Smart Supply Chain Risk Assessment in Intelligent Manufacturing

 

Abstract:

This paper identifies risk factors in the smart supply chain and develops a risk assessment index system for reducing potential losses in intelligent manufacturing. Specifically, based on the supply chain operation reference theory, we investigate performance indicators and characteristics of the smart supply chain in intelligent manufacturing. A conceptual model is developed to identify risks of the smart supply chain, and a questionnaire is designed to measure risks of the smart supply chain in intelligent manufacturing. Using the hierarchical clustering analysis, an improved risk assessment model is derived with 22 risks based on 814 valid sample data. Moreover, the information entropy weight method is used to compute the risk weights for the smart supply chain in intelligent manufacturing. The risk weights are further verified by simulation and proved that these risk factors and risk weights have high practical efficiency. Theoretical and practical implications are also presented.

 

KEYWORDS: risk assessment, smart supply chain, intelligent manufacturing, entropy weight

1. Introduction

With the development of emerging information technologies, digitization and its impacts on markets, production conditions, and intercompany interaction force companies to adapt these technologies continuously in order to be sustainable and competitive [1]. Meanwhile, the global manufacturing industry has entered an era of information explosion. In order to gain the competitive advantages of product and service innovation, manufacturing enterprises need to change the original mode of operation management and synchronize with the development of emerging technologies to meet customer’s demands [2-3]. As a developing country, many manufacturing enterprises are facing complex information exchanges and massive information decision-making tasks. Hence, the manufacturing system of an enterprise needs to be not only flexible, but also intelligent. Under this context, intelligent manufacturing has become an inevitable trend for the modern manufacturing industry in China. This advanced manufacturing mode is also called the manufacturing mode for the 21st century, which will influence the future economic development [4].

With the rapid changes in market demands and increasingly fierce global competition, many problems are emerging in the manufacturing industry in China such as weak innovation abilities, irrational industrial structures, scarce resources, severe environments and rising labour costs. In the light of these problems, the “Made in China 2025” strategy is proposed as an action plan to transform traditional manufacturing into intelligent manufacturing. Intelligent manufacturing in China becomes the new round of industrial revolution, and the breakthrough and main direction for innovation. Such transformation is the effective way to enhance the global competitiveness of Chinese enterprises.

The foundation for intelligent manufacturing is the smart supply chain (SC). The increases in both demand fluctuation and pressures for cost reduction while satisfying customer demands on products and services in intelligent manufacturing have put a premium on smart supply chain management [5]. The goal of smart supply chain management is to realize the intelligence and innovation of the complete manufacturing value chain. From the perspective of manufacturing information management, it would be difficult for the traditional supply chain management models to cope with complex data in an intelligent manufacturing environment. At the same time, t is also a challenge for enterprises to use traditional methods to provide personalized services for consumers in intelligent manufacturing. Hence, enterprises need a “customer-centred” smart SC to deal with complex service information in business. From the perspective of the global value chain of manufacturing industry, manufacturing activities in an intelligent manufacturing environment are no longer initiated by manufacturing enterprises, but rather by the end users. This results in the whole value chain being changed from an enterprise-driven model to a user-driven model. To achieve truly intelligent manufacturing, enterprises should not only develop intelligent products, apply intelligent equipment, build intelligent factories, carry out intelligent R&D and management and promote intelligent services, but also integrate equipment, production lines, factories, products, suppliers and customers into a whole. This puts high demands on the smart SC management in intelligent manufacturing, which should have a holistic view on information sharing and collaboration, as well as accurate and effective customer services. The smart SC requires more intelligent capabilities in the supply chain in respond to dynamic market decision-making. “Made in China 2025” essentially requires the SC to be intelligent at all levels including manufacturing infrastructures, products/equipment, manufacturing process, manufacturing management, personalized service customized and personalized services recommended. Without the support of a smart SC, intelligent manufacturing will be a mere change of the production pattern, with no innovation and upgrade in the business model [6].

The SC is a supply and demand network where raw materials and manufactured products are sold from suppliers to customers and finally to consumers. Since there are all kinds of parties participating in the multi-level and multi-faceted cooperation, SC is also a complex carrier of logistics with information, capital and business flows, which makes it vulnerable to both the external and internal environments; hence, the smart SC risks emerge. To improve the reliability and sustainability of the smart SC, it is necessary to analyse these risk factors [7]. Risks exist everywhere objectively, especially in an environment with rapid technological changes. The risks in the smart SC in intelligent manufacturing are higher than those in traditional manufacturing. Many Chinese manufacturing enterprises have long been accustomed to the production-oriented development model, with little understanding of the intelligent development model and insufficient experiences in this industrial transformation. Chinese enterprises will inevitably encounter many new problems and risk when moving to intelligent manufacturing. In addition, China's market environment is relatively complex, with these characteristics such as rapid development, changing policies, chaotic market order, short-term orientation, excessive competition, and so on. Hence, it is highly necessary to study how to conduct effective risk management in order to improve the sustainability and reliability of the smart SC and promote intelligent manufacturing development.

Due to the lack of research on investigating the risk factors for smoothly and effectively transforming form traditional manufacturing into intelligent manufacturing, this paper aims at developing a risk assessment system for the smart supply chain in intelligent manufacturing. First, a literature review is conducted on intelligent manufacturing and enterprise intelligent management in Section 2. Section 3 proposes a conceptual model based on risks of the smart SC in intelligent manufacturing. Section 4 presents the methodology including development of a questionnaire and data collection. Section 5 develops an assessment model showing weights of risk indexes. Discussions, implications and conclusions are shown in Section 6.

2. Literature Review

2.1. Intelligent manufacturing

Intelligent manufacturing has attracted wide attention from governments around the world. In April 2013, Germany first proposed Industry 4.0, aiming at improving the intelligence of the manufacturing industry. Since the 1990s, the National Science Foundation of the United States (NSF) has been funding many research projects on intelligent manufacturing, including intelligent decision-making in manufacturing, intelligent multi-agent collaboration, concurrent intelligent design, and intelligent automation of logistics transmission, etc. The projects funded have covered most of the research areas in intelligent manufacturing [8]. Furthermore, the programs of “Advanced Manufacturing Partnership” announced by U.S. President Barack Obama and supported by “Smart Manufacturing Leadership Coalition” (SMLC) intend to accelerate the modernization of factories to gain more competitiveness through the application of advanced information technology and automation technology in manufacturing [9]. To keep up with western developed countries, in 2015, China proposed the strategy of “Made in China 2025”, which focuses on the innovation and development of the manufacturing industry. It calls for strengthening the industrial basic capacity and promoting the level of intelligent manufacturing, with an aim to upgrade the intelligence of the manufacturing industry within three decades [10].

At present, intelligent manufacturing is a hot research topic. Most of the relevant studies focus on two aspects, intelligence of manufacturing equipment and products and intelligence of manufacturing process and management, of which the latter is being paid more and more attention to. Xiong Y.L. [11] elaborated on intelligent manufacturing from the perspective of industrial cross integration, and summarized the scope of intelligent manufacturing: intelligent manufacturing technology, equipment, systems and services and various kinds of intelligent products. Zhu J.Y. [12] proposed that the intelligence of enterprise management is more important than that of machinery and equipment.

2.2. Enterprise intelligent management

2.2.1. Intelligent manufacturing system

In 1990, the Ministry of International Trade and Industry of Japan proposed the concept of enterprise intelligent manufacturing system for the first time [13], with the purpose of overcoming the limitations of traditional manufacturing systems. It highlights the intelligent decision-making function of the integrated system. Currently, the concept of “manufacturing” in intelligent manufacturing is an extension of the original manufacturing definition. It not only refers to the traditional processing and technology, but also involves the design, organization, supply, sales, scrapping and recycling of products, so IMS intelligent activities exist throughout the whole life of a product. Conceptually, intelligent manufacturing mainly consists of intelligent manufacturing technologies and intelligent manufacturing systems. It is the integration of IMS, SC system, logistics system, and other service systems [14]. An intelligent manufacturing system consists of intelligent manufacturing models, intelligent production and smart products [15]. Zhang Z.J. (2015) pointed out some deficiencies in the infrastructures, policy support, enterprise collaboration and expertise in the SC system of intelligent manufacturing, and put forward some countermeasures for constructing this SC system [16]. Prickett et al. studied the application of intelligent equipment and technologies in the production process mainly from the perspective of manufacturing science, including distributed numerical control systems, flexible manufacturing systems and wireless sensonetwork, to support the intelligence of production [17]. Ruiz and et al [18] introduced the multi-agent system in the simulation of the production process to meet the new requirements of the intelligent manufacturing environment. Aiming at unlimited production capacity, the manufacturing execution system (MES) is used to optimize the workshop capacity through job scheduling and integrate workshop resources into the ERP system to promote flexible manufacturing [19-20]. Fang and Li (2017) described the composition of the intelligent logistics system and related technologies, and then put forward the countermeasures to intelligently connect and integrate both internal and external logistics processes of an enterprise [21].

2.2.2. Intelligent manufacturing services

From the perspectives of information utilization, service pattern, manufacturing resource management and system integration, specific manufacturing service modes have been proposed for intelligent manufacturing, such as manufacturing grid [22], service-oriented manufacturing [23], cloud manufacturing [24] and social manufacturing, etc. Most of these manufacturing modes are applicable only at the workshop level and focus on the encapsulation and sharing of manufacturing resources at the beginning of life (BOL) and the services matching in the manufacturing process. However, it has been rarely discussed how to integrate data and knowledge at every stage of the SC to enhance the adaptive and self-determining capabilities of the entire manufacturing process and the full lifecycle management.

