Construction of Economic Management Performance Model of Mining
Enterprises under the Background of Supply-side Reform
XINHUI WANG
Lyceum of the Philippines University,
Manila Campus,
Manila,
PHILIPPINES
Abstract: - The proposal of supply-side structural reform measures has ushered in new development
opportunities for mining enterprises in the bottleneck period. As a result, a model for evaluating the economic
management performance of mining firms at the new performance level must be established. This paper
suggests an assessment model for the economic management of mining firms against the backdrop of supply-
side reform in light of this. This study reasonably incorporated financial and non-financial performance
indicators, constructed the Economic Management Performance (EMP) evaluation index system of mining
enterprises, and created an economic management performance evaluation model of mining enterprises based
on the BP neural network and analytic hierarchy process. The study selected the relevant data of five mining
companies A, B, C, D, and E from 2017 to 2022 as the research object, verified the effectiveness of the model,
and analyzed the performance evaluation results of the companies. The research results show that the model
constructed in this study can evaluate the economic management performance level of enterprises within a
reasonable range (the mean relative error is 1.98%). Since 2017, the comprehensive performance level of these
five mining companies has gradually declined. But thanks to the supply-side reform, the comprehensive
performance has gradually recovered after 2022 and among the five mining companies, company A has always
been at the performance level way ahead. Overall, the model developed in this research has strong operability
and practicability and can be utilized more effectively to forecast the mining industry’s potential for future
growth.
Key-Words: - Supply-side reform; Mining enterprises; Performance evaluation; BP neural network; Analytic
hierarchy process
Received: March 8, 2023. Revised: August 29, 2023. Accepted: September 13, 2023. Published: September 22, 2023.
1 Introduction
For the development of China’s social economy,
mineral resources are an important material basis,
and the mining industry is also in a relatively front-
end position in the social industrial chain. It is one
of China’s most significant basic industries, [1].
Since the founding of New China, China has been
one of the greatest mineral resource nations in the
world because of its fast-growth mining sector,
making outstanding contributions to the sustained
and stable development of China’s national
economy, [2]. In recent years, due to the impact of
the soft landing on China’s economy, mining
enterprises have encountered development
bottlenecks. After the supply-side structural reform
measures were put forward, mining companies
actively responded to the call and began to remove
outdated production capacity and reduce inventory.
The mining market gradually restored the balance
between supply and demand, [3], [4]. In today’s era
where challenges and opportunities coexist, it is
particularly important to establish a new economic
management performance evaluation model after
major mining companies have passed a series of
tests such as industrial upgrading and resource
integration, [4], [5]. In view of this, this study
constructs the economic management performance
model of mining enterprises under the background
of supply-side reform. It is hoped that the model
constructed in this study can effectively
comprehensively assess the degree of mining
businesses’ EMP, to determine their future
prospects. It also provides some reference value for
the policy formulation of relevant government
departments.
2 Related Work
Regarding the construction of the economic
management performance model, there are already
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2066
Volume 20, 2023
many scholars doing research on it. Researchers
such as Hall studied the foundation for performance
management focusing on policy results, and they
employed structural equation modeling to quantify
the effect of municipal tax policies on economic
development performance. The experiment
measured the impact on the economy by the
underlying structure of growth in property values
and new home construction. The final experimental
results showed that the choice of taxation could
have an important impact on economic development,
[6]. Researchers began their studies with the
connection between national public health spending,
the logistics performance index, renewable energy,
and ecological sustainability. They also used the
structural equation model to examine how the four
factors affected the performance of economic
development. The final experimental results showed
that utilizing renewable energy responsibly would
enhance environmental and economic development
outcomes. At the same time, the increase in National
spending on public health and poor environmental
performance would cause damage to the healthy
growth of the economy, [7]. Researchers
investigated the function of blockchain technology
in circular economy practice and its influence on
ecological and environmental performance and used
the least squares structural equation for modeling
this topic. The experiment selected the data of 404
companies operating in cross-border supply chains
located in China and Pakistan for empirical
research. The final experimental results showed that
blockchain technology and circular economy can
stimulate the environmental performance and
financial performance of multiple companies, which
verified the validity of the model, [8]. Several
studies integrated machine learning into the
discussion of enterprise performance management.
The experiments measured the accuracy and
interpretability of machine learning algorithms by
discussing machine learning usage in the enterprise.
