Smart Manufacturing for the Space Sector: Integrating Cyber-Physical Systems Across Processes and Facilities

1. Introduction

The Fourth Industrial Revolution has introduced a paradigm shift in manufacturing, characterized by highly connected, flexible, and intelligent production ecosystems. This transition is driven by advanced digital technologies such as cyber-physical systems (CPS), Internet of Things (IoT), artificial intelligence (AI), cloud computing, and Big Data Analytics, which enable the seamless integration of physical and digital domains [1–3]. These technologies bridge operational technologies (OT) and Information Technologies (IT), creating reconfigurable and adaptive manufacturing systems that serve as the foundation of Smart Manufacturing (SM) [4–7]. Among these enabling technologies, CPS play a pivotal role by embedding sensing, computation, and actuation into industrial processes, thereby allowing real-time monitoring, self-awareness, and adaptive control [8, 9]. The five-tier CPS architecture – ranging from smart connection to cognition and configuration, as shown in Figure 1 – provides a robust framework for self-aware machines, extending “closed-loop smartness” from individual components to the entire production value chain. This evolution echoes the earlier concept of Smart Structures, where sensors and actuators were integrated within feedback architectures to provide adaptability and self-functionality [10–12]. In fact, Smart Structures can be regarded as archetypes of CPS, and their transition into manufacturing settings underpins the development of Space Smart Factories [13–15]. The space sector provides a particularly fertile ground for these innovations. Traditionally dominated by low-volume, highly customized production, the industry is now facing the challenges of the New Space Economy: large constellations of small satellites, shortened production cycles, and increased commercial competition [16–18]. This transformation, often referred to as Space 4.0, is reshaping global supply chains and enabling broader participation of private actors, start-ups, and research institutions [19–21]. The European Space Agency (ESA) and NASA have launched initiatives such as Design 2 Produce and the Space Technology Mission Directorate (STMD) to promote advanced manufacturing, digital twins, and additive manufacturing for space applications [22, 23]. Smart Manufacturing methodologies are increasingly being applied to manufacturing, assembly, integration, and testing (MAIT) processes in space systems. Case studies include automated production of composite sandwich panels for mega-constellations [24], digital transformation of facilities into Smart Factories [25], and virtual testing approaches for spacecraft qualification [26]. These examples demonstrate how digitalization and CPS integration enhance visibility, traceability, and responsiveness, ultimately improving efficiency and scalability while reducing costs. Despite rapid progress, significant challenges persist. Many factories have yet to achieve full vertical and horizontal integration, and real-time decision-making capabilities remain limited [27]. Furthermore, the adoption of smart technologies in space manufacturing must carefully balance performance improvements with capital investment and certification requirements [28]. Structured frameworks, such as those based on the Acatech digital maturity model, are therefore essential to guide the stepwise transition toward smart, adaptive, and resilient production systems [29]. In this context, the present chapter introduces a comprehensive methodology for integrating CPS and Industry 4.0 technologies into the space manufacturing domain. Building on the conceptual lineage from Smart Structures to Smart Manufacturing, the chapter presents a framework for digital transformation of MAIT processes. Through selected industrial case studies – including automated sandwich panel production, Smart Facility Management Systems (SFMS), and virtual testing – the chapter illustrates how the space sector can embrace Industry 4.0 principles to meet the scalability and reliability demands of the New Space paradigm [24, 30, 31].

2. A structured framework for the integration of MAIT processes into the space industry

The integration of MAIT processes into the digital transformation of the space sector requires a structured framework that goes beyond the mere adoption of isolated technologies. A stepwise methodology demonstrates how space manufacturing can evolve from traditional sequential workflows into interconnected cyber-physical ecosystems [31, 33–35]. The proposed framework is articulated through the following complementary layers:

1. Process selection and characterization: data collection and process analysis.

   This first step involves identifying the most representative and complex MAIT processes in the considered industrial environment, assessed against criteria such as criticality, lead time, defect rate, and potential for scalability [33]. A detailed analysis of the production processes is performed. During this phase, process bottlenecks and critical issues are identified, and the nature of every single process step is considered, identifying the manual and automated operations. Finally, relevant metrics for performance measurement are selected, and Key Performance Indicators (KPIs) are defined to guide the monitoring and evaluation efforts.

