Internet-based intelligent and sustainable manufacturing: developments and challenges

Abstract

In recent years, the nations of the world have presented the development strategy of manufacturing. Manufacturing is the foundation of a country. However, with the increasing global resource constraints and market heterogeneity, the variety of individual demands, and the long-term goals of sustainable development, with the support of emerging information and communication technologies such as Internet, cyber-physical system, Internet of Things, cloud computing, and big data, industrial value creation is causing a paradigm shift in manufacturing. This paper studies a range of new manufacturing paradigms and presents a state-of-the-art survey of published works. It explores the corresponding current manufacturing concepts, technologies, framework features, application effects, resource optimization, and future challenges in these new paradigms. The integration of various manufacturing paradigms is also analyzed. Through this survey, the developments of these new manufacturing paradigms are explained and the future prospects are also discussed.

1 Introduction

In recent years, the nations of the world have presented the development strategy in manufacturing. In March 2012, the USA has taken the lead in proposing construction of “National Network for Manufacturing Innovation (NNMI)” [1]. In November 2011, the German government has proposed the “Industry 4.0” strategy in the published high-tech strategy 2020 [2]. In May 2015, China’s State Council has promulgated the Made in China 2025 [3], planning a strategy to raise the country’s manufacturing power. Manufacturing is the driving force of economic development and plays a key role in the creation of social value [4, 5]. Manufacturing paradigm has evolved from a single species of craft production, to mass production to multi-variety and small batch of flexible production. With the increasing personalized demands and market complexity, the rapid updating of emerging information and communication technologies (ICTs) such as cyber-physical system (CPS), Internet of Things (IoT), cloud computing, and big data, and the sustainable development of society, all of these market requirements and technology changes drive the innovation of manufacturing paradigm. Manufacturing is heading toward intelligence, real-time, interconnection, servitization, greenization and globalization, and personalization. The emergence of these features leads to the shift in manufacturing paradigm to intelligent manufacturing (IM), cyber-physical production systems (CPPS), industrial Internet of Things (IIoT), cloud manufacturing (CM), sustainable manufacturing (SM), global manufacturing (GM), and mass customization (MC) [6,7,8,9,10].

From the foregoing discussion, we can see that national strategies, social value creations, and technological advances are driving the transformation of the manufacturing paradigm. Moreover, with the consumption of manufacturing resources and the imbalance between manufacturing resources and capacities, manufacturing enterprises are faced with the shortage of or idle resources. Therefore, resource optimization in manufacturing sector plays a vital role in the updating of manufacturing paradigm. For example, manufacturing resource combinatorial optimization is a crucial way to solve resource shortage [11]. It can enable the limited manufacturing resources to be allocated reasonably, and make a variety of manufacturing resources to be utilized fully. Manufacturing enterprises can provide better products or services to the market at lower cost and shorter life cycle by optimizing the use of manufacturing resources. Therefore, a number of optimization algorithms, heuristic algorithms, and intelligent optimization algorithms have been applied in different manufacturing paradigms.

Bonvillian [12] discussed advanced manufacturing policies and paradigms for innovation in detail, emphasizing innovative manufacturing, especially technological innovation which is the predominant factor in economic growth. The ongoing shift in value creation is initiated by the continuous innovation of manufacturing paradigm, while the new technology is the driving force of innovation of the manufacturing paradigm [13]. Manufacturing paradigm innovation is facing new opportunities and challenges because of the rapid development of emerging information and communication technologies such as Internet, cyber-physical system, Internet of Things, cloud computing, and big data. It is more advantageous to solve the problem in the process of manufacturing paradigm innovation by combining the emerging information and communication technology with industrial technology and management technology.

Therefore, this paper makes a review regarding the current manufacturing paradigm innovations and optimization of manufacturing resources in the Internet-based environment. Review of existing studies either analyzes past and future trend in manufacturing from the perspective of the development, or explores a new manufacturing paradigm of a particular technology. Under the background of Internet, literature of modern manufacturing and its future development as a whole is relatively rare, and the combination of advanced manufacturing and resource optimization needs to be constantly updated.

This paper summarizes the current literature on the advanced manufacturing paradigms and optimization methods of resource utilization, from theoretical and practical developments of emerging information and communication technologies and the Internet platform, into six aspects: (1) intelligent manufacturing, (2) cyber-physical production Systems, (3) industrial Internet of Things, (4) cloud manufacturing, (5) sustainable manufacturing, (6) global manufacturing, (7) mass customization. The aim of this paper is to be a reference source for research on the state-of-the-art and resource optimization methods of manufacturing.

To achieve this goal, section 2 of this paper is assigned to clarify the scope of our analysis and to describe the research methodology. The above six aspects are explained in sections 3–9, respectively. At the end of each section, we summarize and discuss the novel manufacturing paradigm of market demand, product characteristics, customer groups, the requirements to enterprises, the concept of manufacturing model, the features and frameworks of manufacturing systems, the integration of advanced information technologies, the optimization methods of resource utilization, and the application and its future challenges of manufacturing paradigms. Finally, section 10 summarizes the whole paper and forecasts the future research direction of manufacturing paradigms and manufacturing resource optimization.

2 Scope and methodology

To clarify advanced manufacturing paradigms and resource optimization methods for manufacturing sector in the internet-based background, this paper presents manufacturing models supported by critical technologies, and discusses the corresponding concepts, market demands, product designs, system development, technology implementation, resource optimization methods, management challenges, application, and development. The previous review focuses mainly on a single manufacturing paradigm. From a holistic perspective, this paper presents new manufacturing paradigms and resource optimization approaches based on the Internet.

In order to have a holistic thinking, we adopt a three-stage qualitative research method (literature search, identification, classification and evaluation) similar to the method used by Gahm [14] in its recent review. In the literature search phase, we search the keywords associated with “manufacturing paradigm,” “resource optimization,” “Internet,” and “emerging information technology” through Elsevier Science Direct, IEEE Xplore, and other core journals. The articles are published in 1995 to February 2020.

In the identification phase, we identify the found articles by reading the title, keywords, and abstract, and then further recognize the enlightening articles and analyze them in full text. To identify relevant papers, several core elements must be considered. First, we focus on advanced manufacturing paradigms including intelligent manufacturing, cyber-physical manufacturing systems, industrial Internet of Things, cloud manufacturing, green manufacturing, sustainable manufacturing, global manufacturing, social manufacturing, and mass customization. Second, we seek the advanced manufacturing technologies that impact the manufacturing paradigms, covering Internet, CPS, IoT, cloud computing, and big data. Third, we look for methods to improve resource efficiency. After the initial identification of articles, we conduct a full-text analysis, and ultimately choose the following advanced manufacturing paradigms: intelligent manufacturing, cyber-physical manufacturing systems, industrial Internet of Things, cloud manufacturing, sustainable manufacturing, global manufacturing, and mass customization. For each identified manufacturing paradigm, we consider range from the demand analysis, product planning, system development, technology implementation, operation management, resource optimization to application and evolution. In this phase, we construct the research framework by identifying it twice. Figure 1 provides an intuitive description of the overall structure of this paper.

Fig. 1

The structure of the article

To enrich the research framework of the previous stages, we have searched and identified the literature related to the corresponding lower level research contents. These newly identified articles are assigned to specified subsections (Table 1). It can be clearly observed that IoT and sustainability in manufacturing industry are a major concern. The annual distributions of the determinate articles are shown in Fig. 2.

Table 1 Reviewed articles and journals

Fig. 2

Paper distribution in international journals and conferences since 1995 (up to February 2020)

3 Intelligent manufacturing

With the fierce global competition, the dynamic requirements of market and the swift development of advanced technologies, manufacturing enterprises not only need to have the capacities of integrating global resources, rapidly responding to the market, but also need to be able to realize resource efficiency. These promote the creation of intelligent manufacturing. Intelligent manufacturing makes personalized products possible and quickly responds to an uncertain environment. Intelligent manufacturing (IM) of “Industry 4.0” is the internet-based manufacturing paradigm which is facing the whole life cycle of products to achieve a generic interconnection. It makes use of next-generation information, artificial intelligence, manufacturing and management technologies to achieve intelligence of product design, manufacture, management and service [15].

3.1 The definition of intelligent manufacturing and intelligent manufacturing system

Davis et al. [16] thought that the dramatically intensified and pervasive applications of networked information-based technologies or “manufacturing intelligence” throughout the manufacturing and supply chain enterprises promoted the emergence of smart manufacturing. Smart manufacturing can achieve real-time understanding, reasoning, planning, prediction, management, and resource sharing and networking of manufacturing systems. It is facilitated by the pervasive use of advanced sensor-based data analytics, computing platforms, communication technology, modeling, and simulation [17]. Dumitrache and Caramihai [18] regarded intelligent manufacturing systems (IMSs) as large pools of human and software agents, with different levels of expertise and local goals, which have to act together. Knowledge management allows agents to communicate or cooperate through the network. The intelligence of IMS is characterized by the self-adaptability, self-awareness, reasoning, and knowledge presentation and processing. The integration of production planning, scheduling, real-time control, and distributed management can be solved through this system.

