A Precondition of Sustainability: Industry 4.0 Readiness

To achieve sustainability and further develop their businesses, manufacturing companies need to find an appropriate response to the frequent changes in customer demands, development of new technologies, and dynamic competition. One of the possible requirements for sustainability is the adoption of the Industry 4.0 paradigm. This paper aims to apply Industry 4.0 general readiness assessment methodology and social network analysis to find out the readiness level of Serbian manufacturing and how digital technologies interplay during the time. The results of this study show that wireless human-machine communication technologies initiated the digital transformation of Serbian manufacturing from non-users to basic readiness level. Secondly, manufacturing companies significantly invested in the Cyber-Physical Production System to increase the level towards high readiness. Finally, manufacturing companies in Serbia should consider investing in capabilities to adapt the Near real-time production control system if aiming to compete with competitors from developed countries. The main contribution of this paper is to show the general readiness level of manufacturing companies for digital transformation in transition economies.


Introduction
Nowadays, manufacturing companies are conducting business in a market that is characterized by the constant and frequent changes in customer demands, development of new technologies, international value chains, and dynamic competition [1]. To achieve sustainability and further develop their business, manufacturing companies need to find an appropriate response to the previously mentioned issues in a way that will increase efficiency, reduce costs, improve agility and flexibility, and upgrade responsiveness and resilience of their production operations [2]. For many manufacturing companies, it seems that the answer lies in the adoption of the Industry 4.0 paradigm [3]. This implies building an interface between the physical and the digital space through the use of Industry 4.0 enabling technologies such as Cyber-Physical Systems (CPS), Internet of Things (IoT), Cloud Computing (CC), Big Data, and Data Science [4][5][6]. The Industry 4.0 ecosystem could provide permanent communication and interaction amongst numerous internal and external objects [7].
Manufacturing companies that wish to adopt Industry 4.0 paradigm into their operations first need to assess the status of their operations (i.e., to determine the readiness and maturity of the company to implement Industry 4.0 concepts) [8]. This represents a starting point for the identification of strengths, priorities, opportunities, potential challenges, and the creation of comprehensive transition plans necessary to overcome multiple barriers on the road to Industry 4.0 concept fulfillment [8]. Although barriers on the path to Industry 4.0 exist, both for manufacturing companies from the developed and developing countries, their properties differ significantly since the characteristics and their sheer number are more substantial in developed countries [9][10][11]. Most of the relevant studies to date, however, Sustainability 2021, 13, 6641 2 of 12 have been from developed countries, and longitudinal research has been significantly lacking. This research intends to shed light on the evolutionary path to adoption of Industry 4.0 concepts of manufacturing companies from developing countries (i.e., Republic of Serbia) as a precondition for sustainability by assessing the readiness index development and key digital technologies. In addition, this study aims to investigate the direct impact of Industry 4.0 on sustainability.
Thus, the question addressed in this research is "What is the readiness level for Industry 4.0 of manufacturing companies, and how do digital technologies interplay it during the time?" To answer this question, we conducted a two-wave longitudinal survey with companies from manufacturing sectors. As expected, the results show that manufacturing companies significantly increased their readiness level for Industry 4.0. As companies progressed along the path, the number of CPS digital technologies that needed to be implemented to increase the readiness level also increased. Moreover, Social Network Analysis shows the density of the companies' networks in 2015 and 2018. Additionally, this analysis shows the impact of digital technologies on the companies according to the firm's size. The application of SNA in Industry research 4.0 gives an innovative view of the digital transformation of the manufacturing sectors.
The remainder of the paper is structured as follows: a literature review is presented in Section 2. Section 3 describes the methodology used in this study. Section 4 presents the research results. In Section 5, research results are discussed. Finally, the paper concludes in Section 6 with identified limitations of the study and suggestions for further research are specified.