2. Technology screening and concept trade-off: digital maturity assessment and Smart Manufacturing roadmap definition.

   The second layer consists of evaluating the suitability of Smart Manufacturing concepts – including Digital Twin, Industrial IoT, and CPS – through trade-off analyses that balance technological maturity, applicability, and expected benefits [33, 34]. This stage borrows from cross-sectoral lessons, with know-how adapted from automotive and consumer industries, where digital traceability and automated quality control are already established [31]. A key step in this phase is to assess the present level of digitalization of the selected process and the envisioned one, once the identified technologies are applied. In this context, an Acatech model assessment [26, 32] (see Figure 2) is conducted to evaluate the technological and digital maturity of the system. This evaluation identifies areas for innovation and improvement in alignment with Industry 4.0 ideas (see Figure 3). The outcomes of this assessment help prioritize and select the SM technologies to be implemented, defining the roadmap for their implementation. The KPIs and measurable data points are emphasized to ensure clarity regarding what will be monitored.

3. CPS Architecture and Implementation

   Once the target process and enabling technologies have been selected, a cyber-physical system (CPS)-based architecture is developed. This architecture includes the sensing layer – composed of embedded and external sensors for structural monitoring – data acquisition pipelines, and advanced analytics for real-time interpretation [34, 36]. It is designed with a focus on modularity and interoperability, enabling seamless integration across different MAIT stations and ensuring scalability for future production volumes. At this stage, two key elements are defined:

   • Networking infrastructure, which collects data throughout the production process, supports data acquisition, categorization, and interpretation. Sensors and non-destructive techniques, integrated within an IoT framework, enable real-time monitoring and quality control, with results visualized through a centralized dashboard.

   • Software infrastructure processes the collected data to support operational decision-making. Interactive dashboards display critical production parameters, allowing for quick and efficient visualization of trends and interdependencies. These real-time insights facilitate the optimization of production parameters in line with defined objectives.

4. AI Integration for decisional support

   Once both visibility and transparency are established, an automated decision-support system can be implemented to achieve progressive levels of process optimization and automation:

   1. Process monitoring: A CPS continuously processes production data, automatically generating reports and issuing alerts to prompt manual intervention when necessary.

   2. Localized process control: AI algorithms predict outcomes and detect potential failures, enabling operators to make proactive adjustments before issues escalate.

   3. Global process optimization: AI leverages real-time data from interconnected stations to dynamically adjust production parameters, maximizing quality, efficiency, and overall throughput.

3. Smart manufacturing applications in the space industry

The effectiveness of the proposed structured methodology is demonstrated through the discussion of two case studies, representative of the current challenges that the space industry sector is facing:

1. Manufacturing of sandwich panels for satellite structures

   The automated production and cyber-physical monitoring of composite sandwich panels enable early detection of process deviations, optimization of key parameters (e.g., adhesive weight, insert alignment), and predictive quality control. This results in higher throughput, reduced scrap rates, and significant lead time savings, supporting scalable production for satellite constellations.

2. Digital transformation of the Linköping production line: From Smart Manufacturing to Smart Facility Management

   The SFMS integrates production assets, logistics flows, and resource planning through a digital backbone, providing real-time visibility of operations. By correlating facility-level KPIs with process data, it enables optimized scheduling, predictive maintenance, and improved coordination, increasing factory efficiency and responsiveness.

The considered case studies can be viewed separately or as two consecutive steps in the realization of a smart industrial environment: the first one at the process level and the second one aiming at a facility-level integration of what happens at the process level.

