Sustainable Frugal Innovation in Cultural Heritage for the Production of Decorative Items by Adopting Digital Twin

Abstract

Throughout history, cultural heritage has accumulated, and is often embodied in monuments, structures, and notable figures. Cultural heritage preservation and management also include digitalization, allowing tangible monuments to be managed as digital inventory with “digital twins”. This provides innovative ways to experience and interact with the real world, in particular by using modern mobile devices. The digitalization of monuments opens new ways to produce decorative items based on the shape of the monuments. Usually, decorative items are produced by craft businesses, family-run for generations, with specialized skills in metal and stone processing. We developed and tested a methodological proposal for frugal innovation: how to produce decorative items with minimal costs based on digital twins, which are particularly in demand in tourism-driven countries like Croatia. A micro-business with three employees, specializing in “metal art,” aims to innovate and expand by producing small-scale replicas of cultural heritage objects, such as busts, statues, monuments, or profiles. A method has been developed to create replicas in the desired material and at a desired scale, faithfully reproducing the original—whether based on a physical object, 3D model, or photograph. The results demonstrate that this sustainable frugal innovation can be successfully implemented using affordable tools and licenses.

1. Introduction

Figureheads—monuments, buildings, well-known individuals, and other creations of art that are ingrained in the memories of generations—are the primary means by which every nation and culture has amassed its cultural heritage (CH) over the ages. According to UNESCO, “Cultural heritage includes artefacts, monuments, a group of buildings and sites, museums that have a diversity of values including symbolic, historic, artistic, aesthetic, ethnological or anthropological, scientific and social significance. It includes tangible heritage (movable, immobile, and underwater), and intangible cultural heritage (ICH) embedded in cultural and natural heritage artifacts, sites, or monuments. The definition excludes ICH related to other cultural domains such as festivals, celebrations, etc. It covers industrial heritage and cave paintings” [1]. Because of their physical character, the material assets that make up a country’s cultural heritage are susceptible to erosion, damage, major changes, and even loss, despite their essential and irreplaceable intrinsic value [1]. Physical monuments that decorate postcards and T-shirts, such as the Eiffel Tower in Paris or the Statue of Liberty in New York, are well-known illustrations that stamp people’s memories all throughout the world, shaping the image of a place or a country. The revitalization and development of cultural heritage and intangible cultural heritage have become a major part of national development policies worldwide, e.g., by adopting a public–private partnership approach to the conservation and reuse of historic buildings [2].

While frugal innovation is regarded as an efficient method of providing low-income clients with sustainable products and services, sustainable business models are a developing topic in the business discourse [3]. Both developed and emerging countries are seeing an increase in the number of frugal innovations. The increasing appearance of frugal innovation in developed markets challenges earlier definitions that often characterized frugal innovation, particularly in the context of emerging markets [4]. Frugal innovation can be described as a resource-constrained solution that fulfills three criteria: substantial cost reduction, a focus on core functionalities, and an optimized performance level. Furthermore, the main dimensions of frugal innovation comprise affordability, adaptability, resource scarcity, accessibility, and sustainability; however, what qualifies as a frugal innovation is still largely up for debate [5]. Studies explore how frugal innovations emerge at the grassroots level and employ novel business models to contribute to sustainable development. Various business model elements are mainly distinguished by value proposition, value creation, and value capture [6]. Frugal innovations ensure that products are appropriate for the local context—meaning they work reliably with the available infrastructure, materials, and user skills. Frugal innovation is often associated with (ecological and social) sustainability because it is characterized by minimizing the use of resources (raw material, production resources, energy, fuel, water, waste, financial resources), it is more affordable, and better accessible than conventional innovations [7].

Digitalization has transformed numerous sectors of the physical world, business, and society by providing alternative representations of reality [8] and providing a wide field for frugal innovations [9]. This is increasingly being used for objects of Cultural Heritage, and the term “Cultural Heritage Digital Twin” was coined [2]. With their continuously improving functionality and decreasing costs, digital devices have expanded beyond the personal domain into a broader sociocultural context [10]. As a result, they offer a method for transdisciplinary approaches (e.g., digital guides) [11]. Consumer demand, consumption norms, and ethical, cultural, and power issues have all changed due to the new cultural practices that have developed between customers and these devices and between devices and markets [12]. In this way, mobile devices work as an entry point (front-end) of the Cultural Heritage’s digital twin. The definition and taxonomy of DT were used from Ref. [8].

Digitalization has also supported conservation and the maintenance of cultural heritage, making physical monuments appear more like inventory items [13,14]. Sketches, casts, photographs, and postcards are just a few of the ways that the desire to reproduce works of cultural heritage has been met over the years. There are numerous methods and applications for these “digital twins” once they have been digitally captured. Similar to industrial practice, such a twin can be used as a constituent in the virtual world, for example, to create a digital experience and digital enlightenment by using augmented reality, digital museums, and virtual walkthroughs (Figure 1). The need or desire to create twin duplicates is now satisfied digitally since computers have taken over as the medium of our day and are creating a vast array of possibilities and scenarios beyond the straightforward copying of artwork [2]. These twins offer a technological foundation for the maintenance of the tangible asset (physical twin) in addition to enlightenment and experience.

However, since the physical twin does not consist of an active, working unit, there is no need for an optimization loop, and the data transfer and synchronization from digital twin to physical twin do not take place [8].

Digital twins of heritage assets can be reproduced in a variety of methods, such as making a physical copy of the authentic object for exhibition, instruction, training, or decorating. The creation of associated decorative items (souvenirs, statues, sacral gifts, bells, chess, relics, crests, and plaquettes) is one potential commercial use. These products are very much in demand and fabricated in regions with significant tourism, like Croatia [15]. Due to unforeseen circumstances, fashion trends, and customer behavior, the market demand for these products is prone to significant variations [16].

