1. Introduction
The industrial landscape is undergoing through a profound transformation, evolving from Industry 4.0 to Industry 5.0. This process was, in the initial phase, primarily driven by the convergence of multiple technologies that reached a point of maturity, initially triggered by the advancements in cyber-physical systems (CPS) and Internet of Things (IoT). These developments have since been augmented by various other so-called “exponential technologies.” Complementarily, political and economic strategies have increasingly emphasized re-industrialization and revitalization of key sectors through digital transformation [
1,
2].
While Industry 4.0 was primarily technology-centric, Industry 5.0 introduces broader priorities, including human-centricity, sustainability, and resilience [
3]. Nonetheless, technology continues to play a central role [
4,
5], with growing emphasis on artificial intelligence (AI), hyper-connectivity among systems, machines, organizations, and people, and the need for robust cyber-security.
This increasing hyperconnectivity reinforces the importance of collaboration among networks of autonomous entities, which is captured by the notion of collaborative network (CN) typically referring to a structure “composed of a variety of entities—organizations people and even smart machines—which are largely autonomous, geographically distributed, and heterogeneous in terms of their operating environment, culture, social capital and goals … that collaborate to (better) achieve common or compatible goals” [
6]. Originally focused on supply chains, extended enterprises, and virtual enterprises, the CN concept has progressively expanded, now encompassing diverse forms and impacting all dimensions of industrial systems. This evolution is also reflected in various roadmaps/strategic research agendas for advanced manufacturing [
6,
7], which increasingly acknowledge the multi-faceted role of CNs.
In this context, the present article offers a brief overview of these developments and emerging trends, along with the author’s views on the main challenges ahead. While updating previous works (e.g. Refs. [
1,
8]), it further aims to offer a synthetic perspective of the potential contribution of CNs to transition towards Industry 5.0.
2. Overview of CN evolution
As technology advances and systems become more integrated and interconnected, diverse models of CNs have emerged. These models represent a variety of organizational forms, support multiple business models, and are facilitated by evolving infrastructures and collaboration platforms. Similar to the classification of industrial revolutions, the development of CNs can be grouped into distinct generations (
Fig. 1), as initially proposed in Ref. [
8].
• CN 1.0—the initial phase of CNs, primarily centered on goal-oriented networks. This stage saw the emergence of dynamic supply chains, extended enterprises, virtual enterprises, and virtual organizations.
• CN 2.0—marked by the conceptualization of the virtual organization breeding environment as a strategic long-term alliance to facilitate the preparedness of its members to engage in dynamic collaborative consortia. It encompasses business ecosystems, industry clusters, industrial districts, and professional virtual communities, and marks the formal introduction of the term “CNs.”
• CN 3.0—emphasizing interaction among multiple CNs. It includes hybrid value system networks, co-creation, and open innovation ecosystems (e.g. digital innovation hubs), and explores complex issues such as multi-level membership, transitions and inheritance across networks, and the rise of multi-supplier service-based models.
• CN 4.0—extending the concept of collaboration to encompass networks of (smart) CPS, including interactions among digital twins representing physical entities. This stage also introduces the perspective of collaborative networks as systems of distributed cognition and highlights their role in support of resilience.
• CN 5.0—reflecting developments along two main directions: ① the emergence of hybrid human-AI CNs, and ② the ability to operate effectively in volatility, uncertainty, complexity, and ambiguity (VUCA) environments. This phase emphasizes advancements in value creation, as well as enhancing resilience, and fostering antifragility within collaborative systems.
As this evolution continues, new forms of CNs will naturally emerge, shaped by evolving needs and contextual demands.
3. CNs in Industry 4.0 and digital transformation
As highlighted in previous works (e.g. Refs. [
1,
6,
9,
10]), CNs serve as a key enabler of Industry 4.0 and broader digital transformation initiatives. Their role spans all dimensions of Industry 4.0, encompassing both the manufacturing/production systems and product/service perspectives (
Fig. 2).