3.1 Manufacturing of a sandwich panel for satellite structures

This case study was carried out as part of the Smart Manufacturing for Future Constellations project, funded by the European Space Agency and coordinated by Sapienza University of Rome, with the participation of Beyond Gravity^© and Thales Alenia Space Italy^© as partners. The focus was on the manufacturing process for composite sandwich panels, which are critical structural elements in small satellites. The goal was to realize a CPS of the process owned by Beyond Gravity^©. In Figure 4, the considered process is represented, and it is clear how it is a complex set of manual and automated operations, with a critical bottleneck in the step involving the Automatic Potting Machine (APM). A proof-of-concept (PoC) of the CPS environment has been realized and tested, considering the APM machine, which, as already mentioned, is a critical point of the manufacturing process.

After a deeper analysis of the process in Figure 4, the selection of Smart Manufacturing tools to be applied has been made, starting from a literature review to highlight which digital tools address the needs considered critical for the selected process. The result of the performed trade-off is shown in Figure 5, where the selected technologies have been highlighted. Clear benefits are envisioned in the application of CPS, IoT, and Digital Twin technologies.

The measurable improvements brought by digitalization are outlined through assessments of the current state (AS IS) and the targeted future state (TO BE) in Figures 6 and 7, showing the envisioned benefits in the production process and performance measurements (KPIs). From the columns regarding the qualitative scale for measuring the performances, it is evident that the digitalization of the process brings a lot of advantages; see [37] for a deeper discussion.

The integration of sensors with an IoT network significantly improves real-time monitoring of critical parameters such as temperature, pressure, and humidity. With the final IT infrastructure, represented in Figure 8, online processing of sensor data inputs is performed through preprocessing, normalization, threshold checks, and monitoring. Processed data is then stored in a data lake, where users are able to have continuous open access, while data is interpreted by a statistical model-based closed-loop of KPI prediction and forecasting and is displayed through a user-friendly visual dashboard (SW platforms to realize such a concept are mentioned in [25, 27–31]).

Real-time collection and analysis of data facilitate performance predictions, while KPIs are measured to assess both industrial and digital improvements in production efficiency. Three levels of upgrades for the system are recognized at the end of the study:

• Process monitoring: this level enables the CPS to automatically process collected data, generate reports, and send alarms, allowing operators to intervene and adjust parameters based on identified issues and trends.

• Small-scale process control: AI algorithms predict process outcomes and signal potential failures, allowing the CPS to act; nevertheless, this cannot adjust process parameters to prevent issues.

• Large-scale process control: the AI-assisted CPS optimizes process parameters for optimal quality and efficiency, performing real-time predictive analysis to plan material flow, predict completion times, and manage maintenance, with multiple stations interconnected.

Once the key parameters of production are monitored, their trends can be predicted, and a logic of predictive maintenance envisioned. In Figure 9, a monitoring logic is shown where the yellow and red colors indicate that the probability of deviation from the optimal value of the process parameters is below or above a critical threshold.

The project ended at the second level of the envisioned monitoring options based on the CPS architecture. The reason is strictly related to the investments necessary to reach a high level of digitalization. Indeed, the higher the goal level of digitalization, the higher the necessity of a facility completely designed for full digitalization, and this will be clear through the next case study discussion.

