Keywords

1 Introduction

Networked production has been increasingly important for companies looking to stay competitive in today’s global marketplace [1, 2]. Initially undertaken due to the requirements of profitability, many manufacturers have been focusing on their core competencies and outsourcing non-core functions to specialised suppliers [1, 3, 4]. The resulting fragmentation made increased digitalisation necessary.

By leveraging technology and digital platforms, companies can connect their manufacturing activities with other organisations, suppliers, customers, and resources, both domestically and globally, to achieve a competitive advantage [5,6,7,8]. Companies do so by accessing a broader range of suppliers and customers, reducing costs, increasing efficiency, and improving the quality of their products when networking. This also allows for greater flexibility in responding to changes in demand, supply chain disruptions, or market conditions and provides the required agility that is so important to counter economic volatility or political instability [9,10,11].

In addition, networked production can create opportunities for innovation and value creation, as companies can work with partners to develop new products or improve existing ones. By sharing expertise, knowledge, and resources, companies can capitalise on each other’s strengths and capabilities to create new business opportunities and improve their competitive position [12, 13]. Companies can thus speed up the product development process and improve time-to-market. This position is also reflected in the concepts of networking in engineering, networked product/service engineering, and collaborative design, respectively [14].

Thus, a significant benefit of networked production is the possibility for companies to move away from traditional sequential or linear process chains and adopt a more dynamic approach to manufacturing and support processes [15, 16]. By connecting their operations with other organisations and resources through digital platforms, companies can access a broader range of manufacturing capabilities, materials, and services. For example, companies can dynamically arrange and release manufacturing and support processes in case production requires a quick scale-up to meet a sudden increase in demand. Additionally, networked production allows and virtually demands that companies share information and collaborate with partners, suppliers, and customers in real-time. This enhances transparency and visibility across the supply chain, which can help to identify and mitigate potential bottlenecks or delays [17].

The functioning of networked production is bound to the management capabilities of the companies involved. Typically, effective management in networked production has required strong leadership and management expertise to ensure that the partners involved work together seamlessly and effectively. Companies needed to establish clear communication channels and protocols, define roles and responsibilities, and establish processes for monitoring and managing performance.

Moreover, the success of networked production depends on the compatibility of integrated, cross-organisational information chains [4]. Network partners need to ensure that they can exchange information seamlessly and securely with their partners, suppliers, and customers. This necessitated the adoption of common standards for data exchange [18]. In addition, data exchange must take place on a trusted basis and therefore requires corresponding communication security measures to protect sensitive information [19].

All this will become even more difficult with a wider expanding network. The increasing complexity of networked production presents a significant challenge to management, which requires new and intelligent approaches to effectively manage and optimise the manufacturing process. While managing networked production would require a holistic understanding of the entire value network involved, including suppliers, customers, internal operations, and other stakeholders, it is hardly possible with traditional corporate management methods based on hierarchical structures [20, 21]. This is also true for the increasingly demanded agile management involving the flexibility and adaptability to respond quickly to changes in customer demand, market conditions, or supply chain disruptions.

To effectively manage networked production, companies can leverage advanced technologies, such as artificial intelligence and data analytics, to develop new management approaches and enhance decision-making. Decentralised Technical Intelligence (DTI) is a new management approach that involves the integration of human and machine intelligence in decision-making processes. DTI allows for the creation of a network of interconnected systems, devices, and agents that operate in a decentralised and autonomous manner, reducing the need for leadership through hierarchy and potentially leading to increased efficiency and productivity [22]. The role of the workforce in this integrated system is crucial, and the human dimension adds a unique quality to the process [23].

To achieve advanced networked production with the help of DTI, a roadmap is necessary, which involves several elements linked to a vision, value promise, and development pathway. The roadmap should include major building blocks, major capabilities, characteristics, and processes typical for the effective implementation of DTI in networked production. By implementing this roadmap, companies can maintain a leadership position in future networked production, which will be crucial for maintaining a competitive edge in the global market.

