Workforce and supply chain disruption as a digital and technological innovation opportunity for resilient manufacturing systems in the COVID-19 pandemic

3.1. Human-centered cyber-physical production systems in times of pandemics

Despite manufacturers significantly increased their investments in digitalization over the last few years (World Economic Forum, 2018) and digitalized manufacturing firms displayed higher resilience and adaptability than manufacturers with lower digital adoption (Okorie et al., 2020), production environments are still inadequate to cope with black swan events and are not as agile and resilient as expected. Over the last few years, Cyber-Physical Production Systems (CPPS) came out as a combination of technological enablers of both physical and digital nature, operating successfully alongside humans (Pinzone et al., 2020). While manufacturing is increasingly automated, there is now a greater demand for more effectively integrating, rather than eliminating, human cognitive capabilities in the loop of production related processes (Krugh and Mears, 2018, Golan et al., 2020). There is general consensus that humans have a central role, as they are the only ones who can govern production systems, address anomalous situations and can provide flexible solutions in case of need (Di Nardo et al., 2020, Fantini et al., 2020). The recent trend is not creating unmanned production facilities, but building human cyber–physical production systems (H-CPPSs) where machines are not designed to replace the skills and talents of humans, but rather to co-exist with, and to assist humans in being more efficient and effective via smart tools and assistance systems (Romero et al., 2016). In this context, the “Operator 4.0” is envisioned as a smart worker, empowered with Industry 4.0 technologies, capable to control decentralized production resources, access intuitively a wealth of knowledge about the factory and receive cognitive support when needed to achieve more effective sociotechnical systems (Emmanouilidis et al., 2019). Recent research has proposed Human-in-the-Loop (HIL) integration in CPPS to empower a workbench with intelligent decision and assistant systems (Costa et al., 2019), to prevent cognitive overload for human operators and enhance decision-making capabilities thanks to a range of automated decision recommendations (Gross et al., 2017), to integrate spatial augmented reality in manual working stations to project technical information on a motorbike engine during a maintenance procedure (Uva et al., 2018). Other applications of intelligent tutoring systems integrated with natural interaction interfaces for operators’ support have been also tested (Longo et al., 2019b). However, research is still primarily confined to onsite operator support and does not encompass workforce management at large or remote working. Further development of methods and technological solutions is required as humans are now being recognized to be a major enabler for the Factory of the Future (Emmanouilidis et al., 2019).

The capacities that a H-CPPS may offer and recent societal challenges of global disruptions, such as the SARS-CoV-2 pandemic, mean that we must re-think how to exploit the maximum potential of the arising “cyber-physical factories” and “digital twin environments”. Two challenges can be identified:

• ▪

  if not properly implemented, high levels of automation and digitalization may result into inflexible and rigid production (Frohm et al., 2008), unable to quickly adapt to changes arising from disruptive events, such as a pandemic. A corollary impact of the pandemic is the need for more flexible human resources (in terms of capabilities and requirements) and production processes to enable more resilient systems. Resilient factories and flexible H-CPPS need to bring together “human competences” and “CPPS autonomy” in order to unlock flexible problem-solving and rapid adaptation to a failure event (Ansari et al., 2018);

• ▪

  the paradigm of socially sustainable workplaces (Hancock et al., 2013) and anthropocentric production systems (Rauch et al., 2020) are today compelled to develop further the topic of sustaining a competitive, healthy and safe workforce. This is particularly true if we match the ageing phenomenon and demographic trends that are affecting the manufacturing workforce (Calzavara et al., 2020) and the devastating impact of the SARS CoV-2 pandemic on older adults. However, while several manufacturers and logistics companies opted for repurposing temporarily their production processes (Okorie et al., 2020) and used Additive Manufacturing (AM) equipment to address critical product shortages (Longhitano et al., 2020, Tarfaoui et al., 2020), very few tested smart technologies to sustain a healthy and safe workforce on the shopfloor (Naughton, 2020, Vincent, 2020, Wuest et al., 2020).

As reported by Kadir & Broberg (2020), in some large companies, the health and safety departments had been involved in the design of innovative digital work, but only after the new digital solutions had been fully developed and implemented. Fantini et al. (2020) argue that emerging needs for work design are not thoroughly addressed by the available studies on CPS-enabled manufacturing.