From existing academic researches on intelligent manufacturing, we can find the biggest difference between this new manufacturing paradigm and traditional manufacturing. Intelligent manufacturing is more intelligent, and realizes more exchanges and cooperation relying on the network. Knowledge rather than data is the object of the intelligent manufacturing process. Thus, intelligent manufacturing is not only the collection of large amounts of data for the production process, but also data analysis and data mining for guiding the production decision-making and application. The knowledge in intelligent manufacturing systems constantly updates through continuous learning, creating new knowledge with existing knowledge.

Zhou et al. [19] proposed that intelligent manufacturing included three paradigms: digital manufacturing, digital-networked manufacturing, and new-generation intelligent manufacturing. Based on the understanding of existing theories and practices, we conclude that intelligent manufacturing consists of intelligent products, intelligent production, and intelligent services. As shown in Fig. 3, the basic framework of intelligent manufacturing and the interrelationship of its constituent elements are described. Intelligent products include intelligent facilities(i.e., sensors, data storage equipment, software, etc.), physical entities (i.e., parts, machinery, etc.), and networking components (i.e., interfaces, wired and wireless network protocols). Intelligent production is based on intelligent manufacturing system as the core, and intelligent factory as the carrier. It forms a complex manufacturing network through vertical integration of internal business and horizontal integration between enterprise value chains, realizing real-time management and optimization of product life cycle. Intelligent services conduct the management, analysis, mining of data generated from intelligent devices, and intelligent production process for intelligent decision making.

Fig. 3

The basic architecture of intelligent manufacturing

3.2 Multi-agent for intelligent manufacturing

The combination of multi-agent theory and production scheduling for intelligent manufacturing is one of the approaches to solve the problem of manufacturing process optimization. Shen et al. [20] investigated the application of agent-based systems in intelligent manufacturing. An “agent-based system” is a loosely coupled network of problem solvers that work together to solve problems that are beyond individual capabilities. Manufacturing scheduling is one of the most difficult problems in all kinds of scheduling problem. The combination of the agent-based approach with manufacturing scheduling method is an important optimization method to increase the productivity and profitability of manufacturing enterprises. In the research of intelligent manufacturing scheduling algorithms, considering the shortcomings of traditional scheduling algorithms in intelligent manufacturing, the framework of intelligent manufacturing system based on multi-agent is proposed, in which a single agent has autonomy, adaptability, and coordination characteristics. The architecture of intelligent manufacturing, the scheduling optimization algorithm, the negotiation processes, and protocols among the agents are described in detail. An example is given to demonstrate the feasibility of the proposed method. It is proved that the implementation of multi-agent technology in IMS makes the production operation much more flexible, economical, and energy efficient [21].

3.3 The integration of intelligent manufacturing and service

Requirements for services constantly increase in social economy, collaborative optimization of products and services has become a concern of enterprises in the stage of product design. Dutra et al. [22] interpreted coalescence of manufacturing and service as a paradigm shift, and provided a new service design framework based on petri nets which made the characteristics of manufacturing and service fuse and changed the design of manufacturing process. In the context that fully integrated enterprises are being replaced by business networks, each participant provides others with specialized services. Giret et al. [23] pointed out that the service-oriented manufacturing system was essential for the enterprise. Implementing the system needs different technologies, standards, functions, protocols, and execution environments. Thus, they presented a framework and associated engineering approach. This method combines the multi-agent system with service-oriented architecture for the development of intelligent automation control and execution of manufacturing systems.

3.4 The challenges of intelligent manufacturing

With the advent of big data era, intelligent manufacturing will face greater challenges. Intelligent manufacturing is required for acquiring, analyzing big data, and using data to make decisions. Tao et al. [24] regarded the realization of data as a valuable commodity, and trusted that industries (manufacturing, energy, transportation, and service) were interested in benefits of big data. Taking intelligent manufacturing as an example, data analysis, data mining and modeling in the production planning, equipment health monitoring, sustainability, and product innovation are discussed. Smart operational control mechanisms not only integrate real-time data from system operations, but also formulate and solve a wide variety of optimal control analysis quickly and efficiently and then translate the results into executable commands. With the token network reference structure, Sprock and McGinnis [25] proposed a conceptual model to express the operational control problem in intelligent manufacturing.

In the investigation of the future trend of intelligent manufacturing, Dumitrache et al. [26] thought intelligent manufacturing rose to maturity since the early 21st century, and predicted the paradigm for the next generation of intelligence manufacturing would originate from Internet-oriented enterprises. The vision and road map for the next generation of sustainable Internet-networked enterprises have a list of challenging qualitative features: social, legal, framework, entrepreneurial, interoperable, customer satisfied, total quality management. Analogously, Jardim-Goncalves et al. [27] presented the challenges and innovations of intelligent manufacturing systems in terms of system frameworks, models, technologies, virtual organisation environments, and servitization. In addition, the emergence of augmented reality (AR) is intensifying the interaction between people and intelligent manufacturing systems. Egger and Masood [28] explored the challenges and directions facing the application of AR to industry.

Distributed autonomous manufacturing system, which is represented by multi-agent manufacturing system and cyber-physical production system, is the main component of intelligent manufacturing system. In particular, the cyber-physical system will be another advanced manufacturing after intelligent manufacturing. They integrate IoT, big data, artificial intelligence, network technology, and virtual manufacturing technology. IoT makes the physical entities in cyber-physical systems have the ability to perceive, collect, and transmit data. Cyber-physical system achieves the integration of the production process and information systems through the seamless connection of physical entities, data, and services. The proposed relationships between IM, CPS, and IoT are shown in Fig. 4.

Fig. 4

The proposed relationships between IM, CPS, and IoT

Therefore, the relationship between intelligent manufacturing, cyber-physical manufacturing, and Internet-of-Things–based manufacturing is interwoven. We can find the core technologies of intelligent manufacturing are comprehensive. It is critical for intelligent manufacturing to realize the integration of Internet, network technology, sensing technology, information and communication technology, manufacturing, and management technology. In this way, the manufacturing industry can accurately capture the demand of diverse markets, quickly respond to changing environment, and achieve intellectualization of product design, manufacturing, management and service, and production decision-making and application. Cyber-physical system technologies which embed IoT, big data, platform technology, virtual information technology, and sensing technology are the key of cyber-physical manufacturing. Cyber-physical systems facilitate the integration of the virtual network and the physical world, and enable the manufacturing industry to not only provide tangible products, but also offer product-related services. Core technologies of Internet-of-Things–based manufacturing are IoT technologies which include sensing technology, network technology, data processing technology, communication technology, artificial intelligence, big data analytics, augmented reality, and virtual reality. IoT technologies enable the interconnection and interaction of smart objects in the physical world by constructing a global network. The following two sections give a detailed insight into cyber-physical manufacturing and Internet-of-Things–based manufacturing.

4 Cyber-physical manufacturing

Integration of manufacturing resources is becoming more and more complicated. Servitization in manufacturing is becoming more and more important on a global scale. Therefore, the purpose of manufacturing industry is not only to provide tangible products, but also to offer product-related services. It requires an open platform to support manufacturing services and to accelerate manufacturing paradigm from traditional goods-dominant paradigm to service-dominant logic. The openness of the internet provides a platform to achieve cyber-physical manufacturing. Lee et al. [29] proposed cyber manufacturing was a transformation, which can translate data from interconnected system into predictive and prescriptive operations to achieve resilient performance. The emergence of IoT, smart analytics, and cyber-physical systems technologies makes raw data into meaningful and actionable operations and enables finding and comprehending the invisible issues. Meanwhile, cyber manufacturing is faced with lack of standards for seamless connectivity and network security and other challenges. Cyber manufacturing requires establishing standards, developing a general platform, so as to achieve rapid decision-making in the big data environment [30]. Cyber-physical systems, as a next-generation intelligent system of intelligence manufacturing, integrate computing, communication, and control, and fully embody the deep integration of the virtual network and the physical world. Therefore, this section does preliminary research on new manufacturing paradigms based on cyber-physical systems.

