Cause Analysis of Coal Mine Gas Accidents in China Based on Association Rules

Abstract

Coal mine gas accidents will cause great economic losses and casualties. It is of great significance to find out the essential causes of coal mine gas accidents and put forward measures to prevent them. In this paper, 110 coal mine gas accidents which occurred in China from 2001 to 2022 are selected to analyze the causes of the accidents by extracting the keywords of human factors, equipment factors, environment factors, and management factors from the accident investigation reports. Firstly, the accident statistical analysis is carried out from three dimensions of factor frequency, accident type, and accident grade. Secondly, the Apriori algorithm is used for data mining to obtain frequent item sets and association rules of coal mine gas accident factors. Finally, the coal mine gas accident cause chains which are obtained by using the association rule. The frequent terms of 9 factors, 23 association rules, and 3 coal mine gas accident cause chains are obtained. The results show that the production of coal mine enterprises by illegal organizations is an important reason for the occurrence of coal mine gas accidents. The lack of good management culture easily leads to habitual violations of personnel and decision-making errors, and then causes equipment problems, reflected in the ventilation system which is not perfect, resulting in gas accumulation. The occurrence of coal mine gas accidents can be prevented to a large extent by preventing the absence of good management culture in enterprise management and the occurrence of illegal production behaviors.

1. Introduction

A comparative analysis of the fatality rate per million tons of various coal mine accidents in China shows that coal mine gas accidents often cause large economic losses and casualties [1,2] In recent years, although coal mine safety production technology has made great progress, the issue of how to prevent the occurrence of coal mine gas accidents is still an important subject of the current research. A large number of historical accident statistics can better reflect the mechanisms of and reasons for the accident. Gao et al. [3] sorted out nearly 170 gas explosion accidents in China and nearly 50 gas explosion accidents abroad, analyzed the accidents from the three aspects of accident location, occurrence time, and types of ignition sources, and obtained ten types of ignition sources that are most likely to cause accidents. Zhao et al. [3] analyzed the tremendous and devastating coal mine gas explosion accidents from 1981 to 2010 from the three dimensions of accident occurrence time, prone provinces, and accident types based on accident statistical analysis and the Granger causality test. Chen et al. [4] analyzed 377 tremendous and devastating accidents which occurred in China’s coal mines from 2002 to 2011 based on accident inducements, and concluded that more than 90% of these accidents were caused by unsafe behaviors of personnel. Montewka et al. [5] obtained the potential relationship between accident causes of different types of marine ship accidents based on the MDTC model through statistical analysis. Wang et al. [6] analyzed the data of coal mine gas accidents in China from 2006 to 2010, and found that the number of accidents and deaths in low-gas mines were more than those in high-gas mines, and the occurrence time of coal mine gas accidents presented exponential distribution. Zhang et al. [7] analyzed serious coal mine gas explosion accidents in China from 1950 to 2015 by using statistical methods, and found that improper ventilation management was the most common cause of gas accumulation.

Some progress has also been made in the analysis of the cause of accidents from a large number of historical accident statistics. Kim et al. [8] proposed the causality model of railway accidents by analyzing the investigation report of railway accidents. Manu et al. [9] proposed a new systematic research method to analyze the causes of construction accidents and put forward corresponding countermeasures. Maiti et al. [10] developed an EEA event assessment method to assess the safety level of coal production by studying coal mine death accidents in India in the past 100 years. Zhang [11] studied and constructed a causal model between life events and human errors of coal miners by using the structural equation method. Niu et al. [12] analyzed the causes of coal mine accidents in recent years and found that most coal mine gas explosion accidents occurred in low-gas mines. Liu et al. [13] used the statistical data of 362 major coal mine accidents in China from 2000 to 2016 to establish the Chinese Coal Mine Human Factor Analysis and Classification System (HFACS-CM), and conducted qualitative and quantitative analysis on the causes of accidents.

However, when determining the cause of accidents, most studies rely on expert experience and previous literature as the main sources, resulting in a lack of objectivity. As an important basis for the cause of coal mine gas accidents, the accident investigation report contains the cause, process, and result of the accident, which cannot be ignored. It should be analyzed as an important report to prevent coal mine gas accidents. At the same time, the occurrence of coal mine gas accidents is not the result of a single factor, but the joint action of multiple factors, and there is usually a potential relationship between these factors. The association rule is one of the data mining methods to reflect the dependence and correlation between things, with good generalization ability, and has become a hotspot of application research in many fields [14]. However, in the field of coal mining, there is little application of association rules and few historical statistics for accident investigation reports. Based on this, this paper combines data mining with theory. The keywords of unsafe factors are extracted from a large number of accident reports and the HFACS framework is modified according to the characteristics of the coal mining industry on the basis of the previous research. The potential relationship between factors is revealed by the association rule method, which provides a better theoretical framework for accident analysis and prevention.

