Introducing the Manufacturing Digital Passport (MDP): A New Concept for Realising Digital Thread Data Sharing in Aerospace and Complex Manufacturing

Abstract

In the current data-driven era, effective data sharing is set to unlock billions in value for aerospace and complex manufacturing and their supply chains by enhancing product quality, boosting manufacturing and operational efficiency, and generating new value streams. However, current practices are hindered by fragmented data ecosystems, isolated silos, and reliance on paper-based documentation. Although the Digital Thread (DTh) initiative holds promise, its implementation remains impractical due to interoperability challenges, security and intellectual property risks, and the inherent difficulty of capturing and managing the overwhelming volume of data in such complex products as a holistic thread. This paper introduces the Manufacturing Digital Passport (MDP), a novel industry-driven concept that employs a product-centric, system-independent digital carrier to facilitate targeted, structured sharing of technical product data across the supply chain. The conceptual contribution of this work is the analytical formalisation of the MDP as a value-oriented carrier that shifts DTh thinking from costly, system-wide interoperability toward an incremental, ROI-driven record of lifecycle data. Rooted in real-world challenges and built on foundational principles of modularity, value creation, and model-based structures, the MDP, by design, enhances traceability, security, and trust through a bottom-up, incremental, use case-driven approach. The paper outlines its benefits through core design principles, definition, practical features, and integration strategies with legacy systems, laying the groundwork for a structured adoption roadmap in high-value manufacturing ecosystems.

1. Introduction

In today’s digital era, data sharing is a competitive advantage to create value through advanced analytics and smarter decisions [1]. Business-to-business data sharing is estimated to contribute an additional USD 285 billion by 2028 to the European Union [2]. Studies also indicate data sharing could unlock USD 100 billion in manufacturing, with 75% of participants exploring its potential to improve operations [3]. Despite this potential, limited data exchange in manufacturing continues to hinder progress, with significant barriers still existing for harnessing advanced analytics [4]. The application of advanced analytics relies on effective data integration across systems to ensure that relevant information is both accessible and actionable across the product lifecycle [5]. This potential for value creation is especially significant in aerospace and complex manufacturing, where even small gains in efficiency, quality, or predictive maintenance can deliver substantial economic and operational benefits, given the high cost of components and long product lifespans [6]. However, the inherent complexity of aerospace products has led to multi-tiered supply chains and fragmented manufacturing ecosystems, resulting in isolated data silos between tiers and even within departments of the same tier [7]. As technological advancements fuel demand for smarter, more autonomous subsystems, this complexity will only grow, expanding tiered and outsourced services and further worsening data and knowledge fragmentation [8].

Despite intensified digitalisation efforts, paper-based documentation remains embedded in the daily operations of aerospace and complex manufacturing [9]. These practices introduce persistent inefficiencies—manual entries delay access, physical records complicate version control, and fragmented documentation hampers traceability across departments and tiers [10,11,12]. This not only results in redundant inspections during part handovers but also prolongs compliance and audit procedures [13]. Beyond operational delays, the reliance on static records constrains the integration of lifecycle data, making it difficult to generate actionable insights or enable predictive capabilities [10]. As product complexity increases, such limitations become more pronounced, further widening the gap between available data and their practical use in decision-making.

To address the persistent challenges of fragmented data and reliance on static records, the Digital Thread (DTh) concept, which refers to a connected data flow that links product information across lifecycle stages (design, manufacturing, and maintenance), has been proposed to enable seamless lifecycle data continuity. Its goal is to connect data across design, manufacturing, and in-service stages, providing a unified view of product information. However, practical adoption in complex manufacturing remains limited. Most implementations are system-centric and rely on costly, high-maintenance interoperability across diverse infrastructures. Variability in communication protocols, data formats, and stakeholder priorities across supply chains makes real-time data flow difficult to achieve on a scale. Security and intellectual property concerns add further constraints, particularly in high-value and safety-critical sectors like aerospace. Moreover, the ambition to capture and use exhaustive operational data is often unrealistic. For instance, a single flight can generate terabytes of data that are rarely integrated or leveraged. As a result, DTh attempts have been largely ad hoc. Daase et al. [14] conclude that no successful DTh implementation has yet been achieved, as documented in the literature.

A promising initiative that could help overcome the interoperability challenges of DTh is the Digital Product Passport (DPP), which provides a product-centric data carrier that bypasses the need for complex system-to-system data streaming. DPPs have been successfully implemented in sustainability-focused industrial sectors, such as textiles, and are mandated in industries like battery manufacturing. The primary focus is specifically on sustainability data, enabling informed consumer decisions; promoting regulatory compliance; and providing product information, including user manuals, maintenance instructions, service details, and guidance for end-of-life processes such as disassembly and disposal. However, by design, the scope of the DPP remains limited to high-level sustainability information and excludes the technical manufacturing details necessary for manufacturing, operational, and service optimisation. This highlights a broader need in complex manufacturing for a structured yet flexible mechanism that enables the targeted sharing of technical data across lifecycle stages, without relying on full system integration.

This paper introduces the MDP as a new concept designed to unlock value through targeted and structured data sharing across the aerospace and complex product lifecycle. The conceptual contribution of this study is the formalisation of the MDP as a product-centric, value-oriented digital carrier that reframes digital-thread thinking from costly, system-wide interoperability to an incremental, Return on Investment (ROI)-driven record of lifecycle data. This concept has emerged from dealing with real-world challenges faced by two aerospace-tier manufacturers where the MDP has been industry-driven, ensuring its applicability and alignment with real-world operational needs. This is nurtured by establishing foundational principles to ensure successful industrial adoption. The key principles include a digital carrier for technical product data that facilitates sharing across design, manufacturing, in-service, and end-of-life stages. The MDP adopts a value-creation record approach, focusing on use cases with proven value, and employs a bottom-up, modular strategy to incrementally build stakeholder interest and justify investments. Central to this is the product-centric carrier, which focuses on product-related data and adopts a system-independent approach, enabled by a top-down data infrastructure capable of integrating data from multiple sources and distributed edges. Additionally, it leverages model-based frameworks and multi-layered record organisation to enhance traceability, security, and trust by design.

The remainder of this paper is structured as follows. Section 2 offers an in-depth review of State-of-the-Art technologies and initiatives in digital manufacturing, with a focus on Digital Twin (DTw), DTh, and DPP, exploring their capabilities and limitations. Section 3 addresses the challenges of current data-sharing approaches within complex, multi-tiered manufacturing ecosystems, concluding the need for a new solution. Section 4 introduces the MDP as a new, practical approach to unlock value through data sharing. It starts by outlining the foundational principles of the MDP, justifying its core design choices in Section 4.1. Section 4.2 defines the MDP, clarifying its scope and purpose. Section 4.3 details the key features of the MDP and explains how they address operational challenges. Section 4.4 demonstrates the integration of the MDP into stakeholder workflows, concluding with a comparative analysis against existing paradigms. Finally, Section 5 discusses the conclusions and future research directions for both academic and practical applications.

2. State-of-the-Art Technologies and Initiatives

Aerospace products consist of hundreds of thousands to millions of intricate parts and components, operating through the integration of multisystem and subsystems to deliver an efficient and reliable performance. With the advent of technologies, smart functions, and autonomy, system complexity is increasing [8]. The demand for higher precision, stringent quality standards, and reliable parts, all at a competitive cost, necessitates the delegation of production to multi-supplier. Tier-1 suppliers deliver complete products, while Tier-2, Tier-3, and lower tiers provide systems, subassemblies, and components [15]. Each tier leverages specialised expertise to produce components that meet efficiency and quality standards. Tiered production chains create silos of expertise and fragmented knowledge, hindering data sharing across the production lifecycle [16]. The long in-service lifespan of aerospace products, coupled with maintenance and service by multiple vendors, exacerbates this fragmentation.

