Naming the Fifth: What the Industry 5.0 Debate Is Actually About

1. Framing the Question

I recently submitted a research article to a leading journal for peer review. One of the major concerns raised by the reviewers related to my decision to frame the study within the context of Industry 4.0. In their view, Industry 4.0 has become somewhat outdated due to the emergence and growing prominence of Industry 5.0. This feedback prompted me to explore the broader debate surrounding the positioning of Industry 5.0 as the successor to Industry 4.0 and to examine how this transition has been perceived and internalized within academic, industrial, and public discourse. The standard critique of Industry 5.0 runs as follows: the four preceding industrial revolutions were driven by technological breakthroughs, whereas Industry 5.0 was announced by a policy directorate; it introduces no new production physics; its enabling technologies are those of Industry 4.0; therefore it is not a revolution but a normative overlay, an “Industry 4.1.”

Each premise is defensible when considered independently. However, the conclusion addresses a question that the nobody is explicitly asking, given that the European Commission does not present Industry 5.0 as a technological revolution. Nevertheless, this question remains highly salient among many stakeholders, particularly researchers, and continues to shape their perceptions, priorities, and decision-making. The January 2021 policy brief states explicitly that Industry 5.0 “should not be understood as a chronological continuation of, nor an alternative to, the existing Industry 4.0 paradigm,” and that it “complements and extends” it.¹² Demonstrating that Industry 5.0 is not a fifth industrial revolution therefore establishes a point its originators concede on the first page.

This does not dissolve the critique; it relocates it. Three questions survive, and they are more interesting than the one they replace:

●        The naming question. If Industry 5.0 is not a revolution, why is it numbered as one? What does the sequential label do that “European industrial policy for a sustainable manufacturing transition” would not?

●        The genealogy question. Is the contrast between bottom-up technological revolutions and top-down policy paradigms historically accurate? This paper argues it is not, and that the discontinuity the standard critique locates at 5.0 occurred earlier.

●        The consequences question. Independent of nomenclature, does the framework distort research funding, industrial investment or competitive position — and what evidence would settle that?

A fourth issue runs underneath all three. The claim that the “real” fifth revolution will be autonomous, machine-driven and human-free is a forecast, not an observation, and it is in tension with a substantial body of evidence on the limits of full automation. That evidence is examined in Section 9.

2. The Sequence Was Invented Backwards

The numbered sequence of industrial revolutions is not a historical record that Industry 5.0 has departed from. It is a retrospective construction, and an unstable one.

No contemporary of the spinning jenny described themselves as living through “Industry 1.0.” The phrase “industrial revolution” entered general English usage largely through Arnold Toynbee’s posthumously published Oxford lectures of 1884, a century after the events they describe. The First Industrial Revolution was named by historians; the Second was identified retrospectively by economic historians in the mid-twentieth century; the Third was named in the 1980s and after. The decimal notation — 1.0, 2.0, 3.0 — is more recent still. It was applied backwards, after 2011, to supply Industrie 4.0 with a lineage.

This matters because the standard critique rests on a comparison between how 1.0 through 4.0 emerged and how 5.0 emerged. But the first three were not announced at all. They were reconstructed after the fact, by analysts choosing where to place boundaries in a continuous process. Comparing a retrospective historical category with a prospective policy programme and finding them dissimilar is not a discovery; the dissimilarity is built into the comparison.

The labels have never had stable referents

The instability is not hypothetical. Consider the Third Industrial Revolution, which the standard account treats as settled — microelectronics, programmable logic controllers, digital computation:

●        That is one usage. Jeremy Rifkin used “Third Industrial Revolution” for something else entirely: the convergence of distributed internet communication with distributed renewable energy.³ Nothing to do with PLCs.

●        In May 2007 the European Parliament issued a formal written declaration endorsing Rifkin’s Third Industrial Revolution as a long-term economic vision for the Union.³ A European institution declared an industrial revolution, top-down, as policy — four years before Industrie 4.0 and fourteen years before Industry 5.0.