3.2. The healthy operator 4.0

In the context of H-CPPSs, the increasing attention to digitally-enabled Occupational Health and Safety aspects (OHS) gave birth to a specific sub-type of the Operator 4.0: the Healthy Operator 4.0 (HO4.0). Envisioned initially as an industrial worker using wearable devices to track health-related metrics (e.g. heart rate or posture) (Sigcha et al., 2018), the HO4.0 concept evolved today towards a holistic perspective on the workforce health management and analytics. Along with smart wearables and ambient intelligence, a new generation of smart exoskeletons, adaptive co-bots and smart personal protective equipment (PPEs) is also emerging (Romero et al., 2018). Prospectively, the integration of autonomous robots, AR and VR, simulation, I-IoT and automation technology, and novel networking technologies such as 5G, will open new possibilities in the next future to sustain a healthier and safer workforce through cognitive e-Health platforms (Zolotová et al., 2020). Sun et al. (2020) envisioned a HO4.0 that gathers the workers’ and environment’s real-time information and fuse them appropriately to develop a digital image of the operator’s behavior, thus enabling what Monostori and Váncza (2020) call a “predictive maintenance of operational staff”. This paradigm can exploit the “caregiver” functionality of a CPPS, theorized by Pinzone et al. (2020), that aims at minimizing the negative impacts of work tasks on the operators’ health while improving their productivity (by providing real-time instructions and searching for information).

Despite a significant number of employees is still fighting the adoption of HO4.0 technologies for multiple reasons (e.g. limited privacy, required upskilling to master them) or their use declines after some initial enthusiasm (Mattila et al., 2013), the recent SARS-CoV-2 pandemic rekindles the concerns on the workers’ well-being and safety in industrial workplaces and fosters the debate on the HO4.0. In the post-pandemic era, we feel that the adoption and acceptance of the HO4.0 technologies will receive a boost as industrial workers will raise awareness on the need to maintain high safety levels at workplaces. This paper looks exactly in this direction.

3.3. Supply chain in times of pandemic

While COVID-19 has unearthed the limitations that pandemic causes on plague the human workforce in the Factory 4.0, it has also highlighted the need for innovative solutions to cope with sudden material shortages or supply chain breaks. During the first COVID-19 wave, several supply chains highly reliant on China’s economy understood the need to avoid over-reliance on any one source or geographical location and started to decouple from China and find other back-up suppliers, wholly independent from them. However, given the large scale impact of a pandemic, even back-up suppliers or subcontractors may be overloaded or unavailable. What can manufacturing firms do when supply chains are interrupted and buffer stocks are lacking?

Research on adaptability and mathematical modeling has been a topic of interest for decades, yet is more important than ever today (Bottani et al., 2019). While few computational studies started to appear in the SC theory, helping decision-makers of high-demand and essential items to make accurate and prompt decisions in designing the revised production plan to recover during a pandemic (Paul & Chowdhury, 2020), recent literature proposes that companies should conduct a comprehensive mapping of their supply chain in order to predict and prepare for disruptions like a pandemic from the supply side (Sheffi, 2020). A transition toward a “capability-based” and “technology-enabled” digital supply network (DSN) can provide the required visibility and agility to face future interruptions and unpredictable black swan events (Ivanov & Dolgui, 2020). This calls for the need of a simulation-based Supply Chain Digital Twin, that companies can use to effectively plan and mitigate risks in the event of disruptions like the COVID-19 (Di Nardo et al., 2020).

Company management can integrate within the industrial organization part of the process which is commonly externalized. In this way, although a lack of flexibility and a further complexity of the organization arises, the system results globally more resilient.

In the I4.0 paradigm technology helps the management introducing some processes which are at the same time flexible and able to substitute the conventional ones, without the need of strong workshop equipment and organization.

More in detail, when material shortages became significant, the Additive Manufacturing (AM) community stepped up and tried to fulfill local demands. Anyone with access to maker equipment, either privately or as part of an organization, contributed to the production of PPE (Mueller et al., 2020). 3D printing was validated as an efficient agile production technology in times of pandemics, supported by the fact that it is relatively cheap to adapt towards changes in design or parameters. Manero et al. (2020) have explored the feasibility of exploiting additive manufacturing, like three-dimensional (3D) printing, to overcome such issues. Utilizing various real-world examples, the authors have illustrated the hypes and hopes of 3D printing and how it can be deployed for counteracting and mitigating the burden generated by COVID-19. Similarly, Salmi et al (2020) have concluded that 3D printing represents a “promising open source solution, especially during emergency situations”.

3.4. Summary of current challenges and gaps

Overall, the above-mentioned streams of literature provide different and complementary viewpoints on manufacturing systems’ resilience, even in relationship with technological systems, but leave gaps. As a matter of facts, there is a lack of guidelines, methods and solutions that can drive the Factory of the Future to cope successfully with black swan events like a pandemic. In particular:

• Gap 1: Research is primarily confined to onsite operator support and does not encompass a holistic view on workforce management or remote working.

The main challenge is to enable an ageing shop-floor workforce with the tools to “bring their work home” when possible and investigate work design structures in the post-pandemic era.

