were selected as the main obstacle factors. The results show that the basic coal reserves (A1), the reserve–production ratio of basic reserves (A2), the balance of social coal stocks at the beginning and the end of the year (A7), the ratio of coal imports to consumption (B2), the urbanization rate (C3), carbon dioxide emissions (D1), and coal consumption for thermal power generation (D2) were the main obstacles affecting the security of the coal supply and demand in China. Other important criteria were the total wholesale profit for coal and related products (A5), railway transportation volume (A6), coal supply and demand ratio (B4), GDP (C1), the proportion of non-coal energy consumption (C5), and the proportion of total power generation hours that come from coal (D3).

(3)

Based on the obstacle degree of each criterion, the obstacle degree for the four subsystems were calculated and the grey model was utilized to predict their obstacle degree for 2020 and 2021. The results showed that, of the four evaluation subsystems, the obstacle degree of the coal supply chain was higher than that of the other three subsystems; over the study period, it decreased initially before increasing. The obstacle degree for the social ecological subsystem has increased in recent years, whereas that of the economy/demographics subsystem has decreased recently. The coal market subsystem had the lowest obstacle degree overall, which has decreased in recent years.

In summary, the security of the coal supply and demand is associated with the coal supply chain (e.g., coal resource endowment, safety production, transportation capacity, social inventory), market factors (e.g., coal prices, external dependence, the relationship between supply and demand), the economy and demographics, energy structure, industrial structure, ecological and environmental protection, and social security. The empirical analysis found that the security level of coal supply and demand in China was not high as a whole over the study period, reaching a warning level in some years. In terms of coal supply security, the coal supply chain subsystem had the greatest influence, and its obstacle degree is increasing. Therefore, measures such as increasing the reserve–production ratio of basic coal reserves, moderately increasing the social stock of coal, and reducing the ratio of coal imports to consumption should be undertaken to enhance the security of the coal supply. In terms of coal demand security, by controlling the total coal consumption and promoting clean and efficient utilization of coal, the reduction in carbon emissions within the coal industry will be promoted, and total coal consumption will be reduced by optimizing the power supply structure to improve the overall coal supply and demand security level.

5.2. Limitations and Prospects

This paper provides a comprehensive evaluation of the security of the coal supply and demand in China, and on this basis, the obstacle factors to the security of supply and demand are further measured and identified, and research results have been achieved to a certain extent. However, there are still several limitations in this study, and more in-depth research and continuous improvement are needed in the future theoretical research and practical application of coal supply and demand security. Firstly, coal supply and demand security is a research category with depth and breadth, and the establishment of a coal supply and demand security evaluation index system is a very complex and systematic process. The risk factors affecting coal supply and demand security should also be further considered in future research. For example, changes in external factors in the coal supply chain, including international external conditions, national energy policies, unexpected natural disasters, and sudden major events, may have different degrees of impact on the security of coal supply and demand. These are the dimensions that can be focused on in future research on the security and risk factors of coal supply and demand. Secondly, this paper constructs a comprehensive evaluation system of coal supply and demand security in China from the national perspective but has not yet conducted a differentiated study of regional coal supply and demand security. The analysis and research on the spatial and temporal variability of regional coal supply and demand security and the influencing factors from the provincial perspective are also of great theoretical research value and practical significance, which is one of the directions that can be promoted in future research work. Thirdly, the research methods can be further expanded—for example, BP neural network [18], DARIA–TOPSIS [55], structural equation modeling (SEM) [56], fuzzy AHP and type-2 fuzzy DEMATEL [57] can be applied or used in combination in research work on coal supply and demand security and the influencing factors. In addition, multiple quantitative analysis methods may be used to conduct a comparative study of the research results, leading to obtaining more extensive research findings and policy recommendations. Fourth, regarding the analysis of obstacle factors, future research work could refer to and introduce the application of the new ranking similarity coefficient method in decision-making problems in an attempt to more effectively address the problem of identifying and ranking barrier factors. In particular, the r_w correlations and WS coefficient of rankings similarity are the representative methods with relative application superiority in the ranking decision. The WS coefficient method depends strictly on the position on which the difference in the ranking occurred, is sensitive to significant changes in the ranking, and is much better suited to comparing the reference ranking and the tested rankings [58].