The same instability afflicts the fifth. Sachsenmeier used “Industry 5.0” in 2016 to denote the industrial application of bionics and synthetic biology.⁴ Özdemir and Hekim used it in 2018 for a big-data and IoT governance agenda.⁵ Nahavandi used it in 2019 for human–robot co-working.⁶ Michael Rada had used it from 2015 for a waste-prevention methodology he called industrial upcycling.⁷ These are four incompatible contents under one label, all predating the European Commission’s adoption of it.

The conclusion is stronger than the one the standard critique reaches, and it cuts in an unexpected direction. Conceptual inflation is real — but it is a property of the numbering convention itself, not a defect introduced by Industry 5.0. The taxonomy was never precise enough to be diluted.

3. The Precedent the Critique Overlooks: Industrie 4.0

The load-bearing claim in the standard critique is that Industry 5.0 inverted the historical pattern: technology first, policy afterwards. On this account, Industries 1.0 to 4.0 emerged bottom-up from enterprises and engineers, with public institutions responding after the fact.

For Industrie 4.0 this is the reverse of what happened.

The term was coined by a working group of the German government’s Industry–Science Research Alliance, established under the Federal High-Tech Strategy. It was introduced publicly in an article by Henning Kagermann (acatech), Wolfgang Wahlster (DFKI) and Wolf-Dieter Lukas — then a senior official at the Federal Ministry of Education and Research — published in VDI Nachrichten on 1 April 2011, immediately ahead of the Hannover Messe.⁸ Chancellor Angela Merkel took up the term in her opening speech at the fair on 3 April.⁹ The working group’s final report, Recommendations for Implementing the Strategic Initiative INDUSTRIE 4.0, appeared in 2013; Plattform Industrie 4.0 was established as the institutional vehicle, with a secretariat provided jointly by the industry associations BITKOM, VDMA and ZVEI.⁸¹⁰ Kagermann and Wahlster later recorded that the initiative’s purpose, formed under the impact of the financial crisis, was to make the German economy more resilient and competitive.⁹

That is: a government research alliance, a national strategy document, a ministry official among the coiners, a head-of-government endorsement within forty-eight hours, a state-backed platform, and an explicit competitiveness rationale. Industrie 4.0 was a policy brand from the outset. One discourse analysis characterizes it as a case of “visioneering” — a vision constructed and promoted by identifiable actors, rather than a bottom-up technological fact subsequently observed.¹¹

This has a direct consequence for the critique. If the objection to Industry 5.0 is that a public institution named an industrial revolution to steer research funding and industrial priorities, then Industry 5.0 is not the exception. It is the second instance of a template that Industrie 4.0 established and that the European Parliament had arguably used earlier still. The genuine discontinuity in how industrial revolutions get named occurred in 2011, and the standard critique passes over it because Industrie 4.0 is the baseline against which it measures.

A weaker version of the original claim does survive: Industrie 4.0 named a technological transition that was independently underway, whereas Industry 5.0 names a set of values. That distinction is real and worth preserving. But it is a difference of degree along a spectrum of policy-driven labelling, not the categorical break between organic and engineered emergence that the standard account asserts.

Table 1. Origins of the numbered paradigms. The transition from retrospective categorization to prospective branding occurs at 3.0–4.0, not at 5.0.

4. What the European Commission Actually Claims

Accurate criticism requires an accurate statement of the position criticized. The Commission’s claims are narrower than the critique usually assumes.

The 2021 policy brief by Breque, De Nul and Petridis, produced by DG Research and Innovation, positions Industry 5.0 around three pillars — human-centricity, sustainability and resilience — and frames it as a shift from shareholder to stakeholder value, using research and innovation instruments to support that transition.¹ It was preceded by a 2020 DG RTD report on enabling technologies and a series of virtual workshops with research and technology organizations, and followed by an ESIR expert group policy brief in January 2022.²¹²

Three features of this deserve emphasis:

●        It is a research and innovation policy instrument, not a claim about the state of manufacturing. It describes a direction for European R&I funding. Criticizing it for failing to identify a shopfloor reality is a category error — it was never offered as a description of one.

