Avainsana-arkisto: industry

An Emerging Technology Challenge: Digital Twins

Mikkel Stein Knudsen and Jari Kaivo-oja:

With Digital Twins, organisations can not only create mirrors of real-world objects and processes but integrate physical and virtual worlds through bidirectional flows of information. With real-time simulations and intelligent algorithms, Digital Twins shifts the focus of data-driven operations from ex-post monitoring to ex-ante predictions and optimization in increasingly complex environments. Digital Twins are said to revolutionize the manufacturing industry, but it may also have major impacts on future studies and the general ways organisations anticipate the future. A Digital Twin development can be a new advanced form of scenario planning.

Finland Futures Research Centre takes part in the project Manufacturing  4.0  (2018–2020) for the Strategic Research Council at the Academy of Finland. Our contribution includes technological foresight related to the identification of promising technologies for the future manufacturing landscape in Finland. The concept of digital twins has featured heavily in our early scanning, as one of the most enterprising new advanced manufacturing technologies.

As the embodiment of Cyber-Physical Systems, Digital Twins has become one of the most hyped technologies of the so-called Industry 4.0. According to a comment in Nature in September 2019, “Digital twins – precise, virtual copies of machines or systems – are revolutionizing industry” (Tao & Qi, 2019). Many major companies already use digital twins, while half of all corporations may use them by 2021 (Gartner). For example, in Finland, Nokia is focusing strongly on Digital Twin challenges.

Digital twins are part of a vast shift in the world’s economy towards ‘mirror worlds’ as a new dimension of human life based on and fuelled by data. The emergence of these mirror worlds will bring about a distinct economy, and require new markets, infrastructure, institutions, businesses, and geopolitical arrangements, according to a recent special report in The Economist.

What are Digital Twins?

In its original form, a Digital Twin is the virtual model of a process, product or service, or ‘a digital representation that mirrors a real-life object, process or system’ (Panetta, 2018). While no general and precise definition of the features and scopes of Digital Twins has been reached (Cimino et al., 2019), a consensus appears of certain characteristics distinguishing ‘real’ Digital Twins from related virtual replications.

Kritzinger et al., 2018 operate with concepts of Digital Models, Digital Shadows, and Digital Twins (see figure 1). What sets Digital Twins apart is the real-time automatic dataflow in both directions between the physical and digital objects.


Fig. 1. From Digital Model to Digital Twin (own representation, after Kritzinger et al., 2018)

Talkhestani et al., 2019 elaborates this further including four necessary features in their definition of a Digital Twin: (1) A Digital Twin has to be a digital representation of a physical asset, including as realistic as possible models and all available data on the physical asset. (2) The data has to contain all process data, acquired during operation as well as organizational and technical information created during the development of the asset. (3) A Digital Twin has to always be in sync with the physical asset. (4) It has to be possible to simulate the Digital Twin of the behavior of the physical asset.

Fig 2. The conceptualisation of Digital Twin.

Summarily, as seen in Figure 2, we conceptualize digital twins as possessing five constitutive features separating them from other virtual models: They need to have counterparts or equivalents in the physical world (be it a person, an organ, a product, a machine, a traffic system, or an organisational environment/infrastructure). They need to provide a fair (= as precise as possible) representation of its equivalent characteristics. They must be updated automatically and continuously (in real-time, or close-to-real-time). It must be possible to simulate the environment of the real-world counterpart on the digital twin (achieving what Qi et al. deem ‘integration between entity DT and scenario DT’). Finally, there must be synchronization directly from the digital twin to its counterpart, so that the physical asset can also mirror new directions or alterations happening virtually to the digital twin.

Digital twins in systems theory and modeling

Until now, digital twins can roughly be separated into two categories (Zhidchenko et al., 2018). Either, they assist with the analysis of very complex systems (like transportation systems), or they provide real-time analysis of relatively small systems (like a vehicle). For complex systems, digital twins are means of providing a safe space for simulation of various potential developments and impacts. The idea of digital twins is thus born out of being a (much cheaper) virtual replacement of the identical twin spacecraft NASA always produced to have suitable test spaces. By using digital twins, it becomes possible to perform simulations using the real-time characteristics of their physical counterparts, even while these are in operation. This has enormous potential to limit operational risks, as well as for optimization issues.

However, the distinction between the two categories of digital twins is increasingly blurred. Digital twins for real-time analysis is being applied to more and more complex systems. If future digital twins, as seems likely, acquire functionalities – enabled by artificial intelligence and machine learning – which allow them to interact with other digital twins in a ubiquitous environment, even digital twins of simple systems will themselves be entities forming complex systems.

