**3. European Research Activities on Digitalization in the Steel Sector**

According to [37], the European steel industry faces important challenges due to cost pressure, regulatory requirements, as well as product and service requirements. For this reason, over the last few decades, it has been involved in several policy activities, R&D projects, and patents in the field of digitalization.

Initiatives related to Industry 4.0 also include the Smart Factory Working Group of the ESTEP Platform, founded in 2008 with the former name of "Intelligent Integrated Manufacturing," which published the first edition of a Roadmap for European Steel Manufacturing in 2009 with a vision up to 2020. The ESTEP Working Group covers a broader range of stakeholders and it consists of plant manufacturers and several European Universities and R &D Institutions. In 2018, a workshop on the concept and operational benefits of the Digital Twins in the steel sector was held in Charleroi [38].

#### *3.1. Digitalization & Enabling Technologies*

Digital technologies are applied in order to improve the flexibility and the reliability of process and to improve the product quality. In addition, they can be applied for monitoring and assessing the environmental performance of processes, improving control of production and auxiliary processes that have an environmental impact, and providing key performance indicators for resources efficiency [39,40]. Some enabling technologies according to the Use Cases of KETS [41] are:


if in the steelmaking plant, existing technologies are enhanced with robots and automation, an improvement of surface quality of the steel products could be achieved [48,49].


• Digitalization of knowledge management. Due to an increasing competitive market, the steel sector has been committed to facing significant challenges in the digitalization. Although this process has already started, further improvement can be achieved. On this subject, the knowledge and experience of the technical staff represents the basis of this improvements. The main barriers about the usage of this knowledge and experience are represented by their heterogeneous distribution over the individual staff members, human obliviousness, and knowledge erosion by leaving staff members.

#### *3.2. Past and Ongoing Research Activities Funded by the Research Fund for Coal and Steel*

In the European steel sector, the most important funding program for the technology development is the Research Fund for Coal and Steel (RFCS) [57]. This funding program concerns aspects related to the innovation in the digitalization of the steel industry as well.

The most active actors among research institutions and steel companies in such projects are: VDEh Betriebsforschungsinstitut BFI, Swerea MEFOS/KIMAB, RINA Consulting-Centro Sviluppo Materiali, Scuola Superiore Sant'Anna, Centre de Recherches Metallurgiques, ArcelorMittal, ThyssenKrupp, as well as Tata Steel and, to some extent, Gerdau and Voestalpine.

The plant manufacturers such as Primetals and SMS Siemag, followed by Danieli (also with its sister company Danieli Automation) are key players to patents and they are seldom involved in those projects.

This study identified 22 RFCS Projects covering aspects of digitalization in the steel industry starting from 2003, considering projects that are ongoing or completed. Figure 1 depicts the identified projects and the enabling technologies that these projects are developing or have developed: In the x-axis, the projects are reported; in the y-axis, the enabling technologies taken into consideration from the project are expressed in percentage.

Among the ongoing projects dealing with the Internet of Things (IoT) systems technology, TrackOpt aims at implementing an automatic ladle tracking system, in order to ensure the tracking of the product from steelmaking via casting to delivery by using a Multi-Objective Optimization (MOO) Framework and big data analytics as well as innovative acoustic sensors.

Quality4.0, NewTech4Steel, Cyberman4.0, PRESED, and Dromosplan are some examples of RFCS Projects related to the Big Data Analytics and Cloud Computing. An adaptive platform is being developed, allowing online analytics of large data streams to realize decisions on product quality and provide tailored information of high reliability in the running Quality 4.0 project. NewTech4Steel is an ongoing project focused on dedicated use cases in the steel industry exploiting all the technological and scientific possibilities offered by the latest technologies concerning data handling and data analysis. Advanced AI and ML-based analytics, also suitable for big data processing, are exploited for process performance monitoring. Cyberman4.0 and CyberPOS deal with CPS. In Cyberman4.0, big-data tools and techniques are applied to merge process and product data in order to forecast quality downgrading, faults, anomalies, and residual life of critical components in order to timely plan suitable and cost-effective maintenance interventions. Cyberman4.0 is also an important example of a Predictive Maintenance approach applied to the rolling area and aims at developing a so-called Integrated Maintenance Model 4.0 (IMM4.0). CyberPOS introduces simulation and verification tools as well as a new IT framework for establishing the feasibility, safety, and benefits of CPPS (Cyber Physical Production System) in the framework of "Steel Industry 4.0 Automation". Moreover, a CPS-based platform for facilities producing long steel products has been developed [58]. PRESED proposed a solution built around Big Data, Feature Extraction, ML, Analytics Server, and Knowledge Management in order to automatically analyze the sensorial time series data.