2. Methods

2.1. Data Source

In order to ensure the reliability and integrity of the accident data, this paper takes 2001 to 2022 as the research period. The data sources are the website of The National Mine Safety Administration and the investigation reports of coal mine gas accidents of the sub-stations of the mine safety supervision bureaus of all provinces (autonomous regions and municipalities).

After statistical sorting, a total of 110 coal mine gas accidents with complete accident investigation reports were selected, covering 20 provinces in southeast, southwest, and northeast China, excluding the possible geographical specificities of the causes of the accidents and ensuring universality. The specific number and location distribution of accident areas are shown in Figure 1.

2.2. Data Preprocessing

The causes of coal mine gas accidents are complex and changeable, and the identification of important factors leading to the occurrence of accidents can effectively prevent the occurrence of accidents [15]. This paper adopts the idea of data-driven methods, extracts the key factors in the coal mine gas accident investigation report, and counts the frequency of each keyword. This paper also summarizes the causes comprehensively from the four aspects of human, equipment, environment, and management. Based on the statistical results of the frequency of the occurrence of keywords, combined with the existing literature and objective facts, 35 keywords with relatively high frequency are selected as research elements. As shown in Figure 2.

(1)

Human factors

The HFACS causation theory model divides human factors into four levels: unsafe behaviors, preconditions of unsafe behaviors, unsafe supervision, and organizational influence [16]. According to the specific accident causes in the coal mine gas accident investigation report, this paper classifies human factors into three unsafe behaviors including skills error, decision-making error, and violation operation, as well as the preconditions of unsafe behaviors, and then divides them into different specific manifestations according to keywords. In order to avoid subjective factors affecting the classification process, the classification of factors in Figure 2 is defined and explained as follows:

A.

Skills error

This refers to the “Missing procedures steps”, “Using inadequate construction technology”, “Using wrong methods/instruments”, etc. The missing procedures steps include failure to take anti-outburst measures, failure to test the effect of anti-outburst, blind operation without eliminating the risk of outburst, failure to measure the gas content in advance, etc. The construction technology is not in place, including the construction strength which is not in place, is not in accordance with the provisions of the design and construction, and equipment (facilities) installation is not reasonable. The use of wrong methods/instruments includes the use of wrong measuring tools, coal mining methods, gas extraction methods, etc.

B.

Decision-making error

This refers to the situation of incomplete understanding and wrong judgment of emergency situations due to the lack of on-site experience or training of personnel, including “Illegally order”, “Inadequate emergency handling”, and “Illegal entry into blind lane and other closed areas”.

C.

Violation operation

This refers to the behavior that the personnel do not abide by the rules and regulations subjectively, which is divided into habitual violation and accidental violation. Habitual violations often occur underground and are often condoned by managers, while operators are accustomed to natural violations, such as failure to carry or wear safety protection equipment, illegal shooting, illegal modification/use of equipment, falsified data parameters, and cross-operation of multiple types of underground work.

(2)

Equipment factor

The objects in underground coal mine are mainly all kinds of working equipment and machines. With the progress and development of coal mining technology in China, the types and quantities of equipment and machines in underground coal mines are gradually increasing, and the unsafe state factors of objects are also increasing. In the process of sorting out and classifying all the equipment factors in the accident investigation report, it was found that there are various equipment factors involved in the specific content of the coal mine gas accident investigation report, and there are many fault types. The machine factors leading to accidents can be divided into missing or damaged parts of the equipment itself, working equipment is not in normal operation state, etc. This article is divided into two categories: “Equipment missing/imperfect” and “Equipment not running/failure”.

A.

Security monitoring system missing/imperfect

This includes insufficient sensor types or quantity, insufficient gas monitor, and damaged safety monitoring equipment.

B.

Gas extraction system missing/imperfect

Including: the ground gas extraction system has not been established, gas extraction drilling construction is not in place, and gas extraction time is not enough. The gas extraction system here includes equipment factors before and during the work, among which, insufficient gas extraction time can be a design defect, but from the overall situation, it should be classified as an imperfect gas extraction system.

C.

Equipment failure

This refers to the long-term and occasional faults of all downhole equipment, such as short circuit, device detonation, cable damage, protection device does not move in time, gas data cannot be uploaded, etc.

(3)

Environmental factors

Environmental factors include: coal mining geology is complicated, geological structure properties are different and so on. Environmental factors will indirectly lead to coal mine gas accidents [17,18,19]. For example, if the spontaneous combustion of coal is strong, coal mines are prone to gas combustion accidents. This is because there is air leakage in the goaf or closed area, and the coal seam will come into contact with oxygen, resulting in spontaneous combustion [20,21]. It is easy to oxidize the coal body and cause gas combustion accidents [22].

Secondly, the working environment of underground coal mine is another influencing factor causing coal mine gas accidents [23,24]. For example, the high noise in underground coal mines can easily affect the psychological state of operators, which may lead to operators making operation errors and to coal mine gas accidents [25]. However, according to the specific content of the coal mine gas accident investigation report, there are few records of accidents caused by the poor working environment. Therefore, this paper mainly analyzes the natural geological conditions.