• Gap 2: Literature on resilient factories and flexible H-CPPS lacks of technology-enabled solutions and work design strategies that are recommended to manufacturing firms to cope successfully with staff deficiencies in black swan events like a pandemic.

The pandemic showed us how companies navigated perilous waters trying to understand what to do to cope with the pandemic effects. The next challenge is not only to understand how to support onsite shop-floor operators with novel 4.0 technologies but also understand how to provide heterogeneous workers with tools and solutions that allow them to fit quickly in the production process and ramp up to their full operational productivity.

• Gap 3. Literature on the Healthy Operator 4.0 is still at an early stage and should be considered from a holistic point of view.

There is a need to sustain a healthy, competitive and safe workforce, but current solutions and approaches seem incompatible with the challenges of a pandemic. Further concepts and solutions must be investigated in the perspective of an enhanced well-being and safety of the workforce.

• Gap 4. Literature on SC digitalization and optimization is mainly focused on its efficiency.

Some events such as a pandemic may have, as a consequence, the supply chain breaking. In this case company have the need to redesign the SC in order to overcome the material shortage.

• Gap 5. Literature on Industry 4.0 presents the main paradigm pillars allowing a firm digital transformation oriented to data generation, analysis and control.

I4.0 has within the seed to increase companies’ resilience even if the literature stresses the role of pillars to increase both efficiency and control capacity. Further efforts on the use of techonologies allowed by the digital ecosystem to increase resilience are required. At the same time, company may show an antifragile behavior, according to the innovation boosting due to the difficulties to manufacture pre-pandemic products.

In the next section, a methodological contribution to address these needs is presented.

5.1. Strategies for a resilient workforce in a pandemic

Starting from an extensive literature review and analysis of documents, companies’ reports, specialized magazines and online news, seven strategies have been identified to allow a resilient workforce in industrial workplaces. The first three strategies directly affect the workforce capacity:

• 1.

  Contingent labor (S1)

In industrial settings facing acute short-term workforce shortages, manufacturers may employ contingent labour to increase their workforce capacity. This is a growing trend as companies worldwide expect to rely much more on individual freelancers or temporary workers from the gig economy to cover production and working capacity in the next future (McKinsey Global Institute, 2020). This includes not only agencies providing temporary workers but also cross-industry talent exchange, which means redeploy temporarily select talent with similar competences from external industries facing reduced demand or shutdown to those facing a shortage of workers. However, getting any temporary workers up to speed quickly comes at a major risk of them not being sufficiently trained. Temporary workers are not familiar with processes and procedures of the “hiring” industry/firm and the learning curve would not be compatible with the need to have the perfect person, made to order, ready for their position and ready to go.

• 2.

  Personnel flexibility (S2)

In case of disruptive events, specific work cells, departments (e.g. purchasing, logistics, production) or facilities may be hit more than others. In this case, firms could redeploy temporarily experienced internal staff from low- to high-burden processes to stem worker shortages arising from COVID-19. This solution would allow the company to rebalance the workforce capacity and cope with delays on some urgent due dates and important customers’ orders. However, even in this case, redeployed workers would need training or upskilling to familiarize with different processes, different materials, different machine tools, etc.

• 3.

  Workforce reduction (S3)

Unlike S1 and S2, short-term reduction of onsite workers may be necessary. The rationale is to temporarily reduce the workforce capacity due to orders plummet or unsafe working conditions on the shop floor. Given to the nature of human work, shop-floor operators are forced to stay home on a (usually paid) leave and cannot “bring their work home”. As a consequence, the company records losses due to production stoppages and inactive workers.

The next two strategies are referred to shift flexibility, that in some cases might allow the entire (or almost) workforce to work onsite safely:

• 4.

  Occasional overtime (S4)

In critical situations where the workforce is stretched thin and it is hard to find other skilled workers, employers may ask selected workers about their availability to do occasional overtime and increase their shifts (or working hours). In this case, although proper overtime compensation and incentives are paid, this strategy stands against the principles of harmonic (or human-centered) innovation, as it would have serious impacts on the employee’s well-being (and indirectly on their productivity).

• 5.

  Flexible and rotating shifts (S5)

The alternative to S4 is to refine production schedules and rotate work shifts flexibly on a, e.g. daily, basis depending on the status of work, the inventory available, changing demand, available workforce and what they are capable to build. That would allow firms to reduce the crowd in the shop floor in case of pandemics and leverage on a workforce that is willing to have rotating shifts or even work overnight when needed (Zucchi et al., 2020). This approach would require not only advanced planning and scheduling processes, but also would impact seriously on the employee’s well-being.

The strategy related to the decrease of number of working hours (or shifts) is here treated as S3, therefore it is not considered separately in this list.

• 6.