**Figure 1.** Research Fund for Coal and Steel (RFCS) Projects and the developed enabling technologies.

The projects Dromosplan, RoboHarsh, and Desdemona are related to the robot-assisted production. The ongoing project Dromosplan aims at using Unmanned Aerial Vehicles (UAV) in steel plants in order

to replace the human intervention in a number of operations related to the monitoring, maintenance, and safety. New sensor data are being produced in this context in order to prove and evaluate the benefits of UAVs in steelworks [59]. RoboHarsh firstly introduced some concepts of human–robot symbiotic co-operation in the steel industry for the development of a complex maintenance procedure [60,61]. In this project, one of the main results is that the operator role is changing, becoming a supervisor, and, therefore, there is no replacement of the worker but a safer and heavyweight operation reduction. Desdemona is another example of development of procedures for steel defect detection by robotic and automatic systems such as UAVs and ground mobile robots.

In vertical/horizontal integration technology, some examples of RFCS projects are DynergySteel and the abovementioned Quality4.0. In DynergySteel, the simulation, decision support procedures, and control tools have been implemented at several steelmaking plants to improve power management capability and power engagement forecasting [62,63].

Simulation and optimization of the production line is another enabling technology, which handles in several RFCS projects, such as GasNet, SOProd, AdaptEAF, Cyber-POS, as well as OptiScrapManage. GasNet developed a simulation tool of the network of process gas and steam including their generation and flows as well as a multi-level strategy for their optimization. ML-based tools and technologies such as Echo-State and FeedForward Neural Networks, as well as advanced optimization approaches (e.g., Mixed Integer Linear programming) have been used in order to improve energy efficiency and environmental sustainability of the steelmaking processes [64–68]. In SOProd Objected-Oriented Programming (OOP), Python language, LabView, MongoDB, and Optical character recognition are some of the technologies adopted to improve product intelligence and autonomous machine–machine and product–machine communication [69]. In this project, a de-central optimization considering a detailed product and process knowledge facilitates a process self-optimization by using individual product properties and processing information of neighboring processes [58]. Advanced methods for process monitoring and control through multi-criteria approach of performances indicators, together with optimization approaches, have been exploited in OptiScrapManage. Laser scanner and acoustic and hyperspectral sensors are some of the used technologies. By considering the properties of the charged materials, AdaptEAF developed an adaptive online control for the Electric Arc Furnace (EAF), by optimizing the efficiency of the chemical energy input by reducing the total energy consumption and improving the metallic yield.

The project TeleRescuer provides an exemplar application of a special Unmanned Vehicle (UV) within a system for virtual teleportation of rescuers to subterranean areas of coal mines.

A method for the collection, representation, storage, and utilization of the human knowledge to exploit it in computer-based applications has been investigated and implemented by the project "KnowDec". Here, a new approach based on the methodology of knowledge-based decision support system has been developed. The operators of the quality department can capture the experiences concerning the approval of slabs and the collected experiences are stored in the knowledge base, useful for decision support and advices in similar cases.

Most of the abovementioned projects, such as Cyber-POS, TrackOpt, Quality4.0, DynergySteel, AdaptEAF, SOProd, Desdemona, PRESED, InfoMap, PlantTemp, and AutoAdapt deal with the self-organizing production technologies. PlantTemp develops an operator advisory system covering the electric arc furnace and casting processes, meeting the target casting temperature, by saving energy and material consumption. AutoAdapt proposes an expandable system, which aims to apply self-learning methods for adapting such automations to new products and plants. Genetic Algorithms (GA), polynomial models, iterative learning control methods, and feed-forward control are some of the used technologies. In InfoMap, a tool for objective interpretation of maps from different devices along the process route, generating concise data suitable for use within automatic control/advisory systems is developed. Here, Convolutional Neural Networks (CNNs) are applied for flatness defects detection and classification [70] (see also Appendix A).

