Estimating the economic feasibility of additive manufacturing: a systematic literature review

Additive manufacturing (AM), which is also referred to as 3D printing, is a technology widely used in industries like aerospace, biomedical, automotive and energy for creating complex, high-quality parts (Ozkan et al., 2024), and while progress in AM methods, materials and processes are expected to enhancing its efficiency and application range (Ozkan et al., 2024), costs are one of the most significant factors influencing the extent to which AM can be applied (Baumers et al., 2016; SME, 2023). According to 2023 industry report (SME, 2023), prototyping and manufacturing aids (jigs, fixtures, tooling) are the current largest use cases for AM, relatively fewer end-use parts manufactured. AM has already shown signs of consolidation in some areas, such as customized health-care products (Cabibihan et al., 2021; Inge et al., 2018; Ozkan et al., 2024) and in the automotive industry, where in addition to prototyping and tooling, AM is extensively used for manufacturing of end use parts, such as cooling vents and valves (Arunkumar et al., 2023). In other industries and contexts, the full potential of AM is still being sought; for example, in construction, AM is still at the pilot level (Khan et al., 2021; Mallikarjuna et al., 2023; Pekuss and García de Soto, 2020), and printing of medicines is still at the experimental level (Amekyeh et al., 2021; Ding et al., 2021).

This study aims to provide a comprehensive view of cost estimation and modeling in the AM context. The practical need for this investigation arose from a research and development project aimed at bringing geopolymer concrete 3D printing to a commercial level. Some participants had an unrealistically strong assumption about the cost competitiveness of the solution. Knowing the state of the art on the subject helps to set the goals of this kind of technology-related development activities at a more realistic level. Cost estimates are needed to support various decisions regarding AM, including those involved in technology investments, equipment investments, manufacturing method selection, service procurement, manufacturing process optimization and optimization of product features.

The fundamental research question of this study is as follows:

This question is answered with a literature review. Studies on AM cost estimation and modeling are categorized based on the decision situations they address; the general themes and related key issues in each category are outlined, and conclusions of previous studies concerning the economic feasibility of AM in different application contexts are summarized and analyzed. The results, a classification of former studies and a synthesis of their conclusions, will increase the understanding of AM’s current and potential future application areas and of the utilization of different cost models to support AM-related decisions.

AM—or 3D printing—is the process in which items are manufactured layer by layer, straight from digital data. AM comprises several techniques that use different methods for material addition, such as the oft-mentioned sintering, extrusion and light polymerization.

Laser sintering and -melting processes are fundamentally similar in a way that polymer or metallic powder is melted selectively in 2D layers with high-power lasers to form a solid part. Several articles in this review discuss selective laser sintering (SLS) (Le Néel et al., 2018; Sharma and Dixit, 2019; Shehab et al., 2018) or selective laser melting (SLM) (Cicconi et al., 2021; Neuberger et al., 2020; Price et al., 2021). Another technique that uses a directed heat source to melt feedstock material is wire and arc additive manufacturing (WAAM), in which the heat energy of an electric arc is used to melt the electrodes and deposit material layers (Yehorov et al., 2019).

Another family of techniques is extrusion-based, in which material is extruded through a printer head. A popular process, particularly with hobbyists, is the fused deposition modeling (FDM) process, also known as fused filament fabrication (FFF), e.g. (Urbanic and Saqib, 2019). Unlike many other AM techniques, FDM is characterized by low investment costs; at the time of writing, lowest price of an FDM 3D printer is less than $100 (All3DP, 2017). In this method, material, typically polymer filament, is extruded through a heated nozzle in overlapping 2D layers, to form a complete part. Contour crafting is a building printing technology wherein concrete-like material is extruded via a printer head attached to a computer-controlled crane or gantry (Zhang et al., 2019b).

