Digital transformation accelerates across manufacturing but integration failures remain major roadblock

The proposed framework shows how digital tools can be organized into a connected system that supports real time performance, long term planning, and strategic decision making. It also offers a structured method that small and large manufacturers can use to plan digital adoption without losing alignment between technology, people, and business goals.

A new wave of evidence shows that industrial digitalization is moving from early technology deployment toward full ecosystem integration, reshaping the structure of modern manufacturing. Mapping this shift, a new study tracks how factories are leaving behind isolated automation tools and adopting intelligent, connected, and increasingly human-centered production systems.

The study, titled Industrial Digitalization: Systematic Literature Review and Bibliometric Analysis, published in Information, brings together findings from 61 peer-reviewed papers released between 2020 and mid-2025. The authors say the landscape is changing fast, driven by advances in artificial intelligence, digital twins, machine learning, cyber physical systems, and the demands of Industry 5.0.

Growing digital pressure on industry

The review shows how digitalization accelerated sharply after 2020, pushed by global supply disruptions, rising competition, and the need for remote operations. Research output grew year after year, and the vocabulary around smart manufacturing expanded with it. The most common themes included smart manufacturing, Industry 4.0, artificial intelligence, machine learning, IoT, cyber physical systems, and eventually Industry 5.0.

The authors report that early work focused on building digital foundations inside factories. This included IoT networks, industrial robots, connected machines, edge and cloud systems, and early uses of big data tools. As the timeline moved forward, research turned toward human-machine collaboration, sustainability, cognitive support systems, and digital ecosystems that spread across supply chains and industrial networks.

Industry 4.0 remained at the center of almost all research, but by 2022 and 2023, new themes such as human-centered design, workforce skills, green digital transformation, and data governance became more common. By 2024, predictive quality systems, sensor fusion for advanced monitoring, digital twin integration, and maturity models for digital readiness dominated the literature. Early 2025 papers showed a stronger push toward the Industrial Metaverse, AI-enhanced digital twins, and systems engineering that links physical and virtual operations.

Across all years, the authors found a clear trend. Research began with narrow applications like predictive maintenance and machine intelligence, then grew into broader questions about how organizations should structure digital adoption, how workers fit into these systems, and how businesses can integrate sustainability into every layer of production.

The rise of a fully integrated digital factory

The study primarily reviews structural changes in digital manufacturing research. The authors identify a steady move away from isolated tools and toward technology stacks that must work together. In the past, factories adopted sensors, robotics, cloud systems, or analytics separately. Now, digitalization requires seamless communication from equipment on the shop floor to planning systems in the cloud.

To describe this shift, the authors propose a full industrial digitalization framework. It works across several layers. The first layer covers the physical systems that gather data and support operations. This includes machines, robotic stations, and human-operated work areas equipped with sensors. The second layer handles edge computing, which allows fast decision making near the source of data. The third layer handles fog computing and acts as the local coordination point for manufacturing execution systems and digital twins. The fourth layer is a cloud environment that drives long term analytics, planning, and enterprise systems such as ERP, CRM, and finance tools. The final layer, called decision support, combines insights from all previous layers into guidance for managers and operators.

The authors state that the strength of this approach is its ability to scale. It allows factories with limited budgets, especially small and medium manufacturers, to adopt digital tools step by step. They can begin with basic sensors, then add edge analytics, then expand toward integrated cloud intelligence as resources allow. The design also supports both manual workstations and fully automated systems, which is important for industries that still depend on skilled labor.

The review notes that this layered approach also fits the needs of Industry 5.0, which aims to keep humans involved in decision making while using AI and automation to support quality, safety, and sustainability.

To test the real-world value of the framework, the authors ran a proof of concept with an electronics manufacturer. The company had already begun automating parts of its lines, but the framework added edge analytics, a more complete execution system, stronger integration between planning and production, and improvements in warehouse tracking. A digital maturity tool called the Smart Industry Readiness Index was used to measure performance. The authors report clear progress across multiple areas, including automation, connectivity, operations, supply chain readiness, and management structure.

The gains were strongest in areas where new sensors, edge AI tools, and coordinated execution systems were deployed. The only area with modest improvement was product lifecycle management, which the authors say remains a weak point across many industries because it requires strong links between engineering and production data.

Barriers slow adoption but new opportunities appear

Although digitalization is gaining momentum, the study highlights major barriers. Many firms still struggle with system integration. Legacy machines often do not communicate well with modern software. Data formats vary widely. Interoperability standards exist but are not yet used consistently.

Cybersecurity is another challenge. As factories connect more machines and rely on cloud services, they face a higher risk of breaches. Older industrial systems were never designed for secure remote access. This creates new pressure on IT and operations teams, and adds costs for monitoring, training, and compliance.

Skills shortages appear throughout the review. Workers need training not only to operate new tools, but also to understand data flows, digital maintenance, and the logic behind AI systems. The authors say many firms underestimate the cultural change needed to support digital transformation. Without leadership support, communication, and training, projects often remain isolated and fail to reach their full potential.

Cost is another barrier, especially for small and medium manufacturers. Sensors, infrastructure, cloud services, and integration tools require investment. Returns on investment are not always immediate. The review notes repeated evidence that smaller firms often postpone digital projects due to budget constraints.

On the other hand, the authors outline major opportunities that are beginning to shape the next phase of industrial digitalization. One major trend is the rise of real time learning systems. Machine learning tools can detect defects, predict failures, and adjust production flows with growing accuracy. Digital twins allow factories to test scenarios before making changes on the floor. Human-machine collaboration tools support operators with insights during complex tasks. Sustainability is also rising as a key driver. Green IoT, energy monitoring, and resource efficiency tools align digitalization with environmental goals.

The most significant opportunity is the formation of connected industrial ecosystems. Instead of isolated digital factories, industries are beginning to build shared platforms and data networks. This allows partners, suppliers, and customers to work together through shared intelligence and integrated planning. Such systems could become vital as industries face supply chain disruptions, energy costs, and global competition.

A new direction for research and industry

The proposed framework shows how digital tools can be organized into a connected system that supports real time performance, long term planning, and strategic decision making. It also offers a structured method that small and large manufacturers can use to plan digital adoption without losing alignment between technology, people, and business goals.

The authors suggest several directions for future work. Research should expand beyond English-language studies and include more findings from emerging economies. It should study underused areas such as full integration between engineering and production systems, advanced edge AI for fast quality control, and stronger links between sustainability and digital planning. Stakeholder views also need deeper study, especially from operators, planners, and maintenance teams who interact directly with digital systems.