  Physical distancing (S6)

Physical distancing (PD) measures are proved to reduce substantially the spread of SARS-CoV-2. To ensure proper PD at the workplace, layout changes or designated one-way pathways may be successful strategies to ensure a resilient workforce in case of pandemic. When this is not possible, some workplaces may adopt 'cohorting' to reduce the number of people each worker has contact with (Shaw et al., 2020). Depending upon the constraints of a given production or logistic process, such work bubbles might be created in a variety of ways, either through physical separation within a workspace or temporal separation via a rotating work schedule (see S5), thus eliminating the overlap between different cohorts. However, high risks of infection still exist: not only PD protocols may not be respected by the workers, but in some cases PD is not even possible (e.g. collaborative assembly work).

• 7.

  Hygiene & health monitoring practices (S7)

Employers need to promote cleaning and sanitation practices, the correct use of Personal Protective Equipment (PPEs) and accelerate the adoption of staff health monitoring and screening procedures (e.g. temperature monitoring, mandatory or recommended ad-hoc testing carried out at periodic intervals). However, such solutions may be expensive and difficult to roll-out at scale (e.g. swab testing).

5.2. A digital toolbox for a resilient workforce

Every form of technology-enhanced cooperation (human–human and human–machine) will be integral part of post-pandemic production systems.

New digital approaches can accelerate the capability-building process and allow employees to develop new skills remotely. Such techniques include the remote delivery of training using e-learning systems or the use of virtual-reality technologies to familiarize operators with new tasks or plant layouts. Augmented-reality systems help shop-floor staff to receive training, advice, and support from remote colleagues. Specialist contractors can use such systems to guide shop-floor staff through machine maintenance or troubleshooting.

The application of digital twins, together with advanced human–machine interaction, automation technology and prescriptive engineering can lead to anticipatory rather than reactive systems capable to stand up to the disruptive impact of black swan events on the workforce. This section of the paper shows how a digital toolbox, defined by the authors and intended as a set of technological principles, tools and solutions inspired by the latest CPS-enabled manufacturing and Operator 4.0 paradigms, allows to overcome the challenges and limitations of the workforce resilient strategies presented in Section 5.1. The digital toolbox recommends three work design approaches for manufacturing and industrial companies who are willing to turn the pandemic in an opportunity to build a resilient workforce, digitalize their processes and innovate in a human-centered perspective, namely:

• i.

  the Plug-and-Play worker (Section 5.3) addressing in particular S1, S2 and S3;

• ii.

  the Remote Operator 4.0 (Section 5.4), addressing S4 and S5;

• iii.

  the Predictive Well-being of Operational Staff (Section 5.5), addressing S6 and S7.

5.3. The “plug-and-play” worker

The pandemic impact on the workforce compelled companies to find people that could replace their actual workforce on the short term. However, the time-horizon of hiring is pretty much linked to the time-horizon of learning. Smart assistance technologies, natural human–machine interfaces and augmented reality have proved to speed up significantly the learning process by allowing workers to fit in a process and be fully operational in no time (Longo et al., 2017). Fitting in and ramping up to desired performance in no time – hence the concept of Plug-and-Play worker – thanks to Industry 4.0 technologies are the requirements for an agile and resilient workforce. Developing skilled employees for an “on demand” age is an urgent priority. We answer here the three hows that people would likely ask (Fig. 4 ):

• 1.

  How many plug-and-play workers are necessary? To forecast and estimate the workforce capacity (as a combination of employees already onsite and plug-and-play workers), simulation-based digital twin models combined with epidemiological models, big data analytics, predictive models and AI-powered prescriptive engineering could support the identification of critical workforce needs. This also includes the estimation of the impact of absenteeism and various levels of workforce availability and productivity on production schedules. AI-powered prescriptive tools can recommend the right amount of plug-and-play workers to hire and how to allocate them based on the expected workload and workplace safety.

• 2.

  How can plug-and-play workers fit in a new production environment and be ready to get onboard? For plug-and-play workers to be able to confront with and fit in an increasingly complex cyber-physical production environment, upskilling and continuous professional development are required to prepare workers. The pandemic teached us the importance of operators with digital skills as a critical requirement for the long-term sustainability and survival of businesses. Virtual reality- (VR-) or augmented reality (AR-) based training or serious game-based simulation can help to keep a “ready-to-go” plug-and-play workforce, that will be able to experience multiple manufacturing and industrial scenarios, experiment with tools and processes, and speed up the learning process even if they are not directly onsite.

• 3.