Personal Digital Twins

Digital twins are also moving out of the manufacturing halls and on its way into, well, you. Signals regarding digital twins for health are already appearing manifold. Last month, The Economist reported on ambitious cardiac-research plans to create digital twins of human patients’ hearts. Digital twins are also potentially important enablers for personalised medicine through “high-resolution models of individual patients that are computationally treated with thousands of drugs to find the drug that is optimal for the patient (Björnsson et al., 2020).

In Finland, novel ideas of personal digital twins also appear in the context of the national AuroraAI-programme, which aims for Finland to ‘enter the AI age in a human-centric and ethically sustainable way’. A YouTube-video from the projects illustrates how you can ‘Let your digital twin empower you’. Futurist Osmo Kuusi has been a key contributor to development (cf. Kaivo-oja et al., 2019). Various technologies in the EU Foresight report “100 Radical Innovation Breakthroughs for the Future – The Radical Innovation Breakthrough Inquirer” indicate the high technical feasibility of digital twins.

Finding Finland’s niches

Manufacturing 4.0 believes Finland has the potential to be a leader in the Digital Twin-revolution, e.g. by developing and applying Digital Twins in strong niche markets. Department of Futures Technologies at the University of Turku has developed world-class digital models utilizing virtual reality/augmented reality, now commercialised through the spin-off company CTRL Reality. Researchers demonstrated their model of a virtual forest at the Finland Futures Research Centre’s Futures Fair in December 2017; this model was recently highlighted in the journal Nature calling for the world to Make more digital twins.

Other projects in Finland related to national strengths involve the environmental impacts of mining and mobile cranes. Usage of Digital Twins is also high on the agenda for the development and modernizations of ports in Pori and Rauma and as an element of the further take-off of the Robocoast-cluster. In Turku, digital twins are integral to the vision of creating a Smart and Wise Turku. Smart City Digital Twins is a major global research and investment area, and Finnish pioneer cities like Turku lead the way. Relying on a massive amount of data collected at high-speed from millions of sensors, there are finally also clear links between the rollout of digital twins and new demands for 5G and 6G networks.

What Digital Twins mean for futures studies?

The rapid trajectory of Digital Twins links with the disciplines of foresight and futures studies in several ways. First, it is in itself a trend to follow, analyse and speculate about the consequences of. It is also an embodiment of megatrends such as digitalisation, personalisation, and altered human-machine interactions.

At the same time, the very idea and definition of a digital twin as a vehicle of simulating the future makes digital twins an inherently futures-related technology. Futures studies-researchers and foresight practitioners must learn to utilize this as an important new tool in their toolbox. As one of its founding ideas, digital twins can provide a “safe simulation environment” to test future novelties, new products, new services, new organizational structures, new medicines, and smart infrastructures. This safety-oriented futures approach may be the most desirable way to use and apply digital twins systems thinking, and it could develop into a great futurist tool. How digital twins can supplement the identification of weak signals would seem like another fruitful avenue of investigation for future futures research.

Simultaneously, the theoretical baggage of futurists and system thinkers may be useful in shaping the field of digital twins. At present, the field seems primarily occupied by engineers and technology optimists, who may not always be aware of potential blind spots in their simulation models. A push might be needed to functionally integrate weak signals and wild cards into the simulations of digital twins. This will be another great topic for futures researchers in the years to come.

Mikkel Stein Knudsen
Project Researcher (M.Sc., Pol. Science), Finland Futures Research Centre, Turku School of Economics, University of Turku    

Jari Kaivo-oja
Research Director, Finland Futures Research Centre, Turku School of Economics, University of Turku.

– – –

The project ‘Manufacturing 4.0’ has received funding from the Finnish Strategic Research Council [grant number 313395]. The project “Platforms of Big Data Foresight (PLATBIDAFO)” has received funding from the European Regional Development Fund (project No 01.2.2-LMT-K-718-02-0019) under a grant agreement with the Research Council of Lithuania (LMTLT).

 

References and additional information:

Cimino, Chiara et al. (2019). “Review of digital twin applications in manufacturing”. Computers in Industry, 113. DOI.

Grieves, Michael & Vickers, John (2017). “Digital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex Systems”. In Kahlen, FJ. – Flumerfelt, S. & Alves, A. (eds.) Transdisciplinary Perspectives on Complex Systems. Springer, Cham. DOI.