●        It disclaims revolutionary status. The “complements, does not replace” formulation is explicit and repeated across Commission materials and the surrounding literature.¹¹³

●        It does not claim technological novelty. The Commission’s own enabling-technologies mapping largely overlaps with Industry 4.0.¹² The Industry 5.0 proposition concerns objectives and evaluation criteria, not new physical mechanisms.

What can fairly be charged is a rhetorical inconsistency, and it is not trivial. The Commission disclaims the revolutionary reading in the body text while retaining a number in the sequence — and numbering carries an implicature that footnoted disclaimers do not reliably cancel. Downstream usage bears this out: a large secondary literature treats Industry 5.0 as a successor revolution regardless of the disclaimer, and Rada, whose 2015 usage the Commission cited, has publicly disputed the appropriation of the term.⁷ The Commission chose a label whose most common reading it then had to spend the document denying. That is a legitimate criticism of communication strategy. It is not evidence of a claim the Commission made.

5. The Formalization and What It Does Not Show

The standard critique often supports itself with an optimization argument. Under Industry 4.0, the objective is broadly economic — a weighted combination of throughput, efficiency and flexibility over a vector of operational parameters χ within the feasible set of cyber-physical production configurations:

max J₄.₀(χ) = α₁·T(χ) + α₂·E(χ) + α₃·F(χ)

Industry 5.0 is then represented as the same system with additional value terms and explicit feasibility bounds — worker wellbeing H, ecological performance S, and resilience R:

max J₅.₀(χ) = α₁T(χ) + α₂E(χ) + α₃F(χ) + β₁H(χ) + β₂S(χ) + β₃R(χ)

subject to   H(χ) ≥ Hₘᵢₙ,   S(χ) ≤ Sₘₐₓ,   R(χ) ≥ Rₘᵢₙ

The intended reading is that since the state space χ and the underlying production mechanics are unchanged, Industry 5.0 adds only scoring criteria — hence “Industry 4.1.”

This inference does not hold, and it is worth being precise about why. In any constrained optimization, changing the objective function and the feasible region changes the optimum. If the constraints bind, the resulting system configuration differs — potentially very substantially — from the one that maximizes J₄.₀. A system designed to a different objective is a different system, even on identical hardware. The relationship is familiar from control theory: identical plant, different cost functional, different closed-loop behaviour. Nobody would say a control system is unchanged because only its cost function was rewritten.

So the formalization, taken seriously, undercuts the argument it is offered to support. It shows that Industry 5.0 proposes a substantive change at the level of system objectives. Whether a substantive change in objectives warrants the word “revolution” is a question about the word, not about the system. The optimization framing cannot settle it, because the framing contains no criterion for what makes a change revolutionary.

Three further caveats apply, and they are the honest limits of this kind of formalization. It is illustrative notation rather than a derived model: no functional forms are specified, and the weights α and β are not estimated from anything. The scalarization of incommensurable objectives into a weighted sum embeds a strong and contestable assumption about tradeability between worker wellbeing and throughput. And the framing treats the weights as exogenous policy parameters, whereas in practice they are the outcome of bargaining between firms, regulators and labour — which is precisely what the political-economy dimension of the question consists of. The formalization should be read as a way of stating the disagreement clearly, not as evidence for either side of it.

6. The Critique That Survives

Stripped of the revolution question and the inverted genealogy, a substantial critique remains. It is the strongest part of the original argument and is stated here at full force.

6.1 The measurement deficit

This is the most robust objection. Industry 4.0 operationalized cleanly: overall equipment effectiveness, cycle time, scrap rate, mean time between failures, first-pass yield. Each is measurable, auditable and translatable into a capital allocation case.