The technological advancements in networked and self-controlling production can offer new opportunities for creating value, but it is important to ensure that these opportunities align with business goals and are economically viable. The business perspective provides a real-world corrective, meaning that any new technological solutions must also make sense from a business standpoint, considering factors such as profitability and feasibility. Thus, it is important to balance technological possibilities with business realities to ensure the long-term success of any innovation.

This article explores the concept of Decentralised Technical Intelligence and its meaning for advancing networked production as a competitive factor. It will also discuss possible building blocks for implementing an effective roadmap for advanced networked production through DTI. Finally, this article will emphasise the need for a business perspective to ensure that any technological advancements align with the economic viability and create real value.

2 Decentralised Technical Intelligence

Decentralised Technical Intelligence (DTI) is a concept that aims to support the deployment of decentralised, autonomous systems with embedded intelligence in manufacturing. It is a response to the management requirements of networked manufacturing and aims to enable decentralised decision-making and autonomous action while reducing centralised planning or making it redundant. It arose from a demand by the ManuFuture European Technology Platform to increase productivity and efficiency in future manufacturing [22, 24].

DTI focuses on several areas to leapfrog performance gains in terms of high performance, high quality, high resource efficiency, high speed, high flexibility, self-optimisation and self-control, to name just a few (Fig. 1). These performance gains stretch over different areas, including ICT technologies, architectures, platforms and standards, high-performance engineering, high-performance manufacturing systems, and high value-added networked production [22, 24].

Fig. 1
A diagram of decentralized technical intelligence presents how various technologies are processed with decentralized control and self-X capabilities including D T I agents to obtain high-performance manufacturing systems with high key performance objectives and sustainability and zero impact goals.

Utilising technologies to enhance universal impact (adapted from Sautter [22])

Full size image

According to the Plattform Industrie 4.0 [25], decentralised intelligence is grounded on the recognition that the European industry’s strength is derived from a system of innovation and commerce fuelled by diversity, heterogeneity, and specialisation. These elements are integral to the European industrial society. A decentralised regime of open and adaptable systems creates optimal conditions for shaping the digital economy while adhering to the principles of a free and socially-oriented market economy [25].

DTI is adopting this position and is developed as a systemic and interdisciplinary approach that represents a key factor for the macro- and multidimensional transformation of manufacturing systems. It builds on previous developments in systematic and automated manufacturing and can be seen as an evolutionary step towards more advanced manufacturing systems (Table 1). Earlier steps included for example control theory and systems engineering, involving feedback loops, artificial intelligence, cyber-physical systems as a connection of informational and electromechanical components, and holonic manufacturing, which incorporates autonomous and collaborative agents creating an adaptive and flexible manufacturing system [22].

Table 1 DTI as the next evolutionary step in industry performance (adapted from Sautter [22])
Full size table

By integrating human knowledge and experience with artificial intelligence and other advanced technologies, DTI aims to create a more intelligent and adaptable manufacturing system that can respond quickly to changing conditions and optimise performance. The concept recognises that humans and machines each have unique strengths and limitations, and that by combining their abilities and expertise, a more effective and efficient system can be created. This requires a multi-agent architecture, where different agents (including humans and machines) work together in a coordinated manner to achieve common goals. The goal is to create a system where each agent contributes to the overall intelligence of the system and helps to create a more efficient and effective manufacturing process.

Consequently, DTI consciously aims for holism and recognises the importance of humans in industrial manufacturing, which puts the concept in contrast to the earlier conceptualisations that were strictly technological in their orientation. DTI recognises that humans play a crucial role in manufacturing and seeks to empower them with the necessary tools and technologies to make informed and effective decisions. By integrating humans into the manufacturing process, DTI enables them to work alongside machines and systems more collaboratively and efficiently.