  How can plug-and-play workers ramp up their operational performance rapidly? In order to be fully operational and work safely in a short time, knowledge-based AI-powered assistance on-the-job, digital work instructions and natural human–machine interfaces are today powerful approaches supporting low-skilled opertors on the job in the perspective of a learning-by-doing principle. Augmented Reality based assistance or intelligent voice-based information retrieval supported by machine vision and ambient intelligence can considerably support task execution, reduce the errors, avoid waste and speed up the decision-making process (thanks to the semantic analysis of how-to manuals or other data sources – including the digital twin of the manufacturing plant – that would compress a large amount of information in an easy-to-understand format). Next to digital support, plug-and-play” also have the possibility to harness an advanced human–human collaboration with remote workers via handheld or wearable technologies, i.e. direct feedback and advice from remote experts (e.g. workers in quarantine) through teleconferencing or hologram based augmented reality.

5.4. The remote operator 4.0

Current circumstances have prompted firms world-wide to quickly adopt remote working or work-from-home structures. However, this is not a viable option in the manufacturing industry (Okorie et al., 2020). Unlike office workers, shop-floor workers (e.g. machine operators, maintenance technicians, quality control experts) cannot “bring their work home”. Despite some avantgarde manufacturing companies started to experiment 24-hour production with fully automated processes and unmanned night shifts (Rico, 2020), human presence is still necessary at the factory floor (cf. Section 3). This real example demonstrated that remote working for shop floor operators is possible though. In particular, shop-floor workers are expected to assume more and more the role of problem-solvers, decision-makers and managers of a robotic and digital workforce. The Remote Operator 4.0 is therefore a new type of worker that carries out work in a symbiotic interaction with technology at any time and at any place. We are living the digital era where technology allows us to “make every place our workplace” and consider our home like an extension of the factory floor (see Fig. 5 ) thanks to immersion technology and the possibility to interact with digital objects in a natural and intuitive manner.

Indeed, there are multiple examples of manufacturing environments where the remote working for shop floor operators is possible (Remote Operators 4.0). First of all, it is worth mentioning that the shop floor operator is a person that is required to work within the shop floor (e.g. at specific workstations) and/or to move around the shop floor. In each manufacturing environment, no matter which is the production type (custom manufacturing, intermittent manufacturing, continuous manufacturing and flexible manufacturing) or the industry categories (e.g. oil & gas, chemical, electronic and electrical equipment, metal industry, aerospace, furniture, etc.), the following are some example of operators take an active role on the shop floor:

• –

  The Production Manager;

• –

  The Manufacturing Cell Manager;

• –

  The Squad Responsible;

• –

  The Maintenance Manager;

• –

  The Quality Manager.

There is no need to elaborate a different approach for these workforce groups. Thanks to 4.0 technologies and CPPS these operators can act remotely for:

• 1.

  Supervise

Supervise can be intended as remote monitoring and situation awareness that is enabled by:

• o

  Industrial Internet of Things (I-IoT, to receive data from shop floor machines and equipment);

• o

  Cloud service (to provide remote access to collected data);

• o

  Ad-hoc network connection (including local WI-FI coverage within the shop floor and high speed internet connection to provide data without latency and delays outside the shop floor);

• o

  machine vision and intelligent cameras (to see machines working productively and onsite operators at their own places);

• o

  web applications and dashboards (to summarize data and get desired information even faster comparing to walking around in the shop floor);

• o

  virtual reality to be physically immersed in 3D Digital Twin of the shop floor (in this case even a virtual “walking around mode” is allowed to provide the operator with the feeling to be in the real shop floor);

• o

  augmented and mixed reality. The Remote Operator 4.0 may define and send information and data to on site operators and support them to solve problems within the shop floor (remote assistance and remote troubleshooting). Onsite operators visualize data thanks to Augmented and/or Mixed Reality Applications This approach is currently revolutionizing the manufacturing as well as the service industry, as external employees (e.g. maintenance responsible) will no longer need to access the facilities. The authors are currently working to a completely new and innovative approach including both on-site and remote operators (this will be proposed as part of a new article).

• 2.

  Understand

Understand can be regarded as remote analysis & Decision making. The data-driven manufacturing paradigm is the key enabling factor of a remote workforce. The combination of big data analytics, artificial intelligence, simulation, predictive and prescriptive analytics, and optimization algorithms opened a whole new paradigm characterized by the possibility to forecast future events, process large amounts of data and receive facts-based recommendations to enhance and speed up the human decision-making process. In this framework, Explainable AI (XAI) for example is the new trend that allows humans to understand what AI is recommending and why. Thanks to computer-based information processing and remote human–human collaboration enabled by teleconferencing tools, the Remote Operator 4.0 will be able to fully understand what is happening in a manufacturing plant kilometers away from his/her location.

• 3.