Industry 5.0’s pillars have no comparable operationalization. “Human-centricity” admits many measurements — ergonomic risk scores, cognitive workload indices, autonomy and job-quality instruments, injury rates, retention — which capture different things and can move in opposite directions. “Resilience” has no agreed metric at all: it may mean inventory buffers, supplier multi-sourcing, recovery time after disruption, or geographic redundancy, and optimizing for one can degrade another. Without agreed indicators there is no way to distinguish a genuine Industry 5.0 implementation from a relabelled Industry 4.0 project, and no way to evaluate whether the policy is achieving anything.

This objection is not merely practical. A framework that cannot be operationalized cannot be falsified, and a framework that cannot be falsified cannot be evaluated as policy. That is a serious defect regardless of whether the underlying values are sound.

6.2 Incentive distortion and narrative compliance

Where funding is conditioned on terminology, terminology adapts. Research groups and firms have an evident incentive to describe existing digitalization work in the vocabulary of whichever programme is disbursing, and reviewers have limited capacity to verify substantive alignment. The predicted result is relabelling without redesign.

The mechanism is plausible and well documented in the general literature on research funding. It should nonetheless be stated as a hypothesis rather than a finding: the writer knows of no systematic study that has attempted to measure the rate of Industry 5.0 relabelling in funded projects, for instance by auditing a sample against declared human-centric or circularity objectives. The bibliometric explosion in Industry 5.0 publications is consistent with narrative compliance, but equally consistent with genuine research interest, and citation counts cannot discriminate between them. This is a testable claim that has not been tested, and it should be presented as such.

6.3 Cost asymmetry and competitive position

The argument is that human-centric and environmental constraints impose costs that competitors operating without them do not bear, producing a structural disadvantage in global markets.

This deserves qualification in three directions. First, the empirical premise is contested: the assumption that ergonomic and human-in-the-loop design is a net cost is exactly what the automation-limits literature in Section 9 disputes, and cobot deployments are frequently justified on quality, flexibility and changeover grounds rather than as a welfare expense. Second, the comparator is inaccurate. It is not the case that competing economies pursue unconstrained automation; China’s industrial programmes incorporate substantial environmental and social objectives, and Japan’s Society 5.0 is explicitly welfare-oriented. Third, the counter-argument from regulatory first-mover advantage — that early standard-setting can become a competitive asset as other jurisdictions converge on it — is at least as plausible as the cost-disadvantage story, and has some support in the EU’s regulatory history.

The honest position is that the net competitive effect of Industry 5.0-type constraints is unknown, that plausible arguments run in both directions, and that treating the disadvantage as established is an ideological rather than an empirical judgment.

7. Comparison with Society 5.0

The contrast with Japan’s Society 5.0, adopted in the 5th Science and Technology Basic Plan in January 2016 and promoted by the Cabinet Office and Keidanren, is instructive on several dimensions.¹⁴

The important observation is not that Society 5.0 is broader in scope, though it is. It is that Society 5.0 is also a top-down, government-authored, numbered framework with no distinct technological base — announced five years before the Commission’s Industry 5.0 and running its numbering on a different object entirely (stages of society: hunting, agrarian, industrial, information, super-smart). Two major economies independently adopted prospective numbered framing for normative technology programmes within a decade. Whatever is going on, it is not a European aberration; it is a general shift in how industrial policy is communicated, and it deserves analysis as such rather than as a single institution’s error.

Table 2. Three numbered frameworks. All three are prospective and state-originated; they differ in scope and in the object being enumerated, not in mode of emergence.

8. Adoption Evidence and Its Limits

The claim that Industry 5.0 has had limited operational impact is probably correct, but the evidence base supporting it is weaker than the confident phrasing in most reviews suggests, and this deserves acknowledgment.

What the literature reliably shows is that the Industry 5.0 publication volume has grown very rapidly since 2021 and that the enabling technologies discussed within it are substantially those of Industry 4.0.¹³¹⁵ What it shows less reliably is the state of practice. Much of the empirical base consists of systematic reviews of other reviews, conceptual frameworks, and multi-criteria decision analyses of expert opinion — including MCDM studies of “critical success factors” that elicit practitioner rankings rather than measuring outcomes.¹⁶ These are legitimate instruments for mapping a discourse. They are weak instruments for measuring shopfloor reality, and a field that studies itself primarily through literature reviews will tend to mistake the growth of its own literature for the growth of its object.