Sautter [22] compares DTI as being similar to our nervous system. The body’s nerves make up part of the main control structure, which provides information on the environment and inner body functioning. This way, the autonomic nervous system controls many body functions and processes in a seemingly automatic way with little or no conscious involvement. Similarly, DTI ensures the smooth functioning and good performance of complex technical systems such as production systems. It allows the production system to execute tasks autonomously in interaction with the environment and keeps the system in optimal condition. DTI components can be used to make the system perform tasks autonomously and optimise its performance without central or conscious control. Therefore, DTI is concerned with self-organisation and self-steering aspects in production systems. Mere artificial intelligence, in turn, is a simulation of human intelligence processes by machines based on advanced data analytics and algorithms that emulate and simulate processes and systems. It is often used as a tool to support human decision-making.

The DTI concept can also be considered a forerunner or prerequisite for the industrial metaverse. The industrial metaverse, as a comprehensive digital twin of every relevant process or artefact in a factory or production network and including the human workforce, will make use of existing and developing technologies such as artificial intelligence and machine learning, extended reality, advanced data management structures and cloud and edge computing. These technologies create an interface between the real and digital worlds. This interface aims at offering fully immersive, real-time, interactive, persistent, and synchronous representations and simulations of complex systems such as machines, factories, cities, and logistics networks in a decentralised way [26]. While specific applications in the networked production realm are still to be explored, possibilities include interactive assembling, involving, for example, network partners such as suppliers and customers, collaborative development of products and services, their design, engineering and testing, product presentations and evaluations and product-related services, sales activities and tracking of material flows, and training of personnel [27].

3 Implications for Networked Production Management

Traditionally, the objectives of networked production included the cutting of logistics costs and high inventories, in particular, by improving the coordination and integration of production and logistics activities across the network, the shortening of lead times, especially by improving the availability of information and the flow of materials across the network to generate quick responses, the enhancement of service levels, for example by improving the coordination, collaboration and understanding between companies and of companies’ needs in the network, and enhancing competitiveness, in particular, by increasing the capacity of industrial companies to operate globally in an agile manner through the network [1, 2, 4, 28, 29]. All of these remain true, and these aspects are joined by more recent perspectives on collaborative design and engineering and the management of services.

Achieving these goals requires collaboration among the companies involved. Therefore, all participants need to see benefits from their effort in collaborating. Although companies will try to maintain their autonomy as actors in the network, emerging technologies enabling DTI are likely to dilute the perception of organisational boundaries in the future. Hence, production will largely take place in non-hierarchical networks where decision-making is decentralised [30, 31]. This requires the development of methods and tools for collaborative planning, management, and optimisation of production resources, as well as distributed planning and scheduling [32,33,34]. Other important tools for management include monitoring production and equipment and their maintenance. Increasingly, tools also need to support the requirements of the circular economy.

It is essential to ensure interoperability between the different technologies and software solutions, business and production processes, and organisational structures [35]. All new methods and tools should follow the principle behind DTI, which allows operating in a decentralised manner so that companies can operate in several production networks simultaneously. Conventionally, these functions have been relying on regular updates to a central database with limited learning capacity built-in and neither with consideration of external data. For advanced management of networked production and its monitoring, decentralised ICT systems are needed [36, 37]. These systems use distributed knowledge and integrate heterogeneous data sources to evaluate the state of monitored systems [14].

Competitiveness is based on various performance objectives such as quality, price, delivery service, responsiveness, and flexibility. With shorter product life cycles and a greater variety of models, manufacturers need to respond quickly to customers’ needs and handle capacities flexibly. The technologies driving DTI allow companies to move towards decentralised operations, take advantage of available resources and, for example, be closer to their markets.