Afterward, the article discussed three enterprise
cases that used machine learning algorithms. The
experiment finally provided an overall summary of
the challenges and opportunities that machine
learning algorithms needed to face when deploying
them in the enterprise, [9]. Researchers proposed
MOPSO to balance the factors of economy, energy,
ecology, coal ore economic benefits, and social
benefits in green coal production. The model was
verified on the DTLZ function, and its effectiveness
was contrasted with that of a number of other well-
known multi-objective algorithms in experiments,
which verified the validity of the model. This
method provided a reference value for the economic
performance management of coal enterprises, [10].
This article chooses to introduce machine
learning algorithms into the enterprise economic
benefit evaluation model. Machine learning
algorithms have also been favored by scholars from
all walks of life in research. To improve the quality
of corporate social responsibility performance
evaluation, researchers such as Li suggested an
enhanced AHP-BP algorithm and included the
algorithm in the CSR performance evaluation
model. The experiment used the model for CSR
performance evaluation in the BP neural network
training stage after introducing expert scoring in the
AHP stage. The final experimental results showed
that the upgraded AHP-BP model performed better
than the traditional BP model, and it could be used
as a good factor for CSR performance evaluation,
[11]. To evaluate the performance of enterprise
personnel, scholars proposed a spatially distributed
data mining algorithm based on the BP network.
The algorithm first constructed spatial network data
in cloud computing and then used the BP network to
classify and identify the mined data features.
Experimental results showed that this method had
higher accuracy in predicting the performance of
enterprise personnel and had better efficiency in big
data processing, [12]. Scholars started from the
development of real estate and developed a CSR
performance assessment methodology that took into
account aspects including financial success,
corporate morality, environmental stewardship, and
social responsibility. The model was realized based
on AHP and fuzzy comprehensive evaluation
method, and it was improved on this basis. The final
experimental results showed that the proposed AHP-
FCE model could provide a good reference value for
CSR performance evaluation, [13]. Researchers
compiled with the influence of national policies and
macroeconomics and proposed a performance
evaluation method for enterprise innovation
capabilities that combined deep learning fuzzy
systems and convolutional neural networks. This
method drew on the traditional performance
evaluation method, and at the same time introduced
an intelligent deep learning algorithm, which was a
relatively innovative enterprise performance
evaluation method. Simulation findings
demonstrated this method’s considerable
applicability and importance to firms resource
optimization strategies, [14]. Researchers
constructed an assessment model based on AHP-
DEMATEL beginning with the variables that cause
coal mine occupational illnesses. The experiment
used the model to construct the coal mine
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2067
Volume 20, 2023
occupational hazard evaluation system and
combined the case analysis. The experimental
results showed that the evaluation’s findings and the
management of coal mining firms’ current
circumstances accord rather well. And this index
had strong versatility and adaptability, and it also
had certain enlightening significance for the
evaluation model of enterprise EMP, [15].
To sum up, the performance model of enterprise
economic management and machine learning
algorithm are recent research hotspots. Although
some researchers have combined the two to conduct
relevant discussions, research on constructing an
enterprise economic management performance
model based on neural networks and other machine
learning methods is still rare. Therefore, this study
incorporated financial and non-financial
performance indicators reasonably, constructed a
mining enterprise EMP evaluation index system,
and built a mining enterprise EMP evaluation model
based on BP neural network and analytic hierarchy
process.
3 Construction of Economic
Management Performance Model of
Mining Enterprises under the
Background of Supply-side Reform
3.1 Construction of Mining Enterprise
Performance Evaluation Index System under
the Background of Supply-Side Reform
Many academics have created a technique for
evaluating the general-sense economic management
performance with an eye on the uniqueness of coal
mining companies. In the past, financial indicators
were often utilized as the basis for the indicators
used to assess economic management success. In the
context of supply-side reform, only using financial
indicators to evaluate the EMP of enterprises can no
longer continue to meet the information needs of
stakeholders. Therefore, it is necessary to use a new
perspective to look at the EMP evaluation of mining
enterprises under the background of supply-side
reform, and then formulate a sustainable, multi-
angle, and all-round economic management
performance evaluation system, [16], [17]. The
concepts of objectivity, science, and systematicity
serve as the foundation for this research, referring to
the “Blue Book of Corporate Social Responsibility
Report”. Meanwhile, based on financial
performance indicators and combined with the
characteristics of the mining industry, some non-
financial evaluation indicators have been
appropriately included. The experiment ultimately
constructed the EMP evaluation index system for
mining enterprises, as shown in Table 1.