Other examples of techniques that use different methods for material transformation are mask image projection stereolithography, in which liquid is solidified in layers by a UV light source (Yang and Li, 2018); binder jetting, in which liquid binding material is used to solidify powdered material (Manoharan et al., 2019); and cold spraying, in which fine powder particles are accelerated in a high-velocity gas stream (Stier, 2014).

There seems to be a consensus in the literature that AM has the greatest potential as a manufacturing strategy for products with a low volume, high degree of customization and increased functionality that can be achieved with optimization (Savolainen and Collan, 2020; Tofail et al., 2018; Weller et al., 2015). The need for customization and the potential for improved functionality depend on the application context. Due to the differences in technologies and application contexts, conclusions about the economic feasibility of AM cannot be generalized across different techniques and application contexts, which is why several cost reviews of AM have been conducted.

The costs of AM have been analyzed in several studies, and a few literature reviews have been conducted on this topic. Review studies focusing on cost estimation have been conducted for AM in general (Costabile et al., 2016; Kadir et al., 2020) and for specific application areas, including surgery (Serrano et al., 2020), the defense sector (Busachi et al., 2017), concrete structures (Khan et al., 2021), metal parts in the aviation industry (Gisario et al., 2019) and the sand mold manufacturing by binder jetting and SLS (Le Néel et al., 2018).

In addition, several authors have sought to provide a general cost model for either AM in general (Ding et al., 2021) or for specific technologies, including multi-jet-fusion technology (Šoškić et al., 2021), direct metal laser sintering (Di and Yang, 2021; Kayacan and Yılmaz, 2019), SLM (Rickenbacher et al., 2013), WAAM (Cunningham et al., 2017), material extrusion (Kampker et al., 2019), contour crafting (Zhang et al., 2019b) and mask image projection stereolithography (Yang and Li, 2018).

Kadir et al. (2020) provide an overview and classification of the costing models developed for AM context. The authors argue that cost models are often targeted at specific technologies and applications, and technologies are discussed from different perspectives, such as production, management and finance. Cost classification techniques can be separated into three separate classes: level-based, task-based and method-based techniques.

Method-based cost models cover the cost elements related to both direct costs of manufacturing (e.g. material, labor, machines and energy) as well as indirect costs (e.g. administration), the goal of this technique being to provide an accurate prediction of the manufacturing costs.

Task-based cost models are typically used from the point of view of manufacturing commissioning. In this case, models include several phases from product development and manufacturing, and distinction can be made between design-oriented and process-oriented cost models. The design-oriented approach typically covers part design and/or redesign as well as process planning and is typically used prior to starting the production, whereas process-oriented models focus only on production processes, to offer a rapid quotation for product unit costs.

Level-based cost models are used from economic and managerial perspectives and can be specified into process-level and system-level models. The process-level models focus on production costs or total costs, whereas the system-level perspective considers services, supply chain and the life cycle of the product.

(Kadir et al., 2020) present that over time, cost models have evolved from process-oriented basic models, focusing on direct production costs, toward more design-oriented and system-level models that use analytical methods (e.g. artificial neural networks and fuzzy algorithms). Also other authors have noted a similar direction of development in cost models. (Costabile et al., 2016) has noted a tendency toward more precise estimations, and concludes that one kind of cost model is valid for different techniques, relevant factors being, e.g. labor, material, machine, warm-up time, power source and energy consumption. (Ding et al., 2021) state that the most recent cost models have extended the remit from costs that are generally understandable for accountants, to also include also “less-structured” costs meaning those that relate to changeover, setup, idling, wastage and quality costs that relate to preventing of process failure and noncompliance. (Ding et al., 2021) present a general cost model that includes the probability of production failures, together with its expected cost effect. Their model is based on the knowledge that it has been shown experimentally with polymeric powder bed fusion, that the costs of process failures can be noteworthy in AM context (Baumers and Holweg, 2019).