Three further considerations temper the “limited impact” verdict:

●        The timeframe is short. Industry 4.0 was announced in 2011 and diffusion was still described as slow and uneven a decade later. Judging a 2021 framework’s operational penetration by 2026 sets a standard that Industry 4.0 itself would fail.

●        Attribution is intractable. Cobot installation, ergonomic monitoring and supply chain diversification are all growing. Whether any of that growth is attributable to Industry 5.0 policy, as opposed to labour shortages, ageing workforces, post-pandemic supply shocks and falling sensor costs, cannot be determined from adoption counts. The observation that these deployments have conventional business cases is often presented as evidence against Industry 5.0; it is equally consistent with the framework having correctly identified where commercial and social interests already align.

●        The instrument may be working as designed. A research and innovation policy operates on funding calls and project portfolios with lags measured in years. Its proximate output is the composition of the research pipeline, not immediate factory practice. Evaluated against its own mechanism rather than against a revolution it never claimed, the appropriate evidence would be a portfolio analysis — which, again, does not appear to have been done.

9. The Autonomy Thesis, Examined

The most substantive empirical claim in the standard critique is prospective: that by defining the fifth industrial era around human-centricity, policymakers obscure the real coming transition to autonomous, machine-driven, lights-out production, and that mandating human involvement will create friction against it.

This claim requires more scrutiny than it usually receives, for four reasons.

9.1 Lights-out manufacturing is not new

FANUC has operated lights-out production at its Oshino complex since 2001, with robots producing robots and lines running unattended for extended periods.¹⁷ Philips has run a near-lights-out razor plant in the Netherlands with human involvement concentrated in quality assurance. The concept is older still: General Motors pursued a “factory of the future” in the 1980s that never reached lights-out status and was widely regarded as a costly failure.

A production model that has existed for twenty-five years, and been aspired to for forty, is not an emerging paradigm being obscured by a 2021 policy brief. The relevant question is not why it is coming but why, having existed for a quarter century, it has remained confined to narrow niches — typically high-volume, low-variance, well-characterized processes such as CNC machining, injection moulding and semiconductor fabrication.

9.2 The most aggressive recent attempt failed publicly

Tesla’s Model 3 ramp was designed around maximal automation. It produced what its chief executive called “production hell”: the company missed successive volume targets, removed a complex conveyor network it had installed, and reverted to manual operations at several stations. In April 2018 Musk stated that excessive automation at Tesla had been a mistake, and that humans are underrated.¹⁸

This is a single case and should not be over-generalized. But it is the most heavily capitalized recent attempt to build a maximally automated assembly plant, undertaken by an organization with no legacy labour constraints and strong ideological commitment to automation. It failed on engineering grounds, not policy grounds. Any argument that human-centric constraints are the binding obstacle to full autonomy has to account for this.

9.3 There is a mature literature on why full automation is hard

Lisanne Bainbridge’s “Ironies of Automation” (1983) identified a structural problem that four decades of subsequent work has largely confirmed. Automating the tractable parts of a process leaves human operators with the residual tasks that resisted automation — which are the hardest ones — while simultaneously degrading, through disuse, the skills and situational awareness those tasks require. Increasing automation therefore tends to increase rather than decrease the demands on the remaining humans, particularly at the moments when things go wrong.

This is an engineering finding, not a value commitment. It suggests that human-in-the-loop design may in some contexts be the correct architecture on reliability grounds rather than a welfare concession — which, if true, substantially weakens the framing of human-centricity as pure regulatory friction. Notably, the standard critique never engages with this literature, and its absence is what allows human-centricity to be read as purely political.