(4)

Management factors

Management factors refer to the factors that cause safety accidents due to wrong or inadequate decision-making at the management level of coal mine safety production enterprises, which is the essential cause of the occurrence of coal mine safety accidents. The article classifies the management factors that appear in the accident investigation report as follows:

• Poor safety technology management
  Including no safety technical measures and safety technical measures are not perfect.
• Flawed area design
  The ventilation partition design is insufficient, the regional anti-penetration design is unreasonable, and the mining and excavation deployment is unreasonable.
• Inadequate implementation of measures

The implementation of gas control measures is not in place, the implementation of blasting system is not in place, and the implementation of anti-intrusion measures is not in place.

2.3. Data Analysis Method

2.3.1. Accident Statistical Method

The samples of 110 coal mine gas accidents are comprehensively counted from three aspects of regional distribution, accident type, and accident grade, and the occurrence regularity of accidents is preliminarily explored.

2.3.2. Association Rule Method

The factors in coal mine gas accident are discrete data, and the occurrence frequency and frequency of factors are not enough to obtain the potential information from them. In order to get the relationship between each factor, the association rule method was selected as a data mining method for analysis.

(1)

Association rule principle

The association rule data mining method is a means to discover potential connections hidden in a large amount of data. These potential connections can reveal the relationship between data, which is more conducive to discovering the occurrence mechanism of events [26].

I for the item set, I = {$i_1,i_2$, … $i_n$}, where $i_n$ called the item, each transaction T is a subset of item set I. All T term sets, like X→Y (X,Y ⊆ I, X ∩ Y = ø) is the implied expression of the association rule.

A.

Support:

This refers to the probability that transaction X and transaction Y occur simultaneously in the item set. In the application of the association rule method, the most important thing is to select the support threshold. The support threshold needs to be set reasonably according to the data situation. If the calculated support is lower than the threshold, it indicates that the occurrence times of the item set {X, Y} in the total item set are too few. Otherwise, subsequent analysis can be carried out. Support is expressed as:

$$Support(X\to Y)=P(X,Y)/P\left(I\right)$$

(1)

B.

Confidence:

This refers to the probability of transaction Y occurring if transaction X has already occurred. More important is the choice of confidence threshold, which reflects the ability of the occurrence of transaction X to affect the occurrence of transaction Y. Confidence is expressed as:

$$Confidence(X\to Y)=P\left(Y\right|X)=P(X,Y)/P(X)$$

(2)

C.

Frequent item sets:

The frequent item set refers to the set of all transactions whose support is greater than or equal to the support threshold. If the support of item set I meets the selected minimum support threshold, item set I is the frequent item set.

D.

Association rules:

The association rules that meet the minimum support threshold and minimum confidence threshold are those selected to be mined.

(2)

Apriori algorithm

Apriori algorithm is a data mining algorithm for mining frequent item sets and association rules. Its purpose is to carry out the two steps of association and mining continuously to find the association rules that meet the threshold of support and confidence.

Firstly, by scanning the database, the count of each item is accumulated and the items that meet the minimum support threshold are collected to find the set of frequent item set 1, denoted as $I_1$. Then, use $I_1$ to find $I_2$ of frequent item set 2, use I to find $I_3$, and so on until frequent k item set can no longer be found. A full scan of the database is required for each $I_k$ found.

(3)

Analysis steps

The frequent item set in this paper is a combination of frequently occurring factors that lead to coal mine gas accidents. The mining results of frequent item sets are closely related to the selection of support threshold. If the support threshold is set too large, there will be too few causes in frequent item sets, which is not conducive to the generation of association rule results. However, if the support threshold is set too small, the number of frequent item sets generated will be too large, and it is not easy to determine the root cause of coal mine gas accidents. In order to prevent improper selection of the support threshold from affecting the conclusion of association rules, this paper determines the most appropriate support degree by comparing frequent item sets obtained from different support degrees. Four support degrees of 0.2, 0.15, 0.1, and 0.05 were selected, and 0.05 was selected as the frequent item set of minimum support mining after comparison of results.

After the frequent item set is obtained, the Apriori algorithm is used to mine association rules. According to the comparison of experience and results, the confidence degree is set as 0.4. One factor in the frequent item set is taken as the front term of association rules, and the other factors are taken as the back term, respectively.

2.3.3. Accident Cause Chains Method

The occurrence of coal mine gas accidents is the result of the interaction and joint action of various factors. According to association rules, a more intuitive cause chain of coal mine gas accidents is constructed, from which three cause chains are extracted for analysis and countermeasures and suggestions are put forward.

3. Results and Discussion

3.1. The Accident Statistics

3.1.1. Factor Frequency Statistics

First of all, the analysis of the occurrence frequency of these accident factors can provide a better basis for the production of association rules. The frequency of occurrence of factor keywords is analyzed and the proportion of these factors in accidents is analyzed. The factor names in Figure 3 and Figure 4 correspond to the label in Figure 2.