  Interact

Working in the factory floor means operating a closed loop system, where process data help the human worker to interact with physical manufacturing objects to regulate and control manually the process. In the future, the Remote Operator 4.0 will be able to use advanced HMIs on their personal devices (e.g. handheld devices or computers) to run the manufacturing process. Not only new interaction paradigms – such as gesture-based or vocal interaction or even tactile feedbacks to simulate the feeling to hold something in the hands – but also Virtual and Augmented Reality, intelligent digital dashboards, high-speed and secure communication enabled by high speed internet connection, will allow the workers to manage a robotic, automated and digital workforce remotely in an efficient and safe manner.

The Fig. 6 show an example of Virtual Reality Environments jointly developed by MSC-LES lab and CAL-TEK (that respectively are the University lab of the lab spin-off company where some of the authors work at). This virtual environment shows a part of the shop floor environment (a warehouse) and the operator (remotely connected) that is carrying out remote monitoring and situation awareness. In this case the Virtual Environment is fed with data representing the entities in the real environment (e.g. the forklifts, the parts being moved, etc.).

The Fig. 7 shows the remote operator (left side) defining information and data to be sent to the onsite operator (right side); the latter is using a Mixed Reality application through Microsoft Hololens©. In this case, the remote operators can send specific information to the onsite operator with the aim of carrying out remote assistance and remote troubleshooting. The Mixed Reality application (see right side of Fig. 7) can be also used in a stand-alone mode by an onsite operator (receiving data locally from the company informative system) or by an remote operator (as explained later on in section 5.6).

5.5. Predictive well-being of operational staff

Smart working (or working from home) surely represents the best pandemic mitigation measure, but is not always applicable: in this case, technology can help ensuring the highest safety level at the workplace. The COVID-19 pandemic represents an incredible opportunity to rethink the HO4.0. With the right hardware, analytics infrastructure and software, organizations can sustain a healthy, safe and competitive workforce.

In the first place, measures to prevent the transmission probability do not only include PD but also the minimization of unnecessary touching of potentially contaminated surfaces and sharing of physical spaces. Increasing digitization and use of personal mobile devices connected to the enterprise local network would allow workers onsite to access a wealth of knowledge without moving around the factory. As spaces as well equipment, machinery and tools are shared among multiple people, sensors and new human–machine interaction paradigms – such as touch-free, voice- or gesture-based interaction (for example using MYO armbands), biometric access (e.g. Face ID) – would considerably reduce the spread of infectious diseases at the workplace. PPEs could also be equipped with intelligence (e.g. connected face shields, smart helmets, etc.) and provide real-time feedback (or even AR content) to the workers.

But what matters the most is that, as personal technology becomes mainstream, employers can gather biometric data and unleash the power of predictive analytics to detect early signs of the disease or of potential ergonomic injuries, reduce health risks, and better manage possible future problems. A solution architecture of a software-as-a-service (SaaS) platform for secure collection, storage, processing, and health data analysis is proposed. As illustrated in Fig. 6, different implementations are possible depending on the requirements (e.g. data volumes, security levels, type of devices, real-time or near-real-time analytics etc.). Data may come from a variety of sources, ranging from biometric data (heart rate, sleep patterns, activity, skin temperature, etc.) collected continuously by wearables and other sources to personal medical records, social media, self-reported health information, and more – all stored and analyzed in an encrypted standards-compliant cloud solution or on-premises data lake. For example, glucose levels and oxygen saturation have been biometric indicators largely used used in clinical analysis to diagnose the COVID-19 disease. In a workplace, a biosensor patch might be employed for early detection of such symptoms (Javaid et al, 2020).

After the data is collected, extracted, transformed, and loaded into the database, an analytics engine can transform data into actionable information. With a combination of historical and real-time analysis, benchmarking, and predictive analytics, and with a link to the digital twin of the manufacturing system, this engine helps managers and operators to detect patterns, track and trace, model risks, visualize which cohorts are at risk and gain valuable real-time insights on workers’ health. Additionally, the platform can launch personalized, event-triggered automated alerts that can encourage operators to respect health practices in the case of pandemic and reduce their risk profiles (e.g. wash their hands or check their PPE). For example, a buzz or a color-coded warning on a smartwatch could let two workers know when they are within 2 m of each other. Analysts and health managers can also access synthesized data in real-time via web-based applications to enhance existing policies and optimize processes and layouts in the factory. This solution would also allow to better plan and forecast workforce capacity needs and immediately deploy a contingent strategy to face the disruption (cf. S1, S2, S3, S4, S5). Finally, we envisage a HO4.0 digital twin platform that will also make it possible to simulate future production scenarios (including exceptional situations, like a pandemic), the impact of certain mitigation measures (cf. Section 5.1) and the corresponding consequence on the well-being of operators. We argue that if a culture of predictive well-being of the operational staff is established, employers will benefit because healthy and serene employees are more productive.

