Mathematical Problems in Engineering

Mathematical Problems in Engineering / 2021 / Article

Research Article | Open Access

Volume 2021 |Article ID 5067092 | https://doi.org/10.1155/2021/5067092

Xin Wang, Bo Luo, "Application of Service Modular Design Based on a Fuzzy Design Structure Matrix: A Case Study from the Mining Industry", Mathematical Problems in Engineering, vol. 2021, Article ID 5067092, 19 pages, 2021. https://doi.org/10.1155/2021/5067092

Application of Service Modular Design Based on a Fuzzy Design Structure Matrix: A Case Study from the Mining Industry

Academic Editor: Sergio Ortobelli
Received18 May 2020
Accepted19 May 2021
Published16 Jun 2021

Abstract

The development of customized service is an important way to transform and upgrade China’s mining industry. However, in practice, there remain problems, such as the slow market response speed of service providers and the contradiction between the large-scale development of service providers and the personalized service needs of service demanders. This paper uses the theory and method of service modular design to solve these problems and explores the process-based service modular design method. Service modular design depends largely on the determination of the relationship between service activities and the reasonable division of modules. However, previous research has rarely made use of modular design methods and modeling tools in the mining service context. At the same time, evaluations of the relationship between service activities relying on knowledge and those relying on experience have been inconclusive. Therefore, this paper proposes a service modularization design method based on the fuzzy relation analysis of a design structure matrix (DSM) that solves the optimal module partition scheme. Triangular fuzzy number and fuzzy evidence theory are used to evaluate and fuse the multidimensional and heterogeneous relationship between service activities, and the quantitative processing of the comprehensive relationship between service activities is carried out. On this basis, the service module structure is divided, followed by the construction of the mathematical programming model with the maximum sum of the average cohesion degree in the module and the average coupling degree between modules as the driving goal. The genetic algorithm is used to solve the problem, and the optimal module division result is obtained. Finally, taking the service modular design of SHD coal production enterprises in China as an example, the feasibility of the proposed method is verified.

1. Introduction

With the transformation of the global industrial structure into a service-oriented economy, all kinds of enterprises have begun to explore ways to provide customized services for customers. As early as the 1980s, in order to achieve sustainable development, the mining industry in Australia, New Zealand, Indonesia, South Africa, China, and other countries was no longer satisfied with one-time mineral product transactions, and began to provide mineral product development services for other mining enterprises, such as equipment maintenance, mine construction, mineral product transportation, mine design, and mineral product mining. This approach has since become one of the most important means to improve the income of the mining industry [1, 2]. Drawing on this literature on field investigations of mining enterprises in Northern Shaanxi, and on studies and news reports of the transformation of mining industry services in Inner Mongolia, Ningxia, and Shandong, this paper finds that the common features across mineral product development services in different countries and enterprises are that service providers take customer satisfaction as their goal and use their own redundant resources, knowledge, and technology. According to the functional needs of the service demander, mine conditions, service content design, resource allocation, business management, and other activities within the limitations of the construction period and cost are integrated into solutions that meet the specific needs of customers, and the essence is customized service.

Because of the implementation of the supply-side reform in China, the mining volume of mineral products is limited, and the requirements for safe production and management are more stringent. The supply and demand of customized services have increased, and economic benefits have been created for enterprises. However, this study finds that the industry as a whole continues to lack norms and standards that service providers are slow in responding to the market, and that there are contradictions in service providers’ pursuit of high efficiency, low cost, and customer personalized demand design.

First of all, although the mining enterprises that provide the business call themselves service providers, the definition of the name, scope, and strategic meaning of the business is neither clear nor consistent. Different enterprises have different names for this business, such as entrusted management, outsourcing, contract mining, specialized services, coal mine trusteeship, and mining services.

Second, the supplier and demander of the service agree on the content, mode, duration, price, and responsibility for the service through negotiation, specify the form and requirements of the deliverable, and finally sign a contract. Therefore, the service scheme is characterized by a high degree of randomness. The service content can be the whole mine trusteeship, or one or more links or processes in the mine development process. Given this lack of standardized business scope, it is difficult for providers to achieve a balance between scale benefit and customer personalized demand.

Third, in each stage of mine development, customer demand continues to change. At the same time, and given the long production cycle of the mining industry, geological conditions, mining time, and other factors continue to change, as does customer demand. When demand changes significantly, service providers need to replan the service process and schedule service resources; this makes it difficult to respond to demand changes quickly.

In this study, the purpose of customization is to meet the personalized needs of customers while maintaining the production efficiency and costs of service providers to achieve rapid market response. The difficulty lies in how to transform the development process of mineral products into several technically feasible specific service tasks, maintain a relatively stable structure over a certain period of time, and combine the tasks into service projects to meet customer requirements. Baldwin and Clark, pioneers of modularity theory, argued that modularity is an effective strategy for organizing and designing complex products or processes [3, 4]. According to modularity theory, complex systems can be decomposed into semiautonomous subsystems that can be designed independently. In recent years, research on service modularization has gradually moved to the forefront of academic interest and has been applied in many industries, and previous studies have shown that the idea of service modularization is the most appropriate theory for addressing the complexities of customized service. The present study argues that we can use the theory and method of service modularization to solve existing problems.

The production object and final product of mining industry are mineral products. The production purpose and process of an enterprise is to mine underground mineral products and then transport them to ground level for processing and sales [5]. From the perspective of service modularization theory, the essence of customized service in the mining industry is the modularization design based on process. In this study, the subprocesses within the process of mineral product development are considered as service products, and process activities with specific functions are identified to form independent service modules. Each module has a deliverable, which is taken as the interface between modules. Because of the adoption of modular technology to promote the standardization of modules, the cost of suppliers can be reduced. At the same time, through the combination of service modules, diversified service products can be formed quickly to meet the needs of specific customers.

This research will help to expand the application scope of service modularization theory, improve the service modularization design research based on process, and resolve difficulties in business model innovation and enterprise transformation in the coal, oil, natural gas, and other industries. In order to achieve service modularization, the basic problems that must be addressed are the determination of the relationship between the process activities and the division of modules.

2. Research Status and Research Ideas

2.1. Literature Review

In recent years, Chinese scholars He, Zhang, Yang, and Liu have used the case study method to summarize the application modes and safeguard measures of several mining enterprises carrying out customized services according to the practice of coal enterprises [69]; however, their work has not fully addressed issues of fairness or included all the relevant concepts, theories, and methods. Research on the transformation of the mining industry from a producer of mineral products to a service provider thus remains in its infancy.