Table 1. Economic management performance
evaluation index system of mining enterprises
Secondary
Indicators of
Financial
Performance
Level 3
Indicators of
Financial
Performance
Non-financial
performance
secondary
indicators
Solvency
(U1)
Asset-
liability ratio
(U11)
Social
Contribution
(U5)
Asset
Current
Ratio (U12)
Net
Operating
Cash Flow
Debt Ratio
(U13)
Energy saving
and
environmental
protection
(U6)
Developing
Capabilities
(U2)
Net profit
growth rate
(U21)
Growth rate
of total
assets (U22)
Operational
Capability
(U3)
Inventory
turnover
(U31)
Safe
Production
(U7)
Accounts
receivable
turnover
ratio (U32)
Total asset
turnover
ratio (U33)
Technology
Research and
Development
(U8)
Profitability
(U4)
equity (U41)
Basic
earnings per
share (U42)
\
Cost
Expense
Profit
Margin
(U43)
As can be seen from Table 1, this study
thoroughly assesses the economic management
performance of mining firms while taking into
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2068
Volume 20, 2023
account a number of variables, including
technological advancement and employee safety.
Moreover, each evaluation index in the table is a
quantitative index, which neutralizes the subjectivity
of the weight assignment of the index. Therefore, it
is reliable and objective and can avoid affecting the
evaluation results.
3.2 Construction of Mining Industry
Performance Evaluation Model based on BP
Neural Network and Analytic Hierarchy
Process
The topology of a BP neural network consists of one
input layer, one output layer, and one or more
hidden layers, [18]. The role of the input layer is to
incorporate the external information data into the
neural network; The hidden layer connects the input
and output layers; The output layer can transmit
information from the hidden layer, which can also
backpropagate the error. A multilayer perceptron
utilizing the BP algorithm is the core of a BP neural
network. BP neural network is a typical forward
network (that is, the input of the previous level is
accepted by the neuron and then output to the next
level, and the information is processed by a simple
nonlinear function). The model structure of the BP
neural network is seen in Figure 1.
... ...... ...
...
x1
x2
xn
z1
z2
zn
Input layer Hidden layer Output layer
Fig. 1: BP neural network model
Figure 1 demonstrates that the basic principle of
the BP neural network model is actually that a group
of input vectors is activated after being stimulated
by neurons after entering the network; Then
transferred to the output layer through the hidden
layer; Finally, the corresponding output is achieved
through the output layer, and a positive propagation
is completed immediately. To complete a
backpropagation from the output layer to the input
layer, the threshold value and weight value are
altered in the opposite direction by assessing the
error between the predicted output and the actual
output. Finally, after constant adjustment, the error
reaches an acceptable range. Then stop the learning
process of the model to realize the mapping between
input and output data. Assuming that the input,
hidden layer, and output nodes in Figure 1 are
respectively
i
x
,
i
y
and
i
z
, and the activation
functions are both
f
, the output of the hidden layer
node is:
i ji i j j
y f w x f net
(1)
In Formula (1),
ji
w
is the connection weight,
which connects the input node and the hidden layer
node;
j
is the hidden layer neuron threshold. The
output of the output layer node is:
i ij j l l
z f v y f net
(2)
In Formula (2),
ij
v
is the connection weight,
which connects the output node and the hidden layer
node;
l
is the neuron threshold of the output layer.
The mean square error function between the
expected and actual output is then:
2
2
1
2
1
2
ll
l
l lj ji i i i
l j i
E t z
t f v f w x












(3)
In Formula (3),
l
t
is the desired output. The
connection weights are derived by this error
function
ij
v
:
1
nkl
k
lj k lj l lj
zz
E E E
v z v z v

(4)
In Formula (4),
E
is a function of
1
k
z k n
,
and only
l
z
is related to
lj
v
, then:
l l l j l j
lj
Et z f net y y
v
(5)
In the Formula (5), let
l
(output node error) be
l l l l
t z f net
. Derive the connection weight
ji
w
through the error function of Formula (3):
j
l
ij
ji l j ji
y
z
EE
w z y w


(6)
In Formula (6), (A)
E
is a function of (A)
1
l
z l n
, one (A)
ji
w
corresponds to one (A)
j
y
,
and is related to (A)
1
l
z l n
. Trial and error are
the technique employed in this study to calculate the
number of hidden layer nodes. To do this, the model
is first trained using the fewest possible nodes,
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2069
Volume 20, 2023
followed by a progressive increase in the number of
training samples as the training proceeds. By
comprehensively considering the time of model
learning and the number of iterations, etc., the
number of hidden layer nodes finally determined in
this study is 8. The constructed economic
management performance evaluation model of
mining enterprises constructed is shown in Figure 2.