From the perspective of service science, Tso and Hu [25-26] put forward intelligent manufacturing services, which mainly consist of product services, production-related technical services, information services, financial and insurance services and logistics services. According to the features and requirements of intelligent manufacturing services, In accordance with the numerous researchers who have stated the need for an integrated and holistic theory of innovation, Cáceres R et al [27] attempts to determine similarities between the processes of innovation in services and manufacturing that support the integrative approach to innovation. Do (2017) introduces an extended product data model that has data objects for bills of processes (BOP), customer orders and manufacturing operation linked to 3D printing services through Internet of things (IOTs) technologies. The extended product data model can support product design, process planning, production planning, and execution of manufacturing operation in a unified single user environment. [28]. However, limit research also showsthat intelligent service has become an important part of intelligent manufacturing.

2.2.3. Management modes of intelligent manufacturing

Cagnin et al (2014) studied the management modes adopted by intelligent manufacturing organizations in different countries or regions under different cultural backgrounds and emphasized the customer-orientation in intelligent manufacturing [29]. From the perspective of management science, Choy and Su studied the smart SC management, intelligent perception of external environment, prediction of production equipment performance and intelligent maintenance and intelligent enterprise management (human resources, finance, procurement and knowledge management, etc.) [30-31]. Based on the existing software and hardware resources and the enterprise intelligent management structure model, they proposed an intelligent management system model based on multi-agent to effectively integrate the enterprise management system with human [32]. Li et al (2017) proposed new models, means, and forms of intelligent manufacturing, intelligent manufacturing system architecture, and intelligent manufacturing technology system, based on the integration of AI technology with information communications, manufacturing, and related product technology [33]. To meet the needs of manufacturing enterprises in materials management, and realize the scientific management of modern enterprise, intelligent manufacturing engineering management system based on bill of material (BOM) was developed and designed by Zhou, and et al [34]. To optimize the allocation of cloud service resources and reduce the cost of manufacturing services, Pan (2016) proposed the optimal allocation of manufacturing resources in the cloud manufacturing environment [35].

However, these studies basically focused on the internal management of enterprises and ignored the fact that product quality is the result of the entire SC environment.

The changes in the global economic and management environment make us have to formulate the logistics and supply chain management (SCM) strategies, to unfold the new era of business collaboration [36]. Complex product manufacturing is especially affected by this development.

2.2.4. Studies on SC management of intelligent manufacturing

SC management plays a crucial role in modern enterprises, especially for smart manufacturing enterprises. It refers to the coordination and integration with upstream suppliers and downstream customers, aiming at optimizing the performance of the entire SC of smart manufacturing enterprises [37]. SC is a strategy network which includes contains a series of business activities related to information flow, capital flow, and logistics flow [38-39], and is a complex system like other nonlinear economic systems [40]. The objectives of supply chain management often need to improve service levels and reduce operating costs at the same time. However, these two objectives are often contradictory. At the same time, uncertainty in the supply chain market environment makes it difficult to adapt the centralized supply chain optimization method to the dynamics of the environment [41]. To solve these problems, it is necessary to enhance the coordination, integration, flexibility, and adaptability of the SC network.

Considering the logistics process in intelligent manufacturing, Li [42] and Qi, and et al [43] combined ERP with the field automation system to smooth the information flow of workshop production and logistics distribution. Through the embedment of RFID (radio frequency technology) into other information systems, logistics tracking and logistics data can be integrated to improve the application of enterprise logistics information [44]. Through integration of the online monitoring system, machine control system, wireless sensor network and sensing technologies, the data acquisition and identification, automatic abnormality diagnosis and adaptive control can all be achieved in the production process [45].

The importance of integrated risk management of SC is increasing, and new threats like Cyber threats are occurred frequently. Therefore, the existing risk-management systems fall too short and cannot match the existing complexity [46]. At present, the research on risk management of the smart SC is unsystematic, which needs to be further deepened.

The importance of integrated risk management of SC is increasing, and new threats from technology, management, market, cooperation, and others are occurred frequently. The existing risk-management systems lag behind the needs of complex risk management.

2.3. Summary

In general, the current research on intelligent manufacturing has become more diversified, multi-perspective, dynamic and multi-disciplinary. Few researches have been done on the risks of the smart SC, but it is quite necessary. Nevertheless, people have some extent understanding on SC risks and its dangers. For example, the risk within a manufacturing supply chain can be classified into different categories that include supply risks, demand risks, catastrophic risks, infrastructure and regulatory risks and bureaucratic and legal risks [47]. Canfield School of Management [48] defined the SC risk as the vulnerability of SC. SC risk usually reduces the operational efficiency of the SC and increases costs, and even leads to the disintegration of SC.

Yang and He (2015) introduced in the SCOR model to identify the risks of the supply chain in five stages and built the corresponding risk evaluation index system using AHP and fuzzy AHP [49]. Chen and Zhao (2013) analysed the risks exposed to automobile manufacturing enterprises on the aspects of the internal operation, cooperating partners and external SC environment, then established the risk evaluation system of the automobile manufacturing supply chain [50]. Taking into account the features of supply chains and complex products, Guo (2017) proposed a specific process for supply risk identification on the condition of the complex product exploitation with the participation of suppliers, and identified five types’ risks: collaborative R&D risk, production operation risk, organizational coordination risk, demand risk and regulatory risk [51]. These available researches of supply chain risk for manufacturing enterprises implies the possibility of research risk on this smart SC.

With the development of intelligent manufacturing in China, there will be more opportunities, challenges and risks ahead, so manufacturing enterprises should pay closer attention to the risks arising from the complex environment of intelligent manufacturing and then conduct more targeted risk prevention.

3. Theoretical Background and Conceptual Model

3.1. Smart SC for intelligent manufacturing

In the intelligent manufacturing environment, the key to gaining competitive advantage for manufacturing enterprises is to build a smart and efficient SC [52].

3.1.1. Definition of the smart SC

As a cooperation-based virtual organization composed of many enterprises, SC has become a real competition indicator between enterprises in the 21st century. The smart supply chain is a technology and management integrated system based on information and intelligent technology to realize intelligence, networking, coordination, integration and automation [53].

The smart SC is a comprehensive integrated technology and management system which integrates Things, Internet, technology and modern SC management theory, method and technology to realize the intelligent, networked and automated SC between enterprises [54]. Luo also emphasized the “smart SC” at Shanghai Conference of Information and Industrial Integration, as an intelligent, networked, automated system, and a system of introducing advanced management theory into enterprises [55]. In short, the various definitions of the smart SC actually show that it should be a system with more information symmetric and contacting in real time between SC members conveniently in information flow, logistics, capital flow, and others, to improve the operational efficiency of SC.

In order to improve the efficiency of information sharing and decision making among SC enterprises, the smart SC needs to adopt information network, intelligent information processing and intelligent decision-making technologies. For this reason, the smart SC has the following unique features compared with the traditional SC.

3.1.2. Characteristics of the smart SC [56-58]

There are some characteristics of the smart supply chain.

(1) Effective integration of advanced technologies. In the smart SC environment, SC managers and operators will take the initiative to systematically absorb all kinds of advanced technologies, including Internet of Things Internet, Internet, artificial intelligence and etc, to achieve innovative changes in SC management.

(2) Visualization and mobility. The smart SC is more inclined to use visual data, such as pictures and videos, and such data can be accessed with intelligent and mobile tools. Owing to these advanced network and information technologies, enterprises in the SC can share information, making the consumer demand forecast accurate and timely. In this way, the bullwhip effect will be alleviated, inventory and production costs reduced, and production efficiency improved.

(3) Information integration. The smart SC can effectively break down the barriers between the information systems of SC members and better integrate and share information within the SC in real time.

(4) Close collaboration. Through the highly integrated information mechanism, enterprises in the smart SC can better acquire information of other members, and grasp the situation inside and outside the SC in a timely manner, and make better adjustments to the collaboration with other enterprises in response to the changes to improve the performance of the SC. In addition, SC companies can also establish a direct cooperation relationship with consumers, so as to satisfy consumer needs to the greatest extent and reduce the total cost of the SC. SC coordination is the key to the SC management of enterprises. The better the coordination, the more effective the SC management will be.

(5) Feasible SC management. Due to information integration and sharing, the vertical and horizontal relationship management of the SC becomes more feasible and convenient, and such SC management based on information flow can in turn make the structure of the smart SC more scalable.  

Obviously, there are many differences between the smart SC and traditional SC in terms of internal structure. The smart SC takes the information network as the “nerve centre” of information transmission in SC. Each enterprise can connect its own system to the information system of the smart SC to ensure the sharing of information and the high integration of information flow in the SC [59]. It can also shorten the enterprises' market response time, minimizes resource consumption as much as possible, improves product quality, and improves the company's ability of decision-making quickly and effectively.

Under this information-based environment, to a certain extent, the application of the smart SC in intelligent manufacturing can improve the operational efficiency of SC. At the same time, the complex operation environment in the smart SC will also bring more unknown risks and disturbances to enterprises. Therefore, it is necessary to study the risks in the SC in order better cope with them.

3.2. Conceptual model of risk framework in the smart SC

3.2.1. Risk Modeling of the Smart SC Based on SCOR Model Theory

(1) SCOR Model and its application in risk analysis of SC

Supply Chain Operations Reference (SCOR) model is developed and supported by the International Supply Chain Council. It is a widely accepted international operation process reference standard for describing and designing the structure of SC.

According to its process definition, SCOR model can be grouped into three levels . The first layer mainly describes five basic processes: planning, procurement, production, distribution and return, which presents the whole supply chain process of all logistics functions for enterprises, also contains competitive performance goals of enterprise. Through the process or performance analysis of SCOR model, enterprise can make basic strategic decisions on some operational performance indicators of SC.