The results of statistical analysis for human and equipment factors are as follows:

It can be seen from Figure 3a that “Habitual violations” occurred 121 times in total, accounting for 32.3%, which is also the factor with the most frequency among the four aspects of the factors, followed by “Missing procedures steps” in skills errors, which occurred 70 times, accounting for 18.7%. “Poor personnel safety awareness” appeared 43 times, which shows that the safety education and training of personnel in most coal mining enterprises, as well as the establishment of personnel safety awareness, are not in place.

In Figure 3b, the factors of “equipment missing/imperfect” is mainly reflected in the safety monitoring system and ventilation system, 43 times and 38 times respectively, accounting for 30.7% and 27.1% of the machine factors; security monitoring. For example, gas accumulation in the roadway is caused by unreasonable ventilation design or failure to open local ventilators, and abnormal concentration of gas is not detected by safety monitoring equipment, so coal mine gas accidents easily occur [27,28]. “Explosion-proof measures not set” and “Gas extraction system missing/imperfect” only accounted for 5.7% of the total equipment factors, which therefor had little reference significance. In addition, “Normal equipment but not running” and “Equipment failure” occurred 20 times and 31 times, respectively, accounting for 14.3% and 22.1% of the total equipment factors, indicating that compared with equipment missing/imperfect, the factor of “equipment failure” is relatively rare.

The results of statistical analysis for environment and management factors are as follows:

Figure 4a shows that in terms of environment, which is the main cause of accidents, gas accumulation accounted for 45.5% of the environmental cause factors, followed by coal seams with outburst danger which accounted for 31.8%. In the process of sorting out the accident causes of the accident report, it is also found that most of the coal mines where coal and gas outburst accidents occurred are identified as having outstanding risks. The occurrence of “Strong spontaneous combustion tendency”, “Abnormal gas emission”, and “Soft coal seam” is less common.

It can be seen from Figure 4b that among all factors, management factors appear the most frequently, indicating that management factors are more important in ensuring coal mine production safety. The hidden dangers of management factors mainly exist in “illegal supervision”, “lack of good management culture”, and “management process loopholes.” Among them, the number of times that ineffective supervision by the government/safety supervision bureau of illegal supervision occurred was 79, accounting for 16.9%; the number of times illegal production organization factors occurred was 68, accounting for 14.6%; the lack of good management culture was reflected in the failure to implement the main responsibility of safety production and the lack of safety education and training work were 9.6% and 10.7%, respectively. The loopholes in the management process are mainly reflected in the inadequate implementation of measures.

3.1.2. Accident Type Statistics

Further statistical analysis of accident types, accident grades, and number of deaths is helpful to better explore the nature of accident causes.

The statistical results for the types of accident and the number of deaths are as follows:

As can be seen from Figure 5, gas explosion (combustion) accidents are the most frequent in terms of the number of occurrences and deaths, with 52 cases occurring in total, accounting for 47.3%, and 984 deaths, accounting for 73.7% of the total. This is mainly because gas explosion is powerful, and easily causes a series of other disasters. If aroused, the dust of the ground can produce a two-dust explosion. Therefore, once a gas explosion accident occurs, there are often more casualties. Preventing the occurrence of gas explosion accidents is of great significance to control the occurrence of casualties and accidents.

There are 42 coal and gas outburst accidents, with 308 deaths, accounting for 38.2% and 23.1% of the total, respectively. Gas asphyxiation (poisoning) accidents are relatively rare, with 16 cases accounting for 14.5% of the total, and 42 deaths accounting for only 3.2% of the total. Among all the coal mine gas accident investigation reports, only one is on a gas poisoning accident, which is a gas poisoning accident in the Xudong Coal industry in the Gansu province in 2015, and the other 15 are gas asphyxiation accidents.

3.1.3. Accident Grade Statistics

The statistical results of the accident grade are as follows:

It can be seen from Figure 6 that coal and gas outburst accidents and gas explosion (combustion) accidents are often categorized as considerable and serious accidents. Gas explosion (combustion) accidents are also tremendous and devastating accidents, which further supports the finding in Figure 5 that gas explosion accidents cause the highest number of deaths. Gas asphyxiation (poisoning) accidents resulted in relatively light consequences, no serious accidents, and especially serious accidents.

As can be seen from Figure 7, the total number of ordinary accidents, considerable accidents, serious accidents, and tremendous, devastating accidents is 18, 58, 23, and 11, respectively, accounting for 16.4%, 52.7%, 20.9%, and 10.0% of the total number of accidents; resulting in 27, 260, 474, and 573 deaths respectively, accounting for 2.0%, 19.5%, 35.5%, and 43.0% of the total number of deaths. Although the number of tremendous, devastating accidents is relatively small, the death toll accounts for nearly half of the total, such as the gas explosion in the Liaoning Sunjiawan Coal Mine in 2005, which killed 214 people and caused heavy casualties and economic losses. It can be seen that coal mine gas accidents in particular need to control the occurrence of tremendous, devastating accidents.