+
20
+++
Input Output
Output layerInput layer
811
BPNN
Fig. 2: Economic management performance
evaluation model of mining enterprises
Figure 2 shows the 20 input samples, 8 hidden
layer nodes, and 1 output layer node of the BP
neural network-based mining enterprise economic
management performance assessment model
developed in this research. The expected output of
the model is determined by the analytic hierarchy
process. The analytic Hierarchy Process (AHP) is an
evaluation method that can take a complex multi-
objective decision-making problem as a whole and
model the research problem hierarchically. It is
characterized by multi-level decomposition of
objectives, then dividing each level into levels with
multiple indicators, and calculating the priority
weight of indicators at each level through the
discriminant matrix, [19], [20]. The EMP evaluation
index system of mining enterprises built in this
study has various levels and covers a wide range, so
it needs to be sorted and summarized using the
analytic hierarchy process, [21]. In this study, the
EMP evaluation index system of mining enterprises
is divided into two first-level indicators, namely
financial performance indicators and non-financial
performance indicators. There are four secondary
indicators under each primary indicator, and the
secondary indicators also include the tertiary
evaluation indicators of the economic management
performance of mining enterprises. Taking the
secondary indicator U2 as an example, the indicator
weight is calculated by constructing a discrimination
matrix. The discriminant matrix is shown in
Formula (7).
11 12 1
21 22 2
12
n
n
m m mn
B B B
B B B
A
B B B
(7)
Refer to matrix A to calculate the weight vector,
use the product Mi of each row of the matrix to
solve the n square root, and obtain the normalized
vector. Based on this
1
1
n
i
i
W
, the weight
coefficient of each index can be obtained. After
obtaining the weight coefficient, the key point is to
maintain the consistency of the final judgment. Use
the formula
CI
CR RI
(
CI
is the consistency index
of the judgment matrix,
RI
is the random
consistency index of the paired comparison matrix)
to judge by taking its ratio. The judgment matrix
passes the consistency test and eventually
determines the weight of each index if
CR
is less
than 0.1. The overall analysis flow chart of the AHP
is shown in Figure 3.
Expert Build judgment
matrix
Calculate single
layer weight
subset
Single layer
consistency
inspection
Calculate total
weight subset
General floor
consistency
inspection
Adopt
Get weight
index
Fig. 3: Overall flow chart of analytic hierarchy
process
According to the logical relationship among
various indicators in the economic management
performance evaluation index system of mining
enterprises constructed in Table 1, the judgment
matrix that needs to be constructed in this study can
be calculated. The quantity and significance of the
judgment matrix are shown in Table 2.
Table 2. Number table of judgment matrix
Judgment matrix
Serial number
Overall performance
A
Financial Performance
B1
non-financial performance
B2
Profitability Level 3 Indicators
C1
Level 3 indicators of solvency
C2
Three-level indicator of operating capability
C3
Level 3 Indicators of Development Ability
C4
Three-level indicator of safety production
C5
Three-level indicators of technology
research and development
C6
Three-level indicators of energy
conservation and environmental protection
C7
Level 3 Index of Social Contribution
C8
Table 2 demonstrates that 11 judgment matrices
must be built for this investigation. Taking the
financial performance index as an example, the
judgment matrix of each evaluation index is
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2070
Volume 20, 2023
constructed. The judgment matrix of financial
performance indicators is shown in Table 3.
Table 3. Judgment matrix of financial performance
indicators
Financial
Performa
nce
Indicators
B1
Profitabil
ity C1
Solven
cy C2
Operati
ng
Capabil
ity C3
Developm
ent
Ability
C4
Weig
hts
Profitabili
ty C1
1
2
3
2
0.425
6
Solvency
C2
1/2
1
2
1
0.229
3
Operating
Capability
C3
1/3
1/2
1
2/3
0.132
3
Developm
ent
Ability
C4
1/2
1
3/2
1
0.212
8
From the expert scoring results in Table 3, the
judgment matrix of financial performance indicators
can be constructed, as follows:
1 2 3 2
1/ 2 1 2 1
11/ 3 1/ 2 1 2 / 3
1/ 2 1 3/ 2 1
B






(8)
Take the judgment matrix B1 of Formula (8) as
an example, use ATLAB R20214a software to
calculate its maximum eigenvalue and eigenvector.