IConSys implemented an Intelligent Control Station, in order to support decision making in rolling and finishing while the I2MSteel project developed a factory and company-wide automation and information technology for an intelligent and integrated manufacturing steel [71]. EvalHD investigated some aspects related to the implementation of Industry4.0 [72,73].

Figure 2 provides a summary of the abovementioned analysis by showing the number of RFCS Projects for each of the highlighted enabling technology. The identified enabling technologies are inserted in the x-axis and for each technology the number of RFCS projects, which takes into consideration such technologies, is reported in the y-axis.

**Figure 2.** Number of RFCS projects by enabling technologies.

#### *3.3. Other European Funding Programs for Digitalization and Low Carbon Technologies for the Steel Sector*

Cost pressure, regulatory, and product/service requirements are some of the European steel industry challenges, which the steel industry has to face. The 7th Framework Program (FP7) (2007–2013) and its successor Horizon 2020 (2014–2020) [74] have been included in addition to the RFCS program by the European Union research and innovation funding program. Most of these projects started between 2014 and 2017. Nevertheless, digitalizing the steel industry started before calling these activities Industry 4.0 [75]. On this subject, some projects, starting in the early 1990s and covering some aspect of digitalization of the steel industry, have been identified, for instance BRICK, OREXPRESS, and TAM. All these projects were funded by EUREKA, which is a pan European network for market-oriented, industrial R&D [76].

As far as the FP7 Projects (2007–2013) are concerned, an example is AREUS, which treated integrated technologies for robotic production systems and robotic manufacturing processes optimization environment. WaterWatt and FACTS4WORKERS are some examples of H2020 projects where digitalization is applied in order to remove market barriers for energy efficient solutions and improve the efficiency for managers and workers within Worker-Centric Workplaces for Smart Factories.

The acronym SPIRE stand for "Sustainable Process Industry through Resource and Energy Efficiency" [77] and refers to a Public Private Partnership (PPP) targeting, within the Horizon 2020 program, the European process industries. DISIRE, CoPro, FUDIPO, MORSE, RECOBA, and COCOP are some SPIRE projects facing digital solutions and with specific demonstration in the steel sector.

Concerning other activities, a project on industry 4.0 has been developed at Dillinger. It is a real-time forecasting project for an "adaptive" Basic Oxygen Furnace (BOF), which "learns" and fine-tunes some settings based on the collected process data [78]. Another project has been led by SSAB and has aimed at making available information and instructions relating to any steel item, regardless of where it is produced. Each link of the production chain can use and accumulate information, by creating a basis for both the circular and platform economy [79].

The circular economy concept promotes the reuse, the refurbishment, and the recycling, maximizing the product life and at the same time keeping products and materials at a high level of utility [80] since an important objective defined in EU Masterplan is a competitive and low-carbon European steel industry [81]. Environmental issues such as CO<sup>2</sup> reduction can have several benefits from the KETs application. Advanced process monitoring and increased quality lead to major efficiency. In the field of CO<sup>2</sup> mitigation technologies, the RFCS and H2020 (2014-2020) programs represent the most important instruments for the EU-funded research projects. Low-carbon steel production requires the development of dedicated technologies.

The current pan-European research for the applicable technologies of CO<sup>2</sup> mitigation is focused on three pathways: Carbon Direct Avoidance (CDA), Process Integration (PI), and Carbon Capture, Storage, and Usage (CCU).

According to Figure 3, several EU Projects have been funded in the Process Integration pathway in order to develop technologies for reducing the use of carbon. For instance, ENCOP dealt with the overall energetic optimization of steel plants [33,82,83]. IDEOGAS focused on injection of reducing gas in the Blast Furnace (BF) and top gas recycling. LoCO2Fe developed a low CO<sup>2</sup> iron and steelmaking integrated process route. The identified low-carbon technologies are inserted in the x-axis and for each technology, the number of projects that takes into consideration such technologies is reported in the y-axis.

**Figure 3.** Number of projects related to low-carbon technologies.

CDA technologies mainly consist of iron ore reduction by hydrogen (produced by H2O electrolysis) and syngas from biomass and Fe reduction by electrolysis. Some examples are HYBRIT, which aims at developing the world's first fossil-free ore-based steel-making technology using hydrogen replacing carbon as reductant, while GrInHy targets the production of a green industrial hydrogen via reversible high-temperature electrolysis designing, manufacturing, and operation of a high-temperature electrolyser.