4.1 Cyber-physical systems and cyber-physical production systems

CPS initiated in 2006 is a collaborative system that collaborates with computational entities which are in intensive connection with the surrounding physical world and its ongoing processes of providing and using, and data-accessing and data-processing services available on the Internet [31]. CPS integrates computing and physical process [30, 32, 33] and provides real-time services [32]. Besides computation devices, embedded sensors and actuators of a system can monitor and coordinate the operations of physical processes in real time [34]. CPS links the physical world seamlessly with the virtual world of information technology and software by using a variety of available data, digital communication facilities, and services [10]. CPS tightly combines its cyber factor and physical factor in distributed computing or grid environments to provide real-time services. In order to obtain precise analysis and useful services, not only the cyber space (CPU, network, storage systems, etc.) and the physical space (location, migration, etc.) need to be considered but also the social space and mental space.

The implementation of CPS in the manufacturing industry has brought about changes in the manufacturing paradigm and has created cyber-physical production system (CPPS) [33]. CPPS consists of autonomous and cooperative elements and subsystems, relying on the latest and foreseeable further developments of computer science, information and communication technology, and manufacturing science and technology, connecting the stages of the production process [31], which are the integration of physical systems and information systems. It achieves the vertical integration and optimization of intelligent production system and the horizontal integration and optimization of the value chain through the collaborative interaction of IoT and internet.

4.2 The hierarchical structure of CPS and CPPS

CPS needs to ensure real-time data acquisition from the physical world and information feedback from the cyber space, and to realize intelligent data management in the initial stage of development. Rahatulain and Onori [35] identified views and viewpoints for describing architecture for cyber-physical manufacturing systems and presented a mapping.

In the driving of recent advances in manufacturing industry, Monostori et al. [31] introduced that CPS maturity model has five levels, namely setting basics, creating transparency, increasing understanding, improving decision making, and, finally, self-optimizing (Fig. 5). Lee et al. [36] proposed 5-level CPS structure as a guideline for implementation of CPS. 5C architecture in CPS consists of smart connection, data-to-information conversion, cognition, cyber, and configuration. The structure above illustrates how the initial data acquisition through analytics transformation to the final value creation by utilizing advanced information analytics and networked machines, making manufacturing perform efficiently, collaboratively, and resiliently.

Fig. 5

The five levels of CPS maturity model [31]

Complete CPS needs cross-domain collaboration, establishment of interoperability standards, flexible, cooperative, and interactive operation. Thus, an open, link-up, adaptable, autonomous CPS is formed [10, 33]. The most important features of CPS are the abilities of intelligent, connectivity, and responsiveness [31].

4.3 The influence scope of cyber-physical manufacturing

CPS is widely applied in transportation, smart home, robotic surgery, aviation, defense, critical infrastructure, etc. [33]. Industrial manufacturers, who are subjected to increasing cost pressure and market volatility, are very concerned about the enhancement of the abilities of making optimized, proactive decisions, competitiveness, and survivability. Intelligent control of collaborative operations on the production of complex products by developing cyber-physical systems is vital. The development of a virtual environment was presented by Chen et al. [37], which adopted mature modeling frameworks through electronics architecture and software technology-architecture description language based on an effective model-based approach. Its meta-model offers a common data specification and semantic basis for information management across product life cycle, models, and tools, both for resource planning and anomaly treatment. Park et al. [32] designed real-time scheduling algorithms based on effective least slack time first and heuristic-effective least slack time first for CPS. The algorithm improves the performance of conventional real-time scheduling algorithm by up to 30%.

Focusing on social aspects and the transformation of production networks from the linkage of single work systems up to the cooperation of geographically dispersed factories of a single company or cross-company cooperation, the socio-cyber-physical systems are designed and applied [38]. Ball et al. [39] developed information-centric manufacturing cyber-physical systems to reduce traditional engineering semi-sequential process and over-reliance on extremely precise requirements, and mediate risk. This development is applied to the defense industry. From the perspective of value creation, Mikusz [10] proposed that the added value of CPS in the industrial context was manifold. In the industrial environment, research on CPS cannot be simply from the perspective of technology, but also from the perspective of business-oriented and customer value creation. Therefore, conceptualization of the industrial software-product-service system is proposed. CPS in manufacturing environment comprises smart machines, storage systems, and production facilities capable of autonomously exchanging information, triggering actions, and controlling each other independently. The involved manufacturing systems are vertically networked with business processes within factories and enterprises, and horizontally connected to disperse value networks.

The production of high-precision, high-value products is accelerated through US and EU manufacturing. CPS enables future generation of such products to design, manufacture, maintain, and serve in a consistent way over their complete life cycle [40]. Based on 11 case studies, Herterich et al. [30] observe maintenance, repair, and overhaul services (MRO) as the most valuable service for manufacturing. The emerging CPS and digitalization technologies facilitate service innovation in manufacturing. CPS describes the integration of computation and physical processes with embedded software consisted of data recording, data evaluating and saving, and global networks connection and service. These new technological capabilities play an important role in improving the efficiency of MRO, minimizing the downtime in an industrial process. They have impacts on all stakeholder groups in service ecosystem simultaneously. Yu et al. [41] applied CPS to blockchain-based shared manufacturing (BSM) and built resource operation blockchain for BSM framework to facilitate a peer-to-peer–based resource sharing paradigm. Ding et al. [42] constructed a digital twin-based cyber-physical production system to fulfill the interconnection and interoperability of a physical shop floor and corresponding cybershop floor. Next, Park et al. [43] designed a digital twin-based cyber physical production system architectural framework. This solution provides technical support for the performance hurdle in the personalized production of a micro smart factory (MSF) and improves the average makespan. Lu and Xu [44] also proposed a resource virtualization framework to crate digital twins for a smart factory. This framework was validated by a case study.

4.4 Threats and future development of cyber-physical manufacturing

As a wide range infiltration of cyber-attacks, and the increasing cyber-security weaknesses in manufacturing systems, CPS is under threat. The security of CPS is rapidly becoming a global focus. Unlike traditional intellectual property theft, cyber-attacks affect the whole process of product life cycle. These attacks not only undermine normal manufacturing, but also pose a risk to human safety such as operators and consumers. Generally, quality problems can be detected by the quality control system, but the impact of malicious attacks cannot. Hence, specially designed cyber-security tools for manufacturing are essential. Wells et al. [45] illustrated the ease and significant threat of implementing attacks. To defend against cyber-attacks, the first step is to educate engineers and designers. Vincent et al. [46] proposed a novel product/process design approach to enable real-time attack to be detected. The proposed approach aims at detecting changes to a manufactured part’s intrinsic behavior through the use of structural health monitoring techniques.

Industrial Internet facilitates the construction of future intelligent manufacturing plants. Service-oriented architectures (SOA) provide available network services through data resources and computational functions. However, whether SOA can be implemented in the manufacturing CPS is still unclear. CPS incorporating ISA-95 standard can identify manufacturing computational requirements. The acquire, recognize, cluster-service oriented architecture (ARC-SOA) provides a topological view of data flow within a field-level manufacturing SOA [47].

Although CPS in manufacturing is initial stage, with the trends of rapid change of market, the personalized demand of products, reuse of resources, and online service, in the future, CPS in manufacturing will integrate with IoT, cloud computing, and big data, etc. [33]. Babiceanu and Seker [48] constructed manufacturing cyber-physical system on the basis of advances in sensor and communication. This work proposed a framework for the development of predictive manufacturing cyber-physical systems, based on cloud manufacturing solutions and big data applications. This system had the functions attaching to the IoT, processing complex events, and designing big data algorithmic analytics.

5 The manufacturing paradigm driven by the Internet of Things

The IoT is an important domain of emerging information and communication technology, and the most valuable technology of CPS [49]. IoT plays a major role in the effective utilization of resources and economic value creation. From the initial concept to the final application of IoT, this process encompasses understanding the basic definition, framework, main characteristics, developing key technologies, dealing with the challenges of technology, security, and application.

5.1 The concept of IoT and IIoT

The term IoT was coined by Kevin Ashton, one of the founders of the original Auto-ID Center at MIT, in 1999. The basic meaning of IoT is “everything can be interconnected through the network” [50]. After that, members of the same team extended the concept to define it as “an intelligent infrastructure for linking objects, information and people through the computer networks, and where the Radio-Frequency Identification (RFID) technology found the basis for its realization” [49]. Until 2005, at the World Summit on the Information Society (WSIS) in Tunis, the formal concept of IoT was proposed by the International Telecommunication Union (ITU). ITU explained that IoT will realize people and things’ interaction in any time and any place with embedding short-distance mobile transceivers into parts and daily objects [51].