SCOR focuses on the management concept of process-centric and horizontal integration. Process integration makes it possible that SC trouble problem can ultimately be reflected in some processes of SC. At present, the SCOR model has become a diagnostic tool for SC problems. Particularly, it has become a theoretical and methodological reference to the study of SC risk. Such as, Ma lin[60] proposed supply chain risk management framework on SCOR by objectives, -culture,process, control mechanism and management information system.  FB Georgise,KD Thoben,M Seifert(2013) [61] , through empirical research, proved that in the era of lean manufacturing, the implementation of SCOR model will improve supply chain performance, reduce supply chain risk and promote enterprises to success. Liang Yixuan(2015) [62] established a risk identification model of SC for B2C enterprises based on SCOR model and characteristics of SC for B2C enterprises. Xu Peipei (2018) [63], on the planning process, procurement process, sales process and distribution process of SCOR model, built a risk identification model for Eternal Asia Company.

It can be seen that SCOR theory has certain feasibility in supply chain risk identification. In the above researches, some of them are generally carried out according to the structural hierarchy of SCOR or the operational process of SCOR, and combined with characteristics of specific SC. This paper considers that SCOR model can provide guidance for the construction and operation of SC. In turn, the performance attributes of SCOR model can become the criterion for meauring the operational status of SC. Based on this, according to the characteristics of the smart SC, from the perspective of reaching the performance attributes standard of SCOR model, this paper innovatively proposes the idea of analyzing the operation elements of the smart SC and the indicators for measuring the operational status of the smart SC.

(2) Operational Factors and Operational Status Indexes of the Smart SC based on SCOR Performance Attributes

SCOR model has five performance attributes: supply chain reliability, supply chain responsiveness, supply chain agility, total cost of supply chain and supply chain resources management [64]. From the  perspective of traditional supply chain around product manufacturing, the reliability of SC indicates the delivery ability of SC correctly and consistently, the responsiveness of SC means   product transactions quickly, the agility of SC shows its competition in response to changes of market and customer needs, costs in SC refers to the operating cost as low as possible, and the resources management attribute of SCOR indicates the efficiency of resource utilization for meeting the needs of enterprise, that is, the allocation efficiency of the tangible and invisible resources (such as human, material , financial, information, intellectual, and other resources.) [65-66].

To some extent, as can be seen that these SCOR performance attributes are consistent with characteristics of the smart SC. That is to say, according to the definition and characteristics of the smart SC, its five typical characteristics are corresponding to the performance attributes of SCOR. At the same time, the mapping relationship between them can be measured by some indicators.

Therefore, the mapping relationship between SCOR performances attributes and the smart SC can be expressed as follows:

Fig. 1. Smart SC operation analysis on SCOR performances attributes and its characteristics

Fig. 1 illustrates that the operational factors of SC, under certain driving strategies, drive SC with certain characteristics to run. Conversely, these factors have the qualities of SC. As a result, the operation status of SC can be measured by some indexes matched with smart SC's characteristics, which are also the concrete manifestation of SCOR performance attributes.

3.2.2. Conceptual model of risk in the smart SC

Risk factors in the smart SC are complex, multi-layered and uncertain. To analyse smart SC risks in a systematic way, the SCOR theory and the smart SC theory have been discussed to result into fig.1. The purpose of Fig.1 is not only to help us clarify the relationship between the operational factors and the operational status indicators for SC. Considering that risk in SC should to be occurred on the foundation of SC operational factors, and will be reflected through the operational status changes of SC, Fig.1 can also be the basis for SC risk analysis as follows:

Firstly, the market risk (equivalent to the living environment of the SC). Members in the manufacturing SC are widely distributed, and their business activities like procurement, manufacturing, sales and logistics are gradually spreading to the whole world, which makes the SC easier to be exposed to the external environmental risks [67]. In spite of the natural, economic and politics factors in the external environment, the risks of the external environment are mainly from the social market. In the face of diverse consumer needs and flexible production driven by products with shortened life cycles, a slightly careless mistake in the market decision making could lead to a SC risk.

Secondly, the technologic risk of the SC (technology is the nerve of a SC). The formation of a smart SC for intelligent manufacturing relies on such advanced technologies as intelligent computing, intelligent information processing, advanced computer technology, advanced SC management technology and the internet, etc. In the environment with rapid technological upgrades and dynamic market, the flexibility and intelligence of SC may not match the new market demands, making it difficult for enterprises to smoothly cooperate with others in the SC and provide consumers with satisfactory products. This can bring risks to the SC in a fiercely competitive market environment. Moreover, the emerging and applying of intelligent information based on IT technology will also bring out some information risk. the information risk (the risk equivalent to the blood in the SC). In addition to logistics and capital flow, information flow is the most important resource flow in a smart SC just like blood flowing through the entire SC. The study on information risk is quite important. In the cooperation between suppliers and demanders, they may deliberately conceal relevant information from each other for their own benefit, which can do harm to the overall interests of the SC. Moreover, due to outsourcing or mergers, intellectual properties or core competence in this information sharing network may also be disclosed to external parties [68]. The expansion of the smart SC can also cause a bullwhip effect. In the networked SC, any risk may spread quickly. Therefore, it is necessary to control the risks at the nodes of the SC.

Thirdly, the service risk (service in SC is the core of SC). With the rapid development of knowledge economy, manufacturing enterprises have not only attached importance to production capacity, but also gradually changed their competition between enterprises from traditional production equipment to manufacturing services [69]. Moreover, the manufacturing industry operation model needs to be transformed into a product service collaborative innovation mode with digital and information-based [70]. In addition to the risk of service before and after sales, Due to the bounded rationality and   opportunistic behaviour of human beings, there are some uncertain obstacles in product-service collaboration [71]. The essence of a smart SC is being “customer oriented”, rather than being “enterprise oriented”. Therefore, every process in the SC has to consider the risk of customer services, including services for cooperative enterprises and for end users. For the former, only if the effective services are provided to all enterprises in the SC, can the win-win situation be achieved [72].

Service quality is the soft power of the smart SC. Once there is any problem with the service, it is easy to trigger a SC risk. For this reason, the service risks of the smart SC are more direct and serious than those of the traditional one.

Fourthly, the risk of SC being taken as a virtual relational network (the relational structure is the skeleton of a SC). Tao (2017) pointed out that the shipbuilding supply chain developed into a complex system with obvious gradual interactive network topology, it is necessary to study the risk of the shipbuilding SC relational network based on cooperative symbiosis from the perspective of complex network [73]. The smart SC shows a high-tech virtual relational network structure for manufacturing high value-added products. This SC not only has many members, but also contains multi-layered and even diversified supply-demand relationships, which makes it a network relational structure. The network structure should be kept stable against the dynamic changes of cooperation to keep SC healthy in the long run, but there are some factors that can easily cause the instability of SC, like bad reputation of members, inadequate risk prevention ability of SC and defects of products or services [74], which are all likely to cause risks in SC.

Finally, the SC enterprise management risk (this risk equivalent to the brain of the SC). This risk is often ignored, but in fact, the management concept, methods and tools of an enterprise can affect the effectiveness of its risk management. The SC management in intelligent manufacturing is even more complex and changeable, so the risk from SC management should also be included in the risk study.

Generally, the above discussions provide a research perspective and a basic idea for the risk analysis of smart SC. In view of the above discuss, the theoretical model of smart SC risks has been given as Fig.2.

Fig.2. Conceptual model of risk in the smart SC

 

In Fig.2, the first layer represents the risk factors in operation; the second layer consists of the primary risk areas; and the third layer contains the secondary risk indicators, i.e. the causes of the secondary risks. In consideration of the setting of the following questionnaire, the risk factors are symbolically identified in the fig.2.

To find out the priority level of and correlation between risk dimension and risk indicator in Fig. 2, this study should conducts empirical data analysis of the questionnaire survey results to test and revise this concept model. Furthermore, the importance of each factor and the relations between them will to be studied. To test the conceptual model, this paper will use statistical analysis methods to study the relevance and importance of risks based on the survey data.

4. Empirical Identification of Risk Factors in the Smart SC

4.1. Questionnaire design and data collection

In order to test the conceptual model, the author extracted and modified questions from existing literatures and designed 43 questions, which cover the market, service, technology, virtual relational structure and management risks of the enterprise SC (Table 1). The answer to each question was set in a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), with 2 (disagree), 3 (neutral), and 4 (agree) in the middle.

 

Table 1. Survey Items

 

Modules

Items

Questions

Market Risks

Product Market Stability: P

MP1:Products in your company have advantages over those from our competitors.

MP2:Customer orders annually from your company is large.

MP3:Your products supply is usually higher than demand from customers.

Market Competitive Capability: C

MC1:The degree of product price competition between your company and your competitors is high.

MC2:The degree of maliciously slander opponents for competition purpose is high.

MC3:Your company’s developing status in your industry is high.

MC4:Your company can provide flexible supply-demand services.

Market Predictability: F

MF1:Your company can frequently adjust products’ production plan according to the forecasting of product sales trends.

MF2:Your company can auxiliary assist product designs according to customers’ demands.

Relational Structure Risks

Relational Structure Feasibility: RU

RU1:The financial dealings between your company and your customers basically establish on online financial business. 

RU2:The supply product quality and supply information can be protected.

RU3:Your supply chain has the capability of automatically providing early warning for shortage of supply sources and automatically supplementing supply sources.

Relational Structure Stability: RS

RS1:Your company has many subsidiaries or outsourced partner companies.

RS2:Your company’s supply of raw material for manufacturing is stable.

RS3:Your company has resource reserve plan and operates according this plan.

RS4:Your company has core technologies that need to be sold.

Coordination Among Companies in Supply Chain: RX

RX1:Your company’s material supplies, logistics cooperation and other cooperative companies have more cases of unreasonable termination of your contracts with them.

RX2:Your company participate in the cooperative business process of supervising cooperative companies.

RX3:Your company has conflicts in ideas or understanding caused by cultural differences with other companies.

Information Technology Risks

Applicability:

JY

JY1:The adoption of information technology in the supply chain in your company helps to reduce your operating costs.

