Industry 4.0 and intelligent predictive maintenance: a survey about the advantages and constraints in the Italian context

Abstract

Purpose

In the context of the Industry 4.0, this paper aims to investigate the state of the art of Italian manufacturing, focusing the attention on the implementation of intelligent predictive maintenance (IPdM) and 4.0 key enabling technologies (KETs), analyzing advantages and limitations encountered by companies.

Design/methodology/approach

A survey has been developed by the University of Parma in cooperation with the Italian Workers' Compensation Authority (INAIL) and was submitted to a sample of Italian companies. Overall, 70 answers were collected and analyzed.

Findings

Results show that the 54% of companies implemented smart technologies, increasing quality and safety, reducing the operating costs and sometimes improving the process' sustainability. However, IPdM was implemented only by the 37% of respondents: thanks to big data collection and analytics, Internet of Things, machine learning and collaborative robots, they reduced downtime and maintenance costs. These changes were implemented mainly by large companies, located in northern Italy. To spread the use of IPdM in Italian manufacturing, the high initial investment, lack of skilled labor and difficulties in the integration of new digital technologies with the existing infrastructure are the main obstacles to overcome.

Originality/value

The article gives an overview on the current state of the art of 4.0 technologies implementation in Italy: it is useful not only for companies that want to discover the implementations' advantages but also for institutions or research centres that could help them to solve the encountered obstacles.

Citation

Stefanini, R., Tancredi, G.P.C., Vignali, G. and Monica, L. (2023), "Industry 4.0 and intelligent predictive maintenance: a survey about the advantages and constraints in the Italian context", Journal of Quality in Maintenance Engineering, Vol. 29 No. 5, pp. 37-49. https://doi.org/10.1108/JQME-12-2021-0096

1. Introduction to 4.0 technologies and intelligent predictive maintenance

Manufacturing is moving toward fully automated and interconnected industrial production (Yang and Gu, 2021), thanks to the Fourth Industrial Revolution, characterized by a series of new technologies that blur the boundaries between the human being, the Internet and the physical world. It is certainly limiting to catalog the 4.0 technologies because it is an evolving world; however, these are some 4.0 key enabling technologies (KETs) considered suitable for the transformation of industrial production: big data and analytics (Bellini, 2018; Yuqian and Xun, 2019), collaborative robots (CRs) and automated guided vehicles (AGVs) (Gillespie et al., 2001; Alcácer and Cruz-Machado, 2019), simulation (Ciminoa et al., 2019), industrial Internet of Things (Iot) (Bortolini et al., 2017; Zhong et al., 2017), cybersecurity (Mozzaquatro et al., 2018), cloud computing (Tsai et al., 2015), additive manufacturing (Dilberoglu et al., 2017) and augmented reality (AR) (Fraga-Lamas et al., 2018; Alcácer and Cruz-Machado, 2019).