The providers of these products are primarily micro-businesses, particularly family-run craft businesses that have been passed down through generations. These businesses specialize in metal processing, especially metal casting and surface treatment, as well as stone processing, producing and selling their goods within their local regions.

One such three-person business, which creates decorative items from metal and stone (including cast brass, aluminium, granite, and marble), aims to expand its product portfolio. By leveraging digital tools, they plan to produce small-scale replicas of cultural heritage artifacts with complex shapes. Until now, the size and complexity, as well as the accessibility of the original, the cost of scanning, and the subsequent process chain to create the mold have presented significant hurdles.

In total, when it comes to producing decorative items that belong to the cultural heritage at a marketable cost, quantity, and quality, and delivering them on the desired deadline, three significant gaps exist in the current product emergence process and its commercial application.

Firstly, the process primarily encompasses manual work (artists and craftsmen) with a low level of digitalization. The product models exist only in the imagination of the artist and craftsman, and are hardly transferable and reusable. The use of software tools is almost not recognized here.

This raises the first research question: How can the product emergence process for decorative items that belong to cultural heritage be improved by using low-cost software tools for capturing and editing the shape of physical objects that are appropriate for a small business? Are professional tools necessary?

Secondly, there is a lack of integration between singular phases in the product lifecycle (“downstream process” within the product lifecycle management [17]), especially related to production, that impairs the sustainability of the entire process. This implies the second research question: How can the process chain be streamlined to improve sustainability (reduce material and energy consumption, scrap, and waste)?

Thirdly, reverse engineering techniques have been known for years to produce replicas of physical objects [18]. The appropriate products for the entire process chain have already been on the market. However, due to their high price, they are not affordable for a craft business.

This raises the third research question: Can the selected approach fulfill the criteria for a sustainable frugal innovation [5]? Can it be expected that the product emergence process for decorative items is improved to achieve a high level of maturity, which can become the foundation for a new business model?

From this perspective, this paper presents a methodological proposal to promote cultural heritage through the production of related decorative items. This proposal includes validation in a use case from practice following a transdisciplinary approach to bring digital twins into widespread use [19]. It considers cultural and educational needs, in addition to various technological requirements and economic considerations. The outline of the paper is as follows. In Section 2, the challenges of digital twins for cultural heritage and the transdisciplinary approach are outlined, followed by the literature review in Section 3. In Section 4, the process chain requirements for frugal innovation are formulated, followed by the description of the conceptual solution in Section 5. In Section 6, the way of a pilot implementation in a use case is explored, followed by a discussion of achievements, barriers, and commercialization in Section 7. Conclusions and outlook are presented in Section 8.

2. Challenges

In the midst of the emerging Fifth Industrial Revolution and an increasingly competitive global economy, organizations must prioritize sustainability, resilience, and agility. Traditional process configurations are now being augmented with cutting-edge technologies and digital tools across the production process to enhance efficiency, flexibility, and productivity, leading to the development of smart manufacturing systems. However, this transformation also brings new challenges, such as the integration of advanced monitoring technologies, cyber–physical production systems, and collaborative production processes [20]. Craft businesses cannot ignore this general development without jeopardizing their existence.

To address the aforementioned three research gaps, our research is based on four key principles that enable the definition of the process chain throughout the product lifecycle. These principles form the foundation for the structure improvement of the product emergence process:

• Principle of continuous process chain (P1)

The product emergence process should be supported by IT tools and models, without gaps, to preserve flexibility and enable instant responses to fluctuations in customer requirements.

2.

Principle of collaboration across stakeholders and disciplines (P2)

The convergence of various stakeholders and disciplines supposes a collaboration that includes a continuous trade-off among primary requirements such as visual perception, technical quality, cost, and timeline.

3.

Principle of sustainable frugal innovation (P3)

The new process chain should provide a resource-constrained solution that fulfills seven criteria of a sustainable frugal innovation: affordability, essential functionality, resource efficiency, environmental sustainability, social inclusivity, scalability and adaptability, and durability and reliability.

4.

Principle of sustainable process definition (P4)

In particular, the new process should contribute to the significant reduction of material consumption, energy, and waste.

Developing a low-cost, adaptable, and user-friendly process chain that considers the particular limitations and needs of a micro-business is crucial to closing the three gaps. This process involves transforming the original (physical object, 3D model, or photo) into a product (replica) that is fabricated of the required material with a specific surface refinement in the desired size range (scale approx. 1 to 5).

A market impact analysis that covers the launching of marketing and alternative commercialization channels (such as licensing to other producers) should be drawn for the intermediate products [21], especially if the generation of the DT offers or requires interaction with the customer.

3. Literature Review

The term “digital twin” in the context of cultural heritage refers to more than just a precise replica of the original artwork. Although it has presented new scenarios on a number of levels, it is far more valuable to all parties involved. DTs have made it possible to repair damage, recreate lost pieces, visit museum halls, print 3D copies, monitor and manage the security of artwork, and gather a vast amount of valuable data that can be used for research and various outputs. However, digital twins have also evolved into works of art themselves [2]. Numerous popular computer-aided technologies are employed for this aim, handling artwork similar to a technical object (Figure 2): digital databases for images and data, 3D scanning and modeling of monuments and objects, internet as mode of storage and communication, virtual and augmented reality for understanding and interaction, edutainment for storytelling via digital content, and mobile apps for on-the-move access.

3.1. Digital Twin of Cultural Heritage

CH sites can be monitored and preserved by using methods like computer vision, deep learning, and artificial intelligence on digital image data. The structural health of CH sites is impaired by defects that spread over time, including weathering, mortar removal, joint deterioration, discoloration, erosion, surface cracks, vegetation, seepage, and vandalism, and can be assessed by appropriate image processing techniques [14].