9.4 The forecast is asserted rather than argued

Claims about closed-loop generative manufacturing, machine-to-machine micro-economics and self-evolving production systems describe capabilities that are, in varying degrees, demonstrated in research settings and rare in production. They may well mature. But three qualifications are needed. Their timelines are unknown and the field’s forecasting record is poor. The binding constraints on physical manufacturing autonomy are substantially mechanical — dexterous manipulation, fixturing, changeover, error recovery on non-nominal parts — and these have improved far more slowly than the cognitive capabilities that dominate current discussion. And the economics are not obviously favourable: full autonomy is capital-intensive and rigid, and rigidity is costly in high-variance, high-mix production, which is much of European manufacturing.

It is entirely reasonable to argue that policy frameworks should prepare for greater machine autonomy. It is not reasonable to present a specific autonomous future as the known real fifth revolution against which policy is measured. Doing so commits the same error the critique attributes to the Commission: declaring in advance what the next industrial paradigm will be.

Table 3. The autonomy thesis assessed. The direction of travel is plausible; the claim that it is imminent, inevitable and policy-obstructed is not established.

10. What Would Settle the Question

A critique gains force when it specifies what evidence would refute it. The following are stated as falsifiable propositions, with the observation that would decide each. None appears to have been tested, which is itself a finding about the state of the literature.

●        Relabelling hypothesis. If Industry 5.0 is narrative compliance, then an audit of funded projects that invoke the term should find no systematic difference in design, instrumentation or measured outcomes from comparable Industry 4.0 projects. Test: matched-sample comparison of funded project deliverables against declared human-centric or circularity objectives.

●        Cost-penalty hypothesis. If human-centric architectures impose a structural competitive penalty, then plants adopting them should show worse unit cost trajectories than matched plants that do not, controlling for product mix and volume. Test: plant-level panel data on cost, quality and changeover performance.

●        Metric-deficit hypothesis. If the pillars are unoperationalizable, then independent attempts to construct human-centricity and resilience indices should fail to converge. Test: inter-instrument correlation across published indices applied to the same facilities.

●        Obscured-paradigm hypothesis. If policy framing suppresses autonomous production, then the rate of lights-out or near-lights-out deployment should diverge between jurisdictions with and without Industry 5.0-type frameworks, controlling for sector composition and wage levels. Test: comparative deployment counts across the EU, Japan, China, Korea and the United States.

●        Policy-efficacy hypothesis. If Industry 5.0 functions as intended as an R&I instrument, then the composition of funded portfolios should shift measurably toward its pillars after 2021 relative to the pre-2021 baseline. Test: portfolio analysis of Horizon Europe calls and awards.

The value of stating these explicitly is that they separate the parts of the critique that are empirical from the parts that are conceptual. The naming and genealogy arguments are conceptual and can be settled by historical evidence, as attempted above. The distortion, cost and obscuring arguments are empirical and currently rest on plausibility rather than measurement. Distinguishing the two is the main thing the existing critical literature fails to do.

11. Conclusion

Industry 5.0 is not a fifth industrial revolution in any sense comparable to the mechanization of production or the introduction of cyber-physical systems. It introduces no new mechanism for manipulating matter or energy, and its enabling technologies are those of Industry 4.0. On this the critical literature and the European Commission agree, which is why establishing the point settles less than is usually supposed.

The more defensible criticism is about naming. A normative research and innovation agenda was given a number in a technological sequence, and the resulting implicature has proved impossible to retract by disclaimer. The cost is real: the label invites the revolution debate that has consumed much of the literature, and that debate crowds out the operational questions — what human-centricity means in measurable terms, whether resilience can be indexed, whether the funding instrument has changed research composition.

But the historical premise usually attached to this criticism does not survive examination. Prospective, state-originated naming of industrial revolutions did not begin with Industry 5.0. It was arguably present in the European Parliament’s 2007 endorsement of Rifkin’s Third Industrial Revolution, and unambiguously present in the German government’s coining and institutionalization of Industrie 4.0 from 2011. Industries 1.0 through 3.0 were not counter-examples of organic emergence; they were retrospective historical categories, assigned decimal labels after the fact to give Industrie 4.0 a lineage. The taxonomy was a branding convention before Industry 5.0 arrived.