3.1. Manufacturing system dimensions
At the manufacturing/production systems level, we can consider three sub-dimensions:
(1) Vertical integration—emphasizing the convergence of physical and digital worlds through CPS, IoT, and the concepts of agentification and digital twins, enabling dynamic, responsive manufacturing ecosystems. At higher organizational levels, the shift is away from large, monolithic systems (e.g. traditional manufacturing execution system (MES) and enterprise resource planning (ERP)) toward federated, collaborative services. This shift supports a redefined view of the enterprise as a multi-layered CN ecosystem, composed of interconnected smart components and human actors.
(2) Horizontal integration—focusing collaboration across the value chain—including suppliers, manufacturers, customers, and recyclers. As manufacturing moves from large, centralized facilities to smaller, distributed production units, networking and collaboration among decentralized entities become crucial. Emerging paradigms such as distributed and social manufacturing highlight the importance of inter-organizational collaboration.
(3) Acceleration of manufacturing—triggered by the rise of “exponential technologies” (e.g., robotics, drones, AI, additive manufacturing, sensing, nanotechnologies, neuro-technologies, mobile computing, and big data analytics), which introduce new actors, work practices, and forms of human-machine interaction (e.g., augmented reality (AR), collaborative robotics, and remote interaction), giving rise to novel collaborative modalities.
3.2. Product and service dimensions
(1) End-to-end engineering—focused on integrating all product lifecycle stages, from design to disposal—which requires both internal and external collaboration. Internally, this entails collaboration across departments such as market research, design, and production. Externally, it involves engaging suppliers and customers in co-design and co-innovation, as well as leveraging social networks to “sense” market trends. Servicing during use and recycling at end-of-life further extend the scope of collaboration beyond the firm.
(2) Smart products—encompassing digital modeling and embedding intelligence into products. For instance, features like memory tags enable traceability and support “stigmergic collaboration” among stakeholders. Furthermore, the fusion of products and services, as in product-service systems, demands multi-enterprise collaboration.
(3) New business models—connectivity and digitalization catalyze new business models such as customer-intimate strategies, extended servitization, and service-dominant logic, “glocal” operations, hybrid value chains combining social and economic goals, and data-driven services. All these models rely on strong multi-stakeholder collaboration.
In addition, a range of transversal challenges cuts across all these dimensions. One notable example is cyber-security in CPS [
11]. The increasing availability of data, fuelled by widespread sensor deployment and systems interconnectivity, leads to broad adoption of bigdata analytics and machine learning [
4,
12].
4. CNs in Industry 5.0
Industry 5.0 emphasizes new goals towards achieving industrial systems that are (more) sustainable, resilient and antifragile, intelligent, and human-centered (
Fig. 3).
The pillar of Sustainability in
Fig. 3 can be well supported by concepts and mechanisms introduced in CN 2.0 and CN 3.0. The other two pillars map well with the concerns and developments of CN 4.0 and especially CN 5.0.
4.1. CNs and sustainability
Sustainability is inherently complex, due to its multi-dimensional nature and the challenge of reconciling often conflicting objectives among diverse stakeholders. In manufacturing, this complexity is reflected in concepts like sustainable manufacturing, industrial symbiosis, and the circular economy.
Effectively implementing these concepts relies on collaboration among involved actors, which underscores the critical role of CNs in enabling sustainable production. For instance, the business ecosystem concept, as in virtual organization breeding environment (VBEs), illustrate shared sustainability responsibility [
6], while circular ecosystems emphasize closed-loop value creation and resource efficiency [
13].
Indeed, across Industry 4.0 dimensions, networks of organizations, individuals, smart machines, and intelligent systems can collaborate toward global efficiency and sustainability. These heterogeneous, autonomous entities pose integration challenges beyond interoperability and control, necessitating environments based on negotiation, contracting, resource sharing, and value co-creation.
Viewing manufacturing systems as networks of (semi-)autonomous components implies distributed responsibility [
6]. Consequently, sustainability must be addressed through a CN lens, emphasizing shared accountability and coordinated efforts among all involved participants.