2.1.1. Theoretical Research on Service Modularization

In 1993, Pine et al. claimed that service modularization is feasible, and was the first to put forward the judgment that both tangible products and intangible services can realize mass customization [10]. Subsequently, Sundbo noted that service modularization is the process of standardizing service products and combining them with customers, explaining that service product standardization can secure a productivity advantage and a customer satisfaction advantage provided by customization [11, 12]. Baldwin and colleagues pointed out that mass customization of services can be realized by adopting the “architecture module” innovation method of product development [3, 4]. Yu et al. [1315] explored the value creation logic of service modularization, and through case studies and empirical studies in three industries found that service modularization is a new way to realize value creation of service customization. The service modularization theory has also been applied in financial services, civil aviation services, manufacturing product services, and other industries and fields [13, 16, 17]. Yan and Jiang conducted a bibliometric study on service modularization, and found that some scholars put forward that the key of service mass customization is the modularization of service process system, which is a clearer entry point for enterprises to implement service modularization [18]. Xia and Xue observed that the modular decomposition and integration of business processes to form different service products can meet different needs within the market [19]. Most recently, Yu and others conducted research from the perspective of division of labor and concluded that service process modularization can be achieved by decomposing the tightly coupled process system into subprocess modules [20].

To sum up, scholars believe that accurate module segmentation of service process can lead to service standardization and generalization, thereby achieving customized service. Nevertheless, the process is generally composed of continuous activities and is not easily divided into service modules. The case of mineral product development is particularly complex, as it consists of multiple heterogeneous activities. Each activity is composed of multiple operations, processes, and resources, and there are complex relationships in multiple dimensions, including function and resource dependence. In order to identify the service activities and their relationship in the process of mineral product development, divide them appropriately into service modules, and combine them into customized service schemes with feasible technology, shorter construction periods, lower costs, and coordinated operations, we need to explore in depth the relationship between service activities and clarify the method of process system division and combination.

2.1.2. Research on the Service Module Partition Method

Early research on the method of service module partition based on process mostly used the service blueprint and quality function methods. Through the research of service modularization literature, Yan and Jiang found that scholars use service blueprint method to form process module by merging and standardizing specific service processes [18]. Service blueprint, house of quality, critical path method (CPM), program/project evaluation and review technique (PERT), graphic evaluation and review technique (GERT), and other technologies are often used to describe the elements and their relationships in the process; however, despite their suitability for serial and one-way processes, their ability to describe the coupling relationship between subsystems is weak, and they are not appropriate for multidimensional integration analysis. Browning argued that a design structure matrix (DSM) can better describe various resources and the interdependence between them, and that it is an effective tool for solving the management problems of complex engineering, particularly in terms of calculation and analysis [21]. In recent years, however, scholars have agreed that clustering DSM to obtain a modular partition scheme is a better method [22, 23]. Yang et al. took the process of a complex R&D project as their research object, proposing a theoretical analysis framework based on DSM tools for measuring, optimizing, and clustering the dependency relationship between subprocesses [24]. Sakao et al. constructed a DSM based on the interaction relationship between service elements and studied the service module division of the customized product service system, but they did not conduct quantitative research on the strength of the interaction relationship [25]. Ying et al. used a self-organizing mapping algorithm for module clustering calculation of numerical DSM, but this has high requirements for dependency data between activities and cannot resolve situations of complex and uncertain dependency [26]. Chen et al. quantitatively calculated the relationship between subtasks on a cloud service platform and mapped it into a DSM to obtain a task composition scheme [27]. Wen and Gao introduced a DSM model to sort and quantify the service activities related to product production, and obtained the optimal functional module partition scheme of an intelligent distribution system [28]. Cheng and Luo used a DSM to express the asymmetric association between product parts, and then applied a fuzzy algorithm to cluster the parts [29]. However, they considered only the information interaction and comprehensive interaction between tasks, ignoring the differences of task interaction strength under different attributes. Jin and Geng took the product service system (PSS) as their research object and used the fuzzy design structure matrix (FDSM) as a module partition tool, but it proved necessary to artificially select the threshold to cluster process activities in order to achieve module partition [17].

In summary, previous research has a number of shortcomings. First, there are two types of research objects: manufacturing and service industries. Scholars have discussed service module partition methods based on product structure and resource types. These methods are inevitably predicated on product structure and do not consider the relationships between service activities. Therefore, they are not suitable for service module partition when the final product has no structure or where there is complex interaction between service activities. Second, module partition must clarify the relationship between the components that form the module. However, there has been little research in this area, and almost no consideration has been given to the relationship between the components of the module in different dimensions. Therefore, there are relatively few quantitative studies on building a comprehensive relationship between service activities in multiple different dimensions.

Third, previous studies have not identified the structure of the service module system or the type of service module, an oversight that is not conducive to research based on the service modularization theory. The concept of modular composition for customized scheme designs does not address the problems of market response speed, service duration, or cost control. Fourth, although scholars have used DSM tools to describe the complex interaction between services, they have seldom discussed the differences of interaction in different dimensions or the processing method of information fusion in multidimensional situations, and they have ignored the fuzzy evaluation information of the association relationship. However, the accuracy of the association information will directly affect the rationality of the service module partition scheme.

Fifth, in the literature on module clustering based on DSM, most studies have adopted the fuzzy clustering method and set the threshold value artificially, which leads to the subjectivity of the partition result. The common way is to first decompose DSM into a block diagonal matrix and then carry out row–column transformation. However, this method has the disadvantage of low efficiency when dealing with the clustering problem for an asymmetric DSM and a large-scale DSM with a large number of row and column elements.

2.2. Research Ideas

In view of the shortcomings of previous research, this paper takes DSM as its basic tool and uses triangular fuzzy number and evidence theory and its algorithm to determine the relationship between service activities and reasonable segmentation of service modules. The research process is shown in Figure 1.

The first step is to analyze the business process, identify service activities, and establish the multidimensional service activity association. From different perspectives, the degree of association between service activities is different, and the evaluation information is uncertain. In order to take account of these situations, this paper adopts DSM, uses triangular fuzzy number theory to express the quantitative results of fuzzy association strength of service activities, and models the DSM of service activities from three dimensions.

In the second step, drawing on fuzzy evidence theory, the evaluation information for different dimensions of the service activity design structure matrix (SER-DSM) is fused to form a comprehensive DSM under multiple criteria.

In the third step, based on the comprehensive DSM and according to the quantitative results of the row and column elements of the matrix, the structure of the design matrix is identified, and the service activities are divided into three types.

In the fourth step, the elements of the structure matrix are designed synthetically for the parts that need clustering calculation, and the mathematical programming model is constructed to maximize the sum of the average cohesion degree of modules and the average coupling degree between modules. The genetic algorithm is used to solve the problem, and the optimal customized service module partition scheme is obtained.

In the final step, the results are applied to the modular design process of customized service in coal mining enterprises to verify the feasibility of the proposed method.

Scholars have defined the concept of service from different perspectives. According to Lovelock et al. and Li, service is a complete process composed of a single activity or a series of activities that has use value and a certain exchange value [30, 31]. Accordingly, this study regards service modularization as the process of modularizing all activities of the whole service process. The minimum unit studied here is service activity, which is the basic construction element of the service module. The service module is composed of a series of interrelated service activities that have more complex and complete functions than single service activities. It should be noted that in the mining industry, each service activity produces a variety of deliverable forms, including physical products, engineering projects, equipment, facilities, materials, and drawings. This paper therefore considers the service activity as having technical independence and functional integrity, including all operations or processes for completing the activity, as well as the necessary resources and services required to complete these activities. We measure the service result and service quality in terms of the deliverability of service module.