The consistency ratio of the calculated results is
0.0038. See Table 4 for the weights of each
indicator. The judgment matrices A, B2, and C1-C8
are constructed in turn according to the way of
constructing judgment matrix B1. The consistency
test is carried out for each interpretation matrix, and
the consistency ratio
CR
is less than 0.1. Therefore,
the index weight distribution in the following table
is obtained through a consistency test.
Table 4. Comprehensive weight of indicators for
economic management performance evaluation of
mining enterprises
First
level
standard
layer (a)
Weig
hts
Secondary
indicator
layer (b)
Weig
hts
Three-level
indicator
layer (c)
Weig
hts
Financial
performa
nce
indicator
s
0.67
Solvency
0.23
Assets and
liabilities
0.26
Asset
current
ratio
0.41
Net
operating
cash flow
debt ratio
0.33
Develop
0.21
Net profit
0.75
ability
growth rate
Growth rate
of total
assets
0.25
Operating
capacity
0.13
Inventory
turnover
0.44
Accounts
receivable
turnover
ratio
0.39
Total asset
turnover
0.17
Profitabilit
y
0.43
Net interest
rate
0.39
Basic
earnings
per share
0.17
Cost profit
margin
0.44
Non-
financial
performa
nce
indicator
s
0.33
Social
contributio
ns
0.16
Commodity
mine output
(million
tons)
0.33
Social
contributio
n value per
share
(yuan/share
)
0.67
Energy
saving and
environme
ntal
protection
0.33
Average
recovery
rate of
mining area
0.37
Comprehen
sive energy
consumptio
n per 10,
000 yuan
output
value (ton
of standard
ore/10,000
yuan)
0.49 _
Comprehen
sive
utilization
rate of
wastewater
0.14
Safe
production
0.15
Safety
production
investment
(100
million
yuan)
0.67
Mortality
rate of
workers
mining a
million tons
of mines
0.33
Technolog
y r & d
0.36
Number of
patents
obtained in
the year
0.67
R&d
investment
as a
percentage
of revenue
0.33
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2071
Volume 20, 2023
The weights of each index in Table 4 have
passed the consistency test, so they have a certain
degree of objectivity. The weights of each index
will be used in the calculation of the evaluation
results.
4 Model Verification and Analysis of
Evaluation Results
Five mining companies are chosen for this research,
and the social responsibility report serves as the
foundation to assure the accuracy and completeness
of the data. Among them, Enterprise A is the largest
worldwide mining enterprise, a massive, all-
encompassing energy company with coal as its
primary fuel source, as well as a land, sea, and
chemical industrial company. Coal mining, washing,
smelting, geological research, and CBM
development make up the bulk of Enterprise B’s
activities. Enterprise B is more focused on coal than
Enterprise-A. Leading coal export company
Enterprise C is primarily involved in the production
of raw coal, mining, operations, manufacture of
equipment, and other enterprises. Currently focusing
on finance and logistics, Enterprise D is primarily
involved in the coal chemical industry, coal mining,
coal processing, and coal-related equipment.
Corporation E is primarily involved in the mining
and trading of coal. It is also a huge, multifaceted
enterprise that also operates in the construction
materials, coal chemical, and power supply
industries.
After collecting certain information on the
official website of enterprises and the online
information publicity platform designated by the
CSRC, 30 data from five mining enterprises A, B,
C, D, and E in 2017-2022 were selected as analysis
samples. 1-6 belong to A mining enterprise sample,
7-12 belong to B mining enterprise sample, and so
on. Set the learning samples of the BP model as 1-5,
7-11, 13-17, 19-23, and 25-29, and conduct training
and testing on the model. The following Figure 4
displays the findings.
Figure 4 demonstrates that the BP model’s mean
square error was less than the target error after 169
steps, demonstrating that the model has a strong
simulation impact. The final actual output value of
the model is shown in the figure to be extremely
near to the predicted actual output value after
training, demonstrating the model’s excellent
accuracy and suitability for performance
assessment. After saving the trained model, input
samples numbered 6, 12, 18, 24, and 30, and Table
5 displays the test results.