CCU technologies concern the different methods for carbon capture based on chemical/biological processes of CO<sup>2</sup> conversion and CO<sup>2</sup> capturing by mineral raw materials. Within the CCU technologies, the focus is the conversion of industrial CO<sup>2</sup> into biofuels dealing with the transformation of CO<sup>2</sup> resulting from the iron, steel, cement, and electric power industries into value-added chemicals and plastics. Thanks to the Carbon4PUR project, it has been demonstrated that industrial waste gases such as mixed CO/CO<sup>2</sup> streams can be turned into intermediates for polyurethane plastics useful for rigid foams and coating. It is also possible to recycle carbon into sustainable and advanced bioethanol, as

shown in the project Steelanol. FresMe and M4CO2 are other projects dealing with a more efficient CO<sup>2</sup> capture.

### **4. The Future of Digitalization in the Steel Sector**

The steel sector, such as the other European industrial sectors, is committed to understanding the logic of digitalization and, consequently, to implement the digital technologies in its production processes. In the digital transformationm the four levers, resulting from researches carried out on key sectors in German and European economies, which are important for effectively implement the digitalization process, are [84]:


Capturing, processing, and analyzing **digital data** can allow better forecasting of process behavior as well as smarter, easier, and faster decision making. The IoT connects devices equipped with sensors, software, and wireless capabilities, coupled with a growing capacity of data collection and storage. This results in new data sources availability to modern analytical technologies, for pre-processing data faster and in a more detailed way. Concerning the steelmaking processes and products, real-time data allow monitoring both of them. In addition, the use of sensors allows checking a single piece along the production chain: Errors and defects can be easily traced back and eliminated. This can lead to a more efficient production. In addition, data availability and ML, by enabling maintenance work to be anticipated and done before something goes wrong, can produce significant improvements in the equipment maintenance that can be scheduled and remotely checked.

The **automation** concerns the combination of traditional technologies and AI- and ML-based approaches, resulting in systems that work autonomously. In the near future, the automation applications will reduce error rates, increase speed, and cut operating costs. In particular, in the steel sector, the automation of production and consumption will be implemented.

The **connectivity** of separate systems (e.g., interconnecting the entire value chain via mobile or fixed-line high-bandwidth telecom networks) allows overcoming the lack of transparency, resulting in process efficiency improvements. This application in industrial plants is based on the interconnection of production systems facilitated by machine-to-machine (M2M) communications. A better connectivity and data sharing applied to the steelmaking processes aim to reduce some problems linked to remote locations and widespread supply chains. In addition, issues due to market fluctuation and potential hazardous working environments can be overcome.

The **digital customer access** allows the direct access to customers through the mobile internet, providing transparency and new services.

The new reorganization of entire industries and the transformation of business models are enabled through the availability of digital data, automation of production processes, interconnection of value chains, and creation of digital customer interfaces. This can allow steel companies to interact with suppliers and customers in a new and better way.

As far as the implementation of Industry 4.0 in the European steel sector is concerned, some previous and ongoing funded European projects will provide significant results on the digitalization in the near future. Some of them will develop cross-sectorial digital solutions, others are mainly prototype applications and demonstrations. On the other hand, some projects funded by the RFCS can provide further results on the real implementation of the digitalization in the steel sector. On this subject, a work including the publicly funded projects, patent analysis, expert interviews, and a qualitative survey of academics and practitioners related to Industry 4.0 in the steel sector has been performed [36]. Results have shown that transformation of the organizational structure of a company represent the main issues. In addition, Industry 4.0 implementations are required in order to achieve economic benefits to

company developments, particularly, improvements in process efficiency and in the development of new business models. Furthermore, Industry 4.0 will improve effectiveness through intelligent support systems for the workforce and the interaction with customers in the organizational domain. In the research approach, the future SPIRE 2050 roadmap, in preparation by the SPIRE Working Group Digital, forecasts an integrated and digital European Process Industry, with new technologies and business models, which will aim to enhance competitiveness and impact for jobs and growth [85]. In the next five years, investments in innovation and digitization will be necessary [37], in order to achieve a level of digitalization to 72% [86]. By providing an example about the future of digitalization in the steel industry, ArcelorMittal group is working in the adoption common platforms and AI algorithms across the whole group and in different business areas [87].