In 1981, Steward proposed DSM as a structural modeling tool to represent the elements and their interactions in a system. McCulley and Bloebaum later established a numerical DSM model, which not only quantitatively describes the dependency relationship between matrix elements but also represents the information transfer direction between elements [32]. Thus, the model has strong problem description ability. According to different application fields, DSM models can be divided into element, team, operation, and parameter types [33]. Element DSM is often used to represent the relationship between the system elements and the system structure. It describes the system structure intuitively and quantitatively but does not need to be refined to specific parameter values, which meets the needs of this study. Therefore, this paper uses an element-type numerical DSM model to describe the structural characteristics of service activities in the process of service activities and their relationships, also known as the SER-DSM. The row and column elements in the model are composed of various service activities, and the values between the row and column elements represent the relationship strength between service activities.

4. Analysis and Modeling of Association Relationships of Service Activities

Identifying all service activities and their relationships is the basis of service module division, which is very important for the design of a customized service scheme. In different dimensions, the direction and strength of service activities are different, and the quantification of their relationship often depends on the evaluations of multiple experts, which constitutes a multidimensional fuzzy fusion evaluation problem. Therefore, it is necessary to analyze the relationship of service activities in each dimension and then integrate them to build a comprehensive SER-DSM model.

4.1. Semantics of Fuzzy Association Relationships in Service Activities

As it is impossible in the mining industry to analyze the relationship of service activities according to product structure, it is very difficult to measure the relationship of service activities. Generally, this is evaluated by means of investigation and expert consultation. However, when the information is fuzzy and incomplete, experts cannot express their evaluation views with accurate values. In this study, a triangular fuzzy number is used as input for fuzzy logic deduction. For any triangular fuzzy number, the form (n1, n2, n3) gives the minimum, most likely, and maximum values of the elements in the triangular fuzzy number, respectively. The algorithm is as follows:

In this paper, the confidence threshold of membership degree is determined using the cut set [34]. The corresponding relationship between triangular fuzzy number and fuzzy semantics is then established and the grades are divided as in Table 1.


Triangular fuzzy numberFuzzy semanticsGrade number

(0, 0.1, 0.3)Weak1
(0.2, 0.3, 0.4)Unimportant2
(0.4, 0.5, 0.6)Common3
(0.6, 0.7, 0.8)Important4
(0.8, 0.9, 1)Very important5

4.2. Identification of Service Activities and Initial DSM Matrix

The core technology and complexity of different service activities are different. In the mining industry, there is a technical and economic connection between service activities. Therefore, according to the principle of technical feasibility and functional independence, the process nodes on the mining business flow chart can be decomposed or merged to obtain the service activity set.

If a business process has n service activities, the whole process can be represented by an N × N DSM matrix, as shown in Figure 2, aggregate . Every node represents a service activity, corresponding to the ith row and ith column in SER-DSM. The element on the nondiagonal line represents the relationship strength between the activities, and is represented by . The line element indicates that the service activity corresponding to the line can support other activities, and the column element indicates the support of other activities required by the column. The 0 element is used to indicate that there is no relationship between the corresponding service activities. The diagonal element has no meaning and is set to 1.

4.3. Classification of Relationship Types among Service Activities

Previous research on service module partition has evaluated the relationship between activities on the basis of product structure, but this method is not suitable for evaluating the relationship between service processes that do not depend on product structure [24, 29, 3539]. Extrapolating from this literature in the context of the characteristics of mining product development process, the present study describes the relationship between service activities on the three dimensions of function, resource, and time (shown in Figure 3), and measures the correlation strength.

In order to simplify the multidimensional model, this paper builds a functional SER-DSM, a resource SER-DSM, and a time SER-DSM and then establishes a comprehensive relationship.

4.3.1. Functional SER-DSM Structure Matrix

The first dimension is the functional dimension, which indicates the relevance of two service activities for achieving new functions or higher-level functions, and can be measured by the necessity or degree of collaboration between them. Taking the coal production process as an example, in order to realize the function of raw coal stripping, tunneling and coal mining must be linked. When the connection between and has an adverse effect on a certain function and the connection between them must be avoided, or when the connection between and cannot achieve other functions, the functional relationship between and is recorded as 0. The relationship between functional association and the triangular fuzzy number is shown in Table 2. Since the functional correlation is symmetric, the functional SER-DSM structure matrix is a symmetric matrix.


Functional relationshipFuzzy semanticsTriangular fuzzy number

The connection between and does not hinder the normal completion of the function.Weak(0, 0.1, 0.3)
The connection between and is beneficial but not necessary.Unimportant(0.2, 0.3, 0.4)
The connection between and is beneficial and the synergy is average.Common(0.4, 0.5, 0.6)
The connection between and is beneficial and the synergy is strong.Important(0.6, 0.7, 0.8)
The connection between si and sj is indispensable for the realization of new functionsVery important(0.8, 0.9, 1)

4.3.2. Resource SER-DSM Structure Matrix

The second dimension is the resource dimension, which represents the degree of resource sharing between two service activities. Based on the theory of resource dependence, resources can be classified [17, 39]. The present study divides resources into three categories: hardware resources are the materials, equipment, venues, and energy needed to carry out service activities; software resources are the knowledge, technology, data, information, and application software that support service activities; and human resources are the personnel needed for internship service activities, such as technicians, managers, and workers. The resource association relationship description and triangular fuzzy number representation for this study are shown in Table 3. If and do not share any resources, the resource relationship between and is recorded as 0. The SER-DSM structure matrix of resources is symmetrical because the type and intensity of service activities and shared resources are the same.


Functional relationshipFuzzy semanticsTriangular fuzzy number

and share only part of the software resources.Weak(0, 0.1, 0.3)
and share hardware and human resources, but the resources are sufficient and can be completely replaced.Unimportant(0.2, 0.3, 0.4)
and share hardware and human resources, which are limited but replaceable.Common(0.4, 0.5, 0.6)
and share hardware and human resources, which are important but replaceable.Important(0.6, 0.7, 0.8)
and share hardware and human resources, which are critical and irreplaceable.Very important(0.8, 0.9, 1)

4.3.3. Time SER-DSM Structure Matrix

The third dimension is the time dimension, which indicates the connection between the execution time of two service activities and can be measured by their execution time and execution results. Unlike processed products, when there is a time correlation between service activities there is a sequence between the description and execution time, and the execution result can create execution conditions (such as information, site, energy, equipment, and tools); for example, the installation of coal electromechanical equipment must be executed before coal mining activities and underground production conditions for coal mining activities can be created.

Generally, scholars agree that there are three time relationships between processes and operations: parallel, sequence, and coupling. Studies have indicated that the implementation of a complex project process has the characteristic of overlapping in time, which is consistent with the survey results of studies of coal production enterprises [24]. Research has also suggested that one activity creates resource conditions for another activity, and that multiple activities can jointly create resource conditions for another activity [40]. The present study identifies a further feature of process activities: for example, activity A may be carried out earlier in time than activity B, but instead of producing intermediate products or semifinished products, activity A only creates the execution conditions for activity B. Therefore, this study assumes that we can measure the intensity of association between activities by creating conditions for other activities.