With the development of technologies in the IoT sector, the concept of IoT has been continuously renewed. In a broad sense, IoT refers to (1) a global network that interconnect intelligent objects by using extended Internet technologies, (2) a set of technologies that support the realization of interconnection, and (3) the new business and market for all applications and services by leveraging such technologies [36, 52]. A more concise view shows that IoT is envisioned as a network wherein smart objects in the physical world interact with each other by means of the Internet. “Smart objects” are abstraction of people or things connected to IoT. They have perception, communication, and computational powers as a result of being embedded in the sensing technology and communication technology. They can collect information and collaborate to complete specific services through access to IoT at any time and any place [49, 51,52,53,54,55]. From the above perspectives, we can conclude that IoT builds on three pillars: (1) “smart objects,” (2) Internet technologies, and (3) the interaction mechanism between them.

The adoption of IoT in industry introduces the term IIoT. It refers to the integration of various collection sensors or control sensors and actuators, ubiquitous technologies, communication technology, artificial intelligence, big data analytics, augmented reality, virtual reality, and security mechanisms [56, 57]. IIoT has been applied to many aspects such as supply chain management, production process optimization, environmental monitoring and energy management, and safety production management.

5.2 The characteristics of IoT

A comprehensive understanding of the characteristics of IoT will help to better apply it to the manufacturing sector and provide new opportunities for production service.

Scalability::

In the network system of IoT , scalability requires to consider at different levels, including from the product level to understand the real needs of end users, from architectural level, object identification level, networking and communication level, and data level to handle the interconnection of objects, from the application and business model to provide dynamic services.

Heterogeneity::

Heterogeneity is the most momentous feature of IoT, including heterogeneous network and device, which means continuing improvement of interoperability standards between communication networks and device access protocols.

Interoperability::

Collaboration between heterogeneous network requires interoperability, such as interoperability between heterogeneous platforms, interoperability between heterogeneous devices, etc. [58,59,60].

Autonomy::

To minimize human intervention, smart objects in IoT need to extract autonomy, that is able to provide self-organization and self-adaptability.

Security::

The IoT provides a secure environment. In addition to ensuring the safety of network communications, but also keeping the user’s privacy from being stolen.

Openness::

Openness refers to the opening of the entrance of IoT. The opening of the entrance consists of two levels, one is the operating system, the chip, the connection protocol from the foundation level; one is the platform, the application, and the service representative application level [49, 52].

5.3 The framework of IoT

To achieve the communication of IoT between human and things in anywhere, anytime, and anyone, to create a new IoT-based business model, to improve the application value of IoT, and to address challenging technical and secure issues, there is a need to identify the existing architecture.

According to different levels of decision-making, the structures of IoT are classified: the overall system architecture, technology architecture, and standard system framework. The overall system architecture plays the role of strategic-level decision-making, with the direction of guidance function. The technology architecture plays the role of tactical decision-making, which is the former implementation of the process method. The standard system framework plays the role of business-level decision-making. It is the operability decision-making in the previous layer. At present, framework of IoT usually refers to the technology architecture, which includes perception layer for data acquisition at the bottom, the middle of network layer for data transmission, the top application layer for data processing, and the throughout public technology.

In the design of IoT architecture, from the routing process, Fersi [61] proposed a new IoT architecture based on Distributed Hash Table protocols to afford the flexibility demand. Huang [62] proposed a new architecture of two-layer distributed hybrid IIOT and its conversion architecture between different networks to achieve seamless access. In dealing with the open problems related to the self-organization and self-adapting systems of interactive devices, Nascimento and de Lucena [55] proposed the IoT framework based on multi-agent systems and machine learning technologies. In the development of business models with IoT technology, Dijkman et al. [54] proposed a business model framework by using the business model canvas.

5.4 The change of manufacturing service driven by the IoT

The innovative visions and excellent characteristics in IoT provide innumerable opportunities for enterprises and customers. At present, in the adoption of IoT for enterprises, there resides three applications:(1)implementing monitor and control by collecting data, (2) analyzing big data generating from the objects of IoT to provide a reference for business decision-making, and (3) achieving information sharing and collaboration between people and things [52, 63]. They are widely applied in various domains, such as smart city, smart community, smart home, intelligent buildings[64], intelligent manufacturing, supply chain management, retail, health, food safety, agriculture, transportation, earthquake warning, environmental monitoring, energy, culture, and tourism. We can be more intuitive comprehending via several important example applications of IoT. Smart home not only can protect the family and reduce energy consumption, but also can improve life quality by using the innovative monitor and control systems of IoT. Users can enjoy personalized healthcare services by utilizing the object of IoT to capture personal health data. Ben-Daya et al. [65] described the impact of IoT on supply chain management. Every node in the supply chain can obtain information in real time, understand the status of the supply chain, and escalate the efficiency of cooperation with the information sharing and collaboration in IoT [51, 52, 54, 55, 63, 66, 67].

With expansion of applications in IoT and the updating of IoT technology, IoT will play a key role in the construction of the business ecosystem. Rong et al. [68] developed the 6C framework (context, construct, configuration, cooperation, capability, change) to systematically analyze the IoT-based business ecosystem. They identified three patterns of IoT-based business ecosystem.

5.5 The challenges of IoT and IIoT

There are endless challenges which are required to address in order to gain substantive development. In the industry, the challenges of IIoT will stem from energy efficiency, real-time performance, coexistence and interoperability, security, and privacy [69]. From sensing technology to network technology, data processing, and security, there are many urgent problems that need to be solved existing in different levels.

Sensing technology is used to collect data. Key sensing technologies constructed IoT are radio-frequency identification (RFID), wireless sensor networks (WSNs), near-field communication (NFC), and Bluetooth technology (BT). These technologies are primarily for the identification and sensing of “smart objects” embedded in computing and communication capabilities. Nevertheless, the management of smart objects in IoT environment is difficult as they are numerous, heterogeneous, and dynamic. Hence, effective and feasible solutions need to be proposed. On the one hand, according to different scenarios, the sensing technology will be required the ability to identify smart objects. On the other hand, smart objects will be required the capacity of self-representation [49, 52, 63].

Network technology is used to transmit data. Data transmission depends on wired technologies, wireless technologies, cellular network, and satellite communication technology. Furthermore, network technologies also involve addressing, routing, end-to-end transmission, gateway, and traffic characterization. Heterogeneous communication technology is one of the most critical technologies to ensure the interaction between objects [49, 52, 63]. The fusion of heterogeneous information in the IoT was discussed in [70]. A novel centralized approach which was used to management end-to-end “group key” was discussed in [71].

A large volume of data is generated from IIoT. Data processing technology is required for efficient and real-time analysis of the results to support decision-makers with valuable information. Service-oriented computing and cloud computing have been used to process the enormous data generated by IoT. However, these technologies must integrate with IoT, resulting in high-quality data. The capacity advantage in processing big data of cloud computing technology has brought new breakthrough opportunities to the IoT. Al-Ali [72] pointed out that IoT and cloud computing were key to the affordance of smart city services.

Botta et al. [73] explored a CloudIoT paradigm which integrated with the IoT and cloud computing. Díaz et al. [74] investigated the integration of the IoT and cloud computing. Mital et al. [75] utilized technology acceptance model, theory of reasoned action, and theory of cloud computing behavior, for exploring the adoption of IoT and cloud computing–based systems in India. Karkouch et al. [76] studied about the quality of data collected in the IoT, then they provided the corresponding evaluation of measurement indicators.

In the far-ranging deployment of the IoT, to reduce the obstacles, the security issues in all levels will not be overlooked. The first is the data entry; Caron et al. [53] pointed out collecting data about individual privacy in the IoT was not completely protected. To ensure that the data are not being accessed optionally, the data access permissions and access control must be set. Conti et al. [77] indicated that authentication, authorization, and access control were critical challenges for the source of data, followed by data transmission network security, which is the most vulnerable to the level of network threats. In addition to network hackers and wireless attacks, are other network security risks exist. In this regard, Ashraf and Habaebi [78] exploited an autonomic taxonomy to reduce the threats in the IoT. Raza et al. [79] designed, implemented, and evaluated a novel intrusion detection system (IDS) for the IoT called SVELTE, with the primary target of detecting routing attacks. Saied et al. [80] proposed a new trust management system (TMS) for the IoT. The system has the function of preventing common attacks. The security at the application level of the IoT involves investment risk, business ethics, and personal privacy. Therefore, the solution to these problems from the design stage to the final stage of application should be taken into account [49, 53].

6 Cloud manufacturing

CM is based on the technology of cloud computing and Internet. It optimizes the configuration for distributed manufacturing resources, and provides on-demand cloud services. CM not only promotes manufacturing enterprises win–win situation through the effective integration of distributed manufacturing resources at the small and medium enterprises, but also reduces the waste of manufacturing resources through the open collaboration and a high degree of sharing between manufacturing resources and services. Some points about CM are discussed in this section.