Taking advantage of available resources via decentralised operations and intense network interactions offers, in particular, opportunities to run make-to-order and engineer-to-order strategies [38, 39]. Both strategies begin their specific operations only after a customer has placed an order. It represents in some way an orientation towards lean management because it consequently aims at saving resources while avoiding producing to stock. On the other hand, the close customer relationship targets customer wishes, flexibility and responsiveness to individual needs and, therefore, reflects the need for customisation [4, 40]. The mentioned strategies will help companies maintain profitability whilst operating with smaller batch sizes and shorter lead times for customer benefit [14].

The technologies associated with decentralisation enable coordination and synchronisation of production and decision-making processes, which are required in non-hierarchical networks [41,42,43,44]. The production strategies mentioned above are most effective when typical production functions such as engineering, capacity planning, order management, manufacturing, and logistics are integrated comprehensively. DTI technologies allow the inclusion of network partners in all information flows and enable sharing of appropriate resources. When leveraging advanced technologies such as AI, machine learning and edge computing to enable decentralised decision-making and coordination of production processes, humans contribute their domain expertise, intuition, and creativity. They possess contextual knowledge and problem-solving abilities and can provide oversight, make strategic decisions and handle complex and non-routine tasks that require creativity, adaptability and critical thinking. AI, on the other hand, can assist humans by automating routine tasks, analysing large datasets, identifying patterns and anomalies and providing recommendations for optimised decision-making. DTI, thus, facilitates the integration of humans and AI through technologies like edge computing, where AI algorithms can be deployed on edge devices, and cloud computing, where AI models can be trained and deployed centrally. This allows for real-time decision-making at the edge and enables seamless coordination and synchronisation across the production network.

Case Study: Production Scheduling Pilot

In the context of production and supply chain management, coordination of processes and their consequences is crucial for maintaining robustness and resilience. The EU research project knowlEdge - Towards Artificial Intelligence powered manufacturing services, processes, and products in an edge-to-cloud-knowledge continuum for humans [in-the-loop] [45] aims to unlock the full potential of AI in manufacturing by developing AI methods, systems and data management infrastructure, with a focus on edge-cloud computing and collaboration between humans and AI. In a production scheduling use case, the project optimises supply chain planning, demand forecasting and production batch optimisation in an industrial environment. DTI technologies are integral to the solution.

Real-time data from production facilities is collected using sensors and data interfaces at the edge, enabling monitoring, analysis and informed decision-making. Machine learning algorithms analyse historical and real-time data to predict demand, optimise production schedules and identify bottlenecks. Process simulation and digital twins are used to simulate and predict future scenarios, allowing proactive adjustments to production schedules.

The integration of AI and data analytics empowers employees to make informed decisions, reducing errors and saving time. The AI system continuously learns and improves, leveraging insights from documented deviations, errors and solutions. User-friendly interfaces facilitate human-centred interaction and visualisation, enabling monitoring, process adjustments and coordination at the edge. This holistic approach enhances production, facilitates proactive decision-making and improves coordination and collaboration between humans and AI agents.

Essentially, this points to smart and collaborative manufacturing, which involves collaborative planning, management and operation of networked production. The approach deliberately utilises the network as a resource for distributing tasks and acquiring means of production and could also be seen as in line with what Jovane et al. call a plug-and-produce network of production services [14]. This way, the network can respond to immediate challenges in a much better way than under a regime of hierarchical planning and management. Thus, the resulting inter-organisational integration of companies reflects the regime of networked production of the future. This becomes particularly obvious when considering the self-organisational and autonomous operations of DTI agents, which, based on predefined process targets, can easily transect organisational boundaries to carry out tasks. Consequently, in a resource constraint world, the focus on networked production and the potential to expand the network when opportunities arise, companies can leverage the strengths of different network partners and achieve a more robust and resilient production, whilst unlocking new opportunities for growth and innovation [46,47,48].