Proceeding from the above analysis, this study divides the time relationship of service activities into parallel, sequence, overlap, cycle, and coupling, and the DSM is shown in Figure 4. A sequential relationship means that the upstream activities precede the downstream activities or create operational conditions for the downstream activities to start; a parallel relationship means that there is no execution time sequence between activities and the execution results do not affect each other; a cyclical relationship means that the upstream and downstream activities are executed in sequence and form a closed loop, and that the execution of needs , and vice versa; a coupled relationship means that the upstream and downstream activities are executed alternately many times to create each other’s operation conditions; and an overlapping relationship means that when the upstream and downstream activities are executed in sequence, the downstream activities begin before the upstream activities are completed. When and have no temporal relationship and the execution results do not affect each other, the value is recorded as 0. Because of the different time relations of service activities, the time SER-DSM is an asymmetric matrix. The semantic and triangular fuzzy number expressions of time association are shown in Table 4.


RelationshipTemporal correlationFuzzy semanticsTriangular fuzzy number

Parallel is executed before , but the result of execution has nothing to do with .Weak(0, 0.1, 0.3)
Overlapping is executed before , and the partial execution result of the first can create conditions for the second.Unimportant(0.2, 0.3, 0.4)
Sequential must precede and create conditions for .Common(0.4, 0.5, 0.6)
Cycle must precede and circulate once to create conditions for each other.Important(0.6, 0.7, 0.8)
Coupled and go back and forth many times to create conditions for each other.Very important(0.8, 0.9, 1)

4.4. Construction of Comprehensive SER-DSM Matrix

The SER-DSM structure matrix is obtained by weighted summation of all dimensions. Although the function SER-DSM and resource SER-DSM are symmetric matrices, the time SER-DSM is asymmetric, and therefore the synthetic SER-DSM is asymmetric. In addition, because the relationship between service activities is evaluated by many experts in different fields according to their experience and understanding, the construction of a comprehensive SER-DSM is a group evaluation problem, which requires aggregation of the experts’ evaluation information. The specific methods and steps used are as follows:Step 1: build the expert evaluation matrix. There are p experts to form a multiperson evaluation group to evaluate the dimensions of association criteria. The evaluators’ evaluation results of service activities under each dimension are expressed by the triangular fuzzy number matrix :where means that under criterion q, evaluator p’s evaluation value of the relationship between and does not need to be standardized when .Step 2: fuse expert evaluation information to obtain the SER-DSM of each dimension. Evidence theory (ET) is suitable for dealing with the uncertainty problem, as it is in line with people’s thinking habits and can solve the evaluation problem based on subjective judgment. At the same time, a triangular fuzzy number can be combined with ET to solve the problem of fuzzy information fusion, and research in different fields has verified the effectiveness of this method [4143]. Therefore, this study uses a fuzzy evidence reasoning algorithm for expert evaluation information fusion.The fuzzy identification framework is composed of N evaluation grades, and denotes the k evaluation grade under criterion q. Its semantics are shown in Tables 14. is the confidence level of for the correlation between and under criterion q. According to formulas (2) and (3), the SER-DSM matrix of each dimension is obtained by fusing the evaluation values of and given by experts under criterion q, and the elements in the matrix are . is the number of experts who evaluate the relationship between and as under the q criterion, and E is the number of experts who participate in the evaluation. The SER-DSM matrix representation of each dimension is shown in Table 5.Note that the basic probability quality of the association between and under the q criterion is , and its residual quality probability is . Then, is divided into residual probability caused by relative importance and residual probability and caused by incomplete information. Let the weight of the q criterion be and satisfy . The weight of each dimension can be determined using the analytic hierarchy process (AHP) and other methods, which will not be discussed in this paper.Step 3: integrate the multiple criteria evaluation information to obtain the comprehensive SER-DSM. In this study, a fuzzy number analytic evidence reasoning algorithm proposed in the literature is used to fuse the consistency of expert group evaluation under multiple criteria [43]. The symbols used are defined in Table 6.Step 4: obtain the triangular fuzzy number of the synthetic SER-DSM and use the mean area method to obtain the clear synthetic SER-DSM matrix.


Evaluation grade123……n

1k = 11…………
k = 2
k = 3
k = 4
k = 5

2k = 11…………
k = 2
k = 3
k = 4
k = 5
…………

n…………1


SymbolDefinition

Scale factor
After fusing the multiple criteria, the correlation between and is the basic probability quality of the kth level
Residual probability caused by relative importance after fusing multiple criteria
Residual probability caused by incomplete information after fusing multiple criteria
After fusing multiple criteria, the relationship and is the belief degree of the second level
Value of comprehensive relation and after fusing multiple criteria
Weight of the qth criterion

5. Structure Division and Clustering Calculation of Service Activity DSM

Using the synthetic SER-DSM matrix, the clustering calculation is carried out to determine which system elements have the potential correlation to form modules, so that each module of the system contains most or even all of the mutual relationships while minimizing the mutual influence and correlation between each module.

5.1. Structure Division of Service Activity DSM

Wise and Baumgartner divided services into four parts [44]. On this basis, taking into account the relationship characteristics of row and column elements in the DSM model, service activities are divided into three categories, as shown in Figure 1.

5.1.1. Independent Activities and Modules

In the integrated SER-DSM, service activities whose row and column elements all have the value 0 are called independent activities. There is no connection between an independent service activity and other service activities, and each independent activity can form an independent module. Since there is no influence on the basic functions, construction period, cost, or market response speed of customer demand, an independent module can be selected by customers themselves and combined with other modules at will; thus, the degree of customization is general. In order to reduce the computing scale, it is not necessary to cluster.

5.1.2. Common Activities and Modules

A service activity whose all row and column elements do not have the value 0 in the comprehensive SER-DSM is called a common activity. This activity is related to all the other service activities and is a shared element in the whole system. Each common activity can form a common module separately. Common modules do not need to be customized, but they are elements that service providers need to match in order to realize or guarantee the basic functions of services, such as management and finance. A common module can be combined directly. They are poorly customized and do not need to be clustered.

5.1.3. General Activities and Modules

After removing independent activities and common activities, we can identify closely related service activities, which are called general activities. They are important factors affecting the scope, schedule, and cost of customized services. These general activities are mainly reflected in the differences of customer needs, and the customer’s mine parameters will affect the cost and duration of these service activities. Therefore, we must consider whether they can be combined quickly. Therefore, it is necessary to introduce clustering calculation and divide it into general modules.

5.2. Clustering Calculation

Clustering based on DSM involves high cohesion and low coupling between modules. Previous studies of the clustering principles of Boolean and numerical DSM have adjusted the influence of the degree of cohesion and coupling on the clustering module by setting weights [26, 45]. However, those studies used Euclidean distance to judge the closeness of the relationship and did not directly take into account the value of the relationship between the elements in the matrix; this increased the number of calculation steps and ignored the direction of the relationship. Accordingly, the present study adopts the principle of high cohesion within modules and low coupling between modules, taking the maximum sum of intramodule and intermodule correlations as the optimization objective.

Although there are many clustering methods, given that the complexity of the clustering calculation in designing the structure matrix increases sharply with the expansion of the matrix scale, this study uses heuristic methods, including the genetic algorithm. The optimization objective function constructed in this paper is expressed as follows:where is the average cohesion degree, is the average coupling degree, and are the weight coefficients, and . In this study, .