Literature Review
Technological innovations such as computerized automation, sensors, and actuators have been used in manufacturing companies for a long time and have enabled manifold benefits in the execution of operations. However, their full potential has not been achieved, due to restrictions in connectivity and integration amongst various technologies used in the manufacturing process [12]. The main challenge has been in the information systems deployed. They have used different communication interfaces and due to this fact, systems have not been able to establish a connection and the subsequent communication necessary for data exchange between them [13]. A possible solution for sustainability arrived with the Industry 4.0 paradigm and the development of its enabling digital technologies (e.g., CPS, IoT, CC). Industry 4.0 has a direct impact on sustainability, both positive and negative. For instance, integration of digital technologies into the manufacturing processes enabled effective connection and communication between technological innovations [14]. However, utilization of Industry 4.0 technologies also triggered increased energy consumption from sensors and machinery [14]. All the pros and cons of Industry 4.0 opened up new questions; for instance, what are the steps on the path towards Industry 4.0 adoption, or how would the implementation of enabling technologies impact their existing operations [3,15].
There is no definitive roadmap for the digital transformation of manufacturing companies [16]. Industry 4.0 is a relatively recent concept among academicians and practitioners from the manufacturing sector and the field is not properly standardized or broad, and lacks clear boundaries [17]. Potential barriers on the path to Industry 4.0 are lack of governmental regulation, high financial investments, the complexity of the technologies, lack of multidisciplinary practical knowledge, internal resistances to organizational changes, and lack of skilled workers [18]. Furthermore, a significant number of manufacturing companies are not certain about the Industry 4.0 concepts, technologies, and necessary actions to maximize eventual benefits and the implications to their businesses to achieve sustainability [19]. In addition, the path towards Industry 4.0 varies notably between developed and developing countries, among different industrial sectors, and among companies within the sectors [1,11,20].
Manufacturing companies, particularly the ones from developing countries, suffer from deficiencies in digital technologies that are an integral part of Industry 4.0. Hence, Sustainability 2021, 13, 6641 3 of 12 they need to gain access to the required infrastructure and "know-how" through technology transfers, contracting research centers, staff training, internal research, and development investments in order to overcome technical challenges arising from updating systems that are already in place [21,22]. A crucial step on the track to the digital transformation of manufacturing companies is to determine their current position with regard to the Industry 4.0 concepts and then subsequently to be able to produce comprehensive transition plans for Industry 4.0 concept fulfillment [23]. Therefore, the use of readiness or maturity assessment models is advised [24] to guide companies and their stakeholders through the transition phase efficiently and systematically [25]. Maturity and readiness are used interchangeably in many cases to describe assessment models. However, for this research, we use the term readiness to signify "the state in which an organization is ready to accomplish a task" [26] in contrast to maturity that signifies "the level of evolution that an organization has accomplished concerning a task" [26]. Thus, readiness differentiates from maturity due to the notion that readiness is assessed before engaging in maturing processes, whereas maturity is assessed from the actual implementation onwards [23].
Since key drivers of manufacturing transition in Industry 4.0 are digital technologies [34], this research examines the utilization of digital technologies as a first step in analyzing the readiness for Industry 4.0. Hence, we used the I4.0 general readiness assessment methodology developed by the Fraunhofer Institute for Systems and Innovation Research ISI [31]. The model applied in this study follows the readiness assessment methodology of the Fraunhofer ISI, which is confirmed in the many studies developed by the partners from the EMS consortium [31]. This model has the aims to assess the current situation in terms of implementation of advanced technologies in the different manufacturing sectors. The methodology combines digital technologies allocated within three technological fields. Therefore, it provides information on the digitalization readiness of individual manufacturing companies.