Following the DTI logic, future production networks will be following a self-organising and self-forming paradigm [22]. Rather than being pure manufacturers of goods, companies will be more and more providers of service, which solve a problem for a user of the manufactured product [49,50,51]. The shift towards a self-forming and decentralised supply chain requires the development of a conceptual framework for a manufacturing-as-a-service (MaaS) approach. This approach involves the provision of value-added services throughout the entire life cycle of a product, including planning, installation, operation, and alterations as well as the recirculation of used product and associated raw materials [52, 53].

In line with the principles of decentralisation and networked production, manufacturing capabilities and resources are provided as on-demand services to companies or individuals. Essentially, MaaS can be facilitated through the integration of technologies such as edge computing, cloud computing and IoT. These technologies enable the connection and coordination of manufacturing resources across a network of distributed nodes.

The decentralisation of activities allows companies to define and execute self-contained service packages independently, supporting the flexibility and adaptability of the network [14]. This means that manufacturing tasks can be outsourced to specialised service providers who offer specific capabilities or expertise. These providers can be in different geographical locations, allowing for flexibility in production and leveraging the strengths of each partner. MaaS may include the integration and coordination of typical manufacturing processes, such as design and engineering, production planning, procurement, assembly, quality control and logistics. Additionally, DTI enables continuous monitoring and control of manufacturing operations, ensuring general visibility, heightened transparency and adaptability throughout the production process.

Additionally, there is a growing demand for “all-inclusive” solutions that incorporate both products and related services, which presents an opportunity for collaboration between larger and smaller companies to design and offer specialised services for new markets [54]. Maas in this context also enables smaller companies or individuals to access manufacturing capabilities that were traditionally available only to larger organisations.

4 Building Blocks and Implementation Roadmap

A technology roadmap is a strategic planning tool that helps organisations to align their technology development efforts with their business goals and objectives [55, 56]. It can also serve as a research agenda for continuous research. This way, the elements of the roadmap, building blocks, required intermediate steps and envisioned outcomes, can be further refined and adapted to new insights. Thus, through the roadmap the major capabilities, characteristics, and processes that are needed to create a successful DTI-enabled and networked production can be understood.

The roadmap for realising advanced networked production with the help of DTI is comprehensive and involves several building blocks, each with its vision, value promise and development pathway. To achieve the desired goals, it is important to have a clear pathway for each building block, divided into near-term (0–5 years), mid-term (5–10 years) and long-term (over 10 years) goals and performances (Fig. 2).

Fig. 2
A block diagram represents different near-term, mid-term, and long-term goals for universal transparency, cooperation to compete, sustainable and circular operations, intelligent control of value networks, and integration with network partners along with respective visions.

DTI roadmap for advanced networked production

Full size image

The first building block, Universal Transparency, aims to enable end-to-end transparency and accountability across value networks, generating added value through comprehensive information available for all involved. The goal is to enable end-to-end transparency and visibility in networked production. An increase in coordination can help to ensure that the different activities within the network are better aligned, which can lead to greater efficiency and better results [57, 58]. This is achieved using various technologies and tools that enable comprehensive availability of information and real-time monitoring of the network’s performance. Universal transparency also promotes customer value by providing more visibility and transparency in the production process, which can help build trust [59]. By building trust, the different partners within the network can better collaborate and achieve their goals together. It also enables social sustainability by promoting accountability and compliance with legal requirements and ensuring that all participants in the network act ethically and responsibly [4]. The near-term goals include progress against sustainability and compliance goals, data management protocols, and information security protocols. Mid-term goals include data analysis for transparency goals involving x-tier suppliers, while the long-term goal is to achieve end-to-end transparency in value chains that extends beyond legal requirements.

Cooperate to Compete is another building block that involves knowledge-based continuous development and network innovation. It involves co-design, co-engineering, and co-production of customised products and services. This building block generates added value by capturing the benefits of collaboration, such as sharing knowledge and resources [60]. The integration of product development into the network is especially important during the early phases of development when collaboration between different actors in the production network can have a decisive influence on competitive factors, such as reducing costs, shorten time-to-market, improve product quality, increase flexibility in production, promote innovation, and develop sustainable solutions [4]. Near-term goals include hybrid systems for knowledge-based engineering and process development, while mid-term goals include inter-factory organisation for enhanced collaborative design and engineering. Long-term goals involve advanced interfaces and methodologies for domain knowledge capturing and situation/context-dependent systems for engineering and process development across factory boundaries.