3.2. Association Rule Statistics

3.2.1. Frequent Item Sets

As mentioned above, the setting of the support threshold is crucial when calculating association rules. In order to ensure the reasonable setting of support, the paper first adjusts the calculation of the support threshold. When trying to set it to 0.5, 0.4, and 0.3, there is no frequent return of item set results, which will lose the value of subsequent research. When the support threshold is set to 0.2, two items are generated, so the paper conducts subsequent research from the support threshold of 0.2.

According to

Table 1. Results of frequent item sets.

Frequent Item Sets	Support
{Habitual violation, Poor supervision of the safety supervision bureau}	0.24
{Habitual violation, Poor supervision of the safety supervision bureau, Illegal production}	0.14
{Habitual violation, Poor supervision of the safety supervision bureau, Gas accumulation}	0.12
{Habitual violation, Gas accumulation, Illegal production}	0.11
{Habitual violation, Poor supervision of the safety supervision bureau, Gas accumulation Illegally production, Ventilation system missing/imperfect}	0.05

, when the support threshold is set to 0.2, the mining result of the frequent item set of coal mine gas accident factors is “habitual violation”, “poor supervision of the safety supervision bureau”. This indicates that the support threshold is set too high and the frequent items are too few. It is not conducive to the generation of association rules.

When the support threshold is set to 0.1, two factors, “Illegally production” and “Gas accumulation”, are increased compared with the support threshold of 0.2, indicating that environmental factors will be returned after the support threshold is lowered.

The support threshold is set to 0.05, meaning that the results of the frequent item in the equipment aspect is more reasonable. If the support threshold is lowered further, there will be too many causative factors, which means the coverage of the frequent item set in the transaction set is very small, therefore the occurrence of factors in frequent item set is very accidental, and the final result lacks feasibility. After synthesis, the support threshold is selected as 0.05 for subsequent analysis. The frequent item sets of coal mine gas accident factors are summarized as “habitual violation”, “poor supervision of the safety supervision bureau”, “gas accumulation”, “illegal production”, and “ventilation system missing/imperfect”.

In the analysis of frequent item set results, it can be found that it is difficult to return the results of environment factors and equipment factors of accidents and the number is small. The reasons may be as follows: Coal mine gas accidents are mostly liability accidents. At present, the investigation reports of coal mine gas accidents mainly focus on liability identification. The direct and indirect causes are generally analyzed for human and management reasons, and the analysis of equipment and environment is less commonly completed, and the records are not detailed enough.

3.2.2. Association Rule

“Habitual violation” is used as the preceding term to obtain the association rule:

As the result of Article 5 in

Table 2. The association rule obtained as “Habitual violation”.

Number	Preceding Term	Consequent	Support	Confidence
1	Habitual violation	Gas accumulation	0.165775401	0.430555556
2	Habitual violation	Lack of safety education	0.171122995	0.444444444
3	Habitual violation	Inadequate implementation of measures	0.176470588	0.458333333
4	Habitual violation	Illegal production	0.176470588	0.458333333
5	Habitual violation	Poor supervision of the safety supervision bureau	0.235294118	0.611111111

, the support degree of “Habitual violation” and “Weakly supervision of the safety supervision bureau” is 0.23, indicating that the probability of accidents caused by both factors is 23.5%. The confidence degree is 0.61, indicating that the probability of ineffective supervision by the safety supervision bureau is 61.1% when habitual violations occur, and the reliability of “Weakly supervision of the safety supervision bureau” is 61%. Similarly, in the case of habitual violations, the probability of “Illegally production” and “Inadequate implementation of measures” is 45.8%, the probability of “Lack of safety education” is 44.4%, and the probability of “Gas accumulation” is 43.1%.

“Weakly supervision of the safety supervision bureau” is used as the preceding term to obtain the association rule:

As shown in

Table 3. The association rule obtained as “Weakly supervision of the safety supervision bureau”.

Number	Preceding Term	Consequent	Support	Confidence
1	Poor supervision of the safety supervision bureau	Gas accumulation	0.139037433	0.426229508
2	Poor supervision of the safety supervision bureau	Lack of safety education	0.144385027	0.442622951
3	Poor supervision of the safety supervision bureau	Illegal production	0.192513369	0.590163934
4	Poor supervision of the safety supervision bureau	Habitual violation	0.235294118	0.721311475

, the probability of “Habitual violations” is 72.1%, “illegally production” is 59.0%, “Lack of safety education” is 44.3%, and “Gas accumulation” is 42.6% when there is “Weakly supervision of the safety supervision bureau”.

“Gas accumulation” is used as the preceding term to obtain the association rule:

As shown in

Table 4. The association rule obtained as “Gas accumulation”.