Data and Sample
This manuscript used data from the European Manufacturing Survey (EMS) [35]. EMS is carried out in the 17 European countries, which is coordinated by the Fraunhofer ISI from Germany [36][37][38]. This research used a data set from the Republic of Serbia, obtained in 2015 and 2018. The research was conducted between manufacturing companies (NACE Rev 2 codes from 10 to 33) with more than 20 employees. We obtained 280 and 240 valid responses in 2015 and 2018, respectively. The response rate in both research rounds was about 34%. Table 1 depicts the distribution of the manufacturing companies by size. The result depicts that about 41.7% of the manufacturing companies in the sample are small, having between 20 and 49 employees, while another 46.9% of the manufacturing companies have between 50 and 249 employees, and 11.4%, more than 250 employees. In the share of the total sample, the highest share belongs to food production, followed by the production of fabricated metal products, except machinery and equipment, and the manufacture of rubber and plastic products, which together have 40.5% of the total sample. Furthermore, Table 2 presents the distribution of the manufacturing companies by sector.

Readiness Assessment Methodology
This research uses Industry 4.0 general readiness assessment methodology developed by the Fraunhofer ISI [31]. The methodology is presented in Figure 1. The result depicts that about 41.7% of the manufacturing companies in the sample are small, having between 20 and 49 employees, while another 46.9% of the manufacturing companies have between 50 and 249 employees, and 11.4%, more than 250 employees. In the share of the total sample, the highest share belongs to food production, followed by the production of fabricated metal products, except machinery and equipment, and the manufacture of rubber and plastic products, which together have 40.5% of the total sample. Furthermore, Table 2 presents the distribution of the manufacturing companies by sector.

Readiness Assessment Methodology
This research uses Industry 4.0 general readiness assessment methodology developed by the Fraunhofer ISI [31]. The methodology is presented in Figure 1. 1. Digital Management Systems: The first field of technology consists of "Software system for production planning and control" and "Product Lifecycle Management Systems". 2. Wireless human-machine communication: In the second field of technology the "Digital Visualization" is combined with the "Mobile Devices". Digital Management Systems: The first field of technology consists of "Software system for production planning and control" and "Product Lifecycle Management Systems".

2.
Wireless human-machine communication: In the second field of technology the "Digital Visualization" is combined with the "Mobile Devices". 3.
Cyber-Physical Production System related processes: The third field of technology is the "Real-time production control system", the "Automation of internal logistics" and "Digital data exchange with customers and suppliers" together.
According to the approach, the manufacturing companies are divided into six levels according to the use of digital technologies:

•
Level 0: Companies with a traditional process that do not use digital technologies Companies that are at level 0 are considered as non-users, at levels 1-3 as companies with basic readiness, and at levels 4 and 5 as high readiness enterprises.

Social Network Analysis
Social Network Analysis (SNA) is the method that investigates the relationship and position of actors in the social structures based on graph theory [39]. The SNA method is often used in the social sciences [40]. However, in the last decade research which investigated manufacturing sectors began to use it [41]. The studies from manufacturing sectors apply SNA to investigate the relationship between companies in the area of cloud manufacturing, logistics, digital servitization, and others [41]. SNA analysis provides a powerful methodology tool from social to technical sciences for investigation relations and ecosystems between the manufacturing sectors [39]. This research investigates the relationship between manufacturing companies and the use of digital technologies in the production process. According to the industry sectors, manufacturing companies are presented with red circles. Authors labeled digital technologies with a combination of letters and numbers ranging from DT1 to DT7 (i.e. DT1-Mobile/wireless devices for programming and operation of equipment and machinery, DT2-Digital solutions for providing drawings, work schedules, or work instructions directly on the shop floor DT3-Software for production planning and control, DT4-Digital Exchange of product/process data with suppliers/customers, DT5-Near real-time production control system, DT6-Systems for automation and management of internal logistics, DT7-Product-Lifecycle-Management-System (PLM) or Product/Process Data Management), and they are presented with blue squares. The snowball method is used for collecting data from surveys as the inputs for eigenvector and density analysis. For the evaluation and impact of the use of digital technologies in the manufacturing sector, eigenvector centrality is used. Furthermore, in the networks, authors take value 0 in the case where there is no network tie between manufacturing firms and use of digital technologies, and take value 1 where there is an existence network tie between manufacturing firms and the use of digital technologies. The SNA graphs present a network for the firms which are connected with digital technologies. Thus, these values are the input for the eigenvector and density analysis of the manufacturing firms.