6.1 The concept and framework of CM

Cloud manufacturing, from the perspective of the Internet platform providing cloud services, was mentioned by Li [81] for the first time. Li held that CM was a platform that utilized network and cloud manufacturing services. It organizes online manufacturing resources (manufactured cloud) according to the needs of users and provides end users with various types of manufacturing services on demand. Cloud service provider (CSP), cloud service demander (CSD), and cloud manufacturing services platform (CMSP) form cloud manufacturing systems. The framework of CM includes five levels, that is, physical resource layer, cloud manufacturing virtual resource layer, cloud manufacturing core service layer, application interface layer, and cloud manufacturing application layer. Taking into account the situation on the dynamic change of customer demand, CM was defined by the National Institute of Standards and Technology (NIST) (2011) as a customer-centric manufacturing mode that exploited on-demand access to a shared collection of diversified and distributed manufacturing resources to form temporary, reconfigurable production lines which enhanced efficiency, reduced product life cycle costs, and allowed for optimal resource loading in response to variable-demand customer-generated tasking [82].

In further study, Li pointed that CM had the advantages of high efficiency and low consumption, network, and agility [83]. Zhang et al. [84] explained that CM was a new manufacturing paradigm that combined emerging information and communication technologies such as cloud computing, IoT, service-oriented technology, and high-performance computing to address bottlenecks in manufacturing applications. At the same time, they discussed the three core components of constructing cloud manufacturing, namely cloud manufacturing resources, manufacturing cloud services, and manufacturing clouds. The constructing of a complete cloud manufacturing framework includes the following: (1) The architecture, standards, and norms for supporting the implementation of cloud manufacturing systems; (2) the way of trading, sharing, and interoperability for manufacturing resources in the background of CM; (3) cloud manufacturing standards, protocols, and norms. Considering the flexibility of the system, Thames and Schaefer [85] introduced the notion of software-defined cloud manufacturing (SDCM), and described the basic architecture of SDCM.

6.2 The core technology of CM

From the perspective of cloud computing, Xu [86] proposed that CM was cloud computing applied to manufacturing. In CM, distributed resources are encapsulated into cloud services and managed in a centralized way. Users can use cloud services according to their requirements. Cloud users can request services ranging from product design, manufacturing, testing, management, and all the other stages of a product life cycle. Wang and Xu [87] proposed a service-oriented system called interoperable cloud-based manufacturing system (ICMS) that encapsulated existing and future manufacturing resources by using the virtual function block mechanism and standardized description to provide a cloud-based environment. Yu et al. [34] regarded CM as an imitation of cloud computing service paradigm, offering on-demand manufacturing services from networks (i.e., cloud-enabled). Manufacturing resources and capabilities can be shared through the Internet. Faced with the problem of the need for efficient resource allocation in cloud computing environment, Singh et al. [88] realized that the agility of the existing network architecture was not supported. For the user, in getting the appropriate service according to the requirements, the cost of the respective resource was also optimized. This research proposed a new agent-based automated service composition (AZSC) algorithm. The proposed framework eliminates user’s headache of finding optimal service provider in any situation and ensures efficient service allocation at the data centers. Helo et al. [89] proposed a shared cloud-based manufacturing execution system solution for processing the needs and challenges of the current management of ERP and distributed manufacturing.

6.3 Optimization of CM

In addition, some scholars studied CM on the basis of the optimization method. Chen et al. [90] indicated CM will unlock the tremendous value in the massive amount of data being generated by the manufactories. They built a multi-objective optimization model to help users make a flexible decision by using the QoS-aware Web service composition (QWSC) method. Chen and Wang [91] considered the CM concept of simulating a factory online by using Web services, used k-means to classify the simulation tasks, and proposed a classifying artificial neural network (ANN) ensemble approach for estimating the required time for a simulation task. In addition to the proposed methodology, six statistical and soft computing methods were applied in real tasks compared with the six existing methods. The experimental results showed that the proposed methodology reduced the estimation time considerably. Luo et al. [92] built a multidimensional information model and proposed a description method for manufacturing capability in cloud manufacturing systems to share manufacturing capability. They verified the proposed method by applying it to a real case. Chen et al. [93] observed that manufacturing resources can be viewed as manufacturing services accessible to clients in CM. Then, they proposed dynamic programming optimization algorithms for job scheduling and adopted a cooperative game to achieve cost savings for clients.

6.4 The future strategy of CM

How would be cloud manufacturing development in the future? Suo and Gao [94] proposed a strategy of building a cloud manufacturing service platform. By using the method of CloudAnalyst, it not only makes the best allocation of resources, but also joins the cost function which is used to determine the best hardware building program of cloud manufacturing service platforms. Li et al. [95] proposed a distributed peer-to-peer network architecture for cloud manufacturing. The security and scalability of the network can be improved by introducing blockchain technology. The proposed methodology was explained based on a case study. Wu et al. [96] explained that cloud manufacturing was a service-oriented, customer-centric, demand-driven manufacturing model. They explored the current state and possible future of CM. They also presented commercial implementations of CM. They claimed that CM would develop in the areas of high-speed, long-distance industrial control systems, flexibility enablement, business models, cloud computing applications in manufacturing, and prominent implementation architectures by comparing the strategic vision and the current state. Adamson et al. [97] presented future trends and directions of cloud manufacturing as well.

7 Sustainable manufacturing based on information and communication technology

Despite a variety of definitions related to sustainable manufacturing, sustainable manufacturing aims to minimize the waste of resources and energy, and the negative effect of the production process on the environment [98, 99]. It fully displays the sustainable idea of industrial value creation and pays more attention to the green of the product life cycle. Deep understanding of sustainable manufacturing and green manufacturing and their mutual links is conducive to improving the use of resources and energy in the production process and reducing the harm to the environment. Ubiquitous information and communication technologies provide immense opportunities for the realization of sustainable manufacturing. Industry 4.0 and sustainable manufacturing complement each other. Machado et al. [100] explored the interrelationship between Industry 4.0 and sustainable manufacturing.

7.1 Covering fields in sustainable manufacturing

In the process of globalization, there is a growing human awareness about sustainable development around the world in the face of increasing demand for products and services, as well as the challenge of declining resources. In addition to the need to reduce the waste of resources and minimize the harm caused by industry, the objectives of sustainable development include the reduction of production costs simultaneously. Thus, industrial value creation is embodied not only in products and services, but also in sustainability. Sustainable industrial value creation encompasses three main sustainable dimensions: social, economic, and environmental (Fig. 6). Wang et al. [101] examined the relation among three sustainable dimensions by establishing a sustainable composite manufacturing practice and a performance framework.

Fig. 6

Three main sustainability dimensions in industrial value [101]

High-level plan around the world is in progress and is bringing about these new manufacturing paradigm, for example, the US “Advanced Manufacturing National Program (AMNP),” the German “Industrie 4.0” plan, the French plan “la nouvelle France industrielle,” and the UK foresight publications on the “Future of Manufacturing.” These changes in manufacturing provide immense opportunities for realizing sustainable manufacturing. From the macroperspective, new business models will have competitiveness in the long run, and provide new services by the use of smart data. The cross-linking of value creation networks offers new opportunities for realizing closed-loop product life cycle and industrial symbiosis. From the microperspective, new manufacturing paradigms have created sustainable manufacturing opportunities for equipment modification, value creation for the organization, organizational change, sustainable design processes, and sustainable product design. In the global sustainable development, researchers put forward sustainable manufacturing technologies and methods under different situations [102].

7.2 Sustainable manufacturing community

In the sustainable value creation of Industry 4.0, sustainable manufacturing will be realized by using the ubiquitous ICT infrastructure [102]. Similarly for the application of ICT, Severengiz et al. [103] presented sustainable manufacturing community (SMC) and elements of an architecture for an SMC. The SMC is a Web-based non-profit platform setting out to collect knowledge conducive to sustainable manufacturing. Community members can formulate their knowledge on the platform as long as they fulfill the structural requirements of a value creation module. Based on the discussion of the expected impact of “Global Drivers (such as population demographics, food security, energy security, community security, etc.),” a large input from the manufacturing industry into all the Global Drivers is required in sustainable development. Humans will continue to play a vital role in the new manufacturing paradigms in Industry 4.0. Siemieniuch et al. [13] discussed a number of these issues by using system ergonomics professional knowledge.