The prediction that the real fifth revolution will be autonomous and machine-driven is the least secure element of the critique. Lights-out production is a quarter-century old and has not generalized; the most aggressive recent attempt to force it failed on engineering grounds; and there is a mature literature arguing that human-in-the-loop architectures are sometimes the reliable design rather than the politically mandated one. To assert a specific autonomous future as the true fifth revolution is to repeat the error being criticized — declaring in advance what the next paradigm must be.

What remains, and it is not small: a framework with sound underlying objectives, a serious and unresolved measurement problem, an untested but plausible risk of terminological capture, and an unfortunate name. Those are grounds for demanding operationalization and evaluation. They are not grounds for concluding that the framework is empty, and the distinction matters for anyone deciding what to do about it.

Note on Sources and Revisions

This version differs from the earlier draft in the following respects:

●        The central thesis was reframed. The earlier draft argued at length that Industry 5.0 is not a technological revolution. Because the European Commission states this itself, the argument was redirected to the naming convention, the genealogy, and the testable consequences.

●        The bottom-up versus top-down contrast was corrected. Industrie 4.0 originated in a German federal working group under the High-Tech Strategy and was institutionalized through state-backed platforms — the closest precedent for Industry 5.0 rather than its opposite. The European Parliament’s 2007 Third Industrial Revolution declaration was added as an earlier instance.

●        The origin of Industry 5.0 was corrected. The term was in use from 2015 with at least four incompatible meanings before the Commission consolidated it in 2021.

●        The optimization argument was re-examined and found to undercut the conclusion it was offered to support; its status as illustrative notation is now stated.

●        The autonomy section was substantially revised against evidence on lights-out manufacturing history, the Tesla ramp, and the ironies-of-automation literature.

●        Counter-arguments were added where the earlier draft asserted contested positions as settled — particularly on competitive disadvantage, on the comparator economies, and on the interpretation of adoption evidence.

●        Sourcing was rebuilt. Citations to a Scribd upload, an ORCID profile page, ResearchGate abstract-request pages and vendor glossaries were replaced with primary documents. One citation in the earlier draft — Kellogg, Valentine and Christin’s “Algorithms at Work,” a study of algorithmic management and worker control — had been used repeatedly to support claims about lights-out manufacturing and autonomous production, which it does not address; it has been removed rather than repurposed.

Works Cited

1. Breque, M., De Nul, L., & Petridis, A. (2021). Industry 5.0: Towards a Sustainable, Human-Centric and Resilient European Industry. European Commission, Directorate-General for Research and Innovation. Publication date 5 January 2021. https://research-and-innovation.ec.europa.eu/knowledge-publications-tools-and-data/publications/all-publications/industry-50_en

2. European Commission (7 January 2021). Industry 5.0: Towards more sustainable, resilient and human-centric industry [news release]. Contains the “complements and extends … not a chronological continuation” formulation. https://research-and-innovation.ec.europa.eu/news/all-research-and-innovation-news/industry-50-towards-more-sustainable-resilient-and-human-centric-industry-2021-01-07_en

3. Rifkin, J. (2011). The Third Industrial Revolution: How Lateral Power Is Transforming Energy, the Economy, and the World. Palgrave Macmillan. On the European Parliament’s May 2007 written declaration endorsing the concept, see also Rifkin, A Smart Green Third Industrial Revolution, European Parliament. https://www.europarl.europa.eu/cmsdata/73881/J%20Rifkin%20-%20A%20Smart%20Green%20Third%20Industrial%20Revolution%20-%20Digital%20Europe.pdf

4. Sachsenmeier, P. (2016). Industry 5.0 — The Relevance and Implications of Bionics and Synthetic Biology. Engineering, 2(2), 225–229.