Cohesion is the association strength among service activities within a module. The larger the association value within a module, the better the clustering effect of the module. There are modules, where is the number of service activities in module , and are any two service activities in module , and is the association value of and . The association value between all service activities in n module is recorded as , and the average value of the relationship in all modules is . The larger the value of , the closer the relationship within the module and the better the cohesion of the module. The formula is as follows:

The coupling degree is the connection between modules. The higher the value, the smaller the connection between modules and the stronger the independence of modules. Suppose modules; , are any two modules; is any service activity in module ; is any service activity in module ; , are the number of service units in module and module , respectively; is the correlation value of service unit in module to service activity in module ; is the correlation value of service unit in module to service activity in module ; represents the relationship between modules and ; and represents the average relationship between modules. The larger the value of , the greater the correlation and coupling between modules and the worse the module separation.

Following analysis methods used in the literature [46], this paper constructs the fitness function by using the lower bound construction method:where is the conservative estimate of the bound of the objective function, , and the fitness function must be positive. Let .

In this study, integer coding is used, and the gene values of each individual in the initial population are generated using uniformly distributed random numbers. The selection operator is roulette selection, and the crossover operator is single-point crossover and the basic bit mutation. In order to determine the value of the parameters, the orthogonal experiment of parameters is designed to obtain the optimal combination [47, 48].

6. Case Analysis

SHD is a large-scale coal production enterprise located in China’s energy-rich areas. Its main business is coal development, production, and operation. The company has the world’s top equipment and great technical strength. In recent years, in accordance with the national requirements of coal production reduction and production limits, some equipment and personnel have remained idle. In order to revitalize these idle resources and increase enterprise efficiency, since 2016 the company has introduced technology and equipment aimed at internal services to the market, going on to provide coal machinery and equipment transportation, coal washing, equipment repair, and maintenance for many coal production enterprises in Inner Mongolia, Shanxi, and other locations, as well as a coal production enterprise in Northern Shaanxi to provide coal mine hosting services. As of the first half of 2020, the service revenue of SHD accounts for about 20% of its total revenue. However, the profit margin is low; there are no systematic service products, and the service items are limited, making it difficult to respond quickly to customer needs.

6.1. Service Unit Identification

Based on the literature [49] and research into SHD, a total of 23 service activities of SHD are identified (see Table 7).


No.Service unit

1Mine design
2Environmental protection engineering
3Administrative engineering
4Hoisting engineering
5Power supply engineering
6Ventilation engineering
7Auxiliary engineering
8Mining
9Drivage
10Supporting and installation of equipment
11Lifting system
12Ventilation and drainage
13Power supply
14Moving upside down
15Coal preparation
16Provisioning
17Coal transportation
18Sales
19Equipment maintenance
20Equipment leasing
21Mine construction and technical consultation
22Technical consultation
23Training

6.2. Calculation of Synthetic SER-DSM Matrix

Five SHD enterprise technical experts and university coal experts were organized to evaluate the relationship between the above service activities on the dimensions of function, resource, and time, and the SER-DSM matrix of each dimension was constructed. The weight was calculated using the analytic hierarchy process, which is not discussed in this paper. The weight of each dimension was set as function weight  = 0.39, resource weight  = 0.24, and time weight  = 0.37. For reasons of space, this paper takes service activities 8 and 9 as examples to illustrate the evaluation semantics and triangular fuzzy numbers reached by the experts according to the multiple criteria given in Tables 24. The results are shown in Tables 810.


ExpertFuzzy semanticsTriangular fuzzy number

1Coal mining and drivage are indispensable for the completion of coal stripping.(0.8, 0.9, 1)
2Coal mining and drivage are indispensable for the completion of coal stripping.(0.8, 0.9, 1)
3Coal mining and drivage are indispensable for the completion of coal stripping.(0.8, 0.9, 1)
4Coal mining and drivage are indispensable for the completion of coal stripping.(0.8, 0.9, 1)
5Coal mining and drivage are indispensable for the completion of coal stripping.(0.8, 0.9, 1)


ExpertFuzzy semanticsTriangular fuzzy number

1Coal mining and drivage share part of the equipment and energy resources, which are critical and sufficient.(0.8, 0.9, 1)
2Coal mining and drivage share part of the equipment and energy resources, which are critical and sufficient.(0.8, 0.9, 1)
3Coal mining and drivage share part of the equipment and energy resources, which are abundant.(0.4, 0.5, 0.6)
4Coal mining and drivage share part of the equipment and energy resources, which are critical and sufficient.(0.8, 0.9, 1)
5Coal mining and drivage share part of the equipment and energy resources, which are critical and sufficient.(0.8, 0.9, 1)


ExpertFuzzy semanticsTriangular fuzzy number

1Coal mining should not precede drivage.(0, 0.1, 0.3)
2Coal mining cannot precede drivage, but both of them circulate many times.(0.8, 0.9, 1)
3Coal mining should not precede drivage.(0, 0.1, 0.3)
4Coal mining cannot precede drivage, but both of them circulate many times.(0.8, 0.9, 1)
5Coal mining cannot precede drivage, but both of them circulate many times.(0.8, 0.9, 1)

According to formula (3), the evaluation level belief degree for each dimension of service activities 8 and 9 was obtained by integrating expert evaluation information, as shown in Table 11. In the table, represents function dimension, resource dimension, and time dimension.



Functional dimension00001
Resource dimension0000.60.4
Time dimension0.40000.6

Then, according to formula (4) and Table 5, the evaluation information from the fusion experts in the functional dimension was calculated, and the evaluation results and confidence of all service activities in the functional dimension were obtained. For reasons of space, this paper shows only part of the content of the functional association relationship evaluation matrix (Table 12).


Service activity number
1……891011……

1……0(0, 0.1, 0.3), 1(0, 0.1, 0.3), 1……
(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0
(0.4, 0.5, 0.6), 0(0.4, 0.5, 0.6), 0
(0.6, 0.7, 0.8), 0(0.6, 0.7, 0.8), 0
(0.8, 0.9, 1), 0(0.8, 0.9, 1), 0

……
8(0, 0.1, 0.3), 11(0, 0.1, 0.3), 0(0, 0.1, 0.3), 0(0, 0.1, 0.3), 0……
(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0
(0.4, 0.5, 0.6), 0(0.4, 0.5, 0.6), 0(0.4, 0.5, 0.6), 0.4(0.4, 0.5, 0.6), 0.4
(0.6, 0.7, 0.8), 0(0.6, 0.7, 0.8), 0(0.6, 0.7, 0.8), 0.2(0.6, 0.7, 0.8), 0.2
(0.8, 0.9, 1), 0(0.8, 0.9, 1), 1(0.8, 0.9, 1), 0.4(0.8, 0.9, 1), 0.4