JY2: The supply chain system in your company provides supportive for your product design, development and production.

JY3:The supply chain in your company can shorten some of your business processes.

Integration: JC

JC1:You are satisfied with the processing speed of the supply chain business in your company.

JC2:The business among departments in your company and the business between your company and the other companies are closely related and integrated.

Advanced: JX

JX1:The supply chain of your company can automatically make some predictive analysis based on operational data.

JX2:The supply chain in your company exists some misunderstanding or not timely in information communications.

JX3:The supply chain of your company provides technology support for supply chain information services.

Service Risks

Service People’s Quality: SP

SP1:The frequency of service personnel being complained about because of their service attitudes is low.

SP2:The service personnel in your company can solve most problems in time or give satisfactory answers.

SP3:The service personnel in your company are familiar with the core business of your company.

Service Quality: SQ

SQ1:Your customers are satisfied with personalized services provided by your company.

SQ2:The supply chain in your company can provide a high level of customer services.

Enterprise’s Profits from Service: SY

SY1:Customers place orders or participate in other business because they receive good services from your company.

SY2:Customers deepen the cooperation behind with your company because they are satisfied with your after-sales services.

SY3:The service capability of your company can be transformed into the company competitiveness to a certain extent.

Management Capability Risks

Cognition Risk: NC

NC1:The risk management staff in your company have Limited awareness of corporate risk.

NC2:The risk management staff in your company have unclear concepts of supply chain risks.

NC3:The risk management staff in your company work chaos.

Identification Risk: NS

NS1:Your company can effectively distinguish between supply chain and enterprise risk.

NS2:Your company has ways to carry out risk identification work.

Prevention and Control Risk: NK

NK1:Your company has insufficient awareness of potential risks, and often starts to handle a risk after it occurs.

NK2: The supply chain in your company can support automatic identification of risks.

NK3:Your company has a feasible risk pre-control plan.

 

The questionnaire to measure these items in Table 1 in this study was modified based on the validated questions in prior studies. The risk constructs in market were adapted from Peng and Wang (2018); Risks in relational structure risks from Long (2014) and Liu (2014); Risks in information technology from Zhou (2014); Risks in service from Liu (2018); And risks in management capability from Jia and Wang (2017).

 Pilot study was conducted on ten managers and senior executives in different companies. Each question in English was translated into Chinese, and then from Chinese to English by a language instructor to ensure the meaning of the questionnaire was the same in different languages. These two versions of the questionnaires were also reviewed by two professionals in academics.

In the large sample study, an online version of the questionnaire was emailed to mainly managers and senior executives from 1000 companies in a large variety of regions in China.  In sum, 860 responses were received, resulting in an 86.0% response rate. To further drop invalid data, we obtained 814 valid responses with 81.4% of valid responses.

Table 2 Sample Characteristics

 

 

Number of      Observations

Percentage (%)

Locations

Nanjing Area

105

14.6%

Zhejiang Area

103

14.3%

Guangzhou Area

52

7.2%

Tianjin Area

103

14.3%

Henan Area

103

14.3%

Sanxi Area

102

14.2%

Shanxi Area

104

14.5%

Hebei Area

8

1.1%

Beijing Area

39

5.4%

Industry

Light-textile industry

338

41.5

Resource processing industry

145

17.8

Mechanical & electronic manufacturing industry

311

38.2

Other industries

20

2.5

Nature of Companies

State-owned enterprise

152

18.7

Collective enterprise

283

34.8

Private enterprise

336

41.3

Foreign-funded enterprise

43

5.3

Scale of Companies

Small

227

27.9

Medium

352

43.2

Large

235

28.9

Your Department

 

Technical department

192

23.6

Procurement department

332

40.8

Financial department

259

31.8

Logistics department

26

3.2

Other departments

5

0.6

Total

814

100

Among the valid responses, 41.5% (338) were received from the light-textile industry, 17.8% (145) from resource processing industry, and 38.2% (331) from mechanical and electronic manufacturing industry. These industries represent the major intelligent manufacturing sectors in China. The nature of companies covers from 18.7% (152) state-owned enterprise, 34.8% (283) collective enterprise, 41.3% (336) private enterprise, and 5.3% (43) foreign-funded enterprise. Besides, the distribution of the number of employees reflects the balance of small (27.9%), medium (43.2%) and large (28.9%) firms. In addition, 23.6% of subjects from the technical department, 40.8% from the procurement department, 31.8% from the financial department, 3.2% from the logistics department, and 0.6% from the other departments. By checking the sample characteristics in Table 2, we see that our sample has a good representation of the population.

To test for common method bias, we use Harman’s single factor (Podsakoff et al. 2003). The results show that there was no significant bias in the collected data set.

4.2. Test of reliability

Reliability refers to the consistency and stability of the survey data results, which is mainly used to test the credibility of the questions in the questionnaire. The greater the reliability coefficient of the questionnaire is, the higher the reliability of the survey results will be [75]. The reliability and validity of the questionnaire are shown in Table 3.

Table 3. Reliability Analysis

First hierarchy

Second hierarchy

Question’ items

Scale average after deleting items

Scaling variance after deleting items

Corrected item and total correlation

Alpha

after deleting items

Reliability coefficient of α

Market risk

(M)

MP

MP1

160.0147

229.188

0.461

0.899

0.834

MP2

160.0737

225.512

0.527

0.898

MP3

160.1020

225.371

0.531

0.898

MC

MC1

160.1290

225.616

0.535

0.898

MC2

160.1302

224.054

0.578

0.897

MC3

160.1855

224.323

0.504

0.898

MC4

160.3882

225.401

0.403

0.900

MF

MF1

160.5283

224.869

0.465

0.899

MF2

160.3317

224.269

0.434

0.899

Relational network risk of SC

(R)

RU

RU1

159.9693

224.598

0.538

0.898

0.896

RU2

159.9582

223.703

0.571

0.897

RU3

160.2998

223.130

0.538

0.898

RS

RS1

160.0651

222.951

0.591

0.897

RS2

160.1462

222.698

0.617

0.897

RS3

160.2113

222.966

0.587

0.897

RS4

160.3071

223.017

0.610

0.897

RX

RX1

160.2039

222.770

0.591

0.897

RX2

160.4128

223.261

0.539

0.898

RX3

160.2052

221.903

0.579

0.897

IT technology risk of SC

(J)

JY

JY1

160.2027

223.323

0.541

0.898

0.844

JY2

160.1327

222.996

0.562

0.897

JY3

160.2064

221.733

0.626

0.896

JC

JC1

159.9681

224.833

0.523

0.898

JC2

160.1278

226.075

0.550

0.898

JX

JX1

160.1007

225.643

0.551

0.898

JX2

159.7039

229.252

0.414

0.900

JX3

160.2666

227.571

0.442

0.899

Service risk in SC

(S)

SP

SP1

160.0749

227.617

0.459

0.899

0.795

SP2

159.8329

235.655

0.154

0.902

SP3

159.9459

236.194

0.102

0.903

SQ

SQ1

159.9005

236.850

0.089

0.903

SQ2

159.8686

237.167

0.068

0.903

SY

SY1

160.1486

236.611

0.086

0.903

SY2

160.2322

235.111

0.097

0.904

SY3

160.1572

235.250

0.095

0.904

Management risk of enterprise' risk

(N)

NC

NC1

160.0909

234.577

0.130

0.903

0.766

NC2

159.9386

235.253

0.181

0.902

NC3

159.9533

235.521

0.166

0.902

NS

NS1

160.0577

233.548

0.191

0.902

NS2

160.0627

234.280

0.150

0.903

NK

NK1

160.3428

232.585

0.171

0.903

NK2

159.9472

235.019

0.185

0.902

NK3

160.1339

234.493

0.137

0.903

The reliability and validity of this SC risk scale are tested. The results in Table 3 indicate good reliability and validity as all alphas are greater than 0.75. The results from good reliability and validity indicate that factor analysis can be performed.

4.3. Correlation analysis

In order to initially grasp the correlation between secondary layer risk factors and the third risk indexes, and to remove the meaningless factors through correlation analysis of SPSS, correlation analysis  results of these factors was shown in Appendix A.

According to the correlation test shown in table 2 and the correlation criteria, on the one hand, it can be found that sig. values of variables in N field are all greater than 0.1, indicating that these variables have lower correlation and need to be eliminated. In the same way, the risk variables correlation in S field are also lower besides SP1 variable; On the other, from the Pearson correlation value, we can see that there are some changes in the relationship between indicator variables and their original dimension variables, which shows that the risk conceptual model given in Fig.2 needs to be further adjusted.

4.4. Improving risk conceptual model of the smart SC

To further optimize the risk model of fig.2, a decision information table will be formed by Pearson correlation value of indexes, and then the Ward's method based on Euclidean distance measure are used, clusters number is assumed to be five, to carry out the clustering analysis.

After data pre-processing, such as data normalized to a maximum of 1, and deleting invalid records (total number of records: 43, effective number of records: 22, discarded number of records: 21), the hierarchical clustering spectrum is as follows:

Fig.3. clustering between indexes and risk dimensions

 

The column number in Figure 3 represents number of index. According to the clustering hierarchy, the optimized conceptual model is shown as follow.

Fig.4. Improved concept risk model of the smart SC

 

Compared with Fig. 2, Fig.4 has two new risk hierarchies, and a product risk dimension has been clustered into. This figure clearly defines the relationships of the remaining 22 indicators after dimensionality reduction and these risk dimensions.

5. Assessment model of the smart SC for manufacturing and results

The concept risk hierarchy model of the smart SC in intelligent manufacturing has been constructed. Then, in order to form a feasible risk evaluation model, we should focus on the study of risk weights.

5.1. Weightss of risk indexes

We perform principal component analysis on the questionnaire data of the remaining valid 22 indicators. The score matrix of the supposed two principal components is shown in Table 4.