#### *4.1. Digitalization Impact on the Workforce*

Significant changes, provided by Industry 4.0 to all aspects of industry structures, also include the workforce dynamics, such as the strategic workforce planning, the right organization structure, developing partnerships, and the technological standardization. The main future directions could be from a human labor-centered production to a fully automated work as well as from monotonous and physical activities to creative ones [88]. Nevertheless, negative changes could also occur, including higher unemployment and widespread workforce de-skilling.

In Germany, in overall sectors, about 23% of the workforce do not have vocational qualifications and in the manufacturing industries, 1.2 million are low-skilled workers. It has been assumed that low-skilled work will be up-skilled, as digitalization will upgrade simple and low-skilled activities and, at the same time, skilled activities will be continuously enhanced [89]. In this context, the industrial low-skilled work will not disappear but rather the level of qualification will steadily rise. It has been foreseen that jobs and skills will be polarized. This thesis consists in the automation of middle-skilled jobs by the use of computers. On the other hand, digitalization increases the productivity of the most skilled jobs, while the low-skilled jobs survive as they cannot be automated. This is because the automated work is concentrated in the middle of the skills distribution [90,91]. In addition, under conditions of digitalization, four development paths for low-skilled work have been identified [90]. The general erosion of low-skilled industrial work and the common idea that simple, routine tasks threatened by the new technologies will probably disappear in the longer term, can be considered only one scenario. In the second path, "upgrading of low-skilled industrial work", a strategy for improving technological product, is paired off with a highly flexible marketing. The third scenario, characterized as "digitalized low-skilled work", shows a high-intensity application of digital technologies and new forms of work (e.g., "crowdsourcing" and "crowdworking") and may also be associated with new forms of low-skilled work. In the fourth scenario, "structurally conservative stabilization of low-skilled work", there is no discernible change in existing employment and organizational structures. The different scenarios show that the potential job losses due to the implementation of the new technologies is controversial. In addition, the consequences for job activities and qualifications are interpreted as the "upgrading" or "polarization" of skills. Nevertheless, significant changes depend on the kind of technology automation and on its implementation process. Consequently, in the medium term, a limited spread of digital technologies is expected [92].

The progressive process of digitalization and automation produce effects and impacts on the employment in the industrial sectors, included the steel industry. In particular, the application of robotics and computerization will increase the creation of new jobs, particularly in IT and data science [14]. For instance, it has been shown that in Germany, only 12% of jobs are endangered by digital automation [93]. In Europe, over 1.5 million net new jobs have been created in the industrial sectors since 2013, with a growth of labor productivity of 2.7% per year on average since 2009, higher than both the US and Korea (0.7% and 2.3%, respectively) [23]. According to the European Centre for the Development of Vocational Training, between 2016 and 2030, over 151 million job openings are expected, with 91% due to the replacement needs (i.e., retirement, migration, movement into other

occupations, or workers temporarily leaving the workplace) and the 9% due to new job openings. In the same period, over 1,750,000 jobs will be opened for ICT professionals [94]. Nevertheless, 2.6 million people worked in skill shortage occupations. During 2013, 47,000 vacancies have been estimated, including 25,600 reported as hard-to-fill by employers and around 23,500 as hard-to-fill, because of the lack of skills required [95]. In this regard, companies should develop their future workforce and adopt new business models and organizational structure, in the perspective of Industry 4.0 [96]. While employees need to be re-skilled, according to the requirements of digital economy, new employees need to be educated, according to the requirements of future jobs and skills. For this reason, the achievement of an up-skilled and re-skilled workforce is possible by implementing training programs, based on digital and business topics. This can be done through a life-long learning approach for addressing digital skills, and continuous training activities represent the key aspects for the companies to achieve a successful future [97]. However, companies also face some issues, such as skill mismatches, that refer to a failure of skill supply to meet skill demand, resulting in stopping the economic growth and in limiting the employment and the income opportunities of individuals [98]. The needs of companies, including the steel sector, are mainly focused on horizontal skillsets instead of high specialized profiles, in order to have a workforce flexible and able to move across multiple tasks. The companies need to have stronger horizontal skillsets rather than highly specialized profiles; in particular, workers with transferable skillsets in order to provide a good level of flexibility and coordination across different departments of their companies. In addition, it becomes increasingly important for companies to have employees who are able to move across multiple tasks and intervene in different areas. In addition, due to current job insecurity, transferable cross-functional skills represent a possibility for a greater security for workers [5]. In the process industries, including the steel industry, although 40,000 jobs have been lost in recent years, due to restructuring [99]; digitalization can provide new flexible skills and a workforce able to fast learn new digital technologies. In this context, cognitive sciences play an important role to provide support, combining awareness and knowledge with advanced control algorithms and optimization [100]. In the Industry 4.0, ICT skills are more important than core skills for employees. In particular, employees should not only have hard-skills, but also soft-skills such as collaboration, communication, and autonomy to perform their jobs in hybrid operating systems. In addition, employees should be able to be adaptable to continuous learning in an interdisciplinary perspective. Concerning engineering, the new education requirements are focused on achieving information and knowledge applicable to the business environment, and different disciplines should be able to work together. Through the design of new integrated engineering programs, the gap between universities and the business environment can be overcome. In addition, working in interdisciplinary teams, realizing interdisciplinary tasks, and providing interdisciplinary thinking represent key aspects for the implementation of Industry 4.0 research areas, such as mechatronic engineering, industrial engineering, and computer science [22].