In this context, also the maintenance field is undergoing significant transformations, in terms of methodologies (Sahal et al., 2021) and available tools (Ji and Su, 2016). Predictive maintenance is based on historic database, models and domain knowledge. It can predict trends, behavior patterns and correlations by statistical or machine learning models for anticipating pending failures in advance to improve the decision-making process for the maintenance activity by avoiding downtime (Morariu et al., 2020). Maintenance was traditionally defined as a set of activities aimed at preserving the plant or machine, maintaining high levels of safety for the operator together with quality, reliability and efficiency (Lee et al., 2017). Nowadays, maintenance is driven by powerful mathematical and statistical models that support maintainers in adopting a new data-driven point of view, called “data-driven maintenance” (Kumar et al., 2018; Zhang et al., 2019). The availability of machine data collected in a robust and reliable way is in constant evolution, thanks to the concurrent development of new tools, such as a cloud-based system (Yuqian and Xun, 2019) useful to control demand, technologies and methodologies which harnesses maintenance policies (Zhang et al., 2017), such as thermal imaging cameras that allow monitoring production processes and help in taking maintenance decisions (Ramirez Nunez et al., 2016), mixed reality troubleshooting methods, which allow identifying the origins of faults and possible solutions (Bottani et al., 2021), radiofrequency systems, consent to collect technical information (Nappi et al., 2021) or drones able to inspect inaccessible or dangerous areas (Zhong et al., 2017). In addition, the development of sensors (which is establishing itself under the name of “smart sensors”) allows the constant connection and interaction between people and things in any place by using a proper connection protocol (Bonfiglioli SpA, 2020), continuous monitoring of the characteristics of the production process, related to product itself, process and machines via online platform (Tancredi et al., 2020). Smart factories began to incorporate a variety of technologies that support operators in carrying out maintenance activities through (1) wearable technologies (Fraga-Lamas et al., 2018), (2) real-time access to machine data and management data (Tsai et al., 2015), (3) intelligent sensors (Kanawaday and Sane, 2017) and (4) technologies for the control of execution. Currently these technologies consent to collect information and real-time data, introducing the paradigm of big data and analytics. The main features of big data are enclosed within the theory of the 4V (Witkowski, 2017): Volume, Variety, Velocity and Veracity. In the manufacturing field, big data refers to a huge amount of multi-source, heterogeneous and real-time data, generated during the production, operation and maintenance phases (Sezer et al., 2018).

Moreover, according to the literature, Industry 4.0 has the potential of growing the industrial sustainability (Giannoccaro et al., 2020; Jambrak et al., 2021; Jagtap et al., 2021; Corallo et al., 2020). However, Italian companies believe that the social and environmental sustainability are often negligible when compared to the advantage of economic sustainability reached thanks to 4.0 KETs (Forti et al., 2020).

Based on these premises found in the literature, this study is born as a collaboration between the University of Parma and the Italian Workers' Compensation Authority (INAIL): the aim is to analyze the state of the art of the manufacturing industry by investigating the implementation of 4.0 KETs and intelligent predictive maintenance (IPdM) in Italian companies. The next section describes the characteristics of the questionnaire that was created for this purpose and submitted to a sample of companies located in Italy. Then, the 70 responses collected are analyzed and results are discussed in the third section. The conclusion of the work will show not only all the benefits and advantages reached by the companies that have introduced KETs and IPdM but also possible solutions to help those companies who encountered obstacles in these implementations.

2. Methodology

2.1 Questionnaire structure

The survey was created with Google Forms and was composed of a total of 31 questions divided in 11 sections (Figure 1). The first section (I) describes the stakeholder involved in the research and the scope of the survey.

The second section (II) contains five questions related to the company profile, in particular the holding location, the market area and the role of the interviewed in the company. Here, the interviewees can submit the company name on their own accord, but their names are not published in this article for privacy reasons.

The third section (III) concerns the current implementation of the I4.0 technologies in the interviewed industry's production environment. It is composed of the question “Did your company introduce new technologies and intelligent methodologies of the I4.0?”: the answer can lead to three different sections according to the flow diagram shown in Figure 1: if the answer is “yes”, then the user will be sent to Section V; if instead the answer is “no”, then the user will be sent to Section X; if it states the inexperience on field, the user will be sent to Section IV i.e. end of the questionnaire.