Train
Validation
Test
Best
Goal
160140120
100
80604020
0
10-5
10-4
10-3
10-2
10-1
167 steps
Mean Square Error (MSE)
The best verification performance is
0.0001954 in step 169
Fig. 4: Training results of BP neural network
Table 5 shows that the final real output of the BP
is remarkably similar to the predicted output after
five prediction samples have been inputted. The
absolute error is almost less than 0.01, the minimum
value of the relative error is only 0.23%, and the
maximum is only 3.11%, which is acceptable for the
EMP evaluation of mining enterprises. It is clear
that the BP model’s assessment results are excellent,
having a low error rate and high efficiency. As a
result, it may be extensively used in the evaluation
of mining businesses’ economic management
performance in the future. Based on the model
constructed, the following research evaluates the
economic management performance of mining
enterprises A, B, C, D, and E from three aspects:
comprehensive performance, financial performance,
and non-financial performance. Figure 5 shows the
results of the comprehensive performance
evaluation.
Table 5. Comparison results of expected output and
the actual output of the BP neural network
Forec
ast
sampl
e
Expec
ted
output
Actu
al
outp
ut
Relati
ve
error
Absol
ute
error
Mea
n
relati
ve
error
Mean
absol
ute
error
6
0.452
3
0.45
86
1.32
%
0.006
2
1.98
%
0.005
17
12
0.332
1
0.33
23
0.23
%
0.000
7
18
0.213
7
0.21
33
0.24
%
0.000
5
24
0.285
7
0.27
21
5.67
%
0.013
7
30
0.147
3
0.15
31
3.11
%
0.004
5
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2072
Volume 20, 2023
0
0.1
0.2
0.3
0.4
0.5
0.6
2014 2015 2016 2017 2018 2019
绩效得分
年份
A B C D E
2017 2018 2019 2020 2021 2022
Performance score
Particular year
Fig. 5: Comprehensive performance evaluation
results of five mining enterprises from 2017 to 2022
Figure 5 shows that the overall performance
levels of the mining firms A, B, C, D, and E are
quite different from one another. Since 2017, the
comprehensive performance scores of the five
mining enterprises have declined year by year, and
the decline is obvious. On the whole, they are at a
low level. However, the comprehensive
performance of mining enterprises D and E began to
show a small upward trend in 2019 and 2022,
respectively. It can be seen that there is still room
for the development of mining enterprises, and it is
inevitable that the performance will decline for
several consecutive years. At the same time, the
comprehensive performance of mining enterprise A
has always been far ahead of the other four
enterprises in 2017-2022. It can also be seen from
the figure that in 2022 the comprehensive
performance scores of mining enterprises A, B, C
and D showed a rising trend. This is mainly due to
the supply side structural reform of mining
enterprises promoted the balance of supply and
demand in the mineral market, which has promoted
the sales volume of mineral products. Therefore, the
operating conditions of the enterprise have gradually
improved. The results of the financial performance
assessment are shown in Figure 6.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
2014 2015 2016 2017 2018 2019
绩效得分
年份
A B C D E
Performance score
Particular year
2017 2018 2019 2020 2021 2022
Fig. 6: Financial performance evaluation results of
five mining enterprises in 2017-2022
The financial performance of mining enterprises
A, B, and D in 2017 can be seen in Figure 6, but the
comprehensive performance scores in Figure 5 are
quite different. This shows that the non-financial
performance scores are the main factors affecting
the EMP of these three mining enterprises in 2017,
as shown by the close financial performance of
these three mining enterprises in 2017. In 2017-
2019, the comprehensive performance score of
mining enterprise C is much higher than that of
mining enterprise E, but the financial performance
score shown in Figure 6 is very close to or even
lower than that of mining enterprise E. It is clear
that the comprehensive performance score of a C
mining enterprise depends on its good non-financial
performance, so C mining enterprises should pay
greater attention to financial management in the
future. The outcomes of the non-financial
performance assessment are shown in Figure 7.
0
0.05
0.1
0.15
0.2
0.25
2014 2015 2016 2017 2018 2019
绩效得分
年份
A B C D E
Performance score
Particular year
2017 2018 2019 2020 2021 2022
Fig. 7: Non-financial performance evaluation results
of five mining enterprises A, B, C, D, and E from
2017 to 2022
As observed in Figure 7, there is no discernible
increase or decreasing trend in the non-financial
performance level of mining firms A, B, C, D, and
E. The reason is that the market cyclical fluctuation
has little impact on the score of non-financial
performance indicators. In addition, the overall
business policies of different mining enterprises in
terms of technology research and development,
safety products, and other aspects have changed
little. Mining enterprise A also has outstanding
performance in non-financial performance, ahead of
mining enterprises B, C, D, and E, followed by
mining enterprises C, B, D, and E. However, from
the perspective of the law of economic and social
development, the non-financial performance of
enterprises should show an upward trend with the
economic development, while the non-financial
performance of enterprises A, B, C, D, and E did not
improve as expected with the economic
development, and even showed a downward trend.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2073
Volume 20, 2023
The above situation indirectly shows that in the face
of survival pressure, mining enterprises cannot take
into account the overall development, thus ignoring
the input in non-financial performance, resulting in
a decline rather than a rise in performance scores.