Results
The analyses begin by presenting practices that are mostly adopted by the manufacturing companies followed by the Industry 4.0 general readiness of the Serbian manufacturing companies. Finally, the results of the SNA analyses with the graphs and eigenvectors are presented. Table 3 presents the level of implementation of digital technologies by the manufacturing companies within the research. Table 3 reported that Serbian manufacturing companies in 2015 implemented the most Real-time production control system (31. lowest utilization in 2018 was reported for the Digital data exchange with customers and suppliers (15.0%). All digital technologies increased the level of use from 2015 to 2018, except the Software system for production planning and control. As for the technology fields, the highest increase was noted in Wireless human-machine communication. The present study also investigated the readiness index of the Serbian manufacturing companies based on Industry 4.0 general readiness assessment. Table 4 portrays the distribution of the Industry 4.0 readiness levels of Serbian manufacturing companies. In 2015, the highest number of manufacturing companies was at Level 0 (49.77%). Results indicate that 43.66 percent of companies had a basic readiness index (Levels 1-3), and only 6.57 percent a high readiness index. In 2018, 30.1 percent of companies reported that they did not use any of the digital technologies studied and fully relied on traditional production processes (Level 0). More than half (51.94%) of the manufacturing companies achieved basic readiness and 17.96 percent a high readiness index. By using the social network analysis, we investigated the level of use and the impact of digital technologies in Serbian manufacturing companies. Our research included seven digital technologies in the production process. Figure 2 presents the use of digital technologies in the manufacturing sector in the Republic of Serbia in 2015.
digital technologies in the production process. Figure 2 presents the use of digital techno ogies in the manufacturing sector in the Republic of Serbia in 2015. The graph presents the use of digital technologies, which indicate no connectives b tween the industry sector of the company and digital technologies. However, it can b argued that similar companies use a DT1 and DT4, which is similar to companies that u a DT3. Until DT2, DT5, and DT7 there are no relations between companies and the use these technologies. Table 5 shows the eigenvector of the use of digital technologies in th manufacturing sector in 2015 for small, medium, and large companies. The results from Table 5 show that the Near real-time production control system, th Digital Exchange of product/process data with suppliers/customers, and the Mobile/wir less devices for programming and controlling facilities and machinery are the digital tec nologies with the highest eigenvector value in the large and medium companies. Add tionally, the small companies have a high eigenvector value for the technologies based o the Systems for automation and management of internal logistics. Digital technologi with the lowest eigenvector value are Software for production planning and schedulin for large companies, Digital solutions to provide drawings, work schedules, or work i structions directly on the shop floor for the medium companies, and Product-Lifecycl Management-Systems (PLM) or Product/Process Data Management for the small comp nies. Furthermore, the density of the network shows the values of 0.155 for the large com panies, 0.181 for medium companies, and 0.211 for the small companies. Figure 3   The graph presents the use of digital technologies, which indicate no connectives between the industry sector of the company and digital technologies. However, it can be argued that similar companies use a DT1 and DT4, which is similar to companies that use a DT3. Until DT2, DT5, and DT7 there are no relations between companies and the use of these technologies. Table 5 shows the eigenvector of the use of digital technologies in the manufacturing sector in 2015 for small, medium, and large companies. The results from Table 5 show that the Near real-time production control system, the Digital Exchange of product/process data with suppliers/customers, and the Mobile/wireless devices for programming and controlling facilities and machinery are the digital technologies with the highest eigenvector value in the large and medium companies. Additionally, the small companies have a high eigenvector value for the technologies based on the Systems for automation and management of internal logistics. Digital technologies with the lowest eigenvector value are Software for production planning and scheduling for large companies, Digital solutions to provide drawings, work schedules, or work instructions directly on the shop floor for the medium companies, and Product-Lifecycle-Management-Systems (PLM) or Product/Process Data Management for the small companies. Furthermore, the density of the network shows the values of 0.155 for the large companies, 0.181 for medium companies, and 0.211 for the small companies.  The graph from the Serbian manufacturing sector in 2018 shows a stronger netwo than in 2015. Moreover, there are more relations between digital technologies and man facturing companies. Nevertheless, there is no similarity between the industry sector the companies and the use of any digital technologies. Results show that manufacturin companies have an individual approach to the use of these technologies. Table 6 presen the eigenvector of the use of digital technologies in the manufacturing sector in 2018 f small, medium, and large companies. The results from Table 6 show that large and medium companies have the highe eigenvector values for the same digital technologies such as the Software for productio planning and scheduling, the Digital Exchange of product/process data with supp ers/customers, and the Near real-time production control system. Additionally, the sma companies have a high eigenvector value for the technologies based on the Digital sol tions to provide drawings, work schedules, or work instructions directly on the shop floo Digital technologies with the lowest eigenvector value are the Mobile/wireless devices f programming and controlling facilities and machinery for large companies, and the Pro uct-Lifecycle-Management-Systems (PLM) or the Product/Process Data Management f medium and small companies. Furthermore, the density of the network shows values 0.423 for large companies, 0.311 for medium companies, and 0.251 for small companies