Kaivo-oja, Jari et al. (2019). ”Digital Twins Approach and Future Knowledge Management Challenges: Where We Shall Need System Integration, Synergy Analyses and Synergy Measurements?”. In Uden, L. – Tinh, IH. & Corchado, J. (eds.) Knowledge Management in Organizations. KMO 2019. Communications in Computer and Information Science, 1027. Springer, Cham. DOI.

Kritzinger, W. et al. (2018). ”Digital Twin in manufacturing: A categorical literature review and classification.”. IFAC-PapersOnLine, 51(11), 1016–1022. DOI.

Lu, Y. et al. (2020). “Digital Twin-driven smart manufacturing: Connotation, reference model, applications and research issues.” Robotics and Computer-Integrated Manufacturing, 61. DOI.

Qi, Q. et al. (2019). “Enabling technologies and tools for digital twin”. Journal of Manufacturing Systems, in press. DOI.

Saracco, Roberto (2019). “Digital Twins: Bridging Physical Space and Cyberspace”. Computer, 52(12), 58–64. DOI.

Talkhestani, B.A. et al. (2019). ”An architecture of an Intelligent Digital Twin in a Cyber-Physical Production System.” at – Automatisierungstechnik, 67(9), 762-782. DOI.

Tao, Fei & Qi, Qinglin (2019). ”Make more digital twins”. Nature. DOI.

Zhidchenko, V. et al. (2018). ”Faster than real-time simulation of mobile crane dynamics using digital twin concept”. Journal of Physics: Conference Series, 1096.

Warnke, Philine – Cuhls, Kerstin – Schmoch, Ultich – Daniel, Lea – Andreescu, Liviu – Dragomir, Bianca – Gheorghiu, Radu – Baboschi, Catalina – Curaj, Adrian – Parkkinen, Marjukka & Kuusi, Osmo (2019). 100 Radical Innovation Breakthroughs for the Future – The Radical Innovation Breakthrough Inquirer. Foresight-report. European Commission. Brussels. DOI.

Cover picture: Pixabay.com

Out of the cages: Here comes the cobots

Mikkel Stein Knudsen and Jari Kaivo-oja:

Forbes, The Guardian, and Financial Times have written about them. The US Department of Commerce lists it as one of 5 Manufacturing Technology Trends to Watch in 2019. Cobots – short for ‘collaborative robots’ – are increasingly entering into industrial manufacturing, profoundly changing the ways in which humans and robots interact.

As one research article puts it, “robots have long left the cages of industrial settings: They work together with humans – collaboratively” (Korn et al., 2018). Smart Cobots are a key technology informing the futures of manufacturing; our research topic in the large Strategic Research Council-project Manufacturing 4.0.

What are cobots?

Collaborative robots differ from traditional industrial robots precisely in the direct interaction with human workers. They are intended to e.g. handle a shared payload without the need for conventional safety cages or separating protective measures. They are generally small, lightweight, mobile and flexible units, and they enable – at least in theory – organisations to leverage the strengths and endurance of robots with the tacit knowledge and agile decision-making skills of humans. Both humans and robots have crucial advantages (Fast-Berglund et al., 2016) – while robots ace repetitive and monotonous tasks, humans remain the most flexible resource in the system. Humans still handling unexpected and unplanned tasks better that their automated co-workers. A human-robotic collaborative approach also proved superior in experimental research settings compared to a similar purely robotic process (Bloss, 2016).

With its focus on flexibility the paradigm of cobots aligns well paradigms of Industry 4.0 – driving at increased automation and increased efficiency in parallel with increasingly flexible production processes, small batch sizes and mass customization.

A sector on the up

Industry forecasts for the near future market for collaborative robots are wildly positive, from global revenues of $7.6 bn in 2027 to the exceptionally optimistic 2019-prediction from the Robotics Industries Association of a $34 billion cobot market by 2026. This will require exponential growth from the current global market of around $600 million in 2018, which in itself was 50% higher than the year before (Sharma, 2019). The academic research output on cobots is also rapidly growing, as the assessment of articles indexed in Web of Science (Figure 1) shows.

Fig 1. Articles indexed in Web of Science with “collaborative robot*” or cobot* as title or keyword (From Knudsen & Kaivo-oja, 2019)

Until now, Finland has not been at the centre of this research. Out of a total of 496 articles in Web of Science published since 2015 (search: 1.1.2019), only 3 are affiliated with Finland. In a ranking of countries based on this data, Finland places 32th. A recent report for the Ministry of Finance in Finland (Rousku et al, 2019) also identified this problem, as well as collaborative robots as a key growth market, asking (p. 46): “Can Finland afford not to take a slice of a market that generates new wealth and new vitality for business and society alike?” A very good question – indeed.