3.1.1 Industrial assessment of the proposed PoC

The CPS integrates real-time data collection through sensor networks, AI-based analysis, and automated or semi-automated decision-making, dramatically enhancing process reactivity, efficiency, and reliability. By unlocking actionable insights hidden in production data, the CPS allows organizations to improve quality, reduce costs, and optimize throughput while preparing production systems to handle the complexity of future spacecraft constellations. The implementation of CPS transforms the production process by enabling continuous monitoring and interpretation of process parameters, early identification of potential deviations, and predictive adjustment of process settings. Through real-time analysis, process outcomes can be predicted with high probability, which allows for preventive actions rather than reactive corrections. For example, if the probability of an insert installation failure is high, the system can discard the insert early, saving time and resources (see Figure 9). CPS can also correlate variables such as insert hole geometry, adhesive quantity, or machine tool conditions with final panel quality, quantify their impact, and adjust process parameters accordingly. These envisioned capabilities reduce process variability, shorten cycle times, improve product quality, and increase overall productivity. Predictive maintenance is also enabled, as the system monitors the health of machine components, allowing potential failures to be anticipated and avoided, maximizing equipment uptime and production capacity. Compared to conventional manufacturing, CPS-assisted Smart Manufacturing introduces a higher level of interactivity, data aggregation, automated reporting, and predictive analytics, turning passive production data into actionable insights. The result is faster reaction times, shorter lead times for both products and entire projects, and improved ability to handle the increasing complexity of large satellite constellations while meeting cost and schedule constraints. To synthesize the impact of the CPS-based proof of concept on the studied production line of sandwich panels for satellites, Table 1 provides a qualitative comparison between the pre-digital baseline and the post-change configuration of the monitored process.

Baseline (pre-digitalization)	Manual data collection, no real-time monitoring, offline quality checks, recurring rework due to undetected deviations.
Post-change (CPS PoC deployment)	Real-time sensor network, automated alerts, improved traceability, early anomaly detection, reduced manual re-inspection tasks.
Example of possible monitored KPIs	Process stability, rework rate, lead time, and resource consumption.
Main outcome	Higher process visibility and lower quality variability enable future predictive control strategies.

Table 1.

Baseline versus CPS-enabled configuration in sandwich panel manufacturing. The table summarizes the qualitative differences between the traditional, manually supervised process and the digitally monitored configuration introduced through the CPS proof of concept. It highlights how real-time sensing, traceability, and early anomaly detection enhance production visibility and reduce rework in aerostructure manufacturing

3.2 Digital transformation of the Linköping production line: From smart manufacturing to Smart Facility Management

The transition toward smart factories is a key enabler of competitiveness in the space sector, where the demand for faster, more flexible, and more cost-efficient production is growing steadily. This second case study focuses on the Linköping production line of Beyond Gravity^©. The project, entitled “Digitalization of Linköping Production Line – Technology Study”, was conducted in collaboration with Sapienza University of Rome and aimed at defining a structured roadmap to transform the facility into a digitally integrated and adaptive production site. The Linköping case study provides a comprehensive roadmap for digital transformation in aerospace manufacturing, and it demonstrates that the combination of:

• Process-level digitalization (via IoT and CPS), and

• Facility-level interconnection (via SFM)

Creates a holistic Smart Factory where the data flows seamlessly, enhancing productivity and quality, and supporting sustainability goals and resilience in the face of supply chain and energy challenges.

3.2.1 Smart manufacturing tools and process analysis

The first phase of the study consisted of a detailed mapping of the production line, identifying five critical phases: bonding, curing, non-destructive inspection (NDI), machining, and final assembly. Through technical interviews with Beyond Gravity^© experts and equipment suppliers (e.g., mTorres for Automatic Fiber Placement machines), the following information was collected; see Figure 10:

• Process parameters (tension, compaction, temperature, curing profiles, machining tolerances, etc.)

• Critical points affecting quality, throughput, or traceability.

• Available machine data and capabilities for integration with Manufacturing Execution Systems (MES).

The analysis revealed a production line with good computerization but limited connectivity, with several manual operations not yet digitally tracked. A set of Smart Manufacturing tools was then selected to address these gaps, as shown in Figure 11:

3.2.2 Maturity assessment with the Acatech model

To evaluate the impact of the proposed Smart Manufacturing tools, the Acatech Industry 4.0 Maturity Model was applied. This model measures transformation across six levels: computerization → connectivity → visibility → transparency → predictive capacity → adaptability. The assessment showed that the current production line is between the first two levels (computerization and partial connectivity); see Figure 12.

By implementing the Smart Manufacturing tools already selected, the production line could reach its full potential; see Figure 13:

• Visibility: end-to-end data acquisition with real-time monitoring.