7.3 The practice of sustainable manufacturing

In an emerging market with a special focus on sustainability, production will be faced with actual challenges. Distributed manufacturing systems (DMSs) are regarded as a promising approach for sustainable manufacturing, largely consisting of the use of decentralized, adaptable, and flexible mini-factories. Such production units, organized in networks, allow production on-demand, therefore reducing transport and emissions [104]. Roberts and Ball [105] collated the knowledge that relates to examples of sustainable manufacturing practice. Based on the creation of a conceptual model, informing the identification of the classification domain, a library was used to promote the realization of sustainability goals for manufacturers. Zhang et al. [106] discussed the sustainability of manufacturing processes in the design phase. A process-oriented information model (PIM) was proposed to integrate the information regarding the sustainable manufacturing product design. Besides considering the performance of service-oriented, sustainability in manufacturing scheduling also should be considered. Mansouri et al. [107] studied energy consumption in shop floor scheduling. They developed a mixed-integer linear multi-objective optimization model to find the Pareto frontier comprised of makespan and total energy consumption. Xu et al. [108] proposed a multi-objective joint model of energy consumption and production efficiency for improving the sustainability of manufacturing equipment services in the job shop. Multi-objective dynamic optimized scheduling of manufacturing service was driven by multi-condition of manufacturing equipment services monitored in real time. Nujoom et al. [109] developed a multi-objective mathematical model in designing the sustainable manufacturing system for minimizing total economic and environmental costs. They used the ε-constraint approach and the LP-metrics approach to obtain the optimal solution through LINGO. The proposed method was tested by a real case data.

7.4 Green manufacturing

Green manufacturing is a sustainable manufacturing paradigm that takes into account the environmental sustainability dimension of manufacturing activities, which applies to a variety of green technologies, such as sustainable green operations, green supply chain management, and remanufacturing. Manufacturing activities of green manufacturing increase the sustainable use of resources, shorten the product life cycle, and reduce production costs [110]. Trentesaux and Giret [111] proposed the generic concept of go-green manufacturing holon to support this major societal and environmental challenge. Zhu and He [112] discussed the green product design issues based on game theory in supply chains under competition, and analyzed how supply chains’ decisions on the “greenness” of products are affected by factors such as supply chain structures, the green product types, and the types of competition. Remanufacturing is one of typical green manufacturing; Xu [113] explained that remanufacturing engineering needed to use core technologies such as information technology, nanotechnology, and biotechnology to save resources and energy, and to protect the environment. Oliveira et al. [114] emphasized the unified and effective management of the recovery from product design to end-of-use, product development, and production. They analyzed the key factors that affect the green product.

7.5 The optimization of green manufacturing

The optimization models and methods, and the optimization of resource and energy for green manufacturing were presented by scholars. In designing the optimal model of the manufacturing system, taking into account the environment aspect of the selection of manufacturing processes (technologies) and input items(components), combining with environmental legislations, Nouira et al. [115] developed two optimization models for manufacturing systems. The greenness of the product in these models was the decisive factor of product demand and price. Concerning the environment, manufacturers are more concerned about reducing energy consumption and environmental pollution. Li et al. [116] considered scheduling problem in green manufacturing enterprises. Luo et al. [117] studied the energy consumption in the production of enterprises, proposed a new meta-heuristic ant colony optimization, comprehensively considering production efficiency and electric power cost, and solving multiple objectives of the production process. Luo et al. [118] examined the role of co-opetition in low carbon manufacturing. Under the cap-and-trade policy, through investigating the pricing and emissions reduction policies for two rival manufacturers with different emission reduction efficiencies, they found higher emission reduction efficiency leads to lower optimal unit carbon emissions and higher profit in both the pure competition and co-opetition models. Simultaneously, compared with pure competition, co-opetition will lead to more profit and less total carbon emissions.

From the perspective of green supply chain management, Ameknassi et al. [119] observed the integration of logistics outsourcing decisions in green supply chain design, with the aim of minimizing both the expected logistics cost and the greenhouse gas, in the context of an uncertain business environment. They proposed a programming model to solve it and obtained the optimal level of logistics outsourcing integration within a decarbonized supply chain, before any further low-carbon investment. Nurjanni et al. [120] proposed a new green supply chain design (GSC) approach for considering environmental and financial issues. This approach incorporates a closed-loop network and a multi-objective optimization mathematical model.

Considering the remanufacturing production planning and inventory control management issues, Alinovi et al. [121] formulated an inventory control model in reverse logistics based on a total cost minimization problem, and getting the optimal return policy which promotes the flow of used items of the remanufacturing. Mahadevan et al. [122] designed three heuristics based on traditional inventory models to determine the production strategy of new product or remanufacturing products. In the case of differences in quality of recycled products, Cai et al. [123] made an acquisition pricing policy and production planning for a hybrid manufacturing/remanufacturing system. Polotski et al. [124] studied the optimal production and setup policies to reduce the production and inventory costs of manufacturing and remanufacturing.

7.6 The barriers of green manufacturing

Renewable resources are gaining attention in green manufacturing. Carbon-based materials from renewable and biodegradable resources will meet future production requirements and promote the competitiveness of advanced manufacturing systems [125]. Thus, the utilization of new materials has become increasingly important and urgent. The unsustainable way that manufacturing enterprises deplete energy and natural resources, and the emission of large quantities of greenhouse gases have led to many economical, environmental, and social problems. Although a growing number of organizations have begun to practice green manufacturing in different ways, there are many barriers in implementing green manufacturing. In order to mitigate these barriers, the prioritization of barriers is essential. Mittal and Sangwan [126] developed a fuzzy TOPSIS multi-criteria decision model to prioritize these barriers from environmental, social, and economic perspectives. The study concluded that lack of awareness/information, technological risk, and weak legislation were the three most important barriers to green manufacturing.

8 Internet-based global manufacturing

Manufacturing paradigm has evolved from mass production, lean production, to mass customization, as manufacturing is heading toward socialization, personalization, servitization, and mass collaboration. Internet platform promotes the global participation in manufacturing. Based on the study of literature, the manufacturing paradigm with the interaction characteristics between social organization and individual or organization and organization in manufacturing activities can be summed up as social manufacturing and global manufacturing. Figure 7 depicts the relationship between global manufacturing and social manufacturing.

Fig. 7

The relationship between global manufacturing and social manufacturing

Social manufacturing is a manufacturing mode which interacts between the organization and the individual about manufacturing activities in the social manufacturing network. Global manufacturing is an extension of social manufacturing, which spans the globe. It can complete an integral global manufacturing network by interacting among the organizations.

8.1 Social manufacturing

Jiang et al. [127] provided a social manufacturing (SocialM) which was a new cyber-physical-social-connected and service-oriented manufacturing paradigm. Organizational logic of SocialM includes four main steps: self-organizing, configuration, operation and collaboration, analysis and relationship management. The core of SocialM is addressed from the configuration operation and management perspectives. This new manufacturing mode would promote socialized resources configuration, social interaction, business collaboration, and all-around production management to accomplish product life cycle efficiently and flexibly. Social manufacturing paradigm supports cross-enterprise manufacturing demand-capability matchmaking through the widespread deployment of ICTs. The large amount of unstructured text data produced in cross-enterprise social interaction media forms the social interaction context of manufacturing services which contains massive manufacturing relationships. In order to enable enterprises to share knowledge in a social interaction context, Leng and Jiang [128] used deep learning approach to extract manufacturing relationships with the context and provided these relationships for decision makers to make decisions. To evaluate the enterprise service matching strategies in a complex social manufacturing environment, Xue et al. [129] constructed a computational experiment–based evaluation framework and applied the method to a real case for validation.

One form of value creation in social manufacturing is open production. Compared with classical factors of production, the paradigm of value creation is shifting, and the value of knowledge and information in the manufacturing process is increasing. ICTs and manufacturing technologies support cooperation among stakeholders, and enable spatially distributed individuals to participate in the process of value creation. The concept of open production enables companies to present their value creation processes, structure, and artifact [130]. Mourtzis et al. [131] considered that employees within companies can collaboratively solve product and production issues through knowledge sharing of social networking. Hence, they developed a knowledge-based mobile app to describe problem-solving instructions and expert opinions. The utilized cloud infrastructure provided a business-oriented social network for employees to facilitate knowledge sharing. Hirscher et al. [132] believed that end users are value creators in social manufacturing. The end user is allowed to participate in the design and manufacturing processes. They studied the new value system and range of business models in the fashion industry that this model has spawned. Rebensdorf et al. [133] considered open community manufacturing (OCM) as an independent, non-commercial organization. OCM aims to use the strategy of open design connecting specialists and volunteers to create sustainable developments via social innovation.

8.2 Global manufacturing

Global manufacturing is an alternative solution for manufacturing enterprises to reduce cost, increase revenue, and improve reliability, whereas it is challenged by the rapidly changing global market environment and complex business environment. Taking into account global manufacturing costs and lead times, Kristianto and Gunasekaran [134] developed a dynamic and multi-domain nonlinear optimization model for global manufacturing. The model integrated product design, manufacturing, and logistics, which complemented pre-contribution related to integrated multi-echelon supply chain design with inventories under endogenous and exogenous uncertainties. In order to handle large-scale models, the solution combined branch and reduce technique and interior point method. The application for a three-echelon forward-reverse global manufacturing network showed that the proposed algorithm can effectively handle large-scale and non-convex problem formulation. With the key objectives of responsiveness, robustness, and resilience (the “Triple R”) in global manufacturing, Kristianto et al. [135] provided a single conceptual model of Triple R by a literature survey, and the method and program for building 3R in global manufacturing, integrating planning, scheduling, real-time optimization, and control.