The need to provide a digital framework for the integration of data from several sources, linked to a single historical object but produced by distinct tools and techniques that are frequently used by distinct research teams and at different epochs, is an emerging difficulty. The documentary data and virtual representation of a heritage asset are included in this digital framework, which is referred to as the digital twin of a heritage asset [13]. By definition and explanation of a Heritage Digital Twin ontology, it is possible to generate different kinds of outputs using the same content and reuse digital resources for the communication and the visualization of cultural heritage in an attractive way through the use of the latest visualization technologies and web applications [22].

The relationship between digital twins and heritage conservation is examined in depth in this review, which divides digital twin techniques into six tiers to address four tasks [23]: data collection [24], visualization [17], operating [25], and utilization [26]. It examines Chinese heritage digitization approaches methodically, considering differences in complexity and goals. It also illustrates typological and temporal traits and the makeup of stakeholders [27]. Likewise, the development of an immersive platform demonstrated an innovative virtual visit method assisted by a real remote guide with e-learning functionality. The solution creates stereoscopic scenes that increase the level of interest, immersion, and the ability to generate emotion and wonder [28].

The uniqueness of a cultural heritage is its most remarkable quality. Three-dimensional scanning technologies are used in contemporary Cultural Heritage documentation to guarantee that an item is protected from environmental degradation, vandalism, and mishaps. When dealing with fractured artifacts, digitization is a necessary step before enabling precise 3D reconstruction. Ref. [29] elaborates a framework that empowers accurate digital reconstruction of chopped or damaged artifacts using ornament stencils obtained from 3D scan data [30]. This approach tackles challenges that are associated with 3D reconstruction processes, such as self-intersections, non-manifold geometry, 3D model topology, and file format interoperability [31]. The resulting 3D model has been integrated within virtual reality (VR), augmented reality (AR), and mixed reality (MR) applications, as well as computer-aided manufacturing (CAM) based on additive manufacturing to facilitate the dissemination of the results [29].

Recent research presents a digital twin-based framework for assessing rockfall hazards at the immediate vicinity of the Rumkale Archaeological Site, a geologically sensitive and culturally significant location in southeastern Türkiye [32]. Using a detailed digital copy of an item of our cultural heritage supports the reconstruction of a damaged or destroyed historic site, too. This allows for the creation of an unaltered prototype material that can be studied by everyone, while also making it simple to reproduce exact copies that ensure museum-quality standards are met and impart the essential and distinctive human artistic character [33].

Based on 3D modeling and digital twins, a framework for protecting intangible cultural assets has been constructed using multimedia technology. This framework seeks to provide an exact reproduction of intangible cultural legacy data by gathering data in multimedia technology, creating a business model, establishing pertinent parameters, and applying matching nodes in digital twin technology [34]. Users can detect three dimensions of twin concepts: matter, space, and time. Because the historical monuments and sites are limited to one dimension (reality) (Figure 3), their digital counterparts are extended both in virtuality and reality (augmented and alternative) [2]. The transition from one dimension to the next presents a particular challenge, which is addressed in this article. In general, DT provides various possibilities to handle CH like any other contemporary material goods.

3.2. Capturing of Heritage Objects (Scan)

Three-dimensional (3D) scanning has become an essential tool across diverse domains, including cultural heritage preservation, healthcare, industrial inspection, and digital manufacturing. Meanwhile, 3D scanning systems have evolved from laboratory-grade equipment to compact, affordable solutions compatible with mobile devices. The primary advantage of these technologies lies in their ability to capture detailed geometric representations of physical objects without physical contact, enabling non-destructive testing, virtual prototyping, and rapid design iteration [35]. Since this is a key step in the process chain generating DT of CH, the characteristics of the scanning process and their evaluation are of particular importance.

In recent years, mobile 3D scanning applications have gained traction as a low-cost alternative to professional systems. Leveraging sensors such as LiDAR and stereo cameras embedded in smartphones and tablets, these applications—like Polycam and 123D Catch—allow users to create accurate 3D models in real time or through photogrammetric processing. Although their accuracy is lower than that of industrial scanners, deviations typically remain within the range of 1–10 mm, which is often acceptable for non-critical applications in architecture, design planning, and healthcare screening. Several studies have confirmed that under optimal conditions, mobile scanners can produce geometrically consistent models suitable for practical use, especially when used for body scans, furniture, or environmental mapping. Moreover, mobile platforms drastically reduce entry barriers to 3D scanning, allowing casual users and professionals alike to digitize real-world geometry with minimal setup and at virtually no cost beyond the mobile device itself [36].

In contrast, desktop and handheld scanners—whether based on structured light or laser triangulation—offer significantly greater precision, often achieving sub-millimetric or even micrometric accuracy. Desktop scanners, such as low-cost turntable-based systems, strike a balance between accuracy and affordability and are particularly useful for digitizing small objects in controlled environments. However, their limitations in scanning volume and sensitivity to lighting make them less flexible than handheld solutions [37].

Professional handheld scanners, particularly those using white structured light or laser triangulation, provide mobility combined with high accuracy and resolution, often within 0.03–0.10 mm. Studies have shown that systems like the Artec Eva and HandySCAN outperform mobile applications in consistency, noise levels, and capture completeness—especially in geometrically complex or reflective surfaces. However, this improved performance comes at a cost, with such systems typically priced between USD 15,000 and USD 60,000, and requiring a higher level of user expertise and computational resources [38]. At the highest tier of the spectrum are industrial-grade scanners that are characterized by their high accuracy, high cost, operational complexity, and lack of portability, making them impractical for general or on-site use [39].

Additionally, research into hybrid approaches has explored the integration of long-range and close-range scanners to combine broad coverage with detailed accuracy. However, these methods can introduce substantial discrepancies—sometimes exceeding 20 mm—between systems, particularly in resolution and registration, underscoring the need for careful alignment and planning in multimodal workflows [40].