The fifth section (V) concerns the degree of implementation of the various 4.0 KETs. It is composed of questions that asked how the company has become aware of the new technologies, the degree of satisfaction in the use of the smart technologies introduced, the advantages reached. The last question of this section is related to the use of IPdM, which is carried out through monitoring via IIoT sensors. As for the previous section, based on the answer given, the pathway will be different: if the company applies techniques and tools for maintenance 4.0 the user will be directed to Section VI, otherwise the interviewed will be led towards Section VII. The sixth section deals with IPdM in detail. Is composed of eight questions, focused on the types of technologies, methodologies, software and smart sensors adopted by the company. Lastly, it is reported a series of options for the eventual consequences noticed after the use of predictive maintenance 4.0, such as reducing costs and accidents, or industrial processes optimization, together with the main hitches encountered such as the lack of knowledge, compatibility issues, lack of experience of the employees and the investment addressed. The seventh section is composed of a unique question, which regards the possible consequences noticed after the implementation of the KETs towards the environmental sustainability: if the consequences stated are optimistic, the user will be directed to section VIII, in the opposite case to Section IX, otherwise to Section XI with the end of the questionnaire. The eighth section is made up of four questions. It regards the technologies that had improved the environmental sustainability and the advantages encountered with their implementation, such as reduction of emissions, waste valorization and waste reduction. Section IX is focused on the various technologies that caused the worsening noticed, by analyzing the criticalities encountered, such as an increase in water and energy consumption or zero waste reduction. Section X is specific for users who stated that their company has not implemented any of the I4.0 technologies. In this section, composed of six questions, were considered the possible reasons for which the company has not still proceeded with I4.0 technologies, the obstacles encountered and the interest in their implementation in a near future. Finally, in the eleventh section, the interviewee concludes the questionnaire and can optionally leave a contact for the forwarding of the results obtained by the survey.

2.2 Reference sample and submission

The reference sample to be interviewed was defined using Kompass database, a free global Business to Business platform to find and contact suppliers of products or services. A non-homogeneous sample was created, considering companies from Northern, Central and Southern Italy, with different sizes and different ATECO code (Italian codification derived from the European NACE industrial standard codification system). Overall, 1,000 Italian companies were found and their name and email contacts were collected in an Excel File. The survey was sent via email to each company: after 4 weeks, the questionnaire was closed and the answers were analyzed.

3. Results and discussion

3.1 Description of the answering sample

Overall, 70 Italian companies answered the questionnaire. The 70 interviewees were located in different areas of the country: 70% was from the North, 17% from the South and 13% from the center of Italy. To classify the company size the numbers of employees and the annual revenue have been considered. Overall, four different sizes were established: a large company has an employee number over 250 and/or an annual revenue greater than 50 million €; a medium company involves between 5 and 249 employees and/or an annual revenue ranging from 10 to 50 million €; a small company is characterized by a number of employees lower than 50 and/or an annual revenue among 2 and 10 million €; a micro company has less than 10 employees and an annual revenue lower than 2 million euros. The collected answers were from a very nonhomogeneous sample from a size point of view, as illustrated in Figure 2.

Regarding the industry sector of the companies, according to the ATECO code, 23 companies operate in the food industry (32.9%), 5 in metal products (7.1%), 6 are service providers (8.6%), 3 in other manufacturing sectors (4.3%). The remaining percentage is equally divided among 33 companies belonging to sectors such as chemicals, electronics, automation and others, with a percentage of 2.9% for each of them. Moreover, 35.7% of respondents holds management role for the whole company, while the rest of the interviewees are departments manager. The 67% of the users chose to share their company name, while the remaining 23 remained anonymous.

3.2 Implementation of 4.0 KETs

More than half (54.3%) of the 70 participating users have already applied smart technologies. This percentage is in turn composed of the 41.4% of companies that currently use some I4.0 technologies and for the 10% that widely use them, while the 2.9% adopted few of them. The remaining percentage has not yet introduced KETs for several reasons, in particular the 31,4% asserts that they will adopt smart technologies soon, while 8.6% states the unfeasibility of adopting them and the 5.7% were inexperienced in this field.

Considering now the sample of 38 companies that have already implemented Industry 4.0 technologies, 18 belong to the food industry sector (47.37%), 5 to the metal products manufacturing (13.16%) and the remaining to various sectors, including plant and machinery manufacturing, paper industry, automotive. However, it is not particularly remarkable that almost the half of the companies using 4.0 technologies are in the food sector, as the initial sample is mostly made up of companies in this branch. However, it has been observed the ratio between the number of companies in each sector that have integrated some or more I4.0 technologies and the total number of industries that belong to the same category. As can be seen in Figure 3, the percentage of smart technologies used in the food industry is around 78%, 33% in the services sector and 100% in the metal manufacturing industry.