5 Conclusions and Recommendations
This study creatively built the EMP evaluation
index system of mining enterprises and the EMP
evaluation model of mining enterprises based on BP
neural network and analytic hierarchy process to
assess the EMP of mining enterprises against the
backdrop of supply-side reform. The study
conducted training and validation of the model by
analyzing relevant data such as the social
responsibility reports of five mining companies A,
B, C, D, and E from 2017 to 2022. The research
results show that the model constructed in this study
has a small relative error (1.98%), and the absolute
difference between the actual and predicted output
is always controlled below 0.01, indicating that the
model has certain practicability. In addition, the
study also analyzed the evaluation results of the five
mining companies. The evaluation results show that
mining company A is always ahead of the other four
companies in terms of performance, and the
economic management performance scores of
mining companies A, B, and D mainly depend on
non-financial performance scores, while Company
C should pay greater attention to its financial
management performance. Although this study has
achieved certain results, due to the relatively small
sample data, the accuracy of the model may be low.
It is hoped that it can be improved in future
research.
Despite the fact that China’s mining companies
are currently on the decline as a whole, this trend
also presents opportunities for mining company
mergers and reorganizations, which accelerates the
process of industrial structure adjustment and boosts
the core competitiveness of Chinese mining
companies. According to the design idea of the
EMP evaluation index system of mining enterprises
proposed in this study, and in light of the difficulties
existing in the comprehensive performance of
mining enterprises, the following proposals are
made. First, in terms of financial performance,
enterprises should correctly control the dynamics of
the macro market, minerals, and related industries;
And learn to be good at capturing policies
applicable to themselves, seize the opportunity of
reform, and explore new development paths.
Secondly, when it comes to technology research,
development, and invention, enterprises need to be
oriented towards reform and innovation, and meet
customer orientation; Meanwhile, it applies cutting-
edge technologies such as big data and the Internet
of Things to comprehensively upgrade core
technologies and operational models. Thirdly, to
effectively prevent and regulate the pollution of
noise, wastewater, and waste gas, as well as to
recycle the slag, it is important to make sure that
businesses have pollution prevention equipment and
processes in the manufacturing process.
The method in this paper uses the trained neural
network to set the weight on a given threshold so
that it can evaluate the comprehensive performance
of any coal mining company, which is more
operable in practical applications. The model
constructed in this paper can be applied to evaluate
the comprehensive performance level of coal
enterprises and predict their future development
potential. Meanwhile, it can also provide some
reference for government departments to formulate
public policies, so it has great application potential.
This article innovatively combines BP neural
network and analytic hierarchy process to
effectively allocate and quantify the weight of each
indicator. The model transforms subjective and
artificial judgments into objective statistical data,
thereby enhancing the rationality, intuition, and
credibility of the analysis conclusions. This research
provides an important reference value for future
researchers to apply intelligent algorithms to the
estimation of enterprise economic benefits.
References:
[1] Zhao L, Dai T, Qiao Z, Sun P, Hao J, Yang Y,
Application of artificial intelligence to
wastewater treatment: A bibliometric analysis
and systematic review of technology,
economy, management, and wastewater reuse,
Process Safety and Environmental Protection,
Vol.133, 2020, pp. 169-182.
[2] Bassetti T, Blasi S, Sedita S R, The
management of sustainable development: A
longitudinal analysis of the effects of
environmental performance on economic
performance, Business Strategy and the
Environment, Vol.30, No.1, 2021, pp. 21-37.
[3] Rehman Khan S A, Yu Z, Sarwat S, Godil D,
Amin S, Shujaat S, The role of block chain
technology in circular economy practices to
improve organisational performance,
International Journal of Logistics Research
and Applications, Vol.25, No.4-5, 2022, pp.
605-622.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2074
Volume 20, 2023
[4] Kovacova M, Lăzăroiu G, Sustainable
organizational performance, cyber-physical
production networks, and deep learning-
assisted smart process planning in Industry
4.0-based manufacturing systems, Economics,
Management and Financial Markets, Vol.16,
No.3, 2021, pp. 41-54.
[5] Levytska S, Akimova L, Zaiachkivska O,
Karpa M, Gupta S, Modern analytical
instruments for controlling the enterprise
financial performance, Financial and Credit
Activity Problems of Theory and Practice,
Vol.2, No.33, 2020, pp. 314-323.