90……(0, 0.1, 0.3), 01(0, 0.1, 0.3), 0(0, 0.1, 0.3), 0……
(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0.2
(0.4, 0.5, 0.6), 0(0.4, 0.5, 0.6), 0.4(0.4, 0.5, 0.6), 0.6
(0.6, 0.7, 0.8), 0(0.6, 0.7, 0.8), 0.2(0.6, 0.7, 0.8), 0.2
(0.8, 0.9, 1), 1(0.8, 0.9, 1), 0.4(0.8, 0.9, 1), 0

100……(0, 0.1, 0.3), 0(0, 0.1, 0.3), 01(0, 0.1, 0.3), 0……
(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0
(0.4, 0.5, 0.6), 0.4(0.4, 0.5, 0.6), 0.4(0.4, 0.5, 0.6), 0.2
(0.6, 0.7, 0.8), 0.2(0.6, 0.7, 0.8), 0.2(0.6, 0.7, 0.8), 0.6
(0.8, 0.9, 1), 0.4(0.8, 0.9, 1), 0.4(0.8, 0.9, 1), 0.2

1100(0, 0.1, 0.3), 0(0, 0.1, 0.3), 0(0, 0.1, 0.3), 01……
(0.2, 0.3, 0.4), 0(0.2, 0.3, 0.4), 0.2(0.2, 0.3, 0.4), 0
(0.4, 0.5, 0.6), 0.4(0.4, 0.5, 0.6), 0.6(0.4, 0.5, 0.6), 0.2
(0.6, 0.7, 0.8), 0.2(0.6, 0.7, 0.8), 0.2(0.6, 0.7, 0.8), 0.6
(0.8, 0.9, 1), 0.4(0.8, 0.9, 1), 0(0.8, 0.9, 1), 0.2

……

Similarly, the resource SER-DSM, time SER-DSM, and function SER-DSM were calculated. The complexity of the calculation process makes it difficult to present it here, and the calculations used to support the findings of this study are available from the corresponding author upon request.

According to formulas (6)–(12), the evaluation matrix of function, resource, and time dimensions was fused to obtain a clear and comprehensive SER-DSM, as shown in Table 13.



10.026700.03830.0270.018200000.030.0300000000000
0.0181000000000000000000000
0.0180100000000000000000000
0.013150010000000.04000000000000
0.018000100000000.10000000000
0.0180000100.10.048000.0600000000000
00000010000000000000000
000000.0115010.1010.10.050.0400.03000000000
000000.024200.210.20.10.150.10.11000000000
0.0180000000.20.1810.130.130.10000000000
0.018000.03910000.10.1180.1100.20000000000
0.01800000.074100.10.1380.1010.10000000000
0.0180000.076000.10.1170.10.120.1210000000000
00000000.10.11300001000000000
00000000000000100000000
000000000.02900.030.0300010000000
00000000000000001000000
00000000000000000100000
00000000.10.06600.060.060.10000010000
00000000000000000001000
00000000000000000000100
00000000000000000000010
00000000000000000000001

According to the analysis in Section 5.1, the numbers of the independent service activities in the comprehensive SER-DSM matrix are 7, 15, 17, 18, 20, 21, 22, and 23. There are no common service activities. For convenience in subsequent calculations, the general service activities were renumbered as shown in Table 14.


No.Service unit

1Mine design
2Environmental protection engineering
3Administrative engineering
4Hoisting engineering
5Power supply engineering
6Ventilation engineering
7Mining
8Drivage
9Support and installation of equipment
10Lifting system
11Ventilation and drainage
12Power supply
13Moving upside down
14Provisioning
15Equipment maintenance

6.3. Module Clustering

Through orthogonal experiment, the results show that the optimal parameter combination scheme is population size n = 100, crossover probability PC = 0.65, and mutation probability PM = 0.01. In this study, the number of iterations (set to 1,000) was taken as the condition for the genetic algorithm to stop running. In the parameter combination scheme, in order to obtain the optimal solution of module partition, 10 experiments were carried out. The hardware configuration for the experiment consisted of an Intel Core i3 processor with 200 GB memory, and the software configuration consisted of Windows 7 operating system and MATLAB version 2016a. The experimental results are shown in Table 15.


TestModuleObjectiveObjective mean/STDFitnessFitness mean/STD

140.4278232120.426725267/0.0014490.932682010.93173/0.001257
240.4280677270.93289476
360.4267234420.9317263
440.4270645270.9320225
540.4259643270.93106777
650.4263220410.93137797
740.424768420.93003221
840.4282277250.93303403
940.4284228320.93320391
1030.423868420.92925439

The optimal service module partition of general service activity was obtained by taking the calculation result corresponding to the maximum fitness value through several iterative experiments, which is the result of Experiment 9.

Figure 5 shows the iteration curve of the fitness function of the process with the highest fitness value. The approximation curve at the lower end of the graph represents the mean value of the fitness function, while the step curve at the upper end of the graph represents its maximum value. Because of the low mutation rate, the convergence image is gentle but the maximum value is obtained quickly.

The service module division scheme of SHD is shown in Table 16.


ModuleService unit

1Mine design
2Environmental protection engineering
3Administrative engineering
4Hoisting engineering, power supply engineering, ventilation engineering, mining, drivage, support and installation of equipment, lifting system, ventilation and drainage, power supply, moving upside down, provisioning, equipment maintenance

From the above results, it is clear that the general activities of SHD can be divided into four modules: mine design (Module 1); environmental protection engineering (Module 2); administrative engineering (Module 3); and activities 4–14 in Table 14, involving the main production business of coal mining systems, such as ground setting and construction, coal mining and tunneling, underground transportation, supply of materials, and maintenance of equipment (Module 4). In SHD, there were only two services: moving and face down services, and coal mine overall trusteeship. The results of this study show that Module 4 categorizes the processes closely related to coal mining into a single module, instead of providing moving and face down services alone, which improves personnel safety.

6.4. Customized Service Model and Application to SHD

After the preliminary division of modules, it was necessary to separate, replace, delete, and add modules to form a structured module system [50]. The necessary operation of the independent public module partition scheme was carried out, and the maintenance services and Module 4 were separated; two independent consulting service activities were merged into one module. Ultimately, there are seven independent modules and five general modules. The customized service model system for SHD is formed as shown in Table 17, and there are five general modules that need to be customized according to customer needs. At the same time, they can be combined according to the personalized needs of customers to form a customized service scheme.


Module typeModule nameService activity

IndependentConstruction of auxiliary facilitiesAuxiliary engineering
LogisticsCoal transportation
LeaseEquipment leasing
Technical consultationMine construction and technical consultation
TrainingTraining
Coal washing and processingCoal preparation

GeneralDesignMine design
Environmental engineering constructionEnvironmental protection engineering
InfrastructureAdministrative engineering
Coal productionHoisting engineering, power supply engineering, ventilation engineering, mining, drivage, support and installation of equipment, lifting system, ventilation and drainage, power supply, moving upside down, provisioning
RepairEquipment maintenance

Because of capacity constraints, SHD had a variety of idle equipment, with a net value of nearly 110 million yuan. In October 2019, SHD learned that SW coal mine was planning to develop a new coal face and contacted them in the hope of providing the necessary equipment. Due to its large capital occupation, SW intends to rent equipment; the company also has equipment maintenance, installation, and other needs. SHD has thus responded quickly to the market and accepted orders from customers. The independent leasing module and the customized maintenance module have been quickly combined to organize staff to carry out business. SW has leased 149 sets of equipment, with a rental fee of 99.94 million yuan and a maintenance fee of 58.3 million yuan. The feasibility of responding to the market using the above research results is thus confirmed.