In particular, the majority of the companies that have already introduced some 4.0 technologies defined themselves as large companies (36.8%), followed by small ones (34%); the remaining are medium-sized (21%), and a few of them are micro (7.9%), as shown in Figure 4.

A significant number of the sample companies making use of 4.0 technologies belong to the geographical area of northern Italy: this is certainly linked to the huge percentage of companies located in the northern rather than southern and center. Figure 5 shows the degree of KETs implementation, rated on a five points scale, where 1 = “not yet implemented” and 5 = “widely used”. Additive Manufacturing (AM) is not so widespread among the interviewees, except for the food sector. On the contrary, big data is more used in media. Regarding AR, the answers show that it is still far from being widely used as well as virtual reality (VR): only a few respondents state that they use them. Regarding CRs, the majority of companies state that the degree of implementation is negligible, but some interviewees introduced them in their industry. Within the companies that indicated a level of use of CRs equal to 4 or 5 and therefore make extensive use of this technology, eight belong to the food sector, two to the manufacture of metal products and one to the glass industry. For Iot and machine learning (ML), even if many companies had not yet implemented these technologies in any way, someone of them indicated a good level of implementation, again in the food sector. AGVs are the less introduced: only 3 participants indicated a degree of implementation of 3 and only one company, which manufactures machinery and equipment for the food industry, the maximum. Cloud computing and cybersecurity registered very non-homogeneous answers, but in media these technologies seem to be more used than the others, in particular in the food sector and in the metal products industry. On the contrary, digital twin is not so used and known, taking into consideration that many companies answered that they do not know if it is implemented.

Another interesting aspect is related to the way in which companies have become aware of Industry 4.0 technologies. According to results, the fiscal and financial incentives set up by the government such as super and hyper-amortization and tax credit were the main tool used by companies for the implementation of smart technologies. Furthermore, the pressures of the marketing and research and development functions within the companies have been promoters of innovation, followed by projects in collaboration with universities or research bodies. Therefore, it will be necessary to focus mainly on economic incentives to motivate also other companies towards this change, or on the knowledge and technology transfer by universities.

3.2.1 Benefits from the introduction of 4.0 technologies

As regards the technologies that have been implemented, 52.63% of companies consider themselves satisfied, while 39.47% are very satisfied, 5.26% are completely satisfied, 2.6% are not so satisfied and finally 0% not satisfied. The participants who confirmed the adoption of I4.0 technologies, were then asked to vote about the benefits they had achieved by implementing such technologies, on a scale of values from 1 to 5 where 1 = “no impact”, 5 = “very high impact” (Figure 6). We can state that the introduction of KETs allowed the increase in quality and safety, the reduction of operating costs, production costs and downtime, a best relationship between the company and suppliers and/or customers and the optimization of decision-making processes.

The Pearson correlation “r” was used to establish the relationship between each technology and each type of benefit achieved. The “I don't know” answers were neglected, as only numerical values are considered by the correlation function. As far as the correlation coefficient is closer to +1 or −1, it indicates a positive (+1) or negative (−1) correlation between the matrices. A positive correlation indicates that if the values in one matrix increase, the values in the other matrix also increase; conversely, a negative correlation indicates the opposite. A correlation coefficient closer to 0 means, on the other hand, that there is no correlation or a weak correlation. The indices obtained are shown through a conditional formatting, which highlights in green the highest correlations (Figure 7). As can be seen, CRs are related to increased availability of equipment, as well as Iot, that improved also the relationship between the company and suppliers and/or customers. Moreover, ML increased quality of products/services, reduced maintenance and production costs but also maintenance planning time.