• Transparency: root-cause analysis based on historical and live data.

• Partial Predictive Capacity: enabling predictive maintenance for AFP heads and curing processes.

3.2.3 IoT and CPS architecture

Building on the results obtained in the first phases of the study, a multi-layer IoT architecture to serve as the backbone of the CPS for the production line was defined. Specifically:

Hardware Layer:

• Sensor networks for temperature, pressure, humidity, vibration, and optical inspection.

• RFID/barcode systems for traceability of components and tooling.

Edge layer:

• Pre-processing and filtering of raw data to enable low-latency alerts.

• Local storage for critical process data before transmission to the cloud.

Cloud and data lake layer:

• Centralized repository enabling historical analysis and machine learning-based predictive analytics.

Application layer:

• Interactive dashboards for KPI visualization, downtime tracking, and anomaly detection.

• APIs for integration with ERP and MES systems.

This architecture is scalable and designed to progressively incorporate advanced functionalities, such as AI-driven process optimization and adaptive control loops.

3.2.4 Smart Facility Management

One of the most innovative outcomes of the study is the extension of digitalization beyond the shop floor to embrace the entire facility infrastructure, leading to the definition of a Smart Facility Management (SFM) system. The SFM leverages IoT, cloud computing, and data analytics to monitor and optimize facility assets that indirectly support production, such as:

• Energy systems such as lighting and power distribution.

• Safety and security systems: access control, surveillance, and fire detection.

• People and space management: occupancy monitoring, flow control, smart parking.

• Waste and resource management: water, recycling, and predictive maintenance of utilities.

The integration of these assets into a single digital platform enables; see Figure 14:

• Real-time awareness of facility status and environmental conditions.

• Predictive and prescriptive maintenance, minimizing disruptions.

• Energy optimization, reducing operational costs, and carbon footprint.

• Dynamic adaptation of spaces based on occupancy and production requirements.

The proposed SFM IT architecture is fully aligned with the CPS approach for production, creating a unified digital twin of the factory. It consists of:

1. Sensing layer: distributed IoT sensors measuring temperature, CO₂ levels, power consumption, and occupancy.

2. Middleware: standardized protocols and a central data warehouse to aggregate and normalize information.

3. Analytics and AI Layer: algorithms for anomaly detection, energy demand prediction, and lifecycle cost forecasting.

4. Visualization layer: dashboards accessible to facility managers and production planners, ensuring a coordinated response to deviations.

By correlating production data with facility data, SFM enables:

• Optimization of energy usage per production batch.

• Early detection of facility issues that might compromise quality (e.g., humidity spikes affecting curing).

• Enhanced sustainability reporting for ESG (Environmental, Social, Governance) compliance.

• Improved worker safety and comfort, contributing to higher productivity.

In this way, SFM becomes a critical enabler of Industry 4.0, turning the plant into a living, adaptive ecosystem where machines, people, and infrastructure co-evolve toward optimal performance. To streamline the implementation of the SFMS, the process is divided into simpler tasks, each aligned with the specific information needs related to smart facility management processes; see Figure 15.

This transition from smart processes to SFM signifies a paradigm shift where not only individual production processes are optimized, but the entire facility is interconnected and managed as an integrated whole. Just as smart processes focus on enhancing efficiency and interoperability within individual production phases, SFMS extends these principles to encompass the entire facility, ensuring that all assets and systems work harmoniously. This holistic approach facilitates seamless vertical integration and comprehensive visibility, ultimately driving superior performance and operational excellence across all facets of the production environment. Based on this, as represented in Figures 16–18, the final IT architecture is organized into three layers: conceptual architecture, detailed architecture, and data warehouse logic. The transition from smart processes to SFM represents a fundamental shift in an approach where not just individual production processes are optimized, but the entire facility is interconnected and managed as an integrated entity. While smart processes aim to boost efficiency and interoperability within specific production stages, SFMS broadens these concepts to the entire facility, ensuring that all assets and systems function in a coordinated way.