Rapidly changing global market environment forces enterprises to adapt to the global manufacturing network. Production master data of the configuration of products, processes, and resources in the global manufacturing network is the basis for design and decision-making of the entire manufacturing network. The data is processed by big data. In the design of the manufacturing network, big data enables integrated manufacturing system design, increasing validity of decisions, high performance, and new analysis options for unexploited potential in the network. Big data can also promote the decision-making process-related issues, analysis, and interaction of the entire manufacturing system [136]. The design and decision-making of global manufacturing networks is becoming more difficult by the dynamics and uncertainties of the business environment. Lanza and Moser [137] proposed a dynamic multi-objective optimization model for global manufacturing networks, which evaluated the impact of business environmental factors and optimized production address allocation and capacity planning in the global design of the manufacturing network. Fuchs [8] investigated the constraints in the global locus of manufacturing. The classical economy showed that the global production efficiency would exceed the loss through the change of manufacturing enterprises. In less constrained cases, global manufacturing could increase opportunities and incentives to innovate.

9 Mass customization based on Internet platform

As the growth of customer personalization demands, the trend that the market provides customers with diversified, customized products is becoming prevalent. In this case, the emergence of mass customization is one of the enterprises’ strong competitive factors. The development of the Internet has created a platform for this paradigm.

9.1 The concept of mass customization

Mass customization (MC) keeps the stability of the process and increases the variety of products simultaneously, compared with the mass production paradigm (MP) with high process stability (i.e., no changes, no modifications are needed during order processing) and a low product variety, as well as one-of-a-kind production (OKP) with high product variety and low process stability (product specifications as well as process specifications change from order to order) [138]. Hart [139] asserted that MC produced varied and personalized customized products and services by the use of flexible processes and organizational structures at the low cost of a standard. MC is also seen as the capability to produce customized products for a mass market [140]. In response to the diversity of customer needs and the requirements of shortening customer waiting time, enterprises provide highly customized products with a short delivery time. MC with high flexibility to design for manufacturing enables the production of multiple types of products with short manufacturing delivery time [141].

9.2 Internet platform for MC

With the fierce competition in the globalized market, the shortened product life cycle, the increased complexity of product service systems, and the management challenges of cost, delivery, and quality of customized products [138, 142, 143], it is imperative to establish a sound manufacturing platform, taking advantage of effective production information and data to support decision makers and to improve customer satisfaction.

9.2.1 Manufacturing networks of MC

Efficient manufacturing network configurations are affected by the volatility of globalized heterogeneous markets in mass customization paradigm. Mourtzis et al. [144] provided a multi-objective decision-making and simulation approach, and multiple conflicting criteria to improve the design and operation of manufacturing networks. Mourtzis and Doukas [143] also discussed the challenges of design and plan for manufacturing networks in a mass customization and personalization landscape. Wang et al. [145] designed a needs-based configurator mechanism to help customers obtain satisfactory product configurations. The proposed method applies a hierarchical attention network to extract the information of customer preferences and needs and matches it with the product attribute specifications.

9.2.2 Product platform of MC

Mourtzis et al. [146] exploited an innovative decentralized manufacturing approach to encourage customers to participate in the design of unique products, and to enable the original equipment manufacturers to efficiently plan manufacturing and transportation activities of highly customized products. The suggested method, which is implemented based on a Web-based platform, consists of (a) the “User Adaptive Design System” (UADS), (b) the “Decentralised Manufacturing Platform” (DEMAP), and (c) the “Environmental Assessment Module” (EAM). To guarantee the validity of the development and maintenance of product platform derived from mass customization, it is necessary to manage complexity and variability models to detect essential, dispensable, and highly incompatible features. Considering the limitation of existing approaches, Heradio et al. [147] provided efficient algorithms to retrieve that information from variability models.

9.2.3 Information and data of MC

The reference models of information systems in existing production and supply chain management cannot meet the requirements of a mass customization driven by dynamic demand. A dynamic reference information model was developed to meet these demands by enhancing information and communications technology of mass customization [148]. Ng et al. [149] developed an analytical framework for the selection of mass customization strategies by product supplier/producer in the supply chain. Based on the analysis of consumption data that comes from Internet Connected Objects (ICO) and IoT, mechanisms of a product supplier/producer’s choice between “tailoring strategy” and “platform strategy” were explained.

9.3 The innovation and application of MC

Mass customization has been applied to many industries because it can provide a wide variety of products and services to meet the specific requirements of particular customers.

The successful implementation of mass customization calls for the identification of the critical factors. Silveira et al. [150] emphasized the following key success factors for mass customization: the existence of customer demand for variety and customization, the appropriate market conditions, the integral value chain; the available manufacturing and information technology, products should be customizable, knowledge sharing in the value chain. Innovation and standardization were considered to be indispensable for mass customization, and were confirmed that they have a positive impact on mass customization capabilities through empirical research [140].

Mass customization is one of the most innovative outcomes of the Internet retailing revolution. The enormous information processing capabilities offer a information foundation for the MC. The establishment of mass customization for internet retailing (MCIR) model was adopted to analyze the processes of Internet retailing key actors (i.e., suppliers, retailers, consumers, and new types of intermediaries) and decision-making influencing factors. A theoretical model of the technical development and market deployment of an innovative Internet retailing application (the “ACTIVE” application) was developed to support information management capabilities through the corresponding interactions implemented among all the participating actors [151]. Yoo and Park [152] explored the emotional factors of consumers in an online mass customization. Jost and Süsser [153] investigated the effects of company–customer interaction in mass customization on manufacturer’ profits and consumer surplus. They found that it is always optimal for the manufacturer to offer customization and consumer surplus is maximized because of the degree of customization chosen by the manufacturer.

The provision of customized service is one of the strong competitive factors in the service industry. The Internet provides a platform for customized services to interact and collaborate closely with customers. The combination of MC and the Internet can provide affordable customization services for the requirements of customers. The results of an exploratory study of 35 IT and marketing managers of banks and insurance companies in the UK showed that MC would become a major competitive strategy in the financial sector, with the Internet as an enabling technology to support MC services [154]. In modern society and Internet environment, service is increasingly concerned by manufacturers. Customers’ specific requirements are more and more desired to be understood. Personalized production meets the requirements of individual customers through customer direct participation in product design [143]. Personalized production pays more attention to the realization of customer value, reflecting the product design of personalization, small batch, and flexibility. Based on the integration of technologies in Industry 4.0, Wang et al. [145] proposed a framework for mass personalization production and explained it through industrial practices and a lab demonstration.

Off-the-shelf products do not meet the diverse requirements of customers. Mass customization is able to provide customers a variety of products and services. Therefore, many industries have begun to adopt this mode. For instance, MC of automobiles has become an increasing concern for automotive manufacturers, a process chain for the production of shoe for mass customized shoes has been developed, flight catering has adopted most elements of MC, and MC of individualized orthotics also has been developed [155,156,157,158]. With the penetration of the green concept, Trentin et al. [159] investigated the interconnectedness of MC and green management on the level of their enabling capabilities.

10 Conclusions and future prospects

This paper discussed manufacturing paradigm revolution caused by deep integration of emerging information and communication technologies and manufacturing technologies. Based on these vital technologies, some new manufacturing paradigms over the past decade are emerging and developing such as intelligent manufacturing, cyber-physical manufacturing systems, industrial Internet of Things, cloud manufacturing, sustainable manufacturing, global manufacturing, and mass customization. The focus was on the concepts, supporting technologies, system frameworks, resource optimization methods, application and influence, challenges, and obstacles as well as future development of corresponding manufacturing paradigm, and integration of different manufacturing modes.

Manufacturing presents new features from various aspects: intelligent manufacturing process by reason of Internet technology, interconnection of manufacturing network by using IoT, manufacturing resource servitization on the basis of cloud platform resource sharing pool, sustainability of manufacturing behavior because of sustainable manufacturing, manufacturing capacity decentralized due to global manufacturing, and differentiation in the manufacturing way of mass customization. Future prospects in the context of manufacturing are summarized as follows.