3.2.2 Industry 4.0 and environmental sustainability

Most of the companies (44.7%) have not noticed any significant changes in environmental sustainability following the use of smart technologies. On the other hand, only 2.6% states a worsening of emissions and an increase in waste, while 34.2% achieved an improvement in terms of lower emissions in water/air/soil and a reduction in waste. The enabling technologies considered most responsible for improvements are CRs, Iot and ML: most users reached the reduction of energy consumption, followed by the reduction of processing waste, water and raw material consumption, CO₂ emissions (Figure 8). However, only one of these companies has carried out Life Cycle Assessment (LCA) studies to demonstrate a lower environmental impact of 4.0 solutions compared to traditional ones: a limit that should be underlined is that it is not possible to know if the other results are only respondents' opinions or are verified numerically with other methods different from LCA. In the future developments, this aspect should be clarified.

3.3 Intelligent predictive maintenance results

Results highlight that 34.2% of companies use some IPdM methodologies, such as monitoring by sensors, machines parameters to diagnose and prevent anomalies and faults, 13.2% do not use it because they use other types of maintenance, 13.2% of users whose did not have enough data to return a precise answer and 36.8% that do not use it at all. Figure 9 shows the sectors of the companies that use some methods and tools of IPdM: more than half are part of the food sector (44%).

Some KETs are used for predictive maintenance purposes, such as big data collection and analytics, Iot and cybersecurity. The sensors that are mostly used are for the position/movement/speed analysis, followed by vibration, optical, acoustic and flow analysis. From a social point of view, after the implementation of predictive maintenance 4.0, few companies in the sample have noticed consequences about employment and company work: in particular, in two companies there has been a reduction in the necessary staff that was replaced by the technologies introduced, and another has seen an increase in the workforce due to the need for skilled personnel.

3.3.1 Benefits from the introduction of IPdM

Another interesting aspect concerns the various advantages observed by companies using smart maintenance techniques and tools. From this analysis it appears that, in general, the highest impacts regarding maintenance planning times and plant shutdowns were found by companies that use, for smart maintenance purposes, technologies related to big data, ML and AGV. Moreover, companies that have shown a positive impact on the quality of the product/service thanks to IPdM use big data, IoT, ML, cloud computing, AGV and CRs; those who instead emphasized the optimization of decision-making processes mainly use big data, IoT and CRs during maintenance strategies. Finally, improvements relating to the increase in equipment availability, the reduction of operating costs, a significant positive impact in customer and supplier relationships and increased productivity have been reached by companies that use big data, ML and cloud computing.

3.3.2 Constraints about the introduction of IPdM

During the process of implementing 4.0 technologies in maintenance activities, companies have also encountered some critical issues: the most impacting is represented by the difficulty in integrating new digital technologies with the existing infrastructure, followed by the absence or scarce presence of competent, experienced and adequately trained human resources (see Figure 10).

3.4 Non implementation of KETs and IPdM

With regard to 32 companies that did not know 4.0 technologies and did not apply them for maintenance activities, the main economic, organizational and social problems were investigated, as well as possible solutions that could help companies to participate in the smart evolution. Based on the results, these companies are part of the food industry sector, or service providers, or work in the trade/production of glass packaging. Others are manufacturing computers and other electronic products, or are involved in maintenance, repair and installation of machinery and equipment and the rest of various sectors, including manufacturing of insulating panels, metallurgy etc. As regards the company size, the small and medium companies found more difficulties in the implementation. The percentage of implementation of 4.0 technologies in the company by geographical area was analyzed (Figure 11): the ratio is high for companies in Northern Italy (61.22% have already implemented some of 4.0 technologies), medium for those of the Center (44%) and low for the companies in the sample belonging to Southern Italy (33%).

The main obstacles have been the lack of knowledge of 4.0 technologies, the absence of expert personnel in the field, the large economic investment that it is necessary to support to proceed with the implementation and the fact that companies are often not yet perceived the need to innovate by introducing smart technologies.