5. Özdemir, V., & Hekim, N. (2018). Birth of Industry 5.0: Making Sense of Big Data with Artificial Intelligence, “the Internet of Things” and Next-Generation Technology Policy. OMICS: A Journal of Integrative Biology, 22(1), 65–76.

6. Nahavandi, S. (2019). Industry 5.0 — A Human-Centric Solution. Sustainability, 11(16), 4371.

7. Rada, M. (2015). Industry 5.0 — From Virtual to Physical [LinkedIn article]. See also Rada’s subsequent published objections to the Commission’s use of the term; cited as a primary claim of priority, not as an independent assessment.

8. Kagermann, H., Wahlster, W., & Helbig, J. (2013). Recommendations for Implementing the Strategic Initiative INDUSTRIE 4.0: Final Report of the Industrie 4.0 Working Group. acatech / Communication Promoters Group of the Industry-Science Research Alliance. Original term introduced in Kagermann, Lukas & Wahlster, VDI Nachrichten, 1 April 2011. https://www.din.de/resource/blob/76902/e8cac883f42bf28536e7e8165993f1fd/recommendations-for-implementing-industry-4-0-data.pdf

9. Kagermann, H., & Wahlster, W. (2022). Ten Years of Industrie 4.0. Sci, 4(3), 26. Includes the authors’ account of the initiative’s origins and of Chancellor Merkel’s 3 April 2011 endorsement. https://www.wolfgang-wahlster.de/wp-content/uploads/sci-I4.0-Ten-Years.pdf

10. UNIDO. What Can Policymakers Learn from Germany’s Industrie 4.0 Development Strategy? Working Paper 22. On the institutionalization of I4.0 through Plattform Industrie 4.0 (BITKOM, VDMA, ZVEI). https://downloads.unido.org/ot/11/71/11712839/WP_22.pdf

11. Pfeiffer, S. (2017). The Vision of “Industrie 4.0” in the Making — A Case of Future Told, Tamed, and Traded. NanoEthics, 11, 107–121. https://link.springer.com/article/10.1007/s11569-016-0280-3

12. Müller, J. (2020). Enabling Technologies for Industry 5.0. European Commission, Directorate-General for Research and Innovation. See also ESIR Expert Group (2022), Industry 5.0: A Transformative Vision for Europe.

13. Xu, X., Lu, Y., Vogel-Heuser, B., & Wang, L. (2021). Industry 4.0 and Industry 5.0 — Inception, conception and perception. Journal of Manufacturing Systems, 61, 530–535. https://www.sciencedirect.com/science/article/pii/S0278612521002119

14. Government of Japan, Cabinet Office. 5th Science and Technology Basic Plan (adopted January 2016), introducing Society 5.0; promoted by Keidanren.

15. Toward Industry 5.0: Mapping technologies, competencies, and research opportunities. Journal of Entrepreneurship, Management and Innovation, 21(4), 2025. https://jemi.edu.pl/vol-21-issue-4-2025/toward-industry-5-0-mapping-technologies-competencies-and-research-opportunities

16. Analysis of critical success factors for Industry 5.0 adoption ensuring sustainability using MCDM tool: a case study. International Journal of Quality & Reliability Management, 43(2), 686. Cited as an example of the expert-elicitation methods that dominate the empirical literature. https://www.emerald.com/ijqrm/article/43/2/686/1311824/Analysis-of-critical-success-factors-for-Industry

17. FANUC Corporation, Oshino plant lights-out operations, in continuous operation since 2001. See also AMT, Lights-Out Manufacturing. https://www.amtonline.org/article/lights-out-manufacturing

18. Musk, E. (13 April 2018), public statement following CBS interview on Model 3 production. Reported in Bloomberg, CNBC and Fortune. https://www.cnbc.com/2018/04/13/elon-musk-admits-humans-are-sometimes-superior-to-robots.html

19. Bainbridge, L. (1983). Ironies of Automation. Automatica, 19(6), 775–779.

20. Toynbee, A. (1884). Lectures on the Industrial Revolution in England. Published posthumously; the principal vector for the general-usage sense of the term.