Despite these advances, studies consistently highlight the growing relevance of mobile scanning solutions, particularly in contexts where cost-efficiency, portability, and ease of use are prioritized over absolute precision. For example, in clinical settings, mobile scanners have demonstrated acceptable deviation levels (≤1.5 mm), proving useful in orthopedic measurement and surgical planning. In architectural documentation and cultural heritage digitization, mobile scanning has enabled rapid and inexpensive data acquisition for projects that would otherwise require costly equipment and expert operators [41].

To support a comparative evaluation of contemporary 3D scanning technologies, the following decision matrix (

Table 1. Decision matrix for selection of scan methods.

Type	Accuracy	Ease of Use	Speed	Portability	Total
Mobile app	2	5	4	5	16
Desktop scanners	3	3	3	4	13
Handheld (standard)	3	4	4	3	14
Handheld (laser)	4	3	4	4	15
Industrial	5	2	3	2	12

) summarizes five representative scanner categories according to four key performance indicators: accuracy, ease of use, speed, and portability. These categories were selected based on criteria frequently cited in the literature as critical for effective application of 3D scanning systems in various domains such as healthcare, manufacturing, and cultural heritage documentation. The total score serves as a composite measure to provide a simplified, yet informative, overview of each category’s overall usability and technical balance.

In the context of cultural heritage preservation and the rise of digital twins as tools for replication and commercialization of tangible heritage assets, the results of the decision matrix reinforce the practical value of mobile and low-cost 3D scanning technologies. While industrial systems deliver superior accuracy, their cost, complexity, and lack of portability make them unsuitable for small-scale artisanal or micro-business contexts. Conversely, mobile applications and desktop scanners, despite their limited metrological precision, demonstrate high usability and sufficient resolution for the digitization of decorative or symbolic artefacts, such as busts and sculptures. This aligns with the needs of micro-enterprises in heritage-rich nations like Croatia, where stone and metal artisans increasingly seek affordable and scalable tools to transform physical monuments into digital models for replication and reproduction. By leveraging mobile scanning systems, such as LiDAR-enabled smartphones, and integrating them into a reverse engineering workflow, small workshops can initiate market entry strategies based on low-cost licensing, customized production, and digital inventory creation. Thus, the decision matrix not only provides a technical comparison, but also supports a broader socio-economic narrative, wherein democratized access to 3D scanning becomes a catalyst for heritage-based entrepreneurship in the digital age.

Digitalization is intended to strengthen the bond between art and the audience rather than substitute it [2]. Thus, it can be understood as a “postponed” modeling process that involves a transition to the “normal state”, which comprises both the real and virtual worlds. Not only can digitalization produce replicas, but they also possess “intelligence” of their own. The objective of producing a tangible copy is just one of the significant facets of digitization. Scanning an artwork allows us to obtain a variety of data that can be applied to a variety of tasks, such as conservation studies or the examination of creative passages involved in that piece [2,8].

3.3. Additive Manufacturing in the Process Chain of Decorative Items

Since a substantial portion of production costs is determined at the design stage, materials and processes must be selected and adapted to control design and development expenses. Small dimensional extensions, a sufficient batch size, and an appealing surface are characteristics of decorative items. Depending on the material, such products are predominantly manufactured by metal casting and injection molding, and possibly by joining. Additive manufacturing (AM) processes are conceivable, but too expensive for metal products. However, AM can be considered to produce molds based on DT. To achieve this, a manufacturer must define a process chain and consider key areas such as knowledge management and semantic interoperability [20].

For a fast generation of molds, FDM (Fused Deposition Modeling), or material extrusion technology, is used as one of the most widespread and accessible methods of additive manufacturing. It is based on the extrusion of thermoplastic filaments through a heated nozzle that deposits the material layer by layer, following a pre-defined digital model. Due to its simplicity and availability, FDM technology finds broad application in various fields, from engineering and medical modeling to education and cultural heritage preservation [42].

One of the key advantages of FDM technology is the variety of available materials. The most commonly used materials include PLA (polylactic acid), ABS (acrylonitrile butadiene styrene), PETG (polyethylene terephthalate modified with glycol), TPU (thermoplastic polyurethane), as well as specialized filaments such as nylon, ASA (acrylonitrile styrene acrylate), and flexible materials like Flexfill. Each of these materials has specific mechanical, thermal, and aesthetic properties that make them suitable for different purposes [43]. For example, ABS is resistant to heat and impact, PETG combines strength and flexibility, TPU and Flexfill enable the printing of flexible objects, while ASA and nylon are known for their resistance to environmental influences and UV radiation. In our case, PLA was chosen for its ease of printing, dimensional stability, biodegradability, and aesthetically uniform surface finish, making it particularly suitable for producing visual and educational replicas of cultural heritage objects. A comparison of the properties of commonly used FDM materials is shown in

Table 2. Comparison of the basic properties of commonly used FDM materials.

Material	Hardness/Strength	Flexibility	Heat Resistance	UV Resistance	Printability	Application
PLA	High	Low	Low (~60 °C)	Low	Very easy	Prototypes, education, models
ABS	High	Moderate	Medium (~100 °C)	Low	Moderately demanding	Technical components
PETG	Moderate	Moderate	Moderate (~80 °C)	Moderate	Easy	Functional parts
TPU/Flexfill	Low	Very high	Low (~60 °C)	Moderate	Demanding	Elastic parts
ASA	High	Moderate	Medium (~95 °C)	High	Moderately demanding	Outdoor applications
Nylon	Very high	High	High (~120 °C)	Moderate	Demanding	Mechanically durable parts

.