7. Conclusion

Using DSM, this study analyzes the modular system architecture modeling and clustering of customized services in the mining industry with a view to resolving the problems of determining service activity correlation and reasonable segmentation of service modules. A number of general conclusions can be drawn.(1)There has been little research on the application of service modularization theory and methods in the mining industry, which is technologically complex and has a single-product structure. The present study helps to expand the application scope of service modularization theory and promote business model innovation in the coal, oil, and natural gas industries.(2)Focusing on the problem of determining the association relationship of service activities in the process of service modular design, and drawing on DSM, the association relationship model between service activities can be constructed from multiple dimensions. Here, the fuzzy measurement of the comprehensive association relationship of service activities in multiple dimensions and with uncertain information was constructed using triangular fuzzy number and fuzzy ET methods.(3)According to the information characteristics of row and column elements in the design structure, the structure of service modules can be divided. On this basis, the mathematical programming model with the maximum sum of average cohesion and coupling degree was constructed, and a genetic algorithm was used to carry out a clustering calculation to obtain the best service module partition scheme. The disadvantage of this method is that the clustering method does not achieve the optimal degree of cohesion and coupling. However, the proposed method does not need prior information (such as clustering center or module partition granularity); after the clustering analysis, it can divide the service modular architecture, thereby determining the boundaries of contract, performance, and responsibility division for customized services in the mining industry. At the same time, it helps service providers to determine the business scope of customized services and to improve the efficiency of their customized service scheme design. It also helps customers to clarify their outsourcing business content and to focus on their own competitive business processes.(4)Taking the SHD coal company as an example, the feasibility of the proposed method has been verified. There are two kinds of service items in SHD. The calculations in this study yield 11 service modules, which can form a variety of customized service items according to customer needs, and solve the problem of rapid market response.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was supported by the National Social Science Foundation of China (Grant No. 19BGL183); the Key Scientific Research Plan of Shaanxi Provincial Department of Education of China (Key Research Base of Philosophy and Social Sciences, Grant No. 18JZ039); the National Science and Technology Support Program of China (Grant No. 2013BAH12F00); the National Natural Science Foundation of China (Grant No. 71273206); the National Soft Science Projects of China (Grant No. 2013GXS4D151); the PhD Initiation Fund of Xi’an University of Science and Technology (Grant No. 2017QDJ066); and the Prosperity and Development Program of Philosophy and Social Sciences in Xi’an University of Science and Technology (Grant No. 2019SY02).