Number	Preceding Term	Consequent	Support	Confidence
1	Gas accumulation	Ventilation system missing/imperfect	0.085561497	0.421052632
2	Gas accumulation	Illegal production	0.128342246	0.631578947
3	Gas accumulation	Poor supervision of the safety supervision bureau	0.139037433	0.684210526
4	Gas accumulation	Habitual violation	0.165775401	0.815789474

, in the case of gas accumulation in the mine working area, the probability of “Habitual violations” is 81.6%, the probability of “Weakly supervision of the safety supervision bureau” is 68.4%, the probability of “Illegally production” is 63.2%, and the probability of “Ventilation system missing/imperfect” is 42.1%.

“Illegally production” is used as the preceding term to obtain the association rule:

As shown in

Table 5. The association rule obtained as “Illegally production”.

Number	Preceding Term	Consequent	Support	Confidence
1	Illegal production	The primary responsibility is not implemented	0.106951872	0.408163265
2	Illegal production	Missing procedure steps	0.106951872	0.408163265
3	Illegal production	Lack of safety education	0.112299465	0.428571429
4	Illegal production	Ventilation system missing/imperfect	0.117647059	0.448979592
5	Illegal production	Gas accumulation	0.128342246	0.489795918
6	Illegal production	Habitual violation	0.176470588	0.673469388
7	Illegal production	Poor supervision of the safety supervision bureau	0.192513369	0.734693878

, when there is illegal production in the mine, the probability of “Weakly supervision of the safety supervision bureau” is 73.5%, the probability of “Habitual violation” is 67.3%, the probability of “Gas accumulation” is 49.0%, and the probability of “Ventilation system missing/imperfect” is 44.9%. The probability of “Lack of safety education” is 42.9%, the probability of “Missing procedure steps” is 40.8%, and the probability of “The primary responsibility is not implemented” is 40.8%.

“Ventilation system missing/imperfect” is used as the preceding term to obtain the association rule:

As shown in

Table 6. The association rule obtained as “Ventilation system missing/imperfect”.

Number	Preceding Term	Consequent	Support	Confidence
1	Ventilation system missing/imperfect	Lack of safety education	0.085561497	0.444444444
2	Ventilation system missing/imperfect	Gas accumulation	0.085561497	0.444444444
3	Ventilation system missing/imperfect	The primary responsibility is not implemented	0.09090909	0.472222222
4	Ventilation system missing/imperfect	Illegal production	0.117647059	0.611111111
5	Ventilation system missing/imperfect	Poor supervision of the safety supervision bureau	0.122994652	0.638888889
6	Ventilation system missing/imperfect	Habitual violation	0.13368984	0.694444444

, under the condition of “Ventilation system missing/imperfect”, the probability of “Habitual violation” is 69.4%, the probability of “Weakly supervision of the safety supervision bureau” is 63.9%, the probability of “Illegally production” is 61.1%, and the probability of “The primary responsibility is not implemented” is 47.2%. The probability of underground “Gas accumulation” is 44%, and the probability of “Lack of safety education” is 44%.

From the above, 26 association rules of 110 coal mine gas accidents were obtained, involving 9 factors in total. Where, 01 is “Illegally production”, 02 is “Weakly supervision of the safety supervision bureau”, 03 is “Lack of safety education”, 04 is “The primary responsibility is not implemented”, 05 is “Inadequate implementation of measures”, 06 is “Habitual violation”, 07 is “Missing procedure steps”, 08 is “Ventilation system missing/imperfect”, and 09 is “Gas accumulation”.

The association rules obtained from the above results meet the support threshold and confidence threshold, but not all the association rules that meet the requirements have application value. Human factors and management factors in coal mine gas accident factors refer to the HFACS causation model, so the causal relationship between them should follow the principle of sequential occurrence in the HFACS model, rather than the reverse order. The hierarchical classification is: organizational factors, unsafe supervision, the premise of unsafe behavior, and unsafe behavior. Among them, 01 and 02 belong to first level, 03, 04, and 05 belong to second level, and 06 and 07 belong to fourth level. At the same time, equipment factors and environmental factors cannot lead to the occurrence of management factors and human factors.

Based on these two principles, the confidence degree that does not meet the requirements is represented by 0, and the confidence degree that meets the requirements is represented by 1, and the final matrix with a modified confidence degree is obtained. The confidence degree in the matrix is the confidence degree of row to column. The matrix is shown in

Table 7. Confidence modified matrix.

	01	02	03	04	05	06	07	08	09
01	0	0.59	0	0	0	0	0	0	0
02	0.73	0	0	0	0	0	0	0	0
03	0.43	0.44	0	0.45	0	0	0	0	0
04	0.41	0	0	0	0	0	0	0	0
05	0	0	0	0.45	0	0	0	0	0
06	0.67	0.72	0.64	0.66	0.73	0	0.65	0	0
07	0.41	0	0.44	0.47	0.58	0	0	0	0
08	0.45	0	0	0.45	0	0.43	0	0	0
09	0.49	0.43	0	0	0	0	0	0.44	0

.

After confidence degree modification, 23 association rules are obtained, and the results are shown in

Table 8. Association rule result table.