FDM technology offers high adaptability in printing, as parameters such as layer height and infill density can be optimized for different types of cultural artifacts, enabling the creation of precise replicas even for complex shapes and fine details. This flexibility makes FDM technology particularly suitable for the research and restoration of cultural heritage, as it allows for sufficient strength and aesthetic credibility for museum and educational applications [44].

4. Process Chain Requirements for Frugal Innovation

The primary objective of this research is to develop a process chain for the generation and exploitation of digital twins of monuments under restrictive constraints specific to micro-businesses’ capabilities (Table 3) while taking specific reverse engineering limitations into account [45]. The widely recognized collection of five essential diametrically opposed requirements was the consequence of input from different stakeholders (customers, owners, cultural heritage officers, and media specialists). Modularity affects only internal processes in order to simplify the process and lower the costs [46]. Thus, the aforementioned characteristics of a frugal innovation (P3) should be demonstrated [47]. It was immediately apparent that the answer would need a digital process up front [31], followed by traditional casting in the end. This can be explained in several ways:

• The business’s limited financial capacity precludes significant capital investment (e.g., manufacturing by 3D printing).
• Insufficient IT competence.
• Challenging time-to-market.
• Strict environmental constraints.

On the other hand, a small, hand-crafted business discovers benefits in terms of flexibility, innovativeness, robustness, and cost structure [48]. We selected investment casting as the primary manufacturing process [49] once more (the cube right above in Figure 3) and combined all other tasks encompassing the inherent process variety into a virtual sub-process, with significant assistance from an external partner [50], in order to consistently overcome the afore-mentioned challenges.

Table 3. Criteria for the composition of a process chain [51].

	Criterion	Value/Range	Impact/Principle
1	Feasibility and competitiveness	Low-cost solution for each step and integration	K.o./P1–P3
2	Duration of the production process	A few days to 2 weeks	High/P1–P2
3	Batch size in a wide range	1–100+	High (cost per unit)/P1–P3
4	Wide range of materials	Brass, aluminium, plaster, plastics	Mid/P2–P4
5	Small amount of rework	Without manual rework	Mid/P3–P4
6	Modularity of the process chain	A module for each work package	Mid/P2–P4

Table 3. Criteria for the composition of a process chain [51].

	Criterion	Value/Range	Impact/Principle
1	Feasibility and competitiveness	Low-cost solution for each step and integration	K.o./P1–P3
2	Duration of the production process	A few days to 2 weeks	High/P1–P2
3	Batch size in a wide range	1–100+	High (cost per unit)/P1–P3
4	Wide range of materials	Brass, aluminium, plaster, plastics	Mid/P2–P4
5	Small amount of rework	Without manual rework	Mid/P3–P4
6	Modularity of the process chain	A module for each work package	Mid/P2–P4

This table highlights that at least three criteria affect each principle for production process improvement. This provides a solid foundation for the initial setup and optimization of the entire process.

5. Conceptualization of a Solution

A solution will be conceptualized to fulfill four principles, starting with the principle of the continuous process chain (P1). The overall (technical) project framework consists of six work packages (modules), in addition to commercialization, which includes all non-technical efforts to complete the customer order (Figure 4). While the manufacturing sub-process (cast model preparation, final casting) as a whole stays largely unaltered, two packages run in the virtual world (scan and 3D model generation—right column), two in the real world (cast model preparation, final casting—left column), and two in the mixed world (prepare scanning and 3D print—mid column). Supposing an appropriate commercial agreement, good teaming, and adequate IT tools, this approach can foster collaboration across stakeholders and disciplines (P2). The remaining two principles (P3, P4) should be implemented by singular measures along the continuous process chain. Modularity should streamline the process chain as a cross-principle criterion.

Because process flexibility is of essential relevance (Criteria 2 and 3 in Table 3), the singular packages should be treated as modules and become interchangeable [52]. The following is a more thorough description of these three parts, as well as the commercialization.

5.1. Manufacturing Technique

Investment casting ought to be the favored technology for producing the final products to maintain the customary workflow of a hand-crafted business [49]. This technique is typically used when manufacturing parts with distinct shapes, small details, and low tolerances. In manufacturing industries where accuracy and complex designs are essential, investment casting is commonly employed. This method can ensure parts with superior surface qualities and offers a wide span of design freedom, which is essential for working with unknown forms. Although it can be more expensive and time-consuming than other casting techniques, it is appropriate for applications where the requirement for precision and fine details outweighs these drawbacks.

Innovation potential is leveraged in the method for preparing molds for investment casting. AM technologies promise significant improvements in terms of speed and accuracy. Hence, an optimal approach selection and operating parameters are crucial for a successful application, considering the specific requirements of a hand-crafted business. Wherever possible, an IT-supported process template should be generated and used [53] to increase reliability and lower the costs.

5.2. Shape Acquisition (Scan)

As explored in Section 3.2, numerous devices based on scanning or photogrammetry are available on the market for shape acquisition. A scanner is costly, sluggish, accurate, and challenging to use. A camera, on the other hand, is inexpensive, simple to operate, and enables speedy operations. However, its accuracy can be a concern [40]. In the meantime, there are a few low-cost or open-source applications available that can create 3D point clouds from a collection of images. This method makes it possible for practically anyone to take pictures of an object of interest anywhere and then forward them for further processing. Consequently, the initial pilot will use images from a consumer camera.

Furthermore, photogrammetry provides a unique opportunity to involve the customers in this process chain from the first step by using their photos, potentially transmitted via the internet [54]. This offers up a new customer group in addition to making communication with the customer easier. Otherwise, this provides an additional potential source of errors in the customer process and is a reason for potential mutual misunderstanding.