Currently, data are often exchanged using paper-based forms, logs, and blueprints, as shown in Figure 1. Although some automated stages have digital representations that allow data to be recorded or stored in dedicated databases within production cells or machine controllers, the exchange often remains paper-based. Even when data are scanned or shared electronically via email or static links, it remains difficult to extract, analyse, or connect them to a specific product, requirement, or data source. At each stage, key design details, production schedules, and quality assurance records are shared in fixed physical formats. This fragmented, paper-centric approach introduces several challenges. Storage and retrieval of vast volumes of documents are costly and demand significant physical space and effort [12]. Version control is prone to errors and inconsistencies, making it difficult to maintain up-to-date records. Furthermore, traceability becomes a significant hurdle, as changes to documents cannot be reliably tracked. Scanned documents, though a step forward, remain static and non-editable. Any updates require manual input or the addition of sperate attachment files, which risks further data fragmentation and loss of traceability.

These paper-based records and fragmented data result in silos of information and incomplete lifecycle visibility, hindering data sharing across the product lifecycle. Figure 1 highlights these fragmented touchpoints, from design and manufacturing to in-service operation, maintenance, disassembly, and second-life services, where data flow is disrupted or entirely absent. Additionally, concerns around data privacy, security, and fears of sharing strategic information further complicate effective data exchange. This fragmentation prevents the extraction of comprehensive insights, limiting the ability to leverage advanced deep-learning capabilities for product improvement, operational optimisation, cost reduction, and innovation.

This section will discuss the State-of-the-Art approaches to addressing these challenges by analysing related initiatives and concepts, including the DTh, Digital Twin (DTw), Digital Shadow (DS), Cognitive Digital Twin (CDTw), and DPP. Given that the DTw is more mature and has already been implemented in industrial applications, it is logical to discuss it first in the following section.

2.1. Digital Twin

The DTw has emerged as an enabler for next-generation digitised manufacturing, transforming past static data analyses into future predictions using dynamic real-time data. At its core, the DTw represents a living model that operates in seamless synchronisation with its physical counterpart, enabling enhanced decision-making, optimised product and processes, and data-driven value creation. Although the DTw concept in manufacturing was shaped during the early 2000s, its adoption accelerated with Industry 4.0, which integrates cyber–physical systems, IoT ubiquitous communication, and cloud computing into industrial operations. Major technology providers such as General Electric, Siemens, Nvidia, and Autodesk have driven advancements in DTw solutions, shaping the field from diverse perspectives. These perspectives are discussed in [17]. ISO 23247-1: 2021 is a widely accepted perspective that defines the DTw in manufacturing as “a fit-for-purpose digital representation of an observable manufacturing element (OME) with synchronisation between the manufacturing element and its digital representation.” An OME could be any physical artifact, process, or behavior on the manufacturing floor (e.g., personnel, equipment, materials, processes, facilities, assets, and systems)” [18]. This definition highlights the DTw as a purpose-specific representation of a product, process, or asset, synchronised with its physical counterpart but not necessarily covering all aspects of the system. Data show that 85% of DTw applications are developed for manufacturing assets, and 11% are developed for factories [19].

In basic reference architecture, a DTw comprises an information model and data-processing capabilities in the digital domain, synchronised with its physical counterpart through a communication medium [19]. Its application has expanded to cover the entire product lifecycle. These models aimed to enable monitoring, optimisation, simulation, and prediction, streamlining processes from the initial design to the end-of-life stage [17]. Key industries, including aerospace and automotive, have embraced these capabilities to enhance product quality and operational performance. Recently, with the growing trend toward servitisation, DTw platforms have emerged to enable service-oriented processes, product tracking, and after-sales services. These platforms offer tailored services based on specific product usage patterns and provide detailed feedback to improve both product and service quality. Additionally, they allow for the integration of outsourced capabilities, such as supplier-driven crowdsourcing, thereby extending manufacturing capacity and offering competitive services to customers [20]. Despite progress, current DTw standards, while foundational, have limitations. Shao [21] reviewed existing DTw standards’ progress, applications, and scope, identifying their potential and limitations, while also emphasising the need for new standards to assist further development and adoption within industry.

Recently, a new concept called Digital Shadow (DS) emerged to enhance the connectivity between the DTw’s virtual and physical domains. DS is a manufacturing-oriented information system that has both macro- and micro-services to stream data, and it includes smart functionality to control and optimise the flow of data. It aims to sustain a high level of information quality during the online operations [22]. Riesener et al. [23] proposed a cross-database information model that helps identify suitable data sources and ensures information quality by matching data-source attributes to the required information. The approach includes a set of indicators to evaluate and select data sources that best meet the needs. Here, DS is viewed as an advanced communication infrastructure with capabilities to enable semantic data reasoning, metadata delivery, and data filtering, adding layers of intelligence to the flow of information. It can be perceived as an advanced manufacturing service bus. Other researchers view DS as an intermediate concept between digital models and DTws [24]. While digital models, such as mathematical models or offline discrete-event simulations, support only manual data flow between physical and digital realms, the DTw enables automatic two-way data flow. DS stands in between, supporting automatic data flow from the physical object to the digital model, but not in the reverse direction [20].

Recently, the Cognitive Digital Twin (CDTw) has emerged to extend the DTw beyond monitoring and prediction and toward perception, memory, reasoning, learning, and goal-directed action. In practice, a CDTw federates physics-based and data-driven models with knowledge graphs and online analytics, orchestrated as macro- and micro-services. These modules interpret context, explain recommendations, and optimise operations under uncertainty during runtime [25]. Contemporary designs embed cognition functions for anomaly diagnosis, causal reasoning, planning, and continual learning. They are supported by semantic interoperability (e.g., ontologies) and persistent histories that act as operational memory and provenance [26]. Rather than a communication and simulation layer, the CDTw operates as an intelligent agent embedded across asset and system lifecycles. It enables human-in-the-loop decision support and safe, closed-loop actuation back to the physical world [27]. Some authors view the CDTw as the natural evolution of the DTw, while others define it as a distinct pattern tightly coupling the DTw with knowledge representation and adaptive control [28]. In contrast to digital models (offline manual flow) and DS (one-way data-to-model), CDTw emphasises two-way explainable decision-making and adaptive optimisation.

The DTw is promised to revolutionise manufacturing, yet the research questions being considered remain superficial, and the implementations are still largely ad hoc [21]. The current practice focuses primarily on monitoring and prediction purposes while being used as decision-making support applications [19]. These limitations stem from challenges such as data noise, filtering, latency in communication, and the fragmented integration of edge and central computing. Moreover, existing DTw frameworks often rely on extensive sensor networks, which, while essential for dynamic data streams, raise concerns over cost, aesthetics, and privacy. Achieving a truly living DTw is further hindered by societal and regulatory barriers to data collection, especially for sensitive products or during in-service stage. In addition, the initial stages of product design lack a physical counterpart for synchronisation, limiting the DTw applicability during early lifecycle phases. This gap underscores the need for robust data records capable of capturing and sharing valuable design decisions, along with their contextual rationale. Another critical gap lies in the creation of vertical silos between DTw models, restricting data sharing not only between individual DTws of assets and system behaviours but also across tier systems of complex product value stream. The Digital Thread (DTh) has been envisioned to break silos and enable data streams, fostering interconnectedness by enabling seamless data exchange across such silo systems.

2.2. Digital Thread

The integration of information streams between design and manufacturing stages has long been a key objective in research and innovation for driving industry. With the advent of IoT and enhanced connectivity of the Fourth Industrial Revolution, where data can be collected and streamed, the focus has shifted toward leveraging data-driven analytics to enhance decision-making, thus improving quality and processes, and creating value [29]. In complex and high-value products, a strategic shift aims to leverage data from product lifecycle stages to drive continuous improvement across design, manufacturing, and operations [30,31]. Over the past two decades, active work has been dedicated to realising data streaming.

A related initiative is the Open Services for Lifecycle Collaboration (OSLC), which focuses on application-to-application integration. Originating around 2008 within IBM’s Jazz platform and now stewarded as a standard by OASIS, OSLC was motivated by the need to replace brittle, proprietary integrations between software tools with an open, extensible approach [32]. Its aim is to enable traceability and impact analysis across tools without forcing data migration into a central repository. OSLC follows an API-centric paradigm, applying Web and linked-data principles to interoperate lifecycle artefacts such as requirements, change requests, test results, and system models [33]. These artefacts are represented as Web resources described in a resource definition framework, with advertised service endpoints, “Resource Shapes”, that formally define their properties, and a lightweight query capability to retrieve matching resources. By linking rather than replicating artefacts, OSLC allows tools to reference, navigate, and reuse information. This API centricity makes OSLC effective for connecting heterogeneous tools and promising cross-tool traceability [34]. However, OSLC remains primarily an integration fabric for linking and discovery. It does not provide a portable, structured mechanism for sharing rich technical product data. While it has proven effective for linking specific lifecycle tools within an organisation, it has yet to demonstrate a scalable, cross-organisation toolchain that can support the full breadth of manufacturing and product-lifecycle data needs [35].

A notable step was taken by Lockheed Martin Aeronautics in using the 3D geometric computer-dimensional design model to directly drive computer numerically control milling or composite programming codes for carbon fibre placement [36]. This approach was further refined and applied to the development of the F-35 Joint Strike Fighter, where it was called a Digital Thread (DTh) [31]. The goal was subsequently expanded to enable informed decisions during in-service operations. By utilising data and insights from the jet design and manufacturing stages and applying advanced analytics, this approach supports continuous learning to enhance its in-service performance and maintenance operations [37]. The idea here is to enable context-aware decision-making, which can be exemplified by a mother interacting with her child. A mother, deeply attuned to her child’s habits, moods, and history, uses this wealth of contextual knowledge to make informed decisions, allowing her to respond and predict thoughtfully rather than simply reacting to immediate behaviour.

Recent statistics indicate an exponential rise in DTh research outcomes from both academic institutions and leading industrial innovation labs, primarily centred on additive manufacturing, DTw, product lifecycle data modelling and simulation, and model-based systems in the aerospace industry [16,38]. These diverse focal areas contribute to varying perspectives on the definition and purpose of DTh. Abdel-Aty and Negri [16] examined existing definitions, concluding that they serve two consensus purposes: linking data and information to provide an integrated view for stakeholders and facilitating data sharing through systematic collection and storage. Zhang et al. [38] observed that despite definitional differences, themes such as authoritative data, data linkage, and model integration are consistent. In this article, we adopt the definition by Bajaj et al. [39], which visualizes the DTh as “an integrated information flow that connects all the phases of the product lifecycle using accepted authoritative data sources”. This broader perspective avoids confining DTh to a model or a framework. Additionally, Daase et al. [14] highlighted that DTh is frequently linked to DTw, with the DTh described as the “the backbone”, “critical for” and “enabler for” DTw. It is evident that DTh and DTw are seen to symbiotically enable robust decision support throughout the product lifecycle.

In the following, we analyse recent works, focusing on featured DTh models and frameworks. For more exhaustive reviews, we refer readers to [14,16,38]. Daase et al. [14] systematically examined the scope of DTh in conjunction with DTw and I4.0, emphasising that DTw focuses on manufacturing stages, while Industry 4.0 covers broader operational aspects like material planning, supply chain, and warehousing. DTh, in contrast, spans the entire product lifecycle, from design to disposal. This study underscores the necessity of inter-systems connectivity and argues that DTh success hinges on DTw, I4.0 integration, and cloud-based technologies. Zhang et al. [30] advocate adopting Model-Based Systems Engineering (MBSE) principles—a methodology that uses formal models to support system design, analysis, and validation across stages—to advance DTh, emphasising the need for the use of surrogate models as an abstraction of native models. A surrogate model is a mathematical approximation aimed at enabling inter-model interactions and reducing the amount of data exchange [40]. Abdel-Aty and Negri [16] propose a conceptual framework for DTh development based on an analytical review. This abstracted framework systematically maps technological data components that can be utilised for data links, and transfer and storage mechanisms. It also highlights a list of standards and infrastructure deemed essential to support DTh advancement. To support this model-governance effort, digital identification (cryptographic digital signature) has been introduced to attach authoritative metadata to MBSE artifacts. It includes author, timestamp, version, and dependency data, supporting consistent reuse, governance, and lifecycle control [41].

Few frameworks have been proposed to advance DTh development, focusing on enhancing data integration, analysis, and lifecycle connectivity. The Lifecycle Information Framework and Technology (LIFT), introduced by Hedberg et al. [42], aimed to integrate data and manage lifecycle information to break down manufacturing silos. Complying with ISO 26324: 2025 [43], LIFT employs a master handler to integrate existing technologies, including an agent-based adapter, service-oriented solutions, and the sematic web, with the aim of enabling seamless data flow and decision support. Its self-learning mechanism links domain-specific data dynamically. Later, in [39], they applied the LIFT within the Syndia platform, testing its ability to stream relational data across different artifacts at lifecycle stages. This approach promises real-time updates, access, and visualisation, with each data point assigned a unique identifier tied to its source, such as an organisation or division.

Ledford et al. [36] propose a framework for mapping and analysing data from diverse sources to advance the digitalisation of systems and DTh. It emphasises capturing data at finer granularity to create individual data threads through a three-step methodology: identifying the functional level, defining data vessels, and establishing the data view. Ramesh et al. [44] present a more operational framework for a cloud-based DTh to support servitisation and mass personalisation. This framework uses multiple DTw agents (as-designed, as-planned, as-built, and as-maintained) to exchange information via a central cloud agent that plans, optimises, and coordinates processes. Serving as an MBSE translator, the authoritative DTh seeks to integrate and coordinate product data across various MBSE models. However, the capability of the mapping and cloud-based frameworks to facilitate data sharing across multi-tier systems with heterogeneous networks and address diverse stakeholder sovereignty constraints remains unclear. Focusing on sharing design rationales to manufacturing, Akay et al. [45] developed a push–pull model to/from Enterprise Resource Planning (ERP) systems by production departments when needed. Large Language Models (LLMs) are tailored for extracting contextual data. Such an in-house model is limited to sharing design data within the boundaries of a single ERP system.

Graph-based models, including network and knowledge graphs, have gained attention in DTh development for their ability to represent and link data while providing meaningful context. Liu et al. [46] developed a graph network to connect manufacturing units, workstation, and production system levels. The model captures physical and virtual entities as edges and nodes, aiming to monitor processes online, optimise them, and detect abnormalities in quality control by comparing as-built and as-manufactured quality specifications. Similarly, Zhang et al. [47] proposed a directed graphs model to enable communication between DTw with physical processes in aircraft assembly, integrating dynamic service agents to monitor, optimise, and adjust processes in real time. Expanding the scope, Hedberg et al. [48] introduced a graph model to represent and contextualise data across product lifecycle. Building on the LSH framework and adhering to RFC standards (RFC 3650–3652) [49] for identifier creation, the approach aims to establish a manageable method by identifying and linking related data points, leveraging cypher queries to exploit the resulting databases. However, a key limitation of graph-based models lies in their dependency on predefined node and edge structures, which are typically tailored for specific purposes, making a universal representation across design, production, and quality stages unfeasible. Kwon et al. [50] developed an ontology with a knowledge graph to merge as-designed and as-inspected data. The as-designed STEP (Standard for the Exchange of Product Model Data) and QIF (Quality Information Framework) are translated into a knowledge graph and linked with semantic web rules to form a queryable Web Ontology Language (OWL) knowledge base. While theoretically robust, OWL ontologies face challenges, including the complexity of unifying knowledge graphs from different models, the risk of propagating inconsistencies from STEP and QIF translations, and the difficulty of maintaining timely responses in complex product environments.

Despite growing interest in DTh frameworks, their implementation in industrial settings remains limited. While the concept is positioned to enable lifecycle-wide data continuity, most practical efforts focus on interoperability between systems. These remain challenging to operationalise due to heterogeneous infrastructures, varying stakeholder priorities, and concerns around data sovereignty. Existing models typically assume system-level integration, which is difficult to maintain across distributed networks and tiered suppliers. As a result, current approaches fall short of enabling secure and practical data exchange, particularly in settings where flexibility, autonomy, and know-how protection are key considerations. As a result, industry interest has shifted toward product-centric approaches that decouple data from direct system integration. One such initiative, the DPP, is discussed in the next section.

2.3. Digital Product Passport

The DPP is a structured collection of product-related data tied to a unique identifier and made accessible electronically via data carriers [51]. Its scope is firmly anchored in promoting sustainability, circularity, and value retention by supporting reuse, remanufacturing, and recycling efforts [52]. A key motivation is to influence consumer behaviour by making product information easily accessible, empowering consumers to make informed choices and enabling regulators to verify compliance. This transparency nurtures fair competition, encouraging the development and adoption of more sustainable products. Building on this vision, the EU has mandated the adoption of DPPs for batteries from 2027 and is expected to extend this requirement to textiles between 2026 and 2030. Although still evolving, early adoption has begun, with some companies piloting DPPs in textile products [53]. In the construction sector, similar initiatives like material passports support circular economy models by cataloguing the material content and sustainability profiles of buildings [54].

Progress in developing and shaping a DPP that fits the vision and appeals to the industry is accelerating. The EU is funding projects to engage stakeholders in defining DPP content and infrastructure. The CIRPASS [51] project proposed an information system architecture centred on semantic interoperability. The recommendations include formalised, standardised data models that integrate high-level ontologies for general product concepts and domain-specific ontologies focused on particular aspects, such as energy labelling. Jansen et al. [55] underscore the technical complexity of implementing DPPs, identifying limitations in current data-sharing infrastructures like data spaces. Similarly, Donetskaya [56] advocates for leveraging data from legacy systems, including Manufacturing Execution Systems (MESs), ERPs, and product data management and quality management systems, as foundational sources of data for structuring DPPs. Conversely, King et al. [57] argue that the DPP cannot be viewed as a single technical system but rather as a system of systems, necessitating a more comprehensive systems engineering approach to create an interconnected DPP ecosystem. Addressing this multifaceted challenge demands coordinated collaboration among regulatory bodies, industry stakeholders, and technical innovators to ensure that the DPP achieves its full potential.

Discussions around the DPP in research also draw parallels with existing initiatives and regulatory frameworks, highlighting both synergies and challenges. For instance, the Asset Administration Shell (AAS) is described by many authors as a digital passport for manufacturing equipment, providing a structured framework for managing resource data [58]. Building on this concept, Plociennik et al. [59] proposed a solution for implementing DPPs using principles derived from AAS. Their approach illustrates how AAS methodologies could inform DPP development, particularly through its capability to represent a wide range of product-related data in structured sub-models, which facilitate semantic and syntactic interoperability across stakeholders and enable both human and machine readability. Furthermore, Adisorn et al. [58] analyse a range of regulatory and conceptual frameworks, including AAS, material passports, chemical hazard regulations, and energy labelling for appliances, identifying overlaps with the DPP concept. This comparison offers valuable insights, such as leveraging established frameworks to standardise data collection, enhance interoperability, and integrate lifecycle information. It also highlights some common challenges, including navigating fragmented regulatory landscapes, addressing data privacy concerns (e.g., supply chain disclosure), and creating effective incentives to drive adoption.

The concept of a Digital Asset shares similarities with the DPP in its aim to establish a persistent, verifiable digital representation of a physical entity throughout its lifecycle. Originating from blockchain and industrial IoT research, the digital asset model emphasises unique identification, ownership, and provenance tracking, often implemented via decentralised registries or tokenisation mechanisms [60]. While this approach excels in ensuring authenticity and facilitating secure transactions between parties [61], its scope is typically limited to static or event-based metadata [62]. Without an extended data model, it struggles to capture the complex, multi-state technical information required in regulated manufacturing contexts [62]. As with the DPP, coupling the digital asset framework with richer semantic models and lifecycle data structures is essential for supporting advanced interoperability and value creation across multi-tier ecosystems.

The DPP demonstrates a product-centric approach that facilitates data exchange without requiring direct system integration, which remains one of the key barriers to implementing the DTh. However, the DPP is primarily designed to support sustainability objectives and therefore includes only general information, such as material composition; supply chain carbon footprint; and guidance on product usability, maintenance, and recycling. These types of data are unlikely to involve sensitive or competitive content, thus making implementation more feasible but also limiting the DPP applicability in technical and operational contexts. For example, while the DPP addresses early-stage concerns such as sustainable materials and durable product design, it does not capture critical details like design choices, engineering rationales, or quality specifications. The absence of such data restricts its value in enhancing traceability, supporting operational optimisation, or improving lifecycle safety in complex manufacturing. To overcome these limitations, this work introduces the MDP, a value-oriented, product-centric data carrier developed to enable secure and targeted sharing of technical information across distributed supply networks. Unlike the DPP, which is often perceived as a regulatory compliance burden [63], the MDP is structured around industrial use cases to deliver measurable ROI and practical value for stakeholders.

3. Why a Manufacturing Digital Passport?

The complexity of aerospace products continues to grow, driven by rapid technological advances, integration of autonomous capabilities, and expanded outsourcing. While multi-tier production chains enable the distribution of specialised work across suppliers, they have also entrenched fragmented data practices and weakened lifecycle visibility. Data sharing remains disjointed, with most technical and quality records exchanged through legacy formats, often paper-based or scanned documents. This creates operational fractions, particularly during part handovers. For example, receiving tiers must reverify product quality due to limited access to trustworthy upstream data. These inefficiencies extend to safety audits and compliance checks, where retrieving reliable documentation is both time- and resource-intensive. Despite the DTh being positioned as a remedy, its reliance on continuous, system-level interoperability has proven impractical. What is needed instead is a product-centric mechanism that aligns how Original Equipment Manufacturer (OEMs) and their suppliers work, and that facilitates secure, purposeful, and step-by-step sharing of technical data across every phase of the product lifecycle.

Multi-tier production systems inherently create silos of expertise and data. Design, manufacturing, and quality information often remain confined within departments or supplier tiers, limiting opportunities for collaboration and integrated decision-making. This is similar to the “over-the-wall” design model, where OEMs hand off designs without accounting for downstream manufacturability constraints. Greater collaboration could improve subsystem quality, reduce costs, and enhance overall product performance. Although modern simulation and analysis tools are capable of handling complex parameters and large datasets, they typically operate in isolation, disconnected from upstream design rationales or in-service feedback. This fragmentation persists across the entire lifecycle, from production to maintenance and recycling, hindering the ability to unlock value from past operations. To address these long-standing issues, a secure, structured mechanism for sharing technical data, aligned with how stakeholders work, is essential to enable targeted lifecycle optimisation and cross-tier value creation.

The absence of trusted quality data sharing, coupled with limited dissemination of expertise and operational insights, drives up quality-inspection costs, extends production lead times and safety audit durations, and prevents the value generation from lifecycle data and shared knowledge. To address these barriers, DTh data streaming has been proposed as a promising approach to enable seamless, secure, and efficient data flow across production departments/tiers and throughout the product’s lifecycle. However, as outlined in Section 2.2, current frameworks and models have seen limited industrial adoption [28]. These system-centric frameworks focus on building interoperability networks to enable data streaming between heterogeneous systems across the product lifecycle, as illustrated in Figure 2. Even so, most of them are aiming for seamless and on-time data streaming between tiers and beyond. Although this sounds theoretically achievable in this digital era, its practical implementation and industrial application have not been observed.

In this context, the thread shown in Figure 2 represents a continuous record of all data generated throughout the product lifecycle. However, capturing such vast amounts of data for complex products is highly challenging. Even if complete records were assumed achievable, leveraging them effectively to train models, extract patterns, or derive insights remains difficult. Daase et al. [14] concluded that there are no documented examples of a fully successful DTh implementation, as defined in the literature, within industry settings. Challenges include diverse infrastructure and communication networks across multi-tier systems, even within a single facility, making the creation of interoperable networks difficult [38]. Moreover, the cost burdens are significant, as OEMs cannot realistically expect lower tiers to adopt compatible networks without prohibitive expense, thereby limiting their flexibility to collaborate with competitors or other industries. Concerns over security, intellectual property, and business secrets further exacerbate these adoption hurdles.

The DPP is a product-centric approach that offers the potential to address the limitations mentioned earlier by enabling data sharing without requiring system interconnectivity or interoperability. This approach avoids the high costs of establishing interoperable networks, alleviates concerns over security and intellectual property, and preserves the independence and operational freedom of individual systems. However, as discussed in Section 2.3, the DPP is primarily focused on sustainability aspects. It carries data such as product footprints, regulatory compliance, supply chain information, and usability or service manuals. By design, DPPs do not encompass critical technical production or lifecycle operational data, such as design rationale, manufacturing parameters, quality specifications, maintenance, or operational history, which are essential for creating value across tier stakeholders. Furthermore, DPPs are often regarded as regulatory obligations rather than practical tools for value generation. This perception makes them less appealing in industries like aerospace, where practicality, operational efficiency, and ROI are the main drivers for adopting new technology. To address these gaps, this work seeks to overcome the limitations of DTh initiatives by developing a product-centric data-sharing inspired by DPPs but tailored to facilitate the sharing of technical and operational data. The main goal driving this new initiative is as follows:

A product-centric data-sharing carrier that bridges silos across the multi-tier product lifecycle, enabling technical data exchange, reducing quality-inspection costs and time, fostering lifecycle optimisations, and creating value while addressing security, know-how, and intellectual property concerns.

The MDP is proposed in this paper as a transformative solution to the data-sharing challenges. Unlike existing approaches, the MDP is a product-centric data carrier that focuses on improving product traceability and safety while significantly reducing the costs associated with quality confirmation and regulatory audits across multi-tier systems. By facilitating technical manufacturing, operational, and maintenance data sharing and leveraging advanced analytics and simulations, the MDP enables the multi-tiers and complex industries to gradually harness the potential of digital and deep-learning technologies. Unlike regulatory-focused tools such as DPPs, the MDP is positioned as a value-oriented carrier designed to capture data that generate foreseen ROI for stakeholders. By recording and sharing actionable data, the MDP provides direct and indirect benefits, encouraging practical adoption for large and small stakeholders in the multi-tier productions and long in-service lifespans.

4. Manufacturing Digital Passport

The development of an industrially adaptable MDP requires clear conceptual framing, given the breadth of enablers needed, including reference schemas, data infrastructure, mediator artefacts, access and security protocols, storage orchestration, data management, and Return on Data Investment (RODI)-based evaluators, among others. This paper addresses that need by defining the foundational principles, scope, and integration logic of the MDP. This framing is necessary to establish technical boundaries and align enabling block developments with verified industrial needs, avoiding the common trap of retrofitting technologies after design. It also clarifies the implications for existing practices and emerging data-sharing paradigms. Our method combines analytical definition and structured mapping of industrial challenges to organise existing knowledge and distil the MDP core requirements, both functional and non-functional, to guide further development and empirical implementation.

The concept for the MDP originated from an aerospace industrial problem within the UK Research and Innovation programme “Digitally Enhanced Low-cost Technology for Aerostructures” [64] project (Ref: 10039976) and funding from the Royal Academy of Engineering Research Chairs programme. A new method was required to improve the quality-inspection handover process between two tier manufacturers, with the objective of eliminating the need for receiver reinspection of parts. Currently, inspections are conducted at the sender facility, and the results are shared through a printed document. The product is then reinspected by the receiver to ensure compliance with specifications. The aim is to develop a method that reduces inspection costs, shortens lead time, and enhances trust through more visibility and traceability. Furthermore, an eye is also on gaining additional benefits by speeding up the plane airworthiness and safety audits. A thorough analysis of the existing tools and literature was conducted to find a solution. However, the analysis revealed a gap in addressing the industrial needs, as discussed in Section 2 and Section 3, and underscored the necessity for new tools centred on these needs while avoiding the concluded implementation obstacles.

4.1. Foundational Principle

To ensure that the MDP addresses real industrial needs and overcomes the implementation barriers discussed in Section 2 and Section 3, five foundational principles were defined. These principles are intended to support practical deployment and encourage adoption across industry. Figure 3 illustrates these pillars, which are discussed below.

I.

A digital carrier for recording and interacting with product data: A record that can electronically represent product information and data in a digital format. It shall be capable of capturing, preserving, and providing access, traceability, immutability, and easy interaction with data. Unlike paper-based records, which are prone to loss, damage, and inconsistency, a digital record should preserve the data in a structured way that systematically ensures accuracy. A digital record reduces long-term costs associated with storage and archiving, which is especially critical for the aerospace industry, where products are complex and have many components and a considerable lifespan. An MDP record should also be traceable, offering tracking, tracing, history verification, location, recorder, approval, and progress of the record, as well as the history of access. Traceability enhances trust in the record as an authoritative source of truth. This not only increases data reliability and usability but also auditability and accountability, which are paramount for rigorous safety standards and tight audits. A record should also be immutable, capable of recording multiple versions of the same data to reflect real-life stages, such as as-designed, as-manufactured, as-inspected versions of a part or product. Even when there is a change in the same data over time, these versions should be recorded without overriding the previous one, when justified. An MDP carrier should facilitate interaction with the data, allowing stakeholders to access, store, and retrieve data on demand. This interaction could be manual, performed by an operator on the shop floor or a line maintenance team at the ramp, or automatic, performed by a production cell DTw or a diagnostic health system. Such interaction eliminates the need for extensive effort and time spent fetching archives and documentation, which are mostly limited to specific locations.

II.

Technical data carrier: A technical data carrier focuses on information related to design, manufacturing, testing, validation, operation, maintenance, logistics, and end-of-life processes, including reuse and recycling. These data support improvements in products (e.g., parts, systems, or subsystems), processes (e.g., design decisions, manufacturing, assembly, or system integration), and operations (e.g., logistics, crew training, and inspections). The utilisation of technical data enhances quality, efficiency, cost-effectiveness, or combinations thereof. In product design, for instance, quality improvements could include adopting lightweight composites for weight reduction, developing heat-resistant engine components, or optimising flight control systems for reduced drag and increased lift. Safety advancements might address sensor reliability or improved tolerances. In processes, enhancements might involve higher-fidelity subsystem simulations, increased machining precision, consistent assembly torque, or better predictive maintenance models. Similarly, operational improvements might include advanced crew training or fuel optimisation strategies. Time optimisation often focuses on streamlining processes and operations, aiming to increase throughput and reduce backlogs. For example, this may involve eliminating non-value-added stages (i.e., repeated inspection steps), minimising downtime waiting for replacement parts, or accelerating approval processes. Meanwhile, cost reduction remains a critical driver. In products, this could mean designing for manufacturability or reducing material waste. In processes, it may involve minimising rework or scrap, while operational savings could come from optimised maintenance schedules or inventory cost reductions.

III.

Value-creation data: Value-creation record focuses on selectively capturing and sharing technical data that generate tangible benefits for stakeholders. Unlike DTh, which aims to record comprehensive product data, the MDP adopts a value-based approach. This avoids inefficiencies seen in indiscriminate data collection, where extracting insights becomes cumbersome, and the cost of cleaning and filtering outweighs potential benefits. Instead, the MDP emphasises data that hold measurable value, ensuring usability, relevance, quality, and cost-effectiveness. Data usefulness stems from their potential to inform analysis and decision-making, often requiring processes like cleaning or integration. Relevance ensures alignment with a specific goal(s), benefitting internal or external stakeholders across multiple tiers. High-quality data (accurate, complete, and timely) are imperative to avoid flawed analyses and costly errors, particularly in such high-value products. Additionally, the cost of data collection, storage, and reuse should be carefully weighed against potential benefits. A practical implementation strategy involves predefined use cases with demonstrable RODI. This modular approach allows stakeholders to accumulate multiple cases over the product lifecycle, where justified, ensuring clear benefits and stakeholder buy-in for each use case.

IV.

Product-centric carrier: A product-centric carrier prioritizes product-related data and adopts a system-independent approach to data access and sharing. It focuses on recording data directly related to the product at any stage of its lifecycle. While primarily concerned with product-specific attributes, a product-centric approach recognises the need to capture certain contextual data to provide a more complete understanding of the features and condition. These may include aspects of the system, machine, or resources. For example, in aircraft wing assembly, product data would include fastener type, torque applied, and wing-section serial number. Contextual data, such as temperature, tool calibration, or operator force feedback, may also be necessary to ensure proper fastening and prevent defects like misalignment or under-tightening. In cases where system or resource data are included, measures must be in place to protect proprietary context, security, and intellectual property. Equally important, the MDP aims to harmonise with existing domain vocabularies used in legacy systems, enabling flexible alignment without enforcing new semantic structures. This preserves current practices while allowing incremental integration into the product-centric record. The second aspect of product-centricity lies in how the MDP approaches data sharing and accessibility. Unlike DTh, which relies on connecting multiple systems to stream and access data threads, the MDP is designed to provide product-centric accessibility. This means that the data recorded in the MDP are accessible independently of the systems in which they were originally generated. This independence ensures that data can be accessed without requiring integration with external systems. Such a principle is essential for eliminating the main barriers to system connectivity and interoperability, which currently hinder effective data exchange.

V.

An MBSE digital record: An MBSE record organises data using structured models rather than text-based formats, enabling consistency, accessibility, traceability, scalability, and efficiency in recording and retrieving data. By adopting MBSE modelling formats and structures, the MDP adopts a shared framework and visual language (e.g., diagrams and simulations), improving understanding across all lifecycle stages and enhancing usability. Leveraging established MBSE standards, particularly within the aerospace industry, ensures data consistency and minimises discrepancies between records. This unified format creates a single source of truth accessible to all relevant stakeholders. Additionally, most MBSE tools and practices offer features that improve traceability, allowing for detailed audit trails that capture changes, approvals, and historical data modifications, ensuring that records remain transparent and verifiable. The efficiency of data storage and extraction is another key advantage. Unlike text-based formats, structured models streamline data organisation, enabling faster recording, retrieval, and analysis. In high-throughput production environments, this architecture also minimises latency by enabling parallel access to data and reduces storage costs by ensuring that only use case-driven, high-value data are recorded. Structured model-based formats further support efficient data management across distributed systems. This efficiency is further enhanced by the ability of DTw and simulation models to access and record data effectively. Well-structured models ensure that relevant product data are easily retrievable and shareable across different systems and organisations without requiring extensive integration efforts. This enhances collaboration between OEMs, suppliers, maintenance teams, and other stakeholders across the product lifecycle. Finally, the model-based approach is inherently scalable and capable of adapting to product evolution and new use cases, ensuring long-term relevance and maximising value generation.

4.2. Definition

The MDP is a new concept designed to be a practical tool for enabling data recording and realising the DTh initiative in industry. It can be defined as “a product-centric, model-based data carrier designed to digitally store and share value-creation data, including technical design, manufacturing, quality, operational, and maintenance information, across multi-tiered production chains and throughout the product lifecycle”. At its core, the MDP decouples data from proprietary systems, enabling more secure, system-independent data records that are both flexible and resilient. It employs an incremental, use case-driven approach that captures only the necessary information at each lifecycle stage, ensuring efficient and purposeful data sharing among stakeholders. Figure 4 presents the MDP-enabled thread data-sharing concept across stakeholders in the product lifecycle, representing the practical aspects of this modular and targeted data management framework.

One of the core concepts of the MDP, depicted in Figure 4, is its system-independent data record, where data are securely stored in the cloud without relying on the specific systems or resources that generate or use them. This eliminates the need for direct system interoperability, addressing major implementation barriers such as cybersecurity risks, know-how protection, network costs, and market flexibility. For instance, certification teams can access inspection data without needing to trace records across different tier manufacturers or operation and maintenance providers. Airlines and maintenance teams can retrieve relevant in-service and maintenance records without being tied to the OEM proprietary platforms. By decoupling data from networks, the MDP potentially reduces the risk of cybersecurity breaches and system failures, ensuring that data recording and sharing do not expose the system to vulnerabilities that could lead to disruptions, production halts, asset compromise, or delayed deliveries. It also ensures know-how protection by design, enabling stakeholders to control which data are recorded, shared, and accessed. This safeguards proprietary knowledge while preventing unauthorised access to sensitive information. The decoupling removes the financial burden of building and maintaining interoperable networks between the stakeholders, which require costly security measures and infrastructure investments to ensure a low-risk and high-security infrastructure. Beyond security and network cost, this approach also enhances stakeholder autonomy, preventing lock-in effects that arise when companies become dependent on proprietary systems or vendor-specific architectures. This independence is particularly important for tier manufacturers, who prioritise retaining flexibility to collaborate with multiple OEMs and partners without contractual restrictions that limit market access and innovation opportunities.

The MDP does not seek to create an exhaustive data thread that records every possible detail across the entire lifecycle of a product, from conceptual design to end-of-life stage and disassembly. Instead, it builds an incremental and use case-driven Digital Thread. It is designed to facilitate targeted data sharing between stakeholders, capturing only the necessary information required for specific applications. Instead of recording all possible data points, the MDP focuses on structuring modular digital records that accumulate justified data cases, where practical benefits can be demonstrated. The incremental concept of the MDP is demonstrated in Figure 4 by the arrows forming the thread at the top of the digital layer, illustrating how data are selectively accumulated across different lifecycle stages. Each arrow represents a specific case where relevant data are captured. For example, a use case depicted as a two-way arrow in the thread illustrates data exchange between design and contract manufacturing. The OEM shares design data with the contract manufacturer, who, in turn, provides quality specification records back to the OEM. This benefits both parties and can also extend to other stakeholders, such as the maintenance or disassembly teams. It can also speed up safety and regulatory audits. This selective approach optimises data storage and management, avoiding excessive data collection, which can be costly and difficult to maintain. Furthermore, it is a confined and measurable case, allowing stakeholders to clearly define and quantify their role in data maintenance, ensuring that data are recorded only when there is a clear justification for their use. This method enables incremental value creation at every stage of the product lifecycle.

Stakeholders in the MDP can include a broad range of participants involved in the product lifecycle, each of whom either generates or requires access to product-related data. These stakeholders may be external, such as multi-tier manufacturers, or internal, including in-house departments, production lines, and production cells. They can be broadly categorised into data owners/creators, who generate data, and data users, who utilise them. Data ownership typically resides with the entity generating the data, the IP holder of the design, or both, depending on contractual agreements. Data users are stakeholders who engage with recorded data in various ways. Primary data users interact with the data directly, integrating them into their services and operations, such as production planning or quality control. Intermediary data users, including AI and analytics providers, process the data to develop models, tools, and insights that support other stakeholders. Passive data users access data solely for inspection, compliance verification, or audit confirmation without actively modifying or processing them. By enabling a broad range of stakeholders and the modular structure, MDP extends the benefits to a wider manufacturing and service ecosystem while ensuring that data records remain purposeful and aligned with stakeholder needs. Given this diversity, developing flexible mediators and mapping tools to interface with legacy systems across the product lifecycle is essential for reducing integration effort and supporting scalable adoption. This is particularly important for stakeholders with limited in-house digital capabilities, where adoption may depend on the ability to align with existing workflows and demonstrate short-term value without major reorganisation.

4.3. Features

This section describes the key features of the MDP that support its practical implementation across different industrial contexts. Building on the foundational principles and definitions presented earlier, this section outlines the key features of the MDP, offering additional design guidance to facilitate successful implementation. These features are derived analytically from system-level requirements and industry needs, guiding practical development and helping practitioners ensure that the MDP effectively realises its overarching objectives. In particular, they consider the complexity of high-value products and the breadth of the manufacturing ecosystem, including multi-tier contract manufacturers, product-lifespan operators, and second-life stakeholders. Figure 5 illustrates these key features, with a focus on data storage structure, accessibility, and the value creation evaluation of potential lifecycle information. It also highlights the trustworthiness, reliability, security, and pragmatism of the MDP, reinforcing its role as a structured and practical data-sharing mechanism throughout the manufacturing ecosystem.

Data-driven solutions have emerged as the central arena where industries compete to unlock new value streams and drive growth. However, to capitalise on data, they must have a clear, assessable value that can be quantified and materialised into true benefits. A logical approach is to focus on high-impact data that drive measurable improvements, reduce risks, or have broad applicability. A systematic approach to defining data value should consider the data’s practicality, from collection and storage to model training and insight implementation. A key step in this process is assessing data utility, which begins by defining use cases and identifying relevant stakeholders. Utility can translate into direct benefits, such as increased throughput, or indirect benefits, such as quality and safety improvements. More importantly, in today’s data-driven economy, data have become an asset, especially for complex products through licensing or resale. Even small optimisations can yield significant financial and operational gains. A cost–benefit analysis helps evaluate the trade-off between expected benefits and the costs of data collection, storage, and processing. A useful metric is RODI, which establishes practical and measurable criteria to ensure targeted, cost-effective data recording. RODI assessment varies in complexity depending on the variety of data points, number of stakeholders, and calculation interval. For instance, a simpler case could involve collecting landing gear-performance data (e.g., vibration, pressure, and temperature) to reduce downtime, minimise flight delays, and lower maintenance costs. In contracts, tracking high-value components to optimise resale value for end-of-life parts is a more complex case. Another critical measure is time to value, which assesses how quickly data-driven insights translate into returns.

Building on value creation, another key feature of the MDP is the practical and efficient structuring of data. Studies conclude that a well-structured database helps overcome data storage and retrieval latency and reduces analysis latency [65]. Layered data structuring is a widely recognised practice for managing complexity and controlling access, allowing different abstraction levels to be tailored to various user needs and privileges. By segmenting information into distinct layers, the MDP can enforce granular access controls so that each stakeholder views only the relevant data in accordance with established authentication rules. At the same time, the MDP accommodates the handling of sensitive or proprietary information through data masking, anonymisation, or complete concealment where necessary, ensuring that any IP remains protected. Figure 5 exemplifies multiple layers of accessibility, ranging from open to shareable and secure. This safeguards data integrity and confidentiality. The multi-layer approach strengthens the immutable recording of transactions in highly secure environments. A key example is distributed ledger technology, such as blockchain, which ensures traceability and has been successfully implemented in applications like Bitcoin. Beyond traceability, a layered data design makes storage more efficient and ensures fast access for machine-learning processes. This structure approach has better data organisation, control, and auditability, which supports compliance with diverse data privacy regulations. This is a key consideration in multinational manufacturing/operational environments where stakeholders must adhere to varying legal and ethical standards.

Another keys aspect of developing a successful MDP is ensuring secure record management and interaction, pragmatic development and usage, and the reliability and trustworthiness of the record. While a multi-layered structure strengthens security through controlled granular access, immutability, and data masking, additional measures are needed to further secure record storage, access, and retrieval. A secure storage infrastructure, complemented by robust backup mechanisms, ensures data resilience, mitigating risks associated with loss or corruption. Encryption during storage, transmission, and retrieval is essential to prevent unauthorised access and data breaches. Decoupling the MDP record from systems enables leveraging existing cloud networks as an additional shield to secure interactions. These measures align conceptually with zero-trust security models, attribute-based encryption, and policy-based control enforcement, offering a baseline for dynamic, context-aware protection without prescribing a fixed security stack.

The pragmatic feature is centred on incremental development without disrupting existing systems. The MDP follows both bottom-up and top-down strategies. The bottom-up approach starts with clear use cases, builds on lessons learned, and gradually expands to more cases. Top-down ensures a unified architecture capable of integrating data from multiple sources and distributed edges. Another pragmatic aspect for shop-floor and on-site adoption is leveraging existing resources (i.e., sensors and infrastructure), thus minimising costs and reducing complexity. User accessibility is another factor: engaging end-users from early design stages and developing user-friendly interfaces that require minimal training and align with existing workflows help overcome practitioner reluctance and drive adoption.

While reliability and trust have been partially addressed through foundational principles and previous features, the MDP must be inherently designed to be both reliable and trustworthy. An active due diligence should be maintained throughout the entire record lifecycle to uphold data integrity and long-term credibility. This is especially crucial in the aerospace industry, where even minor discrepancies can have catastrophic consequences for human lives and severely damage product reputation. Maintaining reliability ensures that the record is accurate, consistent, and available when needed. This is achieved by focusing on data quality and accessibility. Highly available record architecture maintains access even during maintenance, while validation rules at the point of entry help uphold data integrity. Redundant storage with regular backups protects against hardware or software failures, ensuring uninterrupted availability. Building on reliability, trustworthiness is established by securing data authenticity, protecting data from manipulation, and ensuring they originate from a trusted source. Trust in the record is reinforced through provenance tracking, transparent documentation, and the certification of data generation and recording processes. Regular audits, for example, of both the data generation and recording processes, will be necessary.

4.4. MDP Interaction

This section discusses the deployment and integration of the MDP into stakeholder workflows while preserving its independence as a cloud-based record. This will be examined by illustrating how the MDP can be accessed and utilised on a manufacturing shop floor by various actors, including operators, production machine controllers, digital models, and enterprise systems. A basic product-record use case will serve to illustrate the purpose and nature of these interactions. This record is the first step of establishing an open-access record of parts and products at the point of manufacture. When the unit leaves the production line, it is assigned a unique manufacturing serial number and linked to a QR code that resolves directly to its cloud-hosted MDP entry. This serial number may also function as a subsidiary identifier under a parent serial number, supporting hierarchical traceability across components and assemblies. The MDP entry captures a concise yet authoritative set of attributes: product family and model code, baseline configuration and key specifications, build date and shift, and the legally accountable manufacturer of record. While modest in scope, this record delivers an immediate RODI, eliminating manual logbooks and enabling lot-level traceability. Since these data already reside in ERP, MES or AAS systems, exporting them to the MDP involves minimal integration effort, making MDP an ideal starting point before scaling to richer use cases. Figure 6 shows how the MDP, stored as a cloud-based record, spans both the physical and digital domains, highlighting interactions among production assets, controllers, and human users. The unique QR code grants authorised actors direct access to the product data.

When the product arrives on the shop floor, different actors can scan its QR code to retrieve precisely the data they need from the cloud-based MDP. For instance, machine controllers can automatically extract the “as-designed” data to set process parameters such as drilling speeds or torque values. Operators can review these parameters for manual verification, ensuring all special requirements and safety checks are in place. Simulation and DTw models may tap into historical product data and relevant sub-assembly records to refine process plans or generate updated work instructions tailored to the specific product. The ERP system may also leverage aggregated product data for high-level decisions such as production scheduling, logistics, and inventory management. During production, the MDP supports operations by enabling controllers, AI-based models, or DTws to refine and adapt process execution in real time. It serves as a dynamic work-in-progress repository allowing production data to be securely stored in a dedicated layer and accessed by other departments or production cells. For instance, if material properties or customer-specific requirements change, the system can instantly retrieve the updated MDP data to tailor work instructions accordingly. This flexibility is more practically useful when extended across multiple production cells or even geographically distributed factories. After manufacturing is complete, actors can record “as-manufactured” and “as-inspected” data back into the MDP. Automated analytics can flag inconsistencies or anomalies, while operators provide manual sign-off for high-stakes quality or safety validations.

These interactions are governed by role-based access controls to ensure that data visibility and editing privileges are tailored to each actor function. Machine controllers retrieve only the technical specifications they require in a compatible format. Digital tools access the data in a standardised and compatible format, accelerating analysis and decision-making processes. Operators can authorise or adjust configurations through a user-friendly interface. Shop-floor gateways and adapters must be designed to ensure that each actor has appropriate access, allowing efficient interactions to harmonise and enhance the operational workflow. The MDP can be embedded in the workflow by tailoring interfaces to match the operator task instruction. For example, a quality inspector verifying as-inspected data for recording can use a dashboard similar to the coordinate ceasuring cachine interface, including visualisation, a summary, and flagged inconsistencies, compared to as-designed data, where applicable.

This shop-floor walkthrough underscores the need for the MDP as a product-centric layer that enhances asset-focused specifications like the AAS and enterprise Product Lifecycle Management (PLM) backbones. The AAS excels at making a machine self-describing within a plant centred on the asset. The MDP instead follows the product, collecting data from every AAS-wrapped resource it meets and remaining linked as the item moves through subsequent plants, service centres, and end-of-life recycling loops. While the AAS gives the shop floor a common language for its machines, the MDP gives the entire value chain a common language for the product itself. Within an organisation, PLM keeps design revisions and bills of materials aligned, yet its integrations with ERP and MES are deep, heavyweight, and often bespoke. In addition, its ambition to catalogue every attribute still misses the local context needed to solve real problems (Cf., Section 3). The MDP deliberately floats above these backbones, lifting only the context-rich, product-specific slices that a given use case demands. In doing so, it makes purpose-fit data effortless to share across partners while sparing teams the heavy lift of custom integrations.

To conclude this section,

Table 1. Comparison insight between MDP, DTh, DTw, and DPP.

Feature	Digital Thread	Digital Twin	DPP	MDP
Definition	System-centric integrated data stream linking all phases of a product lifecycle.	Digital replica synchronised with an OME (physical asset, process or behaviour) counterpart.	Structured collection of product data focused on sustainability and circularity.	Product-centric data carrier for technical and operational lifecycle data.
Purpose	Enable data flow between systems to inform lifecycle decisions.	Predict, simulate, and optimise product or processes performance.	Promote sustainability and regulatory compliance	Create value by sharing technical data across multi-tier production systems and in-service networks.
Synchronisation	Dynamic, sequential near real-time data streaming	Dynamic, real-time data streaming.	Static data record	Semi-dynamic, sequential data record
Data Coverage	All product related data spanning design to disposal.	All OME model related data.	Product sustainability data, certifications and user guidance.	Technical value creation data spanning design to disposal.
Scope	Entire product lifecycle across systems.	Specific OME.	Product sustainability and regulatory compliance.	Entire product lifecycle across systems.
Implementation Approach	Top-down, holistic integration of product lifecycle systems.	Top-down and bottom-up (more successful).	Top-down, wide product sustainable data integration	Bottom-up, modular, and use-cases driven Top-down data architecture
Cost Effectiveness	High investment in network infrastructure and maintenance.	Moderate to high, depending on model size and complexity.	Moderate, requiring initial setup of a carrier repository.	Cost-effective, justified by the use case value proposition.
Security and Privacy	Concerns over network breaches and IP sharing.	Less security concerns, mostly vertical withing the same system.	No exposure to sensitive data, concerns about revealing product supply chain.	Secured and authenticated with multiple access layers.
Scalability	Low (complex governance).	Low (asset focus).	High (within circular economy).	High (across manufacturing ecosystems).

provides a comparative analysis of DTh, DTw, DPP, and MDP. The table highlights their distinct characteristics, implementation aspects, and practical implications to provide a clear understanding of how they function. It underscores the MDP advantage as a product-centric, value-driven data carrier that is conceptualised to enable data sharing in complex product ecosystems. MDP addresses the gaps in existing frameworks by enabling secure, structured, and scalable technical data sharing across multi-tier production systems. MDP offers a cost-justified, modular approach that can be aligned with real-world manufacturing workflows without imposing excessive implementation burdens. Its semi-dynamic data recording ensures traceability and enables lifecycle optimisation while maintaining security through controlled access layers.

5. Conclusions and Outlooks

This paper demonstrates that system-centric and holistic data sharing remains difficult to realise and use to materialise value in aerospace and other complex-manufacturing ecosystems. This is primarily due to high interoperability costs, intellectual property and security concerns, and the collection of extensive data volumes of more data than can be effectively exploited. The proposed MDP tackles these barriers by shifting the focus from networks to the product itself: a system-independent digital carrier that records only the data that demonstrably create value for stakeholders across the lifecycle. As a conceptual contribution, this paper defines the foundational scope, guiding principles, and system-level requirements of the MDP to support the structured development of its enabling technologies. By framing the MDP against real industrial needs, it establishes clear technical boundaries and design priorities to guide enabling blocks’ development and related technological advancement. Future work will address these developments through industrial case studies, prototype development, pilot integrations, and a quantitative evaluation of the benefits and potential of the MDP across representative use cases.

Building on the five foundational pillars—digital carrier, technical-data capable, value-creation focus, product-centric access, and MBSE structuring—tailored to address industry-specific needs, the MDP offers a pragmatic, bottom-up path to create new data-driven value. It decouples records from proprietary systems, safeguards know-how through layered access controls and supports incremental, use case-driven adoption, so stakeholders can start with a high-impact application and scale as further benefits emerge. The inclusion of features such as layered access control, secure cloud-based storage, and role-based interaction further strengthens the industrial viability of the MDP. Using the basic product record, the MDP illustrates practical interaction with multiple shop-floor actors, including operators, machine controllers, and enterprise systems. A comparative analysis with the holistic DTh, DTw, and the DPP demonstrates the MDP’s selective and product-centric advantages, allowing stakeholders to distinguish its potential. This positions the MDP concept as a low-risk, high-value digital enabler for stakeholders in aerospace and multi-tier manufacturing ecosystems.

The MDP neutralises many of the traditional barriers to data sharing, realising its full potential as an industry-wide backbone necessitates addressing several interdependent challenges. The integration overhead associated with mediators for legacy systems remains a significant barrier, particularly for small- and medium-sized enterprises, as even a streamlined digital passport requires interface adapters, schema alignment, and change-management processes before the first value case can be operationalised. To alleviate this burden, future work will focus on developing a reference toolkit and mediator roadmap that leverages lightweight schema extraction techniques, including the use of LLMs to accelerate mapping from MES/ERP structures to product-centric records.

Maintaining data integrity and availability over the product lifecycle, which is considerably long in aerospace, for example, is equally critical. Techniques such as immutable versioning, redundant storage, and hash-anchoring may offer mitigations. However, their total cost of ownership and the feasibility of technology migration and application for such technical data sharing are worth investigating [66,67,68]. In parallel, it is worth investigating concrete security baselines by integrating zero-trust architectures, policy-as-code frameworks, and attribute-based encryption models. Another consideration is reducing the adoption threshold for stakeholders with limited digital infrastructure or data engineering capacity. While the MDP avoids introducing new operational norms or requiring deep restructuring, uptake can still be constrained if the perceived return on adoption is not evident, especially for downstream suppliers or service providers focused on tightly scoped tasks. Lowering integration friction, aligning with existing workflows, and clearly demonstrating early-stage value are therefore key enablers to encourage broad participation.

Semantic alignment is another burden. Without shared reference ontologies, the MDP risks recreating silos within its own boundaries, underscoring the need to converge with standards such as ISO 23247-1: 2021 [18] and AS9100D [69], and to incorporate lessons learned from the AAS and the DPP development efforts. A robust, high-level governance framework is essential to develop an interoperable schema capable of managing stakeholder-driven vocabulary evolution, leveraging existing sub-models. Future research will explore the integration of legacy system vocabularies into the MDP architecture. Balancing data sovereignty with oversight obligations remains a governance challenge. Fine-grained, role-based encryption models are essential to protect IP, while auditable access for regulators must be ensured [70]. Techniques such as policy-as-code and zero-knowledge proofs present promising avenues for exploration [71,72]. Selecting only data that yield a positive RODI for intensive data use cases remains a challenging optimisation problem. Adaptive heuristics and machine-learning classifiers hold promise for real-time RODI prediction and dynamic throttling of data capture [73]. Finally, while the MDP must scale technically, it must also scale socially. Acceptance among MDP interactors hinges on designing MDP widgets to align with existing legacy systems (e.g., MES, and AAS). Translating this strategy across diverse stakeholders and interactors remains challenging, especially when a human-centric approach is considered.

Addressing these challenges sets a rich agenda for future work. Prototyping the MDP in live production and in-service data-sharing scenarios will be key to quantifying RODI. Cost-benchmarking studies can help de-risk initial pilots and expose implementation hurdles through empirical evidence. Such efforts will also lay the foundation for developing RODI assessor. In particular, high-throughput environments will require a deeper investigation into scalable data management, storage orchestration, and latency-sensitive access. Future work will evaluate storage cost-effectiveness, data-access responsiveness, and system performance under varying data loads to ensure MDP usability at an industrial scale. This tool will evaluate and predict data-value use-cases across the product lifecycle and beyond, critical for industrial adoption. Establishing a data architecture capable of storing distributed, multistate, and layered data under a multi-tier governance framework is critical. Such a framework must allocate liability, IP rights, and audit responsibilities across OEMs, Tier-n suppliers, and regulators. Pilot implementations, supported by proofs and third-party assurance audits, can demonstrate how commercial secrecy and regulatory transparency can coexist. Research into persistent, tamper-evident storage, such as distributed ledger technologies, and lifecycle cost modelling is essential to ensure that the MDP remains trustworthy and economically viable at scale. Comparative studies on hybrid cloud-edge retention strategies will be vital to guarantee long-term data persistence beyond the lifecycle of commercial platforms.