Association Rule	Confidence
Illegal production => Poor supervision of the safety supervision bureau	0.73
Illegal production => Lack of safety education	0.43
Illegal production => The primary responsibility is not implemented	0.41
Illegal production => Habitual violation	0.67
Illegal production => Missing procedure steps	0.41
Illegal production => Ventilation system missing/imperfect	0.45
Illegal production => Gas accumulation	0.49
Poor supervision of the safety supervision bureau => Illegal production	0.59
Poor supervision of the safety supervision bureau => Lack of safety education	0.44
Poor supervision of the safety supervision bureau => Habitual violation	0.72
Poor supervision of the safety supervision bureau => Gas accumulation	0.43
Lack of safety education => Habitual violation	0.64
Lack of safety education => Missing procedure steps	0.44
The primary responsibility is not implemented => Lack of safety education	0.45
The primary responsibility is not implemented => Inadequate implementation of measures	0.45
The primary responsibility is not implemented => Habitual violation	0.66
The primary responsibility is not implemented => Missing procedure steps	0.47
The primary responsibility is not implemented => Ventilation system missing/imperfect	0.45
Inadequate implementation of measures => Habitual violation	0.73
Inadequate implementation of measures => Missing procedure steps	0.58
Habitual violation => Ventilation system missing/imperfect	0.43
Missing procedure steps => Habitual violation	0.65
Ventilation system missing/imperfect => Gas accumulation	0.44

.

3.3. Accident Cause Chains

The cause chains of coal mine gas accidents can be obtained by connecting the factors in the association rules. Three cause chains are obtained and further reflect the deep relationship between the causes of coal mine gas accidents.

3.3.1. Management Factors Cause Chain

In coal mine gas accident cause chain 1 (Figure 8), the implementation of measure is the direct cause of the accident. The main responsibility of work safety is not implemented and this is the main reason leading to the implementation of measures not being in place.

The primary responsibility is not implemented and this is the main reason leading to the inadequate implementation of measures, and at the same time, it will also lead to the lack of safety education. This is because the primary safety responsibility contains “security system formulation” and “safety in production inputs”, so if it is not implemented, the safety production system of coal mine enterprises is not perfect or the safety production investment is insufficient, meaning the theoretical provisions and material support of prevention are not enough.

When coal mining enterprises appear in illegal organization production, there is a 73% possibility that the safety supervision bureau has ineffective supervision; when the safety bureau’s supervision of coal mining enterprises is lax, coupled with the management of coal mining enterprises’ “Focus on production, Light safety”, the wrong concept easily appears in cross-border mining, production without licenses, and other illegal production behavior. Production by illegal organizations, ineffective supervision by the safety supervision bureau, and failure to implement the responsibility of the main body for work safety easily lead to the failure of safety education and training, but the failure to implement the responsibility of work safety accounts for 45%, indicating that this is the most reliable reason. The production of illegal organizations and the safety supervision bureau’s poor supervision will lead to each other, which shows that the management and accident prevention of coal mine enterprises are inseparable from the local government, safety supervision bureau, and the enterprise itself. Therefore, in terms of management in general, illegal supervision will lead to a lack of good management culture, and ultimately lead to loopholes in the management process.

3.3.2. Management and Human Factors Cause Chain

When coal mine gas accidents cause chain 2 (Figure 9), management factors and human factors are closely related to accident causes, and the cause chain is complex. It can be seen that when the safety education is not in place, the primary responsibility of the safety production is not implemented, and the implementation of measures are not in place, which will lead to habitual violations and omissions of procedures. Among them, improper implementation of measures is most likely to lead to habitual violations and missing procedure steps, with a confidence rate of 73% and 58%, respectively. The inadequate implementation of measures means that the theoretical knowledge or practical experience of safety measures is too little, the overall cognitive level of personnel is low, and decision-making errors are easy to occur, so violations and omissions of operation steps easily occur. At the same time, omission of procedural steps may lead to habitual violations by 65%. This is because when operators are not familiar with operational procedures and technical operations, the omission of procedural steps will occur. In the long run, such omission will form habits, produce fluke psychology, and eventually develop into habitual violations.

3.3.3. Comprehensive Factors Cause Chain

When coal mine gas accidents cause chain 3 (Figure 10), gas accumulation is an important cause of coal mine gas accidents, and the cause of gas accumulation is the ventilation system being missing/imperfect. Due to the imperfection or non-operation of the ventilation system, the gas in the working area cannot be discharged in time and thus accumulates. People’s habitual violation and omission of procedural steps will cause problems in the ventilation system, and the reason for the habitual violation and omission of procedural steps is that the measures are not implemented in place. Therefore, in terms of comprehensive factors, loopholes in the management process will lead to decision-making errors and illegal operations, which will lead to the “Ventilation system missing/imperfect” in equipment factors, leading to the occurrence of gas accumulation.

3.4. Countermeasures and Suggestions

According to the coal mine gas accident cause chains combined with the actual coal mine gas accident case, the following countermeasures and suggestions are put forward:

Strictly perform the duties of coal mine safety supervision and strengthen the inspection and guidance of coal mine production enterprises in the area under their jurisdiction.