The development of intelligent manufacturing brings smart products and intelligent production ways. In the global and dynamic market environment, the overall goal of intelligent manufacturing is to respond quickly to market changes, to optimize product design and manufacturing process, to reduce energy consumption, and to achieve recycling. Manufacturers who adopt an intelligent manufacturing strategy have shifted from the focus on the provision of products to the provision of services, which has put forward higher requirements for the resources and business integration capabilities of enterprises. Intelligent manufacturing enterprises use intelligent manufacturing systems to solve the issues of production planning, scheduling, control, and decision-making. The intelligent manufacturing system presents synergistic, real-time, dynamic reconfigurable, and super-flexible features. It also has the ability of self-organization, self-learning, and self-adaptiveness. The focus of future research will be on improving the flexibility, agility, and reconfiguration of manufacturing process through advanced intelligent manufacturing systems and intelligent production methods based on the Internet and other technologies.

CPS is the essential format of intelligent manufacturing, which makes the virtual cyber and physical manufacturing fully integrated. On the one hand, the big data generated by cyber-physical production systems can help realize the customization and personalization of all aspects of the manufacturing process (i.e., design, planning, production, and distribution), and real-time monitoring and maintaining manufacturing activities through the record of all aspects of the manufacturing process. On the other hand, the objects, data, and services in the manufacturing activities are used as the carrier of CPS, which expands the application value of CPS. CPPS combines wireless sensor network technology, embedded technology, intelligent analysis technology, advanced communication technology, and IoT. Therefore, CPPS has ample data, fast response capability, and intelligent interconnection. It is an open, autonomous, interactive, flexible, and complex system. It can not only solve the problems of planning, scheduling, control, and coordination of manufacturing, but also integrate production, logistics, and service value chain and solve the problem of product life cycle management. In addition, this system can create value through data collection and analysis in manufacturing activities. Not only in the field of manufacturing but also healthcare, smart city, intelligent transportation, smart home, financial service, aviation, and defense is CPS widely adopted. CPS will gain more and more attention and be widely used in the industry. Some challenges of management, such as seamless connection standard management, big data management, and security management, should be considered.

The IoT, which combines advanced sensor technology, communication technology, Bluetooth technology, wired technology, and wireless technology, and complements each other with the Internet, cloud computing, and big data, facilitates the interconnection of physical and virtual space by the widespread deployment of distributed devices embedded with computing, identification, communication, and sensing capabilities. Based on IIoT, we can obtain higher quality optimization services in customer relationship management, warehousing management, production management, and supply chain management. However, the issues of how to manage data and security, how to protect customer privacy, and how to connect IoT need be a concern of both entrepreneurs and researchers. In regard to technology, hardware, intelligent equipment, to computing, communication and identification technology, data management, and distributed intelligent technology need to constantly improve. In the aspect of security, privacy security and network security are important issues of concern. On the economic front, the IoT creates economic benefits, but at the same time must take into account the equipment deployment costs. Only once these technical problems and security issues are resolved can the IoT better serve the manufacturing sector and other fields.

The awareness that manufacturing is not only provide physical products, but also virtual services has been recognized by more and more people. Therefore, the sharing of manufacturing resources will be a significant way for enterprises to provide on-demand services. Through virtualization of the physical manufacturing resources, CM shares the virtual resources as a service in the cloud service platform, online trading use rights, and offline conducting production. It fully reflects the flexibility and scalability of the manufacturing system. It is more general in a distributed manufacturing environment. Cloud computing, as a key technology for CM, provides shared pools (i.e., networks, servers, applications, and storage) of computing resources. Compared with intelligent manufacturing, cloud manufacturing requires greater processing of big data. In addition, the IoT, cloud terminal technology, and cloud manufacturing trust technology (trust access, publication technology, trust Web technology, and trust operation technology) ensure the maturity and security of the cloud manufacturing process.

Some issues must be considered in the cloud manufacturing operation management process: (1) cloud manufacturing resource service definition encapsulation, virtualization, publishing process management, real-time dynamic automatic searching, discovery, and dynamic matching mechanism; (2) service delivery and spread way to cloud manufacturing service provider, and the management, authorization mechanism for cloud users (including cloud provider and cloud requestor); (3) the dynamic construction, deployment, and decomposition of cloud manufacturing task, the coordination scheduling optimal configuration for source and service; (4) the comprehensive management of access, publishing, organization and aggregation, management, and scheduling for the cloud service of cloud manufacturing service provider, the efficient and dynamic methods of establishment, aggregation, and storage. With the development of shared economy, the integration of manufacturing resources will be a new form of corporate value creation. Enterprises that adopt cloud manufacturing to provide better manufacturing services and gain higher business benefits, need to consider the following: (1) Cloud manufacturing resource cost and profit distribution and business cooperation mode. (2) The cost composition, pricing, bargaining, and operational strategies of cloud service and the corresponding means of electronic payment and security management. (3) The dynamic construction, management, and execution of business process in cloud manufacturing mode. (4) The credit management mechanism for each party (i.e., cloud manufacturing service providers, cloud service demander, cloud manufacturing system brokers). (5) Operational cost optimization of cloud manufacturing service platform. Figure 8 depicts the overall vision of CM.

Fig. 8

The overall vision of cloud manufacturing

To confront the issues of environmental pressures, resource consumption in manufacturing, especially for developing countries, manufacturing enterprises have to consider effective use, recycling, regeneration of resources, and reduction of the harm to the environment. Green manufacturing, which is under the premise of ensuring the quality, function, and cost of the product at first, considers the environmental effect and resource efficiency that traditional manufacturing does not emphasize. Green ideas of green manufacturing refer to the whole life cycle of product design, production, logistics, use, recycling, demolition, and reuse through green manufacturing. For promoting the successful practice of sustainable manufacturing, especially green manufacturing, researchers have had to integrate advanced technologies in manufacturing, environment, ecology, and information disciplines and optimize the performance of manufacturing systems to reduce the environment hazards and energy consumption of manufacturing processes, to improve productivity. In the future, customer value is not only achieved through products or services, but also reflected through the responsibility of the ecological environment. Under the sustainable manufacturing environment, manufacturing enterprises need to respond flexibly to the dynamic requirements of customers, to reduce production costs, to consider the ecological environment responsibility of enterprises, to choose to use green raw materials, to form closed-loop product life cycle, to decrease pollutant discharge in the production process, and to create new business models.

Between any organization and people, as well as between organization and individual, succeed to link in the Internet age, breaking the traditional relationship between them, promoting the cooperation between enterprises, and communication between enterprises with customers. More and more manufacturing enterprises would like to take globalization strategy because of the rapid development of the global economy. The production of big data and manufacturing knowledge generated in global manufacturing is shared in the global manufacturing network, and these data and knowledge can be translated into valuable information through being processed. Global manufacturing is a global way of production that integrates global product design, manufacturing, and logistics resources. It is also an open production method that facilitates the collaboration among stakeholders. In the era of Internet, the integration of information and communication technology and manufacturing technology can improve the participation of individuals and organizations in global manufacturing and social manufacturing, forming a bottom-up value creation model. The mass consumption of resource and energy in the manufacturing sector drives global manufacturing enterprises to endeavor in the value creation of sustainable manufacturing with the technical support.

In today’s market circumstances, each customer has more alternative products and services. In this case, customers pay more attention to emotional experience and a variety of demands, which makes providing personalized products and services to customers and meeting different market segments become the important part of investigation and analysis for manufacturing enterprises. In practice, enterprises integrate various resources on the basis of the individual requirements of customers, such as flexibility, multi-functionality and dynamic reconfiguration of production resources (i.e., raw materials, parts, equipment), supply chain members, and so on. MC is customer-centric product design, manufacturing, sales, and service. In view of the shortening of product life cycle and the increasing product complexity, senior managers need to consider the idea of overall system optimization, advanced information technology, and management technology, to provide low-cost, high-quality, customized products and services. MC integrates the enterprises, customers, suppliers, and the environment. It embodies the flexibility, robustness, responsiveness, and modularity of product design and production process. However, in the further development and application, enterprises need to take into account the cost, delivery, quality management, customer relationship management, information management, innovation, and green management.

From the aforementioned investigations, we can find that the challenges in advanced manufacturing include technology, management, and environment. Specially, the big data in the manufacturing system needs more sophisticated data technology. Cyber-attacks in manufacturing systems lead to the development of security technologies. Management challenges in advanced manufacturing involve cost-efficiency and real time to make enterprises more profitable and customers more satisfied. Environment challenges in advance manufacturing make stakeholders seek to use eco-friendly raw materials, reducing resource waste and environmental pollution. The future value creation of manufacturing is no longer just a matter of manufacturing resources, but a combination of manufacturing resources, data resources, and service resources. This process can be implemented on the Internet. It should be noted that much more work has been accomplished about the subject of this paper that has not been included in this review because it falls outside of the scope of our survey. We hope that our review of the advanced manufacturing literature provides readers with an updated view of various practical aspects of manufacturing technology, information technology, and management technology.