[6] Hall J L, Kanaan D Z, State tax policy,
municipal choice, and local economic
development outcomes: A structural equation
modeling approach to performance
assessment, Public Administration Review,
Vol.81, No.3, 2021, pp. 459-474.
[7] Khan S A R, Zhang Y, Kumar A, Zavadskas
E, Sreimikiene D, Measuring the impact of
renewable energy, public health expenditure,
logistics, and environmental performance on
sustainable economic growth, Sustainable
Development, Vol.28, No.4, 2020, pp. 833-
843.
[8] Khan S A R, Razzaq A, Yu Z, Miller S,
Industry 4.0 and circular economy practices:
A new era business strategies for
environmental sustainability, Business
Strategy and the Environment, Vol.30, No.8,
2021, pp. 4001-4014.
[9] Lee I, Shin Y J, Machine learning for
enterprises: Applications, algorithm selection,
and challenges, Business Horizons, Vol.63,
No.2, 2020, pp. 157-170.
[10] Cui Z, Zhang J, Wu D, Cai X, Wang H,
Zhang W, Chen J, Hybrid many-objective
particle swarm optimization algorithm for
green coal production problem, Information
Sciences, Vol.518, 2020, pp. 256-271.
[11] Li W, Xu G, Xing Q, Lyu M, Application of
improved AHP-BP neural network in CSR
performance evaluation model, Wireless
Personal Communications, Vol.111, No.4,
2020, pp. 2215-2230.
[12] Liang W, Li T, Research on human
performance evaluation model based on
neural network and data mining algorithm,
EURASIP Journal on Wireless
Communications and Networking, Vol.2020,
No.1, 2020, pp. 1-14.
[13] Li W, Xu G, Zuo D, Zhu J, Corporate social
responsibility performance-evaluation based
on analytic hierarchy process-fuzzy
comprehensive evaluation model, Wireless
Personal Communications, Vol.118, No.4,
2021, pp. 2897-2919.
[14] Abudureheman A, Nilupaer A, He Y,
Performance evaluation of enterprises’
innovation capacity based on fuzzy system
model and convolutional neural network,
Journal of Intelligent & Fuzzy Systems,
Vol.39, No.2, 2020, pp. 1563-1571.
[15] Cheng L H, Cao D Q, Guo H M, Analysis of
coal mine occupational disease hazard
evaluation index based on AHP-DEMATEL,
Archives of Environmental & Occupational
Health, Vol.76, No.7, 2021, pp. 372-384.
[16] Khan O, Daddi T, Iraldo F, The role of
dynamic capabilities in circular economy
implementation and performance of
companies, Corporate Social Responsibility
and Environmental Management, Vol.27,
No.6, 2020, pp. 3018-3033.
[17] Tuffour J K, Amoako A A, Amartey E O,
Assessing the effect of financial literacy
among managers on the performance of small-
scale enterprises, Global Business Review,
Vol.23, No.5, 2022, pp. 1200-1217.
[18] Zhou Y, Zhou W, Lu X, Evaluation index
system of green surface mining in China,
Mining, Metallurgy & Exploration, Vol.37,
No.4, 2020, pp. 1093-1103.
[19] Inshakova A O, Frolova E E, Rusakova E P,
Kovalev S, The model of distribution of
human and machine labor at intellectual
production in industry 4.0, Journal of
Intellectual Capital, Vol.21, No.4, 2020, pp.
601-622.
[20] Ortiz-Barrios M, Cabarcas-Reyes J, Ishizaka
A, Barbati M, Rueda N, Jesús G, Zambrano
C, A hybrid fuzzy multi-criteria decision
making model for selecting a sustainable
supplier of forklift filters: A case study from
the mining industry, Annals of Operations
Research, Vol.307, No.1, 2021, pp. 443-481.
[21] Wu Y, Li X, Liu Q, Tong G, The analysis of
credit risks in agricultural supply chain
finance assessment model based on genetic
algorithm and backpropagation neural
network, Computational Economics, Vol.60,
No.4, 2022, pp. 1269-1292.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2075
Volume 20, 2023
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
The author contributed to the present research, at all
stages from the formulation of the problem to the
final findings and solution.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
No funding was received for conducting this study.
Conflict of Interest
The authors have no conflicts of interest to declare
that are relevant to the content of this article.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.180
Xinhui Wang
E-ISSN: 2224-2899
2076
Volume 20, 2023