For example, on 10 May 2017, a serious gas combustion accident occurred in the Yandefang Coal Mine in Shanziwei, Longyan City, Fujian Province. The expiration date of the safety production license of this coal mine was 2016, so the mine was suspended from 2016. The absence of safety protection measures and the illegal organization of the workers into mining operations resulted in the accident.

When the coal mine gas accident causes chain 1, poor supervision by the safety supervision bureau and illegal production by coal mining enterprises are the essential causes of accidents. As a result of this, the safety supervision bureau should take responsibility and strengthen safety management. At the same time, the safety supervision should strictly investigate and punish illegal mining behavior and cross-border mining behavior of coal mine production enterprises operating without a production license, and ask the coal mine to stop production and rectify their enterprise within a time limit, so as to eliminate hidden safety risks in the embryonic stage. The management of coal mine production enterprises should strictly abide by the relevant laws and regulations of the state, set up correct values, put safety in production first, and should not blindly pursue output and interests at the expense of safety protection.

Earnestly implement the main responsibility of enterprise safety production and strengthen safety education.

For example, there was a gas explosion in the Goutou Coal Mine in the Yunnan province on 22 September 2019, a gas explosion in the Kongjiagou Coal Mine in the Guizhou Province on 13 December 2017, and a large coal and gas outburst accident in the Zhushanchong Coal Mine in the Hunan Province on 8 December 2017. In view of such situations, coal mining enterprises should earnestly implement the provisions of the “Regulations on Reporting and Investigation of Production Safety Accidents” and report coal mine accidents truthfully and in a timely manner.

Illegal supervision will lead to a lack of good management culture in the daily operation of coal mine production enterprises. First of all, good management culture must: have a perfect safety management organization; be equipped with enough professionals; not have personnel without a certificate or a certificate that does not match the job; not illegally hire personnel; formulate practical emergency rescue plans for accidents, and conduct regular emergency training for personnel; ensure adequate investment in production safety, material, and financial security; and strictly follow incident reporting procedures when accidents occur. In addition, coal enterprises need to, on the one hand: strengthen safety education and training of personnel, to pay attention to workers’ safety and in turn pay special attention to the worker’s coal mine gas accident prevention education. On the other hand, lectures, competitions, and other forms of training should be held regularly to popularize and assess the theoretical knowledge of coal mine gas accidents, and safety publicity and education should be carried out in the form of radio, television, and short videos, so that employees can consciously abide by safety rules and regulations and carry out production operations in strict accordance with the operation rules.

Ensure that the relevant measures of operation are put in place to improve the gas control level.

For example, a large coal and gas outburst accident occurred in the Liaoyuan Coal Mine in the Shaanxi province on 22 June 2021. The direct cause of the accident was that the underground outburst prevention measures were not implemented in place and the gas extraction time was not enough. The coal mining operation was carried out without eliminating the danger of outburst at the accident site.

Before mining, protruding coal seam, “four-in-one” outburst prevention measures must be implemented. For gas control measures, it is necessary to carry out targeted scheme design according to different geological conditions of different coal mines. Unreasonable control measures cannot achieve the expected effect of gas extraction, and it is likely to lead to accidents.

4. Conclusions

This paper selects 110 coal mine gas accidents in China from 2001 to 2022. Firstly, the regional distribution, accident type, and accident grade are analyzed statistically. Secondly, the keywords of human factors, equipment factors, environment factors, and management factors are extracted from the accident investigation reports. Finally, the Apriori algorithm is used for data mining to obtain frequent item sets and association rules of coal mine gas accident factors, and the cause chains of coal mine gas accidents are obtained by association rules. The following conclusions were drawn:

(1)

From the perspective of accident statistical analysis, the keyword analysis of factors in 110 accident reports shows that “Habitual violation” is the most frequent factor in human factors, “Security monitoring system missing/imperfect” is the most frequent factor in equipment factors, and “Gas accumulation” is the most frequent factor in environmental factors. Among the management factors, the most frequent factor is “Weakly supervision of the safety supervision bureau”. Among the accident types, gas explosion (combustion) accounted for the largest number of deaths and accidents. From the grade of accidents, especially gas explosion (combustion).

(2)

Frequent item sets and association rules of coal mine gas accident factors are generated. The support degree is set to 0.05 and the confidence degree is set to 0.4. The frequent item set is obtained after mining the keywords of factors, which contained 9 factors in total, including “habitual violation”, “poor supervision of the safety supervision bureau”, “gas accumulation”, ‘illegal production”, and “ventilation system missing/imperfect”. Mining the association relation between the factors and establishing a matrix to exclude the association rules that do not conform to reality, 23 coal mine gas accident association rules are obtained.

(3)

The causes chain of coal mine gas accidents are obtained. By analyzing the relationship of association rules, three causes chains are extracted and generated, and the main causes of coal mine gas accidents are obtained by analyzing the cause chains, and relevant prevention, control measures and, countermeasures are put forward based on accident cases.