5.3. Policy Advice

• Plan the layout of coal production and improve the quality of the coal supply: It is important to optimize the layout of coal production and development, to comprehensively consider the national coal resource endowment conditions, regional economic development characteristics, coal market consumption demand, coal transportation channel capacity, ecological environment carrying capacity and other factors, and to promote the intensive development of coal resources. Furthermore, measures should be undertaken to improve the concentration of coal production and promote coal production in the five major coal production and supply bases of Shanxi, Shaanxi, western Inner Mongolia, eastern Inner Mongolia, and Xinjiang with good resource endowment conditions. Other recommendations include controlling the total amount of coal production, optimizing coal stockpiles, developing an advanced and high-quality production capacity, eliminating backward production capacity, promoting the transformation and upgrade of the industrial structure, improving the quality of coal production and supply, and reducing the conflict between coal production capacity and market supply and demand [59,60,61].
• Improve the coal reserve system and strengthen the emergency coal supply: Coal reserve capacity building determines the length of the coal security interval, and as reserve capacity increases, the length of the coal security interval shortens, and coal consumption tends to develop steadily [62]. According to the time–space fluctuation law of coal demand and the distribution of existing reserves, we recommend that the spatial layout of the coal reserve bases be scientifically optimized, and the effective connection between coal reserves and the construction of coal production, supply, storage, and sales infrastructure should be strengthened [63]. Additionally, it is important to rationally plan and design the site selection of reserve projects so that they have significant strategic locational advantages, such as being adjacent to major transportation channels, and a strong trans-regional transfer capacity throughout the country to realize a stable, secure, and continuous supply of coal to core areas, important cities, key industries, and power enterprises. It is also important to enhance the coal supply security against trans-regional, systematic, and comprehensive risks.
• Diversify the sources of imports and enhance international coal acquisition: China should establish cooperative partnerships with major coal exporting countries, expand the breadth and depth of its international coal trade and industry cooperation, promote the diversification of its coal import sources, enhance its international coal resource acquisition capacity, and improve the national coal supply. Furthermore, it should strengthen cooperation with Russia, Mongolia, Indonesia, and other countries with rich coal resource reserves, establish overseas coal production and supply bases by means of cooperative exploration and development, invest in overseas acquisitions, increase the production and supply of imported coal, and enhance the supply security of imported coal through direct channels [64]. Finally, it is important to build a national market-oriented procurement platform for imported coal and conduct centralized procurement of high-quality international resources to effectively supplement the supply channels of domestic coastal resources, promote the optimization of imported coal types and quality, and stabilize the domestic coal market.
• Promote green development, clean and efficient utilization of coal, and energy conservation and carbon reduction: We recommend that the following measures should be undertaken: improving the level of green coal mining, washing, and processing; improving the efficiency of coal logistics and transportation; optimizing the structure and mode of coal utilization; and promoting the whole industrial chain and life cycle of coal mining, processing, transportation, and utilization to achieve clean and low carbon emissions [60,65]. Other recommendations include integrating superior resources, increasing the investment in scientific research, promoting the use of carbon capture, usage, and storage (CCUS) technologies and conducting a pilot demonstration, speeding up the implementation of key technologies, developing complete sets of equipment for the clean and efficient utilization of coal (e.g., efficient coal-fired power generation, new generation coal-to-oil, modern coal chemical industry, and intelligent energy technology), and promoting energy conservation and emissions reduction in major coal consuming industries [66].
• Optimize the power supply structure and promote smooth power reform: The power industry is still the dominant industry in China in terms of coal consumption, and thermal power generation is currently the primary power generation method [67], so the restructuring of the power supply is imperative. For this purpose, it is necessary to accelerate the transformation, upgrade, elimination, and/or renewal of units that are subcritical and below, promote the upgrade of ultra-supercritical coal-fired generating units to new coal-fired generating units with a higher efficiency (e.g., high and low reheat units and 700 °C ultra-supercritical units), and promote the continuous reduction of coal consumption for thermal power generation [68]. Furthermore, it is important to vigorously promote technological innovation, transformation, and upgrades in the coal–electricity+ field and solid waste-coupled power generation. Taking advantage of a power generation system consisting of active large-capacity coal-fired generating units would promote the transformation and upgrade of fuel preparation, boiler combustion, and environmental protection treatment systems and equipment and promote the research and development, pilot application, and promotion of coal-fired coupled power generation technologies that utilize coal-fired biomass such as agricultural and forestry waste, domestic garbage, and municipal sludge to improve the power generation efficiency of biomass; promote solid-waste consumption, recycling, energy conservation, and carbon reduction; and synergistically improve the overall energy efficiency of the energy system [62].
• Strengthen the overall control of coal consumption and its growth. For this purpose, it is important to rationally balance the competition and cooperation relationship between coal and other fossil-based and clean energies and promote the coupled development of coal and other energy sources. Furthermore, we recommend optimizing the energy consumption structure of major coal terminal consumption industries, promoting coal reduction and coal restrictions in major coal terminal consumption industries such as steel, building materials, chemicals, and cement, promoting clean energy substitutes for coal such as electricity or gas, and increasing the energy consumption of new and renewable energies. Other measures include improving the supporting policies and management mechanisms for total coal consumption control, including fiscal and taxation financial support, differentiated control, coal unit consumption management, coal-savings index trading, differentiated electricity prices, and strengthening statistics [69].