5.6. Use cases

This use case presents some mockups and prototype applications designed and developed at the Modeling & Simulation Center – Laboratory of Enterprise Solutions (MSC-LES) of the University of Calabria, and later industrialized and commercialized by CAL-TEK S.r.l., a Spinoff company of the University of Calabria. They show how the plug-and-play worker, the remote operator 4.0 and the healthy operator 4.0 are being implemented in a collaborative effort with companies toward envisioning the future of work. Independently from the shop-floor level (where a variety of factory automation and network technologies are deployed) and from the location of data (i.e. external cloud or on-premise data lakes), the three types of workers here discussed can use a multi-device system front-end to interact with the factory. The application and device layers of the architecture depicted in Fig. 7 shows the different possibilities in terms of hardware and technological solutions that could be used to allow the three types of workers to connect and interact with the factory “at any place and at any time”.

An example of industrial machine’s innovative interface (that is over-imposed on the real machine) is proposed in Fig. 8 for plug-and-play workers and remote operators 4.0. Functionalities of the interface include:

• i.

  fast and intuitive access real-time data regarding the machine status (e.g. a crown wheel can be used to track the operation progess percentage);

• ii.

  easy access to a wealth of knowledge about the machine itself (e.g. images, videos, audio), both recorded in real-time through a camera-microphone-sensor system (e.g. machine prognostics and health management, PHM, data) onsite and from past manufacturing activities usually not directly available at the workplace onsite or in a remote location;

• iii.

  intelligent consultation of general information and how-to manuals of the machine;

• iv.

  run and control the machine remotely (e.g. start a new program, set up the machine, etc.);

• v.

  access a set of VR, AR, and MR-based training tools, that enable the users to learn how to interact with the machine using its digital twin.

Three scenarios can be then envisioned for a pandemic workforce or for the workforce of the future (Fig. 9 ):

• i.

  onsite plug-and-play workers can use advanced gesture-based interface to interact in an intelligent manner with the machine (Fig. 9.a). Legacy HMIs will be replaced in the future with AI-powered “personal” assistants that will allow plug-and-play workers to “establish a dialogue” with factory machines and be assisted throughout their activities onsite with the support of Augmented and Mixed Reality.

• ii.

  shop-floor operators can shift to remote working (or “smart working”) thanks to the possibility to access the same HMI even from home through secure networks (Fig. 9.b). The next generation of a pandemic workforce or the workforce of the future will be able to monitor and control the machine sitting at their home desk thanks to the use of handheld mobile devices (e.g. smartphone, or tablets) but especially head-mounted displays for virtual and mixed reality (e.g. Hololens, Oculus Rift or HTC Vive) that would provide the operators a higher level of immersion.

• iii.

  digital twin technology can even further empower the remote workers, who will be able not only to access the interface, but also visualize in a VR- or MR-powered environment the cyber-twin (i.e. the exact digital copy) of the physical machine. To do so, ambient intelligence, advanced sensors, machine vision technology, audio systems and machine health data will allow to reproduce exactly in the cyber environment the machine.

We are living the digital era where technology allows us to “make every place our workplace”: for example, a Mixed Reality application powered with ambient intelligence, motion tracking, environmental understanding capabilities provided by the Google’s ARCore enable a natural and intuitive interaction with digital objects in any environment, from your home (Fig. 9.c) to even a parking lot (Fig. 10 ). The shop floor workers of the future can place the digital twin of a machine wherever they want, make annotations or add information in a way that integrates seamlessly with the real world. They can move around and view the digital twin from any angle and interact with it like they would do in the real shop floor (see Fig. 11, Fig. 12 ).

It is worth mentioning that demonstrators and prototypes presented in this section are just some of the solutions developed by the authors and they have not to be regarded as a one-time effort for this specific article. They have been developed over the years and systematically updated to include new features and functionalities (according to new results obtained by authors in different research projects and to the technological evolutions that have characterized some of the Industry 4.0 pillars, e.g. Extended Reality, Cloud, web and mobile applications, etc.). These continuous research activities have finally led to the identification of the three types of workers presented in this article (the plug-and-play worker, the remote operator 4.0 and the healthy operator 4.0).

Additional information about the research projects in which demonstrators and prototypes presented above have been developed, industries involved, their operating sectors, the as-is situation before the implementation and the target situations can be found in the following research works published by the authors (such information are not reported in details in this article because they are already available in the literature and for a matter of space). However, for the sake of completeness, a summary is reported below:

• –

  Longo et al. (2017): solutions based on the use of AI powered Personal Assistant and Augmented Reality are presented. The solutions have been developed under two different funded R&D projects: the SISOM project and the SG-ICT project. Operating Sector: manufacturing companies working in the area of sliding bearings production. Situation as is before implementation: no use of AI personal assistant and Augmented Reality. Targeted situation after implementation: enabling a smart operator concept and improving training capabilities. Benefits as well as hurdles are described in the reference.