Cobots may provide answers to megatrends

One of the reasons the future could be bright for collaborative robots is that they can answer to a number of different societal megatrends. As the research paradigm on cobots matures and moves away from strictly technological concerns, these links between societal drivers and cobots should be explored in much further detail.

An example, already prominently suggested in the literature, is that cobots may reduce ergonomic challenges and improve occupational safety and health e.g. in factory settings. By reducing the physical workload for workers, cobots can also enable work environments more responsive to older employees – a highly significant advantage given the changing demographics of labour markets across most industrialized nations.

Key global trends to 2030
(from ESPAS, 2015)
Potential role of cobots
A richer and older human race characterised by an expanding global middle class and greater inequalities. Enabling inclusive labour markets more responsive to older employees, employees with disabilities.

Providing a work environment more responsive to human factors, ergonomic and OS&H concerns.

A more vulnerable process of globalisation led by an ’economic G3’. ‘Bringing manufacturing back home’; cobots as enabler of competitive manufacturing in high-cost environments.
A transformative industrial and technological revolution. A ‘gateway into factory automation’, enabler of semi-automated manufacturing choosing select elements of Industry 4.0 for optimized production process.
A growing nexus of climate change, energy and competition for resources. Improved resource efficiency, enabler of circular economy and remanufacturing
(Sarc et al., 2019; Huang et al., 2019).
Changing power, interdependence and fragile multilateralism.

In addition, collaborative robotics will be at the absolute forefront of the development of human-machine interactions, which will help shape important parts of our lives in the coming decades. Unlike most of our everyday interaction with machine learning-algorithms, our interaction with cobots has a distinct physical – see, feel and touch – element to it.

We therefore believe that understanding the topic of cobots, envisioning their deployment, and exploring both preferable and undesirable futures of and with cobots must be prominent future research topics.

Fig 2. Current frontiers of cobot research (based on Knudsen & Kaivo-oja, 2019).

Figure 2 shows some of the current frontiers of cobot research and technology, based on our initial literature review. For each of these pillars many research questions are rapidly arising, and they deserve our attention. Because robots are moving out of the cages and into a space near you.

Industrial robots have traditionally worked separately from humans, behind fences, but this is changing with the emergence of industrial cobots. Industrial robots have traditionally worked separately from humans, behind fences, but this is changing with the emergence of industrial cobots. To sum up, emerging cobot issue requires more attention in the field of Industry 4.0/Manufacturing 4.0. Cobots, or collaborative robots, are robots intended to interact with humans in a shared space or to work safely in close proximity. Service robots can be considered to be cobots as they are intended to work alongside humans. This “cobot approach” is very promising, because it focus on human-robot interaction from the beginning of industrial process planning. Typically, sensors and software are needed to assure good collaborative behaviour.

Summary

It is important to note that cognitive aspects and cognitive ergonomics are highly relevant for new digitalized work life. The IFR (Institute for Occupational Safety and Health of the German Social Accident Insurance) defines four types of collaborative manufacturing applications: (1) Co-existence Cobots: Human and robot work alongside each other, but with no shared workspace, (2) Sequential Collaboration Cobots: Human and robot share all or part of a workspace but do not work on a part or machine at the same time, (3) Co-operation Cobots: Robot and human work on the same part or machine at the same time, and both are in motion and (4) Responsive Collaboration Cobot: The robot responds in real-time to the worker’s motion.

All these types of cobots provide interesting possibilities and challenges for Industry 4.0/Manufacturing 4.0 activities. Are we ready to face these challenges?

Mikkel Stein Knudsen
Project Researcher (M.Sc., Pol. Science), Finland Futures Research Centre, Turku School of Economics, University of Turku    

Jari Kaivo-oja
Research Director, Finland Futures Research Centre, Turku School of Economics, University of Turku.

***

The project Manufacturing 4.0 has received funding from the Finnish Strategic Research Council [grant number 313395]. The project “Platforms of Big Data Foresight (PLATBIDAFO)” has received funding from European Regional Development Fund (project No 01.2.2-LMT-K-718-02-0019) under grant agreement with the Research Council of Lithuania (LMTLT).

***

References

Picture copyright Universal Robots A/S, case Hofmann