Although these previous studies provide tools for classifying cost modeling studies and cost models and describe their evolution, they lack a deeper discussion on the use of different cost estimates and models. Former studies have not been classified based on their applicability in decision-making related to AM, nor have reviews endeavored to summarize conclusions about the economic feasibility of manufacturing technologies across different application areas. This study analyzes the cost estimation and modeling of AM, particularly from the perspective of different decision-makers, and summarizes the conclusions concerning the economic viability of AM in different contexts.

Generally, management decisions are often divided on strategic, tactical and operational level (e.g., Ivanov, 2010; Kontio et al., 2013; Shivakumar, 2014). Aaker and Mascarenhas, (1984) define that strategic management involves identifying emerging environmental threats and opportunities, predicting their potential future impact and developing appropriate organizational responses. According to Kontio et al., (2013) tactical decision-making instead, focuses on medium-range plans, schedules and budgets, addressing resource allocation and performance monitoring of organizational subunits like teams and divisions. It involves setting policies, procedures and objectives, whereas operational decision-making focuses on daily, routine decisions. However, it is not entirely self-evident which decisions are considered strategic in practice. Shivakumar, (2014) identifies two dimensions for distinguishing strategic decisions from non-strategic ones: the degree of commitment and the scope of the firm. The degree of commitment reflects the reversibility of a decision, including costs like sunk costs and opportunity costs. The scope of the firm pertains to choices involving products, services, activities and markets.

In this study, literature is categorized based on the AM-related decision-making scenarios addressed. The classification is based on the widely recognized division into strategic, tactical and operational decisions (Figure 1). Another dimension considered is the scope, which defines whether the focus is on specific products, or on broader perspective.

Strategic decisions in AM revolve around whether to invest in or adopt AM technologies in the first place. This involves evaluating the future opportunities of these technologies. General cost models, industry reviews and successful case studies showcasing the use of AM are particularly useful in assessing future potential.

Decisions can be regarded rather tactical than strategic if they focus on the practical implementation of an already selected strategy, that is purchasing equipment or services. For instance, choosing which AM technology to use or determining how to integrate it with traditional manufacturing processes falls under this category. In such scenarios, technology-specific cost models and comparing differences between methods becomes essential.

Operational decision-making focuses on resource efficiency in daily level AM decisions. These decisions emphasize efficiency, learning and optimization in day-to-day operations, including manufacturing processes and related decision-making, such as design and planning. In practice, this involves models, methods and tools for daily decision-making needs.

In this study, a systematic literature review was used to comprehensively analyze existing findings on the economics of AM. This method was chosen as it allows for evidence-based conclusions by integrating findings from multiple studies, thereby enhancing the validity and reliability of the results. (Page et al., 2021; Thomé et al., 2016) Conducting a systematic literature review included the following steps:

• stating the keywords;

• compiling articles through search engines;

• classifying the articles according to predetermined criteria; and

• analysis and discussion.

Figure 2 presents the phases of selecting records for analysis. A keyword search from within the abstract and citation databases Scopus (193 hits) and Web of Science (212 hits), was performed in March 2022.The used databases cover the most central journals in this subject area, that offer high-quality articles. Articles, reviews and conference papers written in English were considered. The topic search string contained either “additive manufacturing” or “3d-printing” coexisting with in minimum one of the following phrases: “cost model,” “cost modelling,” “cost estimation,” “cost estimate,” “cost analysis,” “economic model,” “economic analysis,” “economic modelling” or “economic evaluation.” Because the records are mainly peer-reviewed, no actual eligibility check was performed.

Figure 3 shows that cost estimations concerning AM are a relatively new topic. Studies have been published at an accelerating pace since 2010 but have experienced a slight decrease in the most recent years.

Studies have been published in a wide range of journals. Although many of the journals have a manufacturing focus, studies have also been published in journals with a sustainability science or medical science focus (Table 1).

After preliminary screening, a general classification of the articles was conducted. Articles were categorized according to the level of decision-making they