5.3. 3D Print of Molds

Using FDM 3D printers and temperature-resistant 3D printing materials, 3D molds for investment casting can be created. Molds can be produced quickly and affordably with 3D printing, but layer lines will result in an insufficient surface finish that needs more sanding. In some circumstances, aluminum tooling may be able to endure 10,000 part runs, but a 3D-printed mold is likely to withstand only 100 shots. Resins are the most often used material for 3D printing molds. They can produce accurate, superior parts with a variety of properties, including hardness, flexibility, heat resistance, and chemical resistance.

The digital models were obtained using photogrammetry [55]. For this purpose, photogrammetry software (e.g., Meshroom 2021.1.0) was used to create a detailed 3D model. The resulting mesh model was then exported in STL format and underwent basic editing in Autodesk^® Meshmixer™ software (3.5.0), which provides tools for repairing topology irregularities, smoothing surfaces, and checking model integrity [56].

For 3D printing, the chosen material was 3D4Makers Facilan C8 PLA filament with a diameter of 1.75 mm. This is a high-quality PLA filament developed specifically for enhanced strength, surface smoothness, and dimensional stability during printing. Unlike standard PLA, Facilan C8 offers improved mechanical resistance, reduced brittleness, and excellent surface finish, making it particularly suitable for models where aesthetic details and precision are crucial. One of its most notable advantages is its excellent layer quality, with smooth transitions and minimally visible print lines, even at higher layer heights. The material also features strong layer adhesion, resulting in better structural cohesion of the printed object. Although it has a lower glass transition temperature compared to some technical materials (approximately 60 °C), it exhibits significantly greater resistance to cracking and deformation than standard PLA. Facilan C8 is also biocompatible and produced from renewable sources, placing it in the category of environmentally friendly materials. For all of these reasons, this filament was important to achieve detailed surface features, a stable large-scale structure, and an aesthetically appealing final appearance without the need for extensive post-processing.

Facilan C8 PLA has already been used in previous studies related to 3D printing of cultural heritage, including the production of a replica of a 13th-century metal spur from the archaeological site of Biskupija [57], and a Roman sarcophagus from the Rižinice site [58]. In both cases, the material proved highly suitable for producing dimensionally stable and visually high-quality models with pronounced surface details, further confirming its appropriateness for demanding museum and presentation purposes.

Initially, it was planned to use the 3D print as a positive for subsequent casting of the bust using traditional foundry methods. Accordingly, the first print attempt was executed using the so-called “Vase mode” setting, a spiral printing method with a single wall and no internal infill. Although this configuration allows for faster printing with reduced material consumption, it proved unsuitable for this application. Therefore, it was decided to proceed with a print using defined infill density, with all parameters optimized to produce a stable and technically usable model.

Based on the defined settings, the estimated total print time was 32 h and 57 min. A standard 0.4 mm brass nozzle was used for printing, providing a good balance between precision and speed. The infill type used was Gyroid, which offers a good compromise between strength and flexibility while optimizing material consumption. The final dimensions of the 3D model were 174.67 mm (X) × 130.62 mm (Y) × 198.92 mm (Z), fully utilizing the print volume of a mid-range printer for the desired object size. However, the long print time sets a limit on time-to-market and costs.

5.4. Copyright and License

Monuments (statue, sculpture, architectural work) in public spaces are protected by copyright as artistic works. The creator holds copyright automatically upon creation, unless it is transferred to a copyright holder [59]. Copyright lasts for the life of the author plus 50 to 70 years (depending on jurisdiction). Regulations vary from country to country [60]. For works of art created before World War II, it can therefore be assumed that copyright has expired.

Commercial use requires a license from the creator. A public institution (e.g., municipality where the monument is situated) can become a partner because it tends to acquire the copyright to have the possibility for re-use in different ways and media. The rules for further protection of intellectual property apply as for any other technical product [61].

In a broader perspective that goes beyond the interests of a single company, key heritage policy frameworks, such as UNESCO’s 2003 Convention for the Safeguarding of the Intangible Cultural Heritage [62] and the International Council of Museums (ICOM) ethical guidelines [63], provide essentials for ensuring that sustainable frugal innovations in cultural heritage are implemented responsibly. With the commitment of public authorities, these frameworks emphasize preservation, authenticity, and community involvement, which are crucial when using digital twins to produce decorative items inspired by cultural heritage monuments.

By following UNESCO’s principles, digitalization efforts can respect the cultural significance of heritage objects while promoting accessibility and inclusivity. The use of digital twins enables craft businesses, particularly small family-run enterprises, to innovate in a cost-effective way while safeguarding original artifacts from physical damage.

5.5. Commercialization Affects Customer Groups

Digital twin can be commercialized as a method (service), as a virtual product, and as a physical product [64]. The main purpose of this research is to explore the process from an existing object to a physical replica. Nevertheless, sales remain a big challenge for a small business. The solution should be commercialized by an interest group, formulated according to the European Economic Interest Group (EEIG) model to ensure the sustainability of the product after the pilot project is completed, and comprises three alternative offerings [65]: the direct delivery model (DDM), the cluster delivery model (CDM), and the research, innovation, and initiatives model (RIIM). On DDM, new customers will receive direct delivery of the merchandise. On CDM, clusters (tourist offices, SMEs, etc.) will advertise and run the entire offering using the channels already in place. On RIIM, the selected outcome will be made available to the research community as open source with full access for research purposes in order to increase its popularity [66].

Table 4. Comparison of delivery models, derived from Ref. [69].