Sustainable and Circular Operations is a building block that focuses on closed-loop cycles, greater efficiencies, and new ways of sourcing materials. The added value is particularly produced through resource efficiency, extended machine operation, and securing new resources. In terms of resource efficiency, the focus is on optimising the utilisation rate of machines and equipment, avoiding downtime and production disruptions and reducing waste and energy consumption [4, 14]. Sustainability-wise, the emphasis is on minimising the environmental impact of production processes by reducing waste and emissions, using alternative sources for raw materials, such as secondary sources, and considering circular economy principles for material and resource flows. Near-term goals include machinery and equipment to design and engineer products for a circular economy. Mid-term goals include hybrid systems for knowledge-based efficiency gains, and machinery and equipment designed and engineered for long service life. Long-term goals involve decentralised management for reduced risks and increased resilience, minimal manufacturing and maximal servicing paradigm for sustainability, and emphasis on secondary raw materials.

Intelligent Control of Value Networks involves decentralised, autonomous management of value networks, comprising various agents including human and artificial agents. This building block generates added value through better anticipation and agility. In a learning cognitive production network, assets in the network are interconnected and can self-optimise. Depending on the context, control is carried out automatically or autonomously within the collaboration between humans and artificial intelligence [61]. Predictions can be made based on data sources and amounts that would be hidden from humans alone [4]. Near-term goals include AI solutions applied in various parts of the network, while mid-term goals involve integrated platforms and architectures for network-level AI. Long-term goals involve human and artificial agents working seamlessly together, and increased capacity for adaptation, anticipation, agility, and resilience due to decentralised management.

Finally, Integration with (New) Network Partners (or their dissolution of redundant relationships) involves on-the-fly integration, separation, and interoperation, federated architectures, and manufacturing-as-a-service [62]. This building block creates added value through overall network performance, operational efficiency, customer satisfaction, and enhanced products. Companies can expand their capabilities and access new markets, assets, technologies and resources. This leads to enhanced network performance and will also involve strong collaboration between large companies and smaller, niche manufacturers [14]. Near-term goals include continued digitalisation of supply chain planning, connecting internal production and planning systems, and digitalisation of product information. Mid-term goals involve integration of factories with suppliers and customers, while long-term goals involve autonomous transaction processes and restructuring, and manufacturing-as-a-service and experience.

The roadmap for realising advanced networked production with the help of DTI is complex considering, for example, the variety of technologies involved and requires a long-term vision. In addition, it is important to recognise and exploit new business models and opportunities for creating value, while keeping in mind the viability of the proposed solutions from a business perspective.

5 DTI Deployment from Business and Organisational Perspective

The industrial evolution towards networked and self-controlling manufacturing concepts enables and generates groundwork for new business models, as well as sets new requirements for the skills and expertise of the personnel. In addition to the need to maximise profits, new challenges such as acting properly in the face of environmental damage cannot be ignored. It is therefore a matter of recognising and exploiting new opportunities for creating value throughout a product’s life cycle. Due to its cross-company nature, networked manufacturing offers the appropriate conditions for this, for example, partnerships for knowledge-based and sustainable extraction of necessary resources, manufacturing and product development, and innovation development. At the same time, new business models must also recognise new forms of work and the development of skills, learning and education.

Digitalization of companies’ and organisations’ processes enables the separation of data and information from their physical/real operational environment. This in turn enables new ways to organise work within and between organisations, which also leads to new business model opportunities. In practice, the so-called cyber-physical systems can be implemented where physical processes are monitored by creating a virtual counterpart of the physical world in which decentralised decisions can be made by communicating and cooperating with each actor in real-time using the Internet and IoT components [63].