Previous research has largely focused on CNs’ contributions to sustainability via horizontal integration dimension, particularly in circular economy, industrial symbiosis, and, to some extent, new business models such as hybrid value chains [
14,
15]. Other areas remain underexplored, signalling the need for further analysis of trends and examples to identify research gaps.
While Industry 4.0 initially emphasized technological advancement, recent developments increasingly target sustainable manufacturing practices—addressing not only economic factors like cost and profit but also social and environmental aspects. However, the literature offers limited insights into how collaboration directly or indirectly enhances sustainability within industrial ecosystems.
Achieving sustainability requires measurable progress, supported by performance indicators tailored to evaluate collaboration’s impact on resilience and sustainability. Though initial efforts exist [
6], well-established key performance indicators (KPIs) remain scarce. Moreover, the selection and weighting of KPIs can influence the behavior of CN members, possibly driving stronger sustainability commitments and long-term engagement [
16], a critical area for further research.
4.2. CNs, resilience, and antifragility
Modern manufacturing systems are increasingly exposed to a broad range of disruptive events that can severely affect their operations. These disruptions arise from sources such as natural disasters, pandemics like COVID-19, climate change, terrorism, economic crises, armed conflicts, political instability, and demographic changes [
17], [
18], [
19]. This turbulent environment is encapsulated in the VUCA acronym. Such events can disrupt both global supply chains and local business ecosystems. To address these challenges effectively, CNs must foster resilience, that is, the ability to absorb shocks and recover, and antifragility, that is, the capacity to adapt and grow stronger post-disruption. Despite being long discussed [
18], the application of these concepts in collaborative, sustainable manufacturing is still limited.
In fact, numerous strategies and mechanisms have been proposed to enhance resilience and antifragility, but they are often fragmented, with inconsistent terminology and disconnected perspectives. A structured overview of mechanisms and capabilities for building resilience and antifragility in collaborative business ecosystems is presented in Ref. [
18], offering a classification of strategies and foundational capabilities. However, further research is needed to assess the relevance and adaptability of these strategies to different industrial business ecosystems, considering their specific contexts and characteristics.
5. CNs in human-centric systems
Industry 5.0 marks a shift from purely efficiency-driven and automated systems toward approaches that prioritize human needs, values, and well-being. This perspective unfolds across several sub-dimensions:
•
Value creation for society. In this view, systems are designed not solely for efficiency or profit, but to generate sustainable, inclusive, and ethical value. It promotes participatory involvement (co-creation) and equitable sharing of benefits and responsibilities. Although several methods for fair value distribution have been proposed [
20,
21], such as game theory, Shapley value, social network analysis, and transaction cost theory, the concept of value co-creation represents a significant paradigm shift that remains underexplored [
22]. A key challenge lies in measuring individual contributions, especially in creative or knowledge-based collaboration.
• CNs, particularly as collaborative business ecosystems, can be instrumental in advancing goals such as job creation, upskilling, small and medium size enterprises (SMEs) support, reducing inequalities, and promoting inclusion and sustainability. In this direction, notable examples include social manufacturing, a model that blends decentralized, customized, and community-driven manufacturing, combining social interaction with smart technologies; and hybrid value chains, which merge traditional business approaches with social innovation, by engaging companies, non-governmental organizations (NGOs) and local communities to tackle social challenges collaboratively.
•
Human-in-the-loop—that combines human intelligence with automated systems to enhance accuracy, adaptability, and ethical decision-making. It emphasizes human-machine collaboration, ensuring human oversight where full automation may be inadequate or risky. Technologies such as AR, VR, and natural language processing are key enablers. This approach echoes the earlier concept of balanced automation systems, which advocated a thoughtful distribution of roles between humans and machines [
23].
• Indeed, Industry 5.0 values not only efficiency and automation but also creativity, ethics, empathy, and critical thinking [
24]. This shift calls for changes in engineering education, preparing professionals to act as system architects, ethical decision-makers, and facilitators of collaboration.