 

Table 4. Score coefficient matrix of component

Extraction method: Principal component analysis

Note: Rotation method: maximum variance method of Kaiser standardized Component rating

 

 

In Table 4, the largest score coefficient for each index will be taken as probability values of these indexes. The following calculating about the information entropy weight of every risk dimension should to be carried out on indexes ‘values.

5.2. Entropy weights of risks of the smart SC

In addition to risk indicators at the bottom of concept risk model and variables directly measured, there are three risk dimensions in Fig.4. To do it, we will select a suitable method to define weights. The general methods mainly include expert scoring method, entropy weight method and composite weighting method combining entropy weight with expert scoring. AS we all know, that he subjective difference of expert scoring is large. Moreover, according to the golden ration of questionnaires number and   experts’ number, if we use composite method, it means that we need to find a large number of representative experts with strong business ability, which is obviously unrealistic. In contrast, this study will employ the method of entropy weight, which is an objective weight measure on index variability.

The entropy weight method is built on the calculation of information entropy. Information entropy is proposed by Shannon to indicate the uncertainty of things occurrence and a measure of disorder in the system [76]. According to the basic principle of information entropy, entropy can be used as an ideal scale in the weight calculation of evaluation index [77-78]. It is built on the idea that the larger the entropy is, the greater the evaluation index variation is, the greater the information amount is, and the greater the index weight would be [79]. Although, the entropy weight method sometimes has false low probability events, more than this, the composite weightings of indexes are more complicated and independent of each other to result into some conflicts inevitably in defining indicator weights is more complicated [80]. So, considering that the study based on the empirical data of the questionnaire will avoid false low probability events, in contrast, the objective and simplistic entropy weight method is more suitable for this study.

Therefore, this paper proposes a method to determine the weights of risk fields and risk indexes based on the information entropy.

5.2.1. Calculating the entropy weight of risk in the smart SC

The calculation of information entropy weight mainly takes two steps. One is the calculation of information entropy, and the other is the calculation of information entropy weight [81-83].

(1) Calculating the information entropy of risk indexes in bottom (namely, in 5-th level)

According to the information entropy theory, the information entropy of factor can be calculated by equation (1).

                           (1)

In which,   , stands for the Boltzmann constant. Here, n represents the number of questionnaire samples, that’s n=814.  ( ), represents the occurrence probability of the j-th index in table 4, and the  is the information entropy of j-th index. 

Considering that the indicators in each evaluation group are more closely related, the formula for calculating the weight on information entropy is set as follows:

                                (2)

Here,  denotes the number of members in each group at the fifth level in fig.4.

According to formula 1 and formula 2, the weights on information entropy of 22 risk indicators can be calculated as follows:

 

Table 5.  Weight on information entropy of indexes in 5-th level

  

 

It can be seen that the larger the cumulative contribution rate based on the eigenvalue, the greater the weight of the indicators, and these weights show a normal distribution, so the weights given of these indicators are more reasonable.

(2) Calculating the entropy weight of dimension risk factor

Since the middle dimension of Figure 4 has three layers, we should calculate the weights of these dimension factors from bottom to top according to the following formula.

                          (3)

And, 

 presents the level number in fig.4.   indicates the  -th dimension factor in  -th level.   presents the information entropy of  -th risk factor in   level.  is the number of members included into in  -th group.

Since the information entropy of IE is directly derived from the accumulation of information entropy of its various factors, which have m factor members, so,  .

Similarly, the equation for calculating the entropy weight of dimension indicators is shown as in following:

      (4)

And, 

 

With the above equation (3) and equation (4), the entropies and entropy weights of the risk factors in dimension layers are resulted as shown in Table 6.

Table 6. Weights on Information entropy of indexes in other levels

 

As can be seen from Table 6, the importance of the third risk indicator is: R12 > R22 > R11 > R21 > R23. The result shows that the service risk and IT risk are the core factors causing risks in the smart SC. In order to test the rationality of these risk factors, the empirical simulation analysis should be carried out as following. At the same time, the internal risk such as R2 is more influential than R1. This shows that the possibility of the SC risk caused by the enterprise itself is greater than the risk brought by other nodes in SC. Therefore, enterprises should pay more attention to the management of their internal risks for the supply chain risk management. That is to say, internal causes are always the root of problem.

5.3. Testing of risk factors effectiveness in Fig.4

Assuming that the risk of the smart SC can be represented by a variable of  . Based on the original questionnaire data, we use the method of information entropy to measure the value of SC risk for each investigating enterprise by the formula (5), which can be taken as the actual value of risk of   for the enterprise. The formula is shown as following.

    (5)

The  represents the score category of risk factors in an enterprise, and   is the total number of value category.   shows the score of 1-5 for each factor in the questionnaire. the 22 present the number of measurable variable in fig.4.

According to the equation, some actual risk values of the smart SC could to be measured. Then, the questionnaire data and the actual risk value of the corresponding smart supply chain, is regarded as the sample data set, after 1000 times of training by BP neural network, the result is simulated as follows.

Fig.5. The results of the prediction value and the actual value for variable SC_r

 

The actual accuracy of the system training is 0.000869. From fig.5, it can be seen that the prediction is more accurate for the risk of the smart SC, which present that the smart SC oriented to intelligent manufacturing is deeply affected by these risk factors shown as in fig.4. This shows that the risk factors identified in this paper have a certain practical significance.

6. Conclusions

The smart SC has been a research hotspot in the field of intelligent manufacturing. To help manufacturers in China explore intelligent manufacturing smoothly, studying the risks in the smart SC is very necessary, particularly under such realistic conditions of few research on SC risks.

Starting from the present situation of Chinese manufacturing, SCOR model and statistical analysis of questionnaires based an improved concept model of risk for the smart SC in intelligent manufacturing has been provided innovatively. This provides effective support for our in-depth study of smart SC risks. Moreover, the method of information entropy weight, has been optimized, which has certain innovation in this study method, to evaluate the risk factors in improved concept model proposed. So, we could to form a risk Identification model which can be used in in risk assessment, risk warning, and other risk management field.

The experimental simulation shows that the prediction accuracy is more precise and the predicted result is more ideal. In some extent, the risk factors proposed in the paper is feasible and effective.

 

 

 

 