From an organisation point of view, the DTI vision brings about changes to the distribution of work between human employees and computer systems. More and more human processing can be automated to be done by smart ICT systems. This provides, on the one hand, an opportunity to increase the efficiency of the organisation, and on the other hand, develop new kinds of data-based services and new ways of connecting customers, partners, and suppliers. Furthermore, the skills and competencies of the personnel must be developed and adapted to meet the new requirements the DTI deployment entails. In general DTI increases the connectivity of the different stakeholders, which means that the actors need to broaden their perspective on their business environment to include partners that are further away from them in the value chain, e.g., they need to consider the operations of their customer’s customer. More systemic thinking is required and skills and competences to realise it. An example of the business opportunities this enables is a condition based maintenance service concept, where a component or sub-system provider is connected to the equipment in use either directly or via OEM’s systems to follow the product performance during its life cycle. All these connections, data availability and security issues need to be understood by all partners in the connected system. This requires enhancement of two kinds of competences: (1) competences to understand the technological system and its operation, and (2) competences to understand the operational environment of the end-product at end-user location.

In the digital world, the DTI envisions new services, such as platforms, that are required to enable the myriad of potential connections of digital data flows between different stakeholders. This digital world forms its market environment where new platform providers as well as traditional business ecosystem actors with different resources, competitive advantages and strategic aims are positioning themselves. The digital marketplace is much more flexible and dynamic than the traditional as digital relationships between actors are easier to establish and terminate. Digital platforms have a specific characteristic of aiming to orchestrate a large number of business actors, i.e., their business is based on connecting other actors. This challenges the traditional strategic planning of organisations rooted in the resource-based view of the firm, which focuses on achieving competitive advantage by owning and controlling resources [64]. These resources should have the so-called VRIN characteristics, i.e., be valuable, be rare, be inimitable, or be non-substitutable. Digital platform business models also challenge the positioning framework of strategic management [3], which focus on competitive advantage by raising structural barriers to entry and increasing bargaining power. The digital marketplace enables more dynamic competition, where, e.g., disruptive innovation from start-ups and entry by firms from different industry sectors can bring completely new competitive pressure to existing business models [65]. Instead of competition between firms, the competition is between business ecosystems consisting typically of one or two central (platform) companies and other firms providing complementary goods and services. In this kind of context, the value capture component of a business model should balance the focal firm’s profits and the profitability of the ecosystem partners [65].

While creating new business opportunities the dynamic digital marketplace creates new challenges as well. Many new digital solutions and services are based on combining data from different sources, i.e., from different stakeholders. The knowledge and capabilities of different stakeholders are combined to deliver a specific offering or solution. This can create a challenging business landscape where the perspective on specific assets and complementary assets may be viewed differently by different stakeholders, especially between platform providers and linked service providers [64]. This can bring about risk, especially to the linked parties of the digital ecosystem around the platform. On the one hand, the platform provider may be enticed to integrate successful solutions provided by the linked parties into its offering. On the other hand, the platform operator may stop providing required data sources to the linked parties, thus destroying their digital business model. Naturally, the platform provider has to consider the risk these kinds of actions can cause to its business reputation [64].

Another challenge posed by the digital integration of businesses is the inclusion of the smaller actors in the value chains and business ecosystems, especially considering their digitalization skills and competencies [66]. SMEs are often an important part of a value chain in the manufacturing sector. Thus, to take full advantage of the DTI their integration is essential. Lack of resources and limited competencies are generic barriers for SMEs to implement new technologies. On the one hand, these barriers can be overcome by introducing scalable technology solutions that are easier to be deployed by SMEs [66]. On the other hand, new services and service providers can be established to support the SMEs in digitalization, from understanding the systemic change it brings to their business to running specific technology services for them [67].