Overall, companies demonstrated to be interested in a future introduction of 4.0 technologies, but they stated to need solutions to introduce 4.0 technologies. It seems that the most crucial aspects for these purposes concern technological skills, followed by managerial and analytical ones, such as training courses for manpower, the individuation of skilled industrial partners, national or regional financing and partnership with a university or research centers on 4.0 projects.

4. Conclusions

The advent of Industry 4.0 allows the transformation of many companies all over the world, thanks to new KETs that can be used not only during production but also during maintenance activities, leading to an IPdM. This study aimed to analyze how much these innovations are already introduced today in the Italian manufacturing industry. A questionnaire composed of 31 questions, concerning issues related to the knowledge and introduction of 4.0 technologies and predictive maintenance, was created and then submitted to a sample of Italian companies, different in size, sector and geographical area. The results of the survey can help to draw the main conclusion of the work. At the Italian state of the art, I4.0 technologies such as big data, CRs, cloud computing and cybersecurity are currently implemented by few companies in the Italian territory, mostly located in the northern part and are large sized. These introductions allowed an increase in quality and safety, a reduction of operating costs, production costs and downtime, a best relationship between the company and suppliers and customers and an optimization of decision-making processes. Some of them believe also to have reached some environmental improvements such as reduction of energy consumption, waste, water and CO₂ emissions, even if not always verified analytically. For what concern the use of IPdM methods and tools, 50% of the interviewees who replied that they have implemented some of the 4.0 technologies, do not use any type of smart sensors in the maintenance field. This result underlines that currently smart predictive maintenance is still far from being widely used and known in Italian companies. Indeed, few of them uses big data analytics, Iot, ML, CRs, cloud computing and digital twin for this purpose, in particular a smart sensor that uses position, movement, speed analysis, vibration, optical, acoustic and flow analysis. The main advantages achieved are the increase in productivity, the reduction of maintenance planning times and plant shutdowns, a positive impact as regards the quality of the product/service provided.

In conclusion, today there are too many Italian companies that have not implemented 4.0 technologies yet because of an excessive initial investment, the lack of experienced staff, the difficult identification of competent external partners and difficulty in integrating new digital technologies with the existing infrastructure. Of course, European/regional/national funding for the implementation of 4.0 technologies and the possible identification of trained industrial partners can be useful to help more companies in this Fourth Industrial Revolution. Of course, once that most companies will have implemented innovative technologies and methodologies, the real competitive factor will be represented by the ability to manage them efficiently and effectively. The future development of the study would carry out a personal survey or focus group with a selected sample of expert interviewees, also investigating deeply the environmental impact of 4.0 KETs, since these aspects have been underlined as a limitation in this current study. Are 4.0 technologies sustainable? Of course, more research should be done to answer this question.

Figures

Figure 1

Questionnaire algorithm pathway

Figure 2

Company size [%]

Figure 3

Use of 4.0 KETs in companies coming from different industry sectors [%]

Figure 4

Implementation of 4.0 technologies in relation to company size

Figure 5

Degree of implementation

Figure 6

Contribution of introduced 4.0 KETs to the achievement of advantages

Figure 7

Correlations between each 4.0 technology and advantages

Figure 8

Environmental improvements reached thanks to 4.0 KETs implementations

Figure 9

Companies using intelligent predictive maintenance for each sector [%]

Figure 10

Constraints and problems encountered during the IPdM implementation [%]

Figure 11

Implementation/non-implementation of 4.0 technologies by geographical area

Funding: The authors did not receive funding from any organization for the submitted work.

Data availability: Data collected in the survey are primary data given by the companies.

Ethics approval: The article involves no studies on human or animal subjects.

Consent for publication: The authors provide their consent to publish this article.

Conflict of interests: The authors have no relevant financial or non-financial interests to disclose.