References

Schiffer, M.; Luckert, M.; Wiendahl, H.H.; Saretz, B. Smart Supply Chain Development of the Equipment Supplier in Global Value Networks. IFIP WG 5.7 International Conference, APMS 2018, Seoul, Korea, Proceedings 2008, Part I, pp. 176-183, doi: 10.1007/978-3-319-99704-9_22.
Zhang W.H.; Zhang X.; Weng Z.M. How do Manufacturing Firms Acquire Knowledge for Service Innovations. Management World. 2010, 122-134.
Jian, Z.Q.; Li, L.; Liu, Y. Review and Prospect of service supply chain integration and its impact on service innovation. Foreign Economies and Management. 2013, 35, 37-46.
Sun, D.y.; Qu, X.M.; Zhang, S.B. Advanced Manufacturing Technology. Beijing: Mechanical Industry Press, First edition, 2000, ISBN:  9787111075394.
Okad, T.; Namatame, A.; Sato, H. An Agent-Based Model of Smart Supply Chain Networks. Intelligent and Evolutionary Systems: The 19th Asia Pacific Symposium, IES 2015, Bangkok, Thailand, Proceedings, 2016, 373-384, doi:10.1007/978-3-319-27000-5_30.
Qiu, F.H. Smart Supply Chain Construction for Intelligent Manufacturing. Logistics & Material Handling, 2017, 3, 92-94, doi: 10.3969/j.issn.1007-1059.2017.03.012.
Tian, C.M.; Du, X.Y. To Study on Risk of Supply Chain and Direct Losses in Manufacturing Field. Value Engineering. 2007, 66-68, doi: 10.3969/j.issn.1006-4311.2007.04.021.
Guo, Q.L.; Zhang, M. An agent-oriented approach to resolve scheduling optimization in intelligent manufacturing.Robotics and Computer Integrated Manufacturing. 2010, 26, 39-45.
President Obama Launches Advanced Manufacturing Partnership [EB/OL].   2011, http://www.whitehouse.gov/the-press—otiqce/20111061241.
Wang, X.W. The Manufacturing Romance that Occurred among the China, United States and Germany [EB/ OL]. 2016, http://news.xinhuanet.com / info / ttgg/2016 - 02/01/ c _135063399.
Xiong, Y.L. Intelligent Manufacturing. Science & Technology Review. 2013, 10, 3-3.
Zhu, J.Y. Significance, Technology and Implementation of Intelligent Manufacturing. Aeronautical Manufacturing Technology. 2013, 23, 30-35, doi: 10.3969/j.issn.1671-833X.2013.23.001.
Li, X.L.; Dong, S.Z. Intelligent Monitoring Technology in Advanced Manufacturing, Science Press in China. 1999.
Liu, J.P.; Ye, B.Y. The Summarization of the Development Trends of Intelligence Manufacturing and Its Application. Journal of Shanghai University (English Edition). 2004, 8(Suppl), 129-133, doi:1007-6417(2004)05-0129-05.
Chen, Y.X. Intelligent Manufacturing System and Its Development. Electronic Technology & Software Engineering. 2018, 15, 253-253.
Zhang, Z.J. Research on Supply Chain System Construction for Intelligent Manufacturing. Brand. 2015, 69-69, doi:10.19373/j.cnki.14-1384/f.2015.11.052.
Frankowiak, M.; Grosvenor, R.; Prickett, P. A Review of the Evolution of Microcontroller-based Machine and Process Monitoring.International Journal of Machine Tools and Manufacture. 2005, 45, 573-582, doi:10.1016/j.ijmachtools.2004.08.018.
Ruiz, N.; Giret, A.; Botti, V.; Feria, V.An Intelligent Simulation Environment for Manufacturing Systems[J].Computers & Industrial Engineering. 2014, 76(C), 148-168, doi:10.1016/j.cie.2014.06.013.
Wu, S.M.; Hu, X.B.; Wang, J.W. FMS production management technology research based on MES seamless integration. Manufacturing Automation. 2016, 11-14.
Guo, Y.L.; Zhao, W.Q. Research and Application of Production Informatization in a Military Enterprise Based on MRP2 and MES System. Heilongjiang Science and Technology Information. 2016, 24, 139.
Fang, D.J.; Li, W. Intelligent Logistics System Construction Supporting Intelligent Manufacturing. Logistics & Material Handling. 2017, 88-92, doi:10.3969/j.issn.1007-1059.2017.03.011.
Tao, F.; Zhang, L.; Nee, A.Y.C. A Review of the Application of Grid Technology in Manufacturing. International Journal of Production Research.  2011, 49, 4119-4155, doi:10.1080/00207541003801234.
Gao, J.; Yao, Y.; Zhu, V.C.Y.; et al.  Service-oriented manufacturing:a new product pattern and manufacturing paradigm. Journal of Intelligent Manufacturing. 2011, 22, 435-446.
Li, B.H.; Zhang, L., Wang, S.L.; et al.  Cloud Manufacturing:A New Service-oriented Networked Manufacturing Model. Computer Integrated Manufacturing Systems. 2010, 16, 1-7.
Tso, S K; Lau, H. ; Ho, John, K.L. Coordination and Monitoring in an Intelligent Global Manufacturing Service System.Computers in Industry. 2000, 43, 83-95,  doi:10.1016/S0166-3615(00)00053-1.
Hu, Y.; Zhou, X.; Li, C.Internet Based Intelligent Service-oriented System Architecture for Collaborative Product Development.International Journal of Computer Integrated Manufacturing. 2010, 23, 113-125, doi:10.1080/09511920903440347.
Cáceres, R.; Guzmán, J. Seeking an Innovation Structure Common to Both Manufacturing and Services. Service Business. 2014, 9. No.3, pp.361–379,     doi: 10.1007/s11628-014-0234-1.
Do, N. Integration of Design and Manufacturing Data to Support Personal Manufacturing Based on 3D Printing Services. The International Journal of Advanced Manufacturing Technology. 2017, 90, 3761-3773, doi:10.1007/s00170-016-9688-8.
Cagnin, C.; T Könnölä, T. Global Foresight:Lessons from a Scenario and Roadmapping Exercise on Manufacturing Systems.Futures. 2014, 59, 27-38, doi: 10.1016/j.futures.2014.01.007.
Choy, K.L.; Lee, W.B.; Lau H.; Lu, D.; Lo, V.Design of an Intelligent Supplier Relationship Management System for New Product Development.International Journal of Computer Integrated Manufacturing. 2004, 17, 692-715, doi: 10.1080/0951192042000237483.
Su, Y.C.; Cheng, F.T.; Hung M.H.; Hung, H.C. Intelligent Prognostics System Design and Implementation. IEEE Transactions on Semiconductor Manufacturing. 2000, 19, 195-207,  doi:10.1109/TSM.2006.873512.
Li, A.X; Li, Y.J; Xu, C.M. Study on the Models Based on Multi-agent of Process Enterprise intelligent Management. Manufacturing Automation. 2006, No.11, 23-28,  doi:10.3969/j.issn.1009-0134.2006.11.007.
Li, B.H.; Hou, B.C.; Yu, W.T; Lu, X.B; Yang, C.W. Applications of Artificial Intelligence in Intelligent Manufacturing: a Review. Frontiers of Information Technology & Electronic Engineering. 2017, 18, 86–96, doi:10.1631/FITEE.1601885.
Zhou, C.Y.; Cao, Q.L. Design and Implementation of Intelligent Manufacturing Project Management System Based on Bill of Material. Cluster Computing. 2018, 1-9, doi:10.1007/s10586-018-1934-4.
Pan, X.Y. Dynamic Optimization Policy of the Cloud Platform for Enterprise Manufacturing Resources. Journal of Technology Economics and Management Research. 2016, 15-19, doi:10.3969/j.issn.1004-292X.2016.08.003.
Lee, Y.H. Smart Logistics and Global Supply Chain Management. Maritime Economics & Logistics. 2016, 18, 1–2, doi:10.1057/mel.2015.13.
Pu, Z.; Jiang Q.; Yue, H.; Tsaptsinos, M.  Agent-based Supply Chain Allocation Model and its Application in Smart Manufacturing Enterprises. Journal of Supercomputing. 2018, No.11. 1-11, doi:10.1007/s11227-018-2536-x.
Chen, F. Information Sharing and Supply Chain Coordination. Handbooks in Operations Research & Management Science. 2003, 11, 341-421, doi:10.1016/S0927-0507(03)11007-9.
Qing, L. Research on Supply Chain Reengineering for Hazardous Chemicals. Chemical Engineering Transactions, 2018, 66, 1495-1500,  doi: 10.3303/CET1866250.
Song, S.L. Application of Gray Prediction and Linear Programming Model in Economic Management. Mathematical Modelling of Engineering Problems. 2018, 5, 46-50, doi:10.18280/mmep.050107.
Lambert, D.M.; Enz, M.G. Issues in Supply Chain Management: Progress and Potential. Industrial Marketing Management. 2017, 62, 1–16, doi:10.1016/j.indmarman.2016.12.002.
Li, Y. Initial Effect of Printing and Dyeing Intelligentization Interview with Chen Zhihua, President of China Printing and Dyeing Industry Association. Textile Science Research. 2015, No.6, 42-43.
Qi, D.H.; Cheng, N.; Shi, H.Y.; etc. The Design and Implementation of the Material Pulling System for Engine Assembly Line. Modular Machine Tool & Automatic Manufacturing Technique. 2016, No.7, 154-157, doi: 10.13462/j.cnki.mmtamt.2016.07.043.
Hao, J.; Ji, W.X.; Zhang K.The Application Research of RFID in Elevator Component Material Management and Tracking. Machinery Design & Manufacture. 2016, No.6, 265-268, doi:10.3969/j.issn.1001-3997.2016.06.071.
Jia, G.Y.; Wang, X.H. Application Research of On-line Monitoring Technology for Cutting Process. Technology Outlook. 2017, 27, 119-120, doi:10.3969/j.issn.1672-8289.2017.04.108.
Göllner, J.; Peer, A.; Rass, S.; Zathurecky, V. Supply Chain Network Risk Analysis and Monitoring for Global Cyber Defense, Handbook of Cyber-Development, Cyber-Democracy, and Cyber-Defense. Springer, Cham,2017, Online ISBN 978-3-319-06091-0, pp.1-23, doi:10.1007/978-3-319-06091-0_33-1.
Abdi, M.R; Labib A.W; Edalat, F.D; Abdi, A. Risks in Manufacturing Supply Chain Incorporating RMS. Integrated Reconfigurable Manufacturing Systems and Smart Value Chain, Chapter 11. Springer, Cham, 2018, Print ISBN: 978-3-319-76845-8, 255-279, doi :10.1007/978-3-319-76846-5_11.
Cranfield school of Management Supply Chain Vulnerability[R]. Cranfield Univenity, 2002.
Yang, Y.; He, Z.W. Study on Risk Recognition and Evaluation of New Energy Automobile Supply Chain Based on SCOR Model. Logistics Technology. 2015, 186-191, doi:10.3969/j.issn.1005-152X.2015.10.050.
Chen, J.L.; Zhao, Y. Expandable Matter-element Risk Analysis of Automobile Manufacturing Supply Chain. Logistics Technology. 2013, 32, 410-412, doi:10.3969/j.issn.1005-152X.2013.03.135.
Guo, A.F. Identification and Assessment of Supply Chain Risk on the Condition of the Complex Product Exploitation with the Participation of Suppliers—Based on the Analysis of Equipment Manufacturing Enterprises. Master's thesis of Zhejiang Sci-Tech University of China, 2017.
Zhang, Y.C. The Change of Intelligent Supply Chain under the Intelligent Manufacturing. Logistics & Material Handling. 2018, 84-86.
Huang, C.C; Ye, C.S; Wang, X.X; Shi, Z.M. Research on the Construction of Wisdom Supply Chain System. Value Engineering. 2018, 121-123.
Zhao, Z.Z.; Wang, F. Top Level Design on Smart Supply Chain Cost Control House System: An Example of Oil and Gas Mining Area. China Soft Science, 2014, 184-192, doi:10.3969/j.issn.1002-9753.2014.08.019.
Zhao, R. Development and Construction of Intelligent Supply Chain. China Market Marketing. 2015, 93-94, doi: 10.13939/j.cnki.zgsc.2015.10.09.
Ahn, K.; Lim, S.; Lee, Y. Modeling for Smart Supply Chain: Interoperability, Visibility, and Sustainability, The 2015 World Congress on Information Technology Applications and Services (World IT Congress 2015) . 
Wu, L.F.; Yue, X.H.; Jin, A; Yen, D.C. Smart Supply Chain Management: a Review and Implications for Future Research. The International Journal of Logistics Management. 2016, 27, 395-417, doi :10.1108/IJLM-02-2014-0035.
Song, H. The Core elements and realization path of the smart supply chain. Logistics & Material Handling, 2015, 58-59, doi: 10.3969/j.issn.1007-1059.2015.12.001.
Wang, W.P. Research on smart supply chain of M company. Master degree thesis of Chongqing University in China. 2005.
Ma Lin. A Study of Supply Chain Integration Risk Management Based on SCOR Mode[J].Commercial Research,2008, (6) :1-5. DOI.10.3969/j.issn.1001-148X.2008.06.001
FB Georgise,KD Thoben,M Seifert., Marcus Seifert.Implementing the SCOR Model Best Practices for Supply Chain Improvement in Developing Countries.International Joumal of u—and e—Service,2013,(6)13-26.
Liang Yixuan. Research on risk identification and evaluation of supply chain in B2C enterprises based on SCOR model. Master Dissertation of Capital University of Economics and Business, 2015.
Xu Peipei. Risk Identification and Prevention of Eternal Asia Co.’S supply chain based on SCOR model. Master Dissertation of Jiangxi University of Finance and Economics,2018.
QU Guoqiang. A Research on Coal Supply Chain Model Based on SCOR. Industrial Engineering Journal, 2016, 19(2):7-12.  doi:10.3969/j.issn.1007-7375.2016.02.002
Li Lei. The Process Improvement of TD Company's Supply Chain based on SCOR Model. Master Dissertation of North University of China, 2017.
Liu Yingli. Research on Supply Chain Performance Evaluation Index Construction of BL Company Based on SCOR Model. Master Dissertation of China University of Geosciences, 2013.  
Thun, J.H.; Hoenig, D. An Empirical Analysis of Supply Chain Risk Management in the German Automotive Industry. International Journal of Production Economics. 2011, 131, 242-249, doi :10.1016/j.ijpe.2009.10.010.
Ma, S.H. Supply chain management [M]. Beijing: Machinery Industry Press in China. 2014, Second Edition, ISBN:9787560996202
Zhang, H.Q; Lu, R.Y. Research of supplier involvement in service innovation process and its impacts. Studies in Science of Science. 2010, 28, 1422-1427.   doi :10.16192/j.cnki.1003-2053.2010.09.021.
Dai, Y.; Suo, C.; Song, H. Research on the Present Situation and Countermeasures of the Collaborative Innovation of Civil-Military Integration. Journal of Chongqing Institute of Technology. 2017, 31, 64-70, doi: 10.3969/j.issn.1674-8425(s).2017.06.010.
Zhao, L.L; Wei, J; Zheng, X.Y. Definition, Classification and Research Framework of Service Innovation Strategy for Manufacturing Enterprises. Foreign Economies & Management. 2012, 34, 59-65, doi:10.16538/j.cnki.fem.2012.09.002.
Dawid, H.; Decker, R.; Hermann, T; Jahnke, H.; Klat, W. ; König, R.; Stummer, C. Management Science in the Era of Smart Consumer Products: Challenges and Research Perspectives.Central European Journal of Operations Research. 2017, 25, 203–230, doi: 10.1007/s10100-016-0436-9.
Tao, Q. Study on Risk Management of China's Shipbuilding Supply Chain Complex Network Based on Cooperative Symbiosis Mode. Shanghai Management Science. 2017, 9, 64-72, doi : 10.3969/j.issn.1005-9679.2017.02.014.
Trkman, P.; McCormack, K. Supply Chain Risk in Turbulent Environments a Conceptual Model for Managing Supply Chain Network Risk. International. Journal of Production Economics. 2009, 119, 247-258, doi: 10.1016/j.ijpe.2009.03.002.   
Peng, Y.C; Chang, W.H.; Shen, Y.H. How to Measure the Reliability of a Questionnaire. Chinese Journal of Hospital Administration. 2004, No.6, 383-384, doi:10.3760/j.issn:1000-6672.2004.06.027
Shannon, C.E; Weaver, W. The Mathematical Theory of Communication. Urbana: University of Illinois Press, 1971, ISBN:  9780252725487.
Meng, X.M; Hu, H.P.Application of Set Pair Analysis Model Based on Entropy Weight to Comprehensive Evaluation of Water Quality.Journal of Hydraulic engineering. 2009, 40, 257-262, doi:10.3321/j.issn:0559-9350.2009.03.001.
Huang, R.; Zhang, F.X. Method of Comprehensively Evaluating Water Quality Based on Optimal Set Pair Analysis. Journal of Yangtze River Scientific Research Institute. 2016, 33, 8-10, doi:10.11988/ckyyb.20150157.
Shi, K.F.; Diao, C.T.; Zuo, T.A; Sun, X.F.; Sun Y.A. Evaluation of Eco-security of Cultivated Land Requisition-compensation Balance Based on Entropy Weight and Matter Element Model, Chinese Journal of Eco-Agriculture. 2013, 21, 243-250, doi: 10.3724/SP.J.1011.2013.00243.
Yang, D.B.; Wang, ZH.ZH.; Ma ,X.M. Ecological Impact Post-assessment of Projects Based on Fuzzy Matter-element Model, Journal of Water Resources Architectural Engineering. 2012, 10, 93-96, doi: 10.3969/j.issn.1672-1144.2012.02.023.
Xue, P. Research on the Evaluation of Community Workers' Training Effect in Baoji City Based on Entropy Method. Dissertation for the Professional Degree of Master of Northwest University in China, 2017.
Peng, B.H.; Wu, B.Y ; Zhou, Y. A Study on the Risk Assessment of Cold Chain Scission Based on Entropy Weight and Extension Decision Model. Journal of Beijing Jiaotong University (Social Sciences Edition). 2017, 16, 110-119, doi: 10.3969/j.issn.1672-8106.2017.01.013.
Lei, X.P. Empirical Research on Supply Chain Risks Evaluation in Chinese Automobile Manufacturing Industry: Analysis Based on Entropy Extension Decision Model. Computer Engineering and Applications. 2015, 51, 264-270, doi:10.3778/j.issn.1002-8331.1306-0344.
Liu Peng. After-sales Service Risk analysis of Railway Signaling Equipment Manufacturers[J]. Railway Signaling & Communication Engineering, 2018, 15(8) :85-87,92
Zhou Si Hua.  The Impact of Information Technology on Supply Chain Risk Management: A Capability-based Perspective[D]. PhD Dissertation of Shanghai Jiaotong University, 2014.
Long Yong-qiang. An empirical study on the relationship between trust, flexible supply and supply performance in supply chain partnership [D].  Master’s Degree thesis of Renmin University of China, 2008.
Liu Qi. Research on the Influence of Relationship Stability on the Performance of Supply Chain——Based on the Fresh Agricultural Supply Chain [D].  Master’s Degree thesis of Ning Bo University,  2014.
Peng Hong-jun,Wang Miao-miao. Risk Evaluation of the Forest-pulp-paper Supply Chain based on AHP and EVM[J]. China Forestry Economy, 2018, (5) :25-29.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879-903.