6 Discussion

This article discusses the use of Decentralised Technical Intelligence (DTI) to improve networked production. While the primary focus is on the network, the transition is built on a multilevel perspective that spans from technical processes on the factory floor to networking and business operations at the upper level. This requires the integration of various micro and macro levels in real-time through both company internal and external networks of information and communication technologies, necessitating international standards.

Consequently, it is essential to develop and integrate a multilevel architecture for control and management, which includes implementing decentralised intelligence through edge-cloud solutions. This integration allows for the effective combination of physical levels with ICT control structures.

This development implies that organisations involved in the networked production will integrate more and their roles may transform into functional specialisation or strategic competencies within the network. This integration leads to a synergistic effect among collaborating companies, resulting in significant efficiency gains and cost savings across different categories of costs. Achieving advanced networked production through DTI requires following the described roadmap and implementing the building blocks of DTI in a way that is transparent, trustworthy, cooperative, sustainable, and intelligent.

The figure (Fig. 3) presented in this article illustrates the approach’s goal to surpass the current fragmented approaches that mainly involve incremental improvements. Despite witnessing progress within the different levels of the production system and the interconnections between these levels, disciplinary boundaries often limit us. With DTI, we can surpass these limitations by enabling autonomous activities of DTI agents, which include both humans and machines. This can lead to various benefits such as self-organisation, self-optimisation, and self-maintenance. All of these actions occur within a predefined performance space, with a focus on situational and contextual requirements.

Fig. 3
A chart represents the progress in networks, factories, production divisions, production systems, production cells, machines, and shop floor processes by implementing advanced networked production using D T I agents.

Progress towards advanced networked production based on DTI (adapted from ManuFuture [24] and Sautter [22])

Full size image

Moreover, leveraging DTI technologies enables a shift away from strict hierarchies towards more flexible and adaptive operational management. With DTI, employees are empowered to make decisions based on their expertise and the real-time information provided by AI systems. This empowerment allows for quicker response times, as employees can take immediate action and adapt to unforeseen events without waiting for approval from higher levels of management.

Thus, DTI involves converging towards heterarchy, which refers to a non-hierarchical organisational structure where decision-making authority and control are distributed among various nodes or agents rather than being concentrated at the top of a hierarchical chain. Heterarchies emphasize collaboration, decentralised decision-making and empowered employees [68].

As a result, DTI facilitates a systemic and transcending approach that dissolves the boundaries of organisations and disciplines, allowing for a fundamental transition in production. DTI catalyses leapfrogging performance, as Sautter [22] describes it. DTI also provides a novel understanding of process quality, flexibility, and resilience, enabling the production system to adapt to changing situations.

From a business implementation perspective, it is essential to develop incentive systems that support the collaboration of the different stakeholders in the network. Each actor should have a clear business value proposition to engage in the activities DTI entails, i.e., collaboration, commitment to common goals, and sharing of knowledge and information. This may require, e.g., further development of pricing methods and mechanisms for knowledge and data.

7 Conclusion

DTI has wide-ranging implications for long-term development in industrial production. The concept has the potential to alter operative management and enhance flexibility, adaptability, robustness, and resilience in production networks.

Understanding the benefits will coincide with a surge in strategic management efforts, research, development and deployment of technologies, which provide an underpinning to the decentralisation of intelligence and agents’ actions and responses.

DTI is an overarching visionary concept for the future of manufacturing. Implementing it is a long-term process that requires a mindset change from all the actors in manufacturing ecosystems. Changing all at once is not viable, instead, incremental development steps in collaboration with relevant actors in the manufacturing networks should be taken.

The roadmap serves as a guiding framework that reflects the need for incremental development in the implementation of DTI and emphasises the importance of taking progressive steps forward while allowing for continuous refinement and adjustment. Stakeholders can align their efforts towards realising the long-term vision of enhanced flexibility, adaptability, robustness and resilience in production networks. This will position them at the forefront of the future of manufacturing.