• –

  Longo et al. (2019a): solution based on the use of VR for improving operators’ preparedness in case of emergency in industrial plants and for monitoring and control through Remote Operators. The solution has been developed under the DIEM-SSP funded project. Operating Sector: Oils and Refineries (specific case study on re-refining of waste lube oil). Situation as is before implementation: no use of VR solutions. Targeted situation after implementation: proved capabilities of VR solutions for operators’ training, possibility to fed a Digital Twin for remote monitoring and control. Benefits as well as hurdles are described in the reference.

• –

  Longo et al. (2019b): a solution for implementing the ubiquitous knowledge concepts in manufacturing systems (knowledge available everywhere, no matter where you are) as part of a Digital Twin. This solution has been internally developed by MSC-LES and CAL-TEK and the case study has been funded through a specific research contract. Operating Sector: manufacturing of turbines and pumps for the Oil & Gas Sector (Baker Hughes General Electric plant, located in Italy). Situation as is before implementation: no use of Digital Twin solutions, limited use of the ubiquitous knowledge concept. Targeted situation after implementation: statistically significant improvement of multiple production and business performances have been achieved. Benefits as well as hurdles are described in the reference.

• –

  Bottani et al. (2021): a solution based on Mixed Reality (through Microsoft Hololens©) and Mobile technologies providing new enabling interfaces for shop floor operators. This solution has been developed as part of the W-ARTEMYS funded project. Situation as is before implementation: no use of Mixed Reality and mobile app solutions. Targeted situation after implementation: positive impact on operator’s productivity, downtimes reductions and employees’ safety enhancement. Benefits as well as hurdles are described in the reference.

6.1. Strategies for a resilient supply in times of pandemic

Production resilience means capacity of the production system to overcome a shock due to unpredictable events. We learned by COVID-19 experience how it is possible, although unexpected, that a production system can be affected by a limitation of the available workforce in the plant and, contemporarily, a break of the supply chain due to some suppliers’ lock-down.

As above described (cf. Section 5), technology would help to redesign the process control allowing some distance-workstations outside the plant, ensuring the same process efficiency and effectiveness.

Supply chain breaks naturarly enlarges the supplier research domain according to the Liquid Supply Chain paradigm recently postulated by some of the authors (Passarelli et al., 2021). But the latter is an external answer to the industry need. At the same time, technology allows also an internal answer.

Industry can move toward two different directions, at least for some production processes:

• i.

  Allow the vertical integration of some processes in order to face an eventual supply chain break.

• ii.

  Invest in alternative technologies which can substitute the traditional ones in case of supply chain breaks.

Both of them increases industry resilience since the organization has more chances to overcome temporary reduction of the supplier production. Actually, vertical integration which means internalization of some manufacturing processes is really costly because requires at least three elements:

• 1.

  Skilled workforce able to execute some specific processes;

• 2.

  Proper equipment;

• 3.

  Enough space inside the plant.

For the above reasons, internalization is usually reserved to very critical processes for which the risk coming from market is really strong. On the contrary, the acquisition of technologies able to substitute other conventional processes, even if slower and more expensive, can be a suitable strategy.

In our common experience we understand how it is possible to obtain different products using diferent processes: a dress can be manufactured by hands instead of machines, a screw car be produced using a forming machine or a lathe, a dish can by produced by a moulding machine but also using a Fused Deposition Method (FDM) additive manufacturing machine. Thus, if vertical integration (i) may upset the industry organization because it may require a huge investment in terms of resources, the use of substitutive technologies (ii) seems to be a suitable solution to increase process robustness in case of supply chian troubles. According to I4.0 paradigm, a massive research on the use of additive manufacturing technology, intended as a digital & unconventional sourcing, instead of the traditional ones is carried out all over the World (Achillas et al., 2017).

On the other hand, what is really disruptive is the use of:

• i.

  COVID-19 pandemic as an extraordinary opportunity of innovation.

In such cases, indeed, supply chain interruption was not easily supported by an internal or external substitution of the previous suppliers. The pandemic acts as an innovation booster, allowing a quick product development to overcome the supply chain weakness. Thus, a dress can be manufactured using the local natural fibres; screws can be substituted by new fastening methodologies; dishes can be replaced by new trays made of a special cardboard.

6.2. Use cases

During pandemic time several published papers described how resilient companies changed their production to manufacture products useful to cope with the disease. Masks for breath protection, components for pulmonary fans, medical dress and devices, spacer fences are just few examples (Fairgrieve et al., 2020). However, here we present and discuss some business use cases showing how companies decided to preserve their production looking at their products after pandemic.