 

 

 

Appendix A. Correlation between risk dimensions and risk indexes.

 

MP

MC

MF

RU

RS

RX

JY

JC

JX

SP

SQ

SY

NC

NS

NY

MP1

Pearson correlation

.666**

.384**

.358**

.369**

.383**

.402**

.356**

.321**

.384**

.220**

-0.017

-0.012

-0.026

0.039

-0.005

Sig.

0

0

0

0

0

0

0

0

0

0

0.622

0.728

0.466

0.268

0.889

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

MP2

Pearson correlation

.880**

.563**

.359**

.374**

.448**

.374**

.388**

.320**

.327**

.193**

-0.033

-0.005

0.058

0.053

-0.007

Sig.

0

0

0

0

0

0

0

0

0

0

0.344

0.882

0.096

0.133

0.844

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

MP3

Pearson correlation

.861**

.599**

.362**

.385**

.427**

.392**

.354**

.318**

.330**

.183**

-0.043

-0.024

0.029

0.051

.084*

Sig.

0

0

0

0

0

0

0

0

0

0

0.224

0.502

0.41

0.144

0.017

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

MC1

Pearson correlation

.645**

.761**

.381**

.416**

.431**

.371**

.377**

.330**

.362**

.179**

-.080*

-0.045

0.012

0.032

0.052

Sig.

0

0

0

0

0

0

0

0

0

0

0.022

0.202

0.723

0.357

0.138

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

MC2

Pearson correlation

.656**

.755**

.390**

.439**

.481**

.404**

.417**

.345**

.343**

.233**

-0.016

-0.007

0.053

0.062

0.054

Sig.

0

0

0

0

0

0

0

0

0

0

0.643

0.843

0.128

0.077

0.126

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

MC3

Pearson correlation

.434**

.779**

.457**

.426**

.404**

.390**

.389**

.318**

.294**

.132**

-0.063

0.01

-0.006

0.019

0.03

Sig.

0

0

0

0

0

0

0

0

0

0

0.071

0.774

0.873

0.596

0.393

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

MC4

Pearson correlation

.300**

.757**

.456**

.315**

.310**

.320**

.319**

.290**

.275**

.113**

-0.06

-0.06

-0.03

-0.015

0.047

Sig.

0

0

0

0

0

0

0

0

0

0.001

0.087

0.089

0.394

0.671

0.184

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

MF1

Pearson correlation

.373**

.594**

.742**

.384**

.391**

.359**

.360**

.325**

.277**

.138**

-0.027

-0.019

-0.006

0.009

0.002

Sig.

0

0

0

0

0

0

0

0

0

0

0.434

0.584

0.861

0.792

0.962

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

MF2

Pearson correlation

.313**

.278**

.795**

.465**

.363**

.419**

.367**

.334**

.360**

.154**

-0.033

-0.011

-0.008

0.044

0.017

Sig.

0

0

0

0

0

0

0

0

0

0

0.346

0.745

0.819

0.214

0.633

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RU1

Pearson correlation

.371**

.393**

.456**

.796**

.524**

.477**

.426**

.369**

.385**

.172**

-.086*

-0.027

0.041

0.023

0.013

Sig.

0

0

0

0

0

0

0

0

0

0

0.014

0.438

0.239

0.506

0.701

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RU2

Pearson correlation

.357**

.421**

.425**

.799**

.506**

.560**

.456**

.393**

.415**

.222**

-0.015

0.007

0.046

0.03

0.001

Sig.