#### *4.2. Digitalization and Economic Impact*

The digital economy can offer new opportunities to companies, including the steel sector. It is important to better understand how digitalization is changing the rules of competition, in order to optimize existing business models and to develop new ones. Due to a growth of the third country imports by 16.3% year-on-year, in the final quarter of 2018, a decrease of the domestic deliveries from EU mills to the EU market compared with the same period of 2017 has been revealed [101]. Economic and steel market outlook 2019-2020 European steel is squeezed between rising import pressure and a depressed home market. The main reason for the weakening of the EU economy in 2018, which will at least persist over the first half of 2019, has been the slowing global economic momentum and the related deteriorating contribution from net trade. A digital economy can be successfully achieved through a pan-European coordination based on a harmonized EU-wide approach. On this subject, different actions have to be implemented and, in particular, it is important to outline common standards at European level as well as to share ideas, knowledge, and experiences. A connected economy needs to

rely on a strong infrastructure, in order to connect plant and machinery in an extensive and secure way. The digital transformation of the European manufacturing sector should be quickly achieved, in order to increase competitiveness and limit the new competitor actions. Reduction of energy and raw material consumption, lower OPEX, and reduction of losses as well increase of product qualities and productivity are the most important factors related to the innovative technologies in Industry 4.0 [102]. In [103], the recorded scrap information is transferred to EAF for the calculation of the optimized and best melting condition thanks to the detection and recording of volume and weight for each layer of scrap in the bucket. The raw materials, in fact, are a crucial factor and reducing their cost is more effective than acting on the transformation cost. According to [104], the main implementation areas in manufacturing are real-time supply chain optimization, human robot collaboration, smart energy consumption, digital performance management, and predictive maintenance. Especially the predictive maintenance, according to [104], can help not only increasing revenues, by reducing the maintenance costs from 10% to 40% and by reducing the waste from 10% to 20%, but also optimizing planned downtime, limiting unplanned downtime, and an estimation of a reduction of the operating cost by 2% to 10% is also foreseen. Moreover, digital technologies and ML can be useful in the metals industries in order to avoid unplanned shut down time to repair or replace key components, since such breakages are extremely costly. By using predictive maintenance methods, actuators can be replaced before they break [102]. The advanced analytics techniques like AI and ML can automatically help for the quality issues defining the basic causes, optimizing the optimal recipes for new products/grades, and by reducing the rejection rate [105]. The tools exploited in [106] facilitate the production planning by adopting AI and ML and help to improve due date reliability improving the overall economic success of the steelmaking company.

#### **5. Conclusions**

Although the steel production is already partly automated, the application of new technologies can further sustain the optimization of its entire production chain. This will allow the steel industry to become smarter in evolving towards Industry 4.0. The implementation of digital technologies, by continuously adjusting and the optimizing the processes online, contributes to improve the flexibility and the reliability of processes, maximizing the yield, and improving the product quality and the maintenance practices. In addition, they also contribute to increase the energy efficiency as well as to monitor and to control the environmental performance of processes in an integrated way.