To summarize the effects of the SFMS implemented at the Linköping site, Table 2 contrasts the initial state of the system with the digitally enhanced scenario enabled by CPS and IoT integration.

Baseline (initial state)	Partially connected production line, limited machine–facility interoperability, and no unified monitoring of resources.
Post-change (roadmap implementation)	IoT-enabled data layer, unified monitoring backbone, and dashboard-based supervision of production and facility assets.
Example of possible monitored KPIs	Machine uptime, energy usage per batch, environmental compliance, operations traceability.
Main outcome	Shift from isolated smart tools to an integrated, factory-wide decision-support architecture.

Table 2.

Baseline versus digitally enhanced configuration of the Linköping facility. The table outlines the qualitative shift from a partially connected production environment to an integrated, CPS-enabled smart facility. It illustrates how unified data collection, IoT-based monitoring, and dashboard-level supervision improve asset visibility, support predictive maintenance, and enable facility-wide optimization.

4. Integration of CPS and AI: Human–machine interaction and challenges

The implementation of CPS and AI transforms the operational ecosystem within the MAIT processes. In Table 3, the key industrial roles, their new tasks, required skills, and decision responsibilities are mapped. This framework highlights the progressive shift from manual control to data-driven and AI-supported decision-making, underlining how human expertise remains central while being augmented by intelligent automation. This human–machine synergy ensures both adaptability and trust in the digital transition of MAIT processes, which has been discussed in the previous sections.

Otherwise, the transition toward fully digitalized and CPS-integrated production lines introduces several cross-domain challenges that extend beyond the simple technical aspects. While Table 3 focuses on how responsibilities evolve within CPS-assisted workflows, Table 4 outlines the structural barriers that organizations must overcome to enable their effective deployment. As shown in Table 4, the main challenges of industrial digitalization in the space sector, along with the corresponding mitigation strategies that can support a successful and sustainable implementation, are presented.

Overall, the joint analysis of role evolution and implementation challenges confirms that the digital transformation of MAIT processes is not solely a technological shift, but a systemic transition that requires coordinated progress in skills, governance, and data infrastructure.

5. Conclusions

In the rapidly evolving space industry, driven by the rise of private companies and the growing focus on mega-constellations of small satellites, the space manufacturing sector must adapt quickly, adopting innovative strategies to improve satellite and launcher production efficiency. This study presents a structured approach for the digitalization and optimization of space industry MAIT processes. Through detailed case studies from the space industry sectors, the approach demonstrates how digitalization can significantly enhance quality, reduce costs, and improve efficiency, offering a practical roadmap for future space manufacturing advancements. Through this chapter, a structured framework for integrating CPS and Industry 4.0 technologies into MAIT processes of interest for the space industry has been presented. Through two representative industrial case studies – automated composite panel manufacturing and Smart Facility Management implementation – the proposed approach demonstrates how digital transformation can be systematically planned and executed in a space industry environment. The study confirms that CPS-based architectures enable higher levels of process visibility, traceability, and predictive capability, translating into tangible gains in productivity and quality. Moreover, the extension of digitalization from process-level monitoring to facility-level management establishes the foundation for truly adaptive and resilient production ecosystems. From a strategic perspective, this dual-layer approach (process  +  facility) represents a significant step toward Space Factory 4.0, where machines, humans, and infrastructure are fully interconnected within a unified digital backbone. This not only supports higher production rates without a loss in production quality but also facilitates sustainability objectives through optimized energy management and data-driven decision support.

Acknowledgments

The presented results have been developed within the framework of the projects Smart Manufacturing for Future Constellations, funded by the European Space Agency, and Digitalization of Linköping Production Line – Technology Study, funded by Beyond Gravity^©. The authors used AI-based tools (specifically, large language models) solely for language refinement and editorial polishing. All scientific content, analyses, and conclusions were entirely developed and validated by the authors, who take full responsibility for the manuscript.