•
Human-
AI collaboration/hybrid teams. AI is already widely adopted across industries, especially for optimization, sensory perception, and machine learning [
25]. A paradigm shift is now underway, from treating AI as a tool to viewing it as a collaborative teammate. Although most early works focused on bilateral setups (one human and one AI agent), as in collaborative robotics and AI copilots, it is becoming increasingly evident that greatest gains can come from integrated human-machine teamwork. Rather than replacing humans, AI augments human capabilities, enabling new forms of collaboration based on shared goals, mutual understanding, coordinated actions, co-learning, and continuous feedback.
• Moving beyond bilateral interaction, the notion of hybrid collective intelligence or hybrid teams, emerges as a framework where multiple humans and intelligent systems collaborate synergistically [
26,
27]. Research in this area often draws on psychology and social sciences to shape the design of effective AI teammates, emphasizing roles, expectations, explainability, and even physical embodiment. These principles influence how AI is embedded in robots, virtual agents, or decision-support systems.
• Key challenges here include developing shared mental models, enhancing explainability, and fostering co-learning. This direction aligns with broader efforts to “reimagine work,” using AI to improve performance and well-being. Recent research emphasizes defining design principles for AI teammates, such as roles, constraints, and expectations, often inspired by social sciences [
28], which sometimes leads to friction between traditional AI experts and these newer interdisciplinary contributions. As understanding of AI’s limits and possibilities matures, cross-disciplinary collaboration becomes essential in designing future intelligent manufacturing systems. Ethical considerations, namely ensuring human oversight, preventing misuse (e.g., disinformation), promoting trust transparency, and embedding human values, remain essential. These aspects align well with Industry 5.0's vision of technology serving broader human goals.
•
Importance of learning from Nature. Nature offers countless examples of collaboration marked by sustainability and efficiency [
29]. Extensively studied in fields like ecology, biology, and zoology, these systems provide rich insights into collaborative mechanisms, role distribution, behavioral patterns, and organizational structures. For instance, the concept of business ecosystems directly mirrors natural ecosystems [
30], while stigmergy, observed in the collective behavior of social insects, has been effectively applied to mass collaboration scenarios [
31]. Another example seeks inspiration in nature for self-organized collaborative communities [
32]. This rich interdisciplinary knowledge offers valuable guidance for designing more efficient, resilient, and sustainable CNs in industry.
6. Conclusions
The adoption of a CNs perspective can significantly advance the realization of Industry 4.0 and Industry 5.0 objectives. Beyond supporting these paradigms, it also enables a rethinking of system design from a novel standpoint, revealing new opportunities for innovation.
To fully leverage the potential of CNs, a robust interdisciplinary and multidisciplinary approach is essential. Contemporary industrial systems must increasingly be understood and treated as socio-technical systems, integrating both technological and human dimensions.
Simultaneously, it is crucial to temper the enthusiasm surrounding emerging technologies, avoiding overhyped expectations. This involves acknowledging the varied maturity levels of technologies and fostering an “intergenerational” dialogue and integration, that is, a coexistence of technologies at different stages of their life cycles.
Ultimately, intelligent manufacturing is, by its nature, networked and collaborative, requiring seamless coordination and sharing among distributed and heterogeneous actors to achieve adaptability, resilience, and innovation.
Besides the adoption of a CN perspective and integration of existing results, several important research directions are identified, including:
• Development of effective hybrid human-AI CNs, properly addressing distribution of roles, supporting mutual understanding and trust, dealing with emotions and ethical principles, while coping with continuous technological evolution.
• Further exploration and validation of collaborative models in support of resilience and antifragility.
• Development of an assessment framework with proper KPIs and responsibility distribution models to leverage collaborative approaches to sustainability.
• Further development of models of value co-creation and fair benefits and responsibility distribution in collaborative contexts.
• Identification of needed skills and education for the new industrial contexts.
PDF
(1916KB)