Characteristics/Model	Direct Delivery Model	Cluster Delivery Model	Research, Innovation, and Initiatives Model
Customer reach	Local/online direct buyers	Wider, collective markets	Niche + premium markets
Costs	Moderate	Shared	High (R&D)
Brand control	High	Medium–Low	High
Collaboration	Low	High	Optional (with R&D partners)
Innovation	Low–Moderate	Moderate	High
Scalability	Limited	High	Medium–High

highlights the comparison of six main characteristics for three delivery models. While DDM reinforces the traditional way of business with moderate costs, low or limited collaboration, innovation, and scalability that preserve brand control, DDM requires high collaboration, and offers high scalability and almost no brand control. RRID poses a special case that combines high costs with high innovation and scalability. Based on these preliminary results, a craft business making decorative items might start with direct delivery to build a customer base. After a period, it could join a cluster model for wider distribution. In parallel, the option exists to use research and innovation to stand out with unique, trend-driven products [67,68].

Additionally, CDM can be used to engage public authorities, which can help build up a library of digital twins of monuments of interest. In the subsequent step, participation in an ecosystem such as Etsy [70] is a conceivable development. However, these presuppose strict conditions that are difficult for a craft business to meet [71].

6. Pilot Application in a Use Case

A medium-sized bust was chosen as an example for the pilot application, which is supposed to demonstrate the minimal viability of this approach. Franjo Tuđman (1922–1999), the first president of the Republic of Croatia, is presented in this bust. He is still highly regarded in the nation and is honored in dozens of monuments of all sizes and compositions. These distinguishing characteristics—size, material, and a few detail features—led to the selection of a bust from Pisarovina, Croatia [72], for this project (Figure 5, left). This object is easily accessible, adequately lit, clearly visible from all sides at eye level, well-maintained, and without surface damage. As such, this object is very suitable for trying out and comparing different capturing options. Additionally, it was anticipated that a product associated with a well-known individual would induce more attention from potential buyers. Finally, the artist agreed to this re-use of her intellectual property.

The most feasible method of acquiring shapes of an outdoor monument was to shoot dozens of ordinary images around the object of interest using a smartphone, as indicated by the ranking in Table 1. This procedure takes ten minutes under ideal settings and is accessible to anyone. Almost everyone can complete this task. Photogrammetry was used to construct the 3D bust model (Figure 5, second from left). Overall, the photogrammetry yields good results, although occasionally, minor problems occur (point cloud outliers) that cannot be attributed to a specific cause. As a result, the CAD system is required to update, adjust, and slightly improve the final model. Moreover, an object with these characteristics does not require the use of a scanner (a day rental charge in the range of EUR 1000).

The most difficult part of this project was the transition from digital to the physical process. Initially, it was agreed to produce the final product at a 1:2 scale of the original. The available 3D printer’s capabilities (workpiece extension) determined this dimension. Two or more pieces would need to be joined after the production in their original size. A mold with a 10 mm shell was made after multiple attempts to print a thin shell workpiece suitable for a lost-wax method failed (Figure 5 mid).

Using the worker’s craft skills, the next steps were completed as usual. Plaster was used to create the initial casting, which was then painted in a brass-like color (Figure 5, second from the right). This is necessary to approve the outer shape of the object. Finally, a brass object (Figure 5, far right) was cast, chemically treated to achieve a specific finish, and mounted on a granite base. The overall weight of the 18-cm bust, including its foundation, is 8.6 kg. A limited series of ten pieces was produced and widely distributed to collect consumer feedback.

Table 5. Results of the pilot project.

	Criterion	Result	Comment
1	Feasibility and competitiveness	Successful use of open-source and low-cost solutions implies additional risks	Needs continuous improvement
2	Duration of the production process	Estimated: 1 week	Heavily depends on points 1 and 5
3	Batch size in a wide range	A ramp-up from 1 to at least 20 is easily achievable	Depends on the capacity of the foundry
4	Wide range of materials	Brass and plaster work well	Aluminium and plastics to be checked
5	Small amount of rework	Unthinkable without manual rework	Needs further process improvement
6	Modularity of the process chain	Partially achieved	Interdependencies are higher than expected

provides a summary of the project’s results and experiences with reference to Table 3. Generally speaking, producing a small batch of ten products with these characteristics is both feasible and possible. However, no requirement could be entirely fulfilled. Such a “trial-and-error” method has numerous drawbacks and ought to be further enhanced, for example, by processing additional objects. Like in any manufacturing process, the overall process design is given special consideration [73]. There are still system breaks in the process chain; therefore, they should be avoided. Despite its high potential for total productivity, the virtual sub-process appears too sophisticated for a micro-business and is hence a candidate for outsourcing [74]. When numerous challenges are overcome, the process chain’s modularity can finally be realized.

The total share of manual work is reduced, but it remains in each step of physical work (column left, Figure 4). Otherwise, it is hard to imagine a craft business without manual work.

7. Discussion

Overall, the approach has provided acceptable and encouraging results. No showstopper was obtained. Each error in the process chain could be resolved with a few additional loops. The identified gaps in the current product emergence process for decorative items that belong to the cultural heritage could be (almost) closed.

7.1. Validation of Key Principles and Proof of Sustainable Frugal Innovation

Firstly, the digital process steps (Figure 4, right column) were realized as building blocks for a digital twin and subsequent utilization in mechanical production. They are a digital bridge between the artist and a craftsman. It is advised that it is possible to employ software tools, either on-site or as a service. In the first attempt, low-cost tools provide at least sufficient results. For more challenging objects, professional tools still seem to be inevitable.

Secondly, digitalization is a prerequisite for integration if good interfaces and interoperability modules are available [75]. While the downstream integration between singular phases in the product lifecycle by exchange of product data has become possible, there is still a gap in the product lifecycle management [76], e.g., how to manage a multitude of parameters for each step and the process chain as a whole [77]. The process chain is made more efficient by the metadata stream as well as the seamless geometry stream [78].

This challenge is expressed by principles P1 and P2. The remaining principles, P3 and P4, are addressed indirectly. Frugal innovation (P3) is proven as indicated by fulfillment of 6 of the 7 criteria (

Table 6. Proof of frugal innovation, derived from Refs. [6,79].