0

0

0

0

0

0

0

0

0

0

0.663

0.834

0.193

0.388

0.974

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RU3

Pearson correlation

.391**

.432**

.455**

.818**

.533**

.542**

.440**

.356**

.358**

.203**

-.074*

-0.035

0.004

-0.058

-0.058

Sig.

0

0

0

0

0

0

0

0

0

0

0.034

0.325

0.92

0.1

0.096

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RS1

Pearson correlation

.446**

.425**

.381**

.494**

.790**

.529**

.530**

.405**

.390**

.176**

-0.049

0.03

0.041

0.026

0.022

Sig.

0

0

0

0

0

0

0

0

0

0

0.165

0.39

0.239

0.463

0.54

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RS2

Pearson correlation

.448**

.457**

.397**

.521**

.831**

.558**

.565**

.444**

.423**

.141**

-0.056

-0.01

0.029

0.019

0.009

Sig.

0

0

0

0

0

0

0

0

0

0

0.11

0.784

0.413

0.581

0.792

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RS3

Pearson correlation

.408**

.391**

.382**

.542**

.818**

.614**

.535**

.375**

.436**

.157**

-.091**

-0.058

-0.002

-0.007

0.035

Sig.

0

0

0

0

0

0

0

0

0

0

0.009

0.096

0.953

0.836

0.316

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RS4

Pearson correlation

.391**

.437**

.433**

.553**

.817**

.672**

.529**

.460**

.469**

.164**

-.088*

-0.063

-0.011

-0.046

-0.02

Sig.

0

0

0

0

0

0

0

0

0

0

0.012

0.074

0.75

0.194

0.567

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RX1

Pearson correlation

.387**

.411**

.417**

.513**

.676**

.764**

.534**

.429**

.433**

.167**

-0.054

0.007

-0.013

-0.033

-0.007

Sig.

0

0

0

0

0

0

0

0

0

0

0.126

0.849

0.703

0.35

0.835

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RX2

Pearson correlation

.407**

.374**

.413**

.478**

.539**

.821**

.413**

.417**

.432**

.213**

-0.057

-0.022

-0.032

-0.022

-0.019

Sig.

0

0

0

0

0

0

0

0

0

0

0.102

0.539

0.356

0.53

0.582

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

RX3

Pearson correlation

.370**

.390**

.401**

.596**

.558**

.840**

.481**

.426**

.450**

.183**

-0.025

-0.019

-0.015

-0.001

0.004

Sig.

0

0

0

0

0

0

0

0

0

0

0.482

0.592

0.663

0.969

0.918

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

JY1

Pearson correlation

.343**

.376**

.384**

.424**

.540**

.441**

.846**

.419**

.401**

.190**

-0.036

0.002

0.031

-0.016

0.016

Sig.

0

0

0

0

0

0

0

0

0

0

0.31

0.954

0.374

0.639

0.652

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

JY2

Pearson correlation

.385**

.411**

.375**

.445**

.526**

.441**

.860**

.388**

.399**

.186**

-0.003

-0.009

0.041

0.03

0.051

Sig.

0

0

0

0

0

0

0

0

0

0

0.934

0.802

0.245

0.399

0.147

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

JY3

Pearson correlation

.411**

.445**

.432**

.511**

.606**

.600**

.813**

.444**

.447**

.231**

-0.068

-0.01

0.028

0.024

0.034

Sig.

0

0

0

0

0

0

0

0

0

0

0.053

0.767

0.425

0.491

0.329

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

JC1

Pearson correlation

.350**

.373**

.375**

.423**

.465**

.464**

.428**

.914**

.530**

.200**

-0.04

-0.022

0.006

-0.021

-0.012

Sig.

0

0

0

0

0

0

0

0

0

0

0.25

0.532

0.867

0.544

0.739

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

JC2

Pearson correlation

.360**

.379**

.397**

.409**

.466**

.480**

.469**

.886**

.650**

.255**

-0.055

-0.032

-0.036

-0.04

0.007

Sig.

0

0

0

0

0

0

0

0

0

0

0.114

0.358

0.305

0.258

0.836

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

JX1

Pearson correlation

.396**

.398**

.382**

.431**

.481**

.474**

.435**

.684**

.761**

.254**

-.073*

-0.032

0.002

-0.026

-0.007

Sig.

0

0

0

0

0

0

0

0

0

0

0.038

0.366

0.946

0.453

0.837

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

JX2

Pearson correlation

.271**

.281**

.272**

.327**

.373**

.371**

.362**

.400**

.757**

.263**

-0.03

-0.011

-0.029

-0.021

-0.05

Sig.

0

0

0

0

0

0

0

0

0

0

0.388

0.756

0.403

0.547

0.153

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

JX3

Pearson correlation

.317**

.279**

.311**

.352**

.371**

.410**

.352**

.424**

.804**

.427**

-0.064

-0.028

-0.014

-0.062

-0.047

Sig.

0

0

0

0

0

0

0

0

0

0

0.067

0.425

0.681

0.076

0.184

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

SP1

Pearson correlation

.320**

.327**

.309**

.373**

.379**

.393**

.358**

.415**

.713**

.604**

-0.062

-0.027

-0.005

-0.043

-0.062

Sig.

0

0

0

0

0

0

0

0

0

0

0.078

0.445

0.883

0.221

0.078

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

SP2

Pearson correlation

.074*

0.032

0.003

0.023

0.013

0.015

0.039

0.019

0.012

.600**

.256**

.161**

.254**

.255**

.254**

Sig.

0.035

0.357

0.936

0.515

0.713

0.669

0.265

0.592

0.725

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

SP3

Pearson correlation

0.016

-0.017

-0.007

0.006

-.084*

-0.037

-0.002

-0.031

-.075*

.613**

.390**

.195**

.253**

.303**

.259**

Sig.

0.65

0.633

0.835

0.854

0.016

0.288

0.948

0.375

0.032

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

SQ1

Pearson correlation

-0.011

-0.044

-0.009

-0.043

-.081*

-0.046

-0.059

-0.039

-.076*

.309**

.813**

.214**

.277**

.296**

.251**

Sig.

0.751

0.212

0.807

0.22

0.021

0.193

0.093

0.271

0.031

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

SQ2

Pearson correlation

-0.053

-.074*

-0.055

-.076*

-0.062

-0.045

-0.011

-0.047

-0.044

.181**

.826**

.228**

.270**

.229**

.250**

Sig.

0.132

0.034

0.114

0.029

0.076

0.196

0.761

0.176

0.209

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

SY1

Pearson correlation

0.006

-0.008

0.011

-0.015

0.004

0.014

0.017

-0.023

-0.041

0.068

.143**

.687**

0.052

.127**

.119**

Sig.

0.865

0.822

0.747

0.662

0.902

0.686

0.626

0.517

0.24

0.053

0

0

0.14

0

0.001

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

SY2

Pearson correlation

-0.028

-0.019

-0.038

-0.002

-0.024

-0.013

-0.008

-0.034

-0.019

.143**

.246**

.797**

.207**

.154**

.162**

Sig.

0.426

0.596

0.284

0.963

0.489

0.72

0.828

0.339

0.588

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

SY3

Pearson correlation

-0.012

-0.049

-0.012

-0.037

-0.044

-0.028

-0.019

-0.013

-0.017

.162**

.225**

.835**

.179**

.167**

.202**

Sig.

0.734

0.163

0.74

0.293

0.212

0.429

0.593

0.711

0.628

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

NC1

Pearson correlation

0

0.011

-0.043

0.03

0.013

-0.033

0.025

-0.006

-0.028

.197**

.256**

.134**

.771**

.325**

.292**

Sig.

0.992

0.748

0.216

0.394

0.703

0.35

0.473

0.866

0.432

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

NC2

Pearson correlation

0.042

-0.01

0.006

0.04

0.017

-0.007

0.065

-0.019

0.007

.203**

.225**

.142**

.734**

.430**

.353**

Sig.

0.234

0.786

0.87

0.25

0.622

0.839

0.062

0.583

0.832

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

NC3

Pearson correlation

0.035

0.01

0.036

0.01

0.009

-0.01

0.001

-0.011

-0.012

.172**

.264**

.180**

.724**

.377**

.328**

Sig.

0.32

0.785

0.307

0.774

0.808

0.786

0.984

0.745

0.742

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

NS1

Pearson correlation

0.05

0.038

0.023

-0.025

0.009

-0.018

0.037

-0.012

-0.02

.232**

.277**

.179**

.449**

.804**

.488**

Sig.

0.153

0.277

0.516

0.473

0.795

0.605

0.292

0.74

0.571

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

NS2

Pearson correlation

0.047

0.009

0.035

0.018

-0.012

-0.019

-0.012

-0.042

-0.058

.194**

.244**

.138**

.359**

.823**

.456**

Sig.

0.184

0.789

0.323

0.613

0.732

0.592

0.729

0.236

0.101

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

NY1

Pearson correlation

0.056

0.047

-0.002

-0.027

-0.005

-0.006

0.016

-0.015

-0.042

.172**

.243**

.188**

.352**

.510**

.833**

Sig.

0.109

0.182

0.944

0.435

0.884

0.875

0.654

0.667

0.229

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

NY2

Pearson correlation

0.045

0.031

0.007

0.021

0.048

0.031

0.032

0.027

0.005

.179**

.219**

.170**

.289**

.419**

.635**

Sig.

0.201

0.384

0.845

0.554

0.168

0.379

0.363

0.436

0.891

0

0

0

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

NY3

Pearson correlation

-0.028

0.052

0.025

-0.026

0.003

-0.036

0.046

-0.008

-0.051

.161**

.224**

.118**

.311**

.363**

.745**

Sig.

0.418

0.139

0.474

0.465

0.923

0.31

0.192

0.811

0.149

0

0

0.001

0

0

0

N

814

814

814

814

814

814

814

814

814

814

814

814

814

814

814

 Type your paragraph here.

KEYSTONE UNIVERSAL

​EDUCATION