Discussion
This research empirically tested the Industry 4.0 general readiness assessment met odology developed by the Fraunhofer ISI [31]. Companies from all 24 manufacturing se tors (NACE Rev 2 codes from 10 to 33) were assessed to answer the research questio The graph from the Serbian manufacturing sector in 2018 shows a stronger network than in 2015. Moreover, there are more relations between digital technologies and manufacturing companies. Nevertheless, there is no similarity between the industry sector of the companies and the use of any digital technologies. Results show that manufacturing companies have an individual approach to the use of these technologies. Table 6 presents the eigenvector of the use of digital technologies in the manufacturing sector in 2018 for small, medium, and large companies. The results from Table 6 show that large and medium companies have the highest eigenvector values for the same digital technologies such as the Software for production planning and scheduling, the Digital Exchange of product/process data with suppliers/customers, and the Near real-time production control system. Additionally, the small companies have a high eigenvector value for the technologies based on the Digital solutions to provide drawings, work schedules, or work instructions directly on the shop floor. Digital technologies with the lowest eigenvector value are the Mobile/wireless devices for programming and controlling facilities and machinery for large companies, and the Product-Lifecycle-Management-Systems (PLM) or the Product/Process Data Management for medium and small companies. Furthermore, the density of the network shows values of 0.423 for large companies, 0.311 for medium companies, and 0.251 for small companies.