The analysis reported in this review paper highlights that the challenge of digitalization consists of the integration of all systems and productions units, through three different dimensions: Vertical Integration (Integration of systems across the classic automation levels from the sensor to the ERP system); Horizontal Integration (Integration of systems along the entire production chain); Life-cycle Integration (Integration along the entire lifecycle of a plant from basic engineering to decommissioning) [30] and the Transversal integration (based on the decisions taken during the steel production chain, taking into account technological, economic, and environmental aspects). The digitalization process also requires jobs based on interdisciplinary teams, tasks, and thinking, to provide interdisciplinary skills. These achievements can be possible by integrating new IT, automation, and optimization technologies. Furthermore, Predictive Maintenance techniques can be implemented by equipment monitoring combined with intelligent decision methods. In this context, the application of Data Mining techniques, also based on ML, can allow anticipating maintenance work and scheduling it. In addition, Knowledge Management is a key factor for achieving improvements in the digitalization process.

The future expectations for the steel industry about digitalization include the optimization and the interactions of the individual production units, within the entire production chain (and beyond). This will allow reaching the highest quality, flexibility, and productivity. Furthermore, the following digitalization applications will represent the most important trends in the future: Adaptive online control, through-process optimization, through-process synchronization of data, zero-defect manufacturing, traceability, and intelligent and integrated manufacturing.

In the coming years, in order to achieve a successful implementation of digitalization, the steel sector has to afford some important challenges, such as the standardization of systems and protocols, work organization and more skilled workers, investments, and research aiming to adopt appropriate frameworks. The implementation of digitalization is expected to generate productivity effects in the industrial sectors and growth in the economy. Concerning the potential consequences of digitalization for industrial workforce, on one hand, new technologies can cause job losses, but on the other hand, higher qualifications can be achieved. Nevertheless, changes will depend on different factors and they are expected to occur in the medium term, leading to some impacts on the industrial workforce. In addition, the steel sector needs to produce within environmental constraints in order to achieve its sustainability. In particular, the pressure of the environmental constraints represents a challenge for the steel sector to implement digital technologies that can help cope with the increasing trend in energy demand and the requirement of adopting low-carbon energy systems. In the coming years, the steel sector should be able to achieve zero waste, zero climate change emissions, and use half its current resources. On this subject, digital technologies can play an important role to enable improvements in sustainability performance, to plan processes in order to better account for demands and opportunities offered by industrial sustainability, and to enable the experimentation with new business models. The transformation of processes for significantly reducing emissions and improving energy efficiency will lead to the circular economy paradigm achievement and, on the other hand, adopting high-performance components, machines, and robots will optimize the materials and energy consumptions.

However, digital transformation and the full implementation of new digital solutions will only be successful if non-technological aspects are also considered in the technological development and implementation, such as framework conditions at European, national, and regional level, market and consumers, human resources, skills, and labor market. These aspects are integrated in the new SPIRE Roadmap 2050. Here, human resources and new (digital) skills especially will play a crucial role for unfolding the potential of new solutions within the companies.

**Author Contributions:** Conceptualization, V.C, T.A.B., and M.M.M.; methodology, V.C. and M.M.M.; validation, V.C. and A.J.S.; formal analysis, T.A.B. and M.M.M.; investigation, T.A.B., B.F., and E.S.; resources, V.C. and A.S.; writing—original draft preparation, T.A.B., B.F., and VC; writing—review and editing, M.M.M. and A.J.S.; visualization, B.F.; supervision, V.C.; project administration, A.S. and V.C.; funding acquisition, A.J.S. and V.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the European Union through the Erasmus Plus Programme, Grant Agreement No 2018-3019/001-001, Project No. 600886-1-2018-1-DE-EPPKA2-SSA-B.

**Acknowledgments:** The research described in the present paper was developed within the project entitled "Blueprint "New Skills Agenda Steel": Industry-driven sustainable European Steel Skills Agenda and Strategy (ESSA)" and is based on a preliminary deliverable of this project. The ESSA project is funded by Erasmus Plus Programme of the European Union, Grant Agreement No 2018-3019/001-001, Project No. 600886-1-2018-1-DE-EPPKA2-SSA-B. The sole responsibility of the issues treated in the present paper lies with the authors; the Commission is not responsible for any use that may be made of the information contained therein. The authors wish to acknowledge with thanks the European Union for the opportunity granted that has made possible the development of the present work. The authors also wish to thank all partners of the project for their support and the fruitful discussion that led to successful completion of the present work.

**Conflicts of Interest:** The authors declare no conflict of interest.