References

  1. L. J. Kirk, “Owner versus contract mining,” in Mine Planning and Equipment Selection, G. N. Panagiotou and T. N. Michalakopoulos, Eds., pp. 437–442, Taylor & Francis, Abingdon, UK, 2000. View at: Google Scholar
  2. B. Luo, J. Zhang, and Z. Li, “Service risk evaluation of the general contract for coal mine production and operation: case study at Shendong Jinjie Coal Mine in China,” Mathematical Problems in Engineering, vol. 2019, Article ID 7845756, 12 pages, 2019. View at: Publisher Site | Google Scholar
  3. C. Y. Baldwin and K. B. Clark, “Managing in an age of modularity,” Harvard Business Review, vol. 1, pp. 75–84, 1997. View at: Google Scholar
  4. C. Y. Baldwin, “Where do transactions come from? Modularity, transactions, and the boundaries of firms,” Industrial & Corporate Change, vol. 17, no. 1, pp. 155–195, 2008. View at: Google Scholar
  5. Department of Statistical Design Administration, National Bureau of Statistics of the people’s Republic of China, Notes to National Economic Industry Classification GB/ T4754-2017, Department of Statistical Design Administration, National Bureau of Statistics of the people’s Republic of China, Beijing, China, 2017.
  6. S. He, “Research on the specialized service mode of coal enterprises in the context of supply side reform,” China Coal, vol. 41, no. 11, pp. 125–128, 2017. View at: Google Scholar
  7. L. Zhang, “Practice and exploration of specialized services for coal enterprises,” China Coal Industry, vol. 375, no. 5, pp. 69–71, 2018. View at: Google Scholar
  8. P. Yang, “Exploration of Shendong mine production combination service mode,” China Coal, vol. 42, no. 5, pp. 5–9, 2016. View at: Google Scholar
  9. W. Liu, Q. Fu, and Y. Luo, “Discussion on outsourcing mode of coal mine production service and application of coal mine trusteeship,” China Mining, vol. 10, pp. 38–42, 2015. View at: Google Scholar
  10. I. I. Pine, B. Victor, and A. C. Boynton, “Making mass customization work,” Harvard Business Review, vol. 71, no. 5, pp. 108–117, 1993. View at: Google Scholar
  11. J. Sundbo, “Modulization of service production and a thesis of convergence between service and manufacturing organizations,” Scandinavian Journal of Management, vol. 10, no. 3, pp. 245–266, 1994. View at: Publisher Site | Google Scholar
  12. J. Sundbo, “The service economy: standardisation or customisation?” The Service Industries Journal, vol. 22, no. 4, pp. 93–116, 2002. View at: Publisher Site | Google Scholar
  13. C. Yu and Z. Wu, ““Modular operation of civil aviation service from the perspective of value creation–based on the perspective of exploratory case analysis,” China’s Industrial Economy, vol. 12, pp. 141–153, 2012. View at: Google Scholar
  14. C. Yu, Z. Wu, and Y. Cheng, “Value creation path of service modularization operation of auto financial service industry – from the perspective of case study,” East China Economic Management, vol. 8, pp. 159–164, 2013. View at: Google Scholar
  15. Y. Zhou, C. Yu, and X. Shen, “Empirical analysis of service modularization level: a case study of Jiangxi Art Photography Museum,” Management Engineer, vol. 24, no. 1, pp. 13–20, 2019. View at: Google Scholar
  16. T. Yan and J. Li, “New service development process model of financial enterprises from the perspective of service modularization: multi case study,” Technical Economy, vol. 2, pp. 47–54, 2018. View at: Google Scholar
  17. Y. Jin and X. Geng, “Process modular design of product service system based on fuzzy DSM,” Industrial Engineering and Management, vol. 141, no. 2, pp. 159–167, 2020. View at: Google Scholar
  18. T. Yan and W. Jiang, “Research context, benchmark and prospect of service modularization: analysis based on foreign literature,” Foreign Economy and Management, vol. 37, no. 1, pp. 43–51, 2015. View at: Google Scholar
  19. H. Xia and Q. Xue, “Research on modular development of service industry,” Shanghai Economic Research, vol. 3, pp. 28–34, 2010. View at: Google Scholar
  20. C. Yu, X. Zhao, and X. Xing, “Analysis of the difference between service modularization and manufacturing modularization,” Journal of Jiangxi University of Finance and Economics, vol. 6, pp. 48–55, 2016. View at: Google Scholar
  21. T. R. Browning, “Applying the design structure matrix to system decomposition and integration problems: a review and new directions,” IEEE Transactions on Engineering Management, vol. 48, no. 3, pp. 292–306, 2001. View at: Publisher Site | Google Scholar
  22. L. Sun, G. Li, Z. Jiang et al., “Service-oriented manufacturing: an advanced manufacturing mode in the 21st century,” China Mechanical Engineering, vol. 18, no. 19, pp. 2307–2312, 2007. View at: Google Scholar
  23. A. Tukker and U. Tischner, “Product-services as a research field: past, present and future. Reflections from a decade of research,” Journal of Cleaner Production, vol. 14, no. 17, pp. 1552–1556, 2006. View at: Publisher Site | Google Scholar
  24. Q. Yang, J. Lu, and S. Kherbachi, “Progress in modeling and optimization for complex R&D project based on design structure matrix,” System Engineering–Theory & Practice, vol. 36, no. 4, pp. 989–1002, 2016. View at: Google Scholar
  25. T. Sakao, W. Song, and J. Matschewsky, “Creating service modules for customising product/service systems by extending DSM,” CIRP Annals, vol. 66, no. 1, pp. 21–24, 2017. View at: Publisher Site | Google Scholar
  26. Li Ying, B. Ying, and Z. Rong, “Module division and evaluation of product design structure based on SOM,” Journal of Wuhan University of Science and Technology, vol. 41, no. 4, pp. 63–68, 2017. View at: Google Scholar
  27. J. Chen, M. Rong, and J. Chu, “Modular reconfiguration and allocation method of collaborative tasks for industrial design cloud service platform,” Computer Integrated Manufacturing System, vol. 24, no. 3, pp. 720–730, 2018. View at: Google Scholar
  28. P. Wen and H. Gao, “Research on function module division of intelligent distribution system based on DSM model,” Logistics Science and Technology, vol. 41, no. 6, pp. 18–25, 2018. View at: Google Scholar
  29. X. Cheng and J. Luo, “Product module partition method considering the relationship between two modules,” Mechanical Design, vol. 36, no. 4, pp. 72–76, 2019. View at: Google Scholar
  30. C. H. Lovelock and J. Wirtz, “Services marketing: people, technology, strategy,” World Scientific Books, vol. 18, no. 5, pp. 922–924, 2011. View at: Google Scholar
  31. B. Li, “Efficiency analysis of service modularization,” Shanghai Management Science, vol. 32, no. 1, pp. 59–63, 2010. View at: Google Scholar
  32. C. McCulley and C. L. Bloebaum, “A genetic tool for optimal design sequencing in complex engineering systems,” Structural Optimization, vol. 12, no. 2-3, pp. 186–201, 1996. View at: Publisher Site | Google Scholar
  33. S. D. Eppinger and T. R. Browning, Design Structure Matrix Methods and Applications, MIT Press, Cambridge, MA, USA, 2012.
  34. X. Liu and K. Zhao, “Multi attribute decision making of triangular fuzzy number based on unknown attribute weight of connection number,” Fuzzy System and Mathematics, vol. 2, pp. 99–110, 2017. View at: Google Scholar
  35. K. Frenken and S. Mendritzki, “Optimal modularity: a demonstration of the evolutionary advantage of modular architectures,” Journal of Evolutionary Economics, vol. 22, no. 5, pp. 935–956, 2012. View at: Publisher Site | Google Scholar
  36. M. Pero, M. Stößlein, and R. Cigolini, “Linking product modularity to supply chain integration in the construction and shipbuilding industries,” International Journal of Production Economics, vol. 170, pp. 602–615, 2015. View at: Publisher Site | Google Scholar
  37. W. Wang and F. Liu, “Task decomposition optimization based on task relation matrix,” Military Operations Research and Systems Engineering, vol. 31, no. 4, pp. 9–14, 2017. View at: Google Scholar
  38. L. Feng, D. Huang, and Z. He, “Process modularization analysis method for complex product development project,” Industrial Engineering, vol. 21, no. 3, pp. 4–13, 2018. View at: Google Scholar
  39. J. Li, W. Zhu, and J. Mao, “The influence of manufacturer customer relationship on the process of manufacturer service: based on the theory of resource dependence,” Business Research, vol. 10, pp. 88–95, 2013. View at: Google Scholar
  40. J. U. Maheswari and K. Varghese, “Project scheduling using dependency structure matrix,” International Journal of Project Management, vol. 23, no. 3, pp. 223–230, 2005. View at: Publisher Site | Google Scholar
  41. J. B. Yang, Y. M. Wang, D. L. Xu, and K. S. Chin, “The evidential reasoning approach for MADA under both probabilistic and fuzzy uncertainties,” European Journal of Operational Research, vol. 171, no. 1, pp. 309–343, 2006. View at: Publisher Site | Google Scholar
  42. J. Jiang, X. Li, L. N. Xing, and Y. W. Chen, “System risk analysis and evaluation based on fuzzy evidence reasoning,” System Engineering–Theory and Practice, vol. 33, no. 2, pp. 529–537, 2013. View at: Google Scholar
  43. W. Jia, Z. Liu, and D. Liu, “Modular design method and application for complex product based on fuzzy correlation analysis,” Journal of Mechanical Engineering, vol. 51, no. 5, pp. 130–142, 2015. View at: Publisher Site | Google Scholar
  44. R. Wise and P. Baumgartner, “Go downstream: the new profit imperative in manufacturing,” Harvard Business Review, vol. 28, no. 1, pp. 89–96, 2000. View at: Google Scholar
  45. J. Pandremenos and G. Chryssolouris, “A neural network approach for the development of modular product architectures,” International Journal of Computer Integrated Manufacturing, vol. 24, no. 10, pp. 879–887, 2011. View at: Publisher Site | Google Scholar
  46. A. Zhu, “Research on fitness function of genetic algorithm,” System Engineering and Electronic Technology, vol. 11, pp. 60–64, 1998. View at: Google Scholar
  47. C. Ding and C. Zhang, “Orthogonal experimental genetic algorithm and its application in function optimization,” System Engineering and Electronic Technology, vol. 10, pp. 57–60, 1997. View at: Google Scholar
  48. L. Wang, J. Cai, and Li Ming, “Optimization of genetic algorithm based on orthogonal experiment,” Journal of Nanjing Normal University Engineering and Technology Edition, vol. 16, no. 2, pp. 81–85, 2016. View at: Google Scholar
  49. Q. Zhang and K. Li, “Study on the life cycle of resource exhausted enterprises: a case study of coal enterprises,” Coal Economy Research, vol. 1, pp. 8–12, 2005. View at: Google Scholar
  50. C. Aoki, Module Era–Essence of New Industrial Structure, Shanghai Far East Press, Shanghai, China, 2003.

Copyright © 2021 Xin Wang and Bo Luo. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Related articles

No related content is available yet for this article.
 PDF Download Citation Citation
 Download other formatsMore
 Order printed copiesOrder
Views119
Downloads290
Citations

Related articles

No related content is available yet for this article.

Article of the Year Award: Outstanding research contributions of 2021, as selected by our Chief Editors. Read the winning articles.