Discussion
This research empirically tested the Industry 4.0 general readiness assessment methodology developed by the Fraunhofer ISI [31]. Companies from all 24 manufacturing sectors (NACE Rev 2 codes from 10 to 33) were assessed to answer the research question "What is the readiness level for Industry 4.0 of manufacturing companies, and how do digital technologies interplay during the time?". Based on the results of the study it can be argued that Sustainability 2021, 13, 6641 9 of 12 the general readiness of Serbian manufacturing companies transformed from non-users in 2015 to basic readiness in 2018. This was mainly triggered by the implementation of Wireless human-machine communication technologies. The reality of many manufacturing companies is to first invest available financial resources in technology such as digital visualization and mobile devices to achieve higher performance. For example, most of the manufacturing companies have a Wi-Fi connection on the shop floor and they might find it strategically convenient, and of great value to rather invest in wireless human-machine communication technologies instead of digital management or cyber-physical production systems. Therefore, manufacturing companies that have not started to incline themselves towards Industry 4.0 [27], could start using digital visualization or mobile devices within the plants to increase the level of general readiness towards Industry 4.0 and move closer to developed countries. For instance, in Germany, only 19% of manufacturing companies use mobile devices [31].
Further, Industry 4.0 has a direct positive impact on sustainability by improving manufacturing innovation [14]. Results indicate that the number of high readiness manufacturing companies tripled from 2015 to 2018 due to an increase in the implementation of Digital data exchange with customers and suppliers. This technology from the field of the Cyber-Physical Production System generated a higher level of general readiness towards smart manufacturing. Also, SNA results indicate that Digital data exchange with customers and suppliers had the highest impact on all other digital technologies in Serbian manufacturing companies, both in 2015 and 2018. This may be due to the fact that manufacturing companies have integrated their operations within digital supply chains. These results are in line with the previous research stating that Industry 4.0 can create unified sustainable production systems [42]. Since almost all digital technologies increased the level of use from 2015 to 2018 within the Serbian manufacturing domain, it can be argued that when all these technologies are put together, it might give opportunities to create a sustainable industrial environment.
Results indicate that the only drop from 2015 to 2018 was in the use of the Software system for production planning and control, from 19.2 to 16 percent, respectively. Compared to findings from manufacturing companies in Germany (67% in 2015), this represents the biggest difference. The reason behind low utilization may be that the manufacturing companies in Serbia had limited capability to adapt such systems due to connectivity requirements, and revision of software architectures [43,44]. SNA results indicate that medium and large companies have the highest eigenvector values for the same digital technologies in 2015 and 2018. Unlike small companies, which have different digital technologies with the highest values such as Digital solutions to provide drawings, work schedules, or work instructions directly on the shop floor in 2018 and Systems for automation and management of internal logistics in 2015. The Near real-time production control system is the digital technology with the highest eigenvector in all three groups of companies. Additionally, the density of the network increased in value from 0.155 to 0.423 for large companies, from 0.181 to 0.311 for medium companies, and from 0.211 to 0.251 for small companies. These results show that manufacturing firms use a higher number of the same digital technologies among the sectors. Moreover, SNA results show that manufacturing firms from the Republic of Serbia are strongly transforming their ecosystem from traditional to digital.

Conclusions
This paper identified a precondition for sustainability by assessing the general readiness of Industry 4.0 of manufacturing companies in Serbia. We defined a research question that was answered with results gained through the application of methodology developed by the Fraunhofer ISI coupled with the SNA method. The general readiness level of Serbian manufacturing companies for smart manufacturing altered from non-users to basic readiness in the period of three years, thus creating an opportunity for a sustainable industrial environment. The main findings of this study are:

•
Wireless human-machine communication technologies initiated the digital transformation of Serbian manufacturing to a basic readiness level. Manufacturing companies found it strategically convenient to invest in digital visualization and mobile devices to accomplish sustainability.

•
Manufacturing companies significantly invested in the Cyber-Physical Production System to increase the level towards high readiness. Digital data exchange with customers and suppliers had the highest impact on two other technology fields (i.e., Digital Management Systems, Wireless human-machine communication).

•
Manufacturing companies in Serbia must consider investing in capabilities to adapt Software systems for production planning and control if aiming to compete with competitors from developed countries. This technology could increase the readiness level towards smart manufacturing.
The major limitation of this paper is that it considered all manufacturing sectors and company sizes. Future research could analyze the general readiness level of each of the manufacturing sectors (NACE Rev 2 codes from 10 to 33), and the readiness levels of small, medium, and large enterprises. Furthermore, it did not take into consideration the influence of economic crises in the previous two years. While the current paper explores the Industry 4.0 general readiness in the period from 2015 to 2018, a new longitudinal approach could give answers on how the coronavirus pandemic has influenced the use of digital technologies.