	Criterion	Achievement (Step 1)	Potential (Step 2)
1	Affordability (substantial cost reduction)	By better modularity enabled by digitalization; reuse by digital twin; reduced scrap	Optimization of mold production by a feasible AM technique
2	Essential functionality	By making molds using AM, the craft business can focus on the core business	Outsourcing of sales; use of a web store
3	Resource efficiency	Reduced scrap (30–50%), material, and energy (10–20%)	Optimization of the AM and casting process
4	Environmental sustainability	Reduced waste, material, and energy (10–20%)	Optimization of the AM and casting process
5	Social inclusivity	N/A	Customer involvement in capturing the object of interest
6	Scalability and adaptability	Reduced time-to-market; higher flexibility; launching of a new offering	Customer involvement in capturing the object of interest
7	Durability and reliability	Optimized performance level using digital twin	Introduction of product lifecycle management [76,77]

).

Reducing waste, energy, and material consumption is an example of a sustainable process definition (P4). The digital twin facilitates an accurate calculation of demand for material and energy and the design of alternatives. This advances the further cost calculation, too.

7.2. Future Directions

Further aspects are marginally affected that go beyond the scope of this research [26] and could become relevant for future activities. They are summarized in clusters organization and human factors, process improvement, and medium-term implications.

7.2.1. Organization and Human Factors

Since previous research has revealed significant differences in skills, culture, and work habits between artisans and scientists, a new approach to communication and collaboration is needed to promote interactive cooperation among stakeholders [80]. Further stakeholders (e.g., technical designers and facility managers) face new organizational challenges, including the need to acquire innovative skills and manage processes that require integrated knowledge from multiple fields [81]. Since the preservation of CH is an inherent task of public bodies, the marketing of this product could be done in cooperation with them. It could be embedded in a broader context of digital business models, as shown in Figure 3 [81].

7.2.2. Process Improvement

Ensuring the accuracy of virtual representations of physical entities [38]—such as their position, shape, size, tolerances, material properties, and assembly details—is particularly challenging for cultural heritage objects [82]. It demands transdisciplinary expertise [83] and highly specialized skills across a variety of fields, including architecture and conservation, among others [84].

Verification and validation methods and technologies should account for the unique characteristics of heritage objects, rather than simply adapting traditional approaches from the construction sector [40]. The data quality plays an important role here. It can be gathered in two directions: operational control of data quality as an autonomous procedure (supported by an application) and data quality as an inherent component of the business process as a whole [85].

Costs and time-to-market could be further reduced with an improved AM process for thin-walled molds. It requires further and costly investigations. A substantial time and cost analysis should be carried out to evaluate the cost–benefits of implementing digital twins [86]. Similarly, a life cycle assessment should be conducted to quantify sustainability metrics [87].

7.2.3. Considering Public Policies

The reuse of public goods in private businesses should be carefully coordinated with public policies and, where possible, integrated into a comprehensive approach. Following ICOM guidelines ensures that replicas and decorative items are produced with ethical transparency, avoiding misrepresentation or exploitation of cultural assets. This is especially relevant for tourism-driven economies, where demand for small-scale replicas of statues, monuments, or traditional motifs is high. Integrating these guidelines into production helps define proper licensing, intellectual property management, and fair benefit-sharing with local communities.

Public institutions responsible for the preservation of cultural heritage can strengthen and simplify this process by digitizing all assets and making them available to interested individuals and companies. By aligning digital twin technology with these heritage policies, micro-businesses can achieve sustainable frugal innovation, balancing economic growth with cultural integrity. This approach fosters innovation, supports local artisans, and strengthens the responsible commercialization of cultural heritage.

7.2.4. Medium-Term Implications

Finally, each manufacturing business must determine how to use digital innovation that will affect not only competitive capabilities such as product positioning, pricing, and marketing, but also protect workers and organizations from possible hazards (

Table 7. Transdisciplinary implications beyond the scope of the project, derived from Ref. [89].

	Criterion	Impact	Potential/Risk
1	Material processing	Time-to-market, costs, and environmental impact	Competition with affordable 3D metal print
2	Information processing	This remains a sensible point that may be improved by the software of the new generation	Seamless integration would optimize the process chain; IT provides an opportunity to interact with customers
3	Workers’ activity	The remaining portion of manual activities must be carefully adjusted with digitalization	Innovations in product promotion, distribution, and pricing also require a managerial revision
4	Societal	Spreading awareness about CH; making CH tangible	Potential for further penetration and promotion as a social innovation [90]

) [88].

8. Conclusions

Digitalization is increasingly acknowledged as an effective tool for the preservation and maintenance of CH. Digital counterparts of public spaces and museums offer virtual visitors an exceptional and immersive experience. Numerous stakeholders across society and the economy can advance from the digital twin approach, particularly when it includes a 3D representation of the actual condition. Additionally, manufacturers of decorative items could significantly streamline their processes by accessing a digital library of relevant objects. Currently, however, they must rely on reverse engineering techniques to create models of cultural assets.

In this paper, we have conducted a proof of concept that the digitalization of CH is possible, and the corresponding digital twin can be used to produce replicas of the original object on a dedicated scale from various materials. This is a sustainable frugal innovation for a small craft business that can expand its offering. A number of important options remain open, mostly regarding the commercialization of this solution.

The next step toward greater efficiency in managing newly created digital information could be the development of an ecosystem based on complementary participants and standards, offering a space for decorative item manufacturers to participate. This would involve the creation of new business models as well as the expansion of existing ones, similar to trends already seen in the manufacturing sector [91]. It would also foster the development of innovative applications by removing concerns about the benefits of digitalization, scanning devices, and data formats. Ultimately, this approach enhances both product innovation and the cultural identity of local communities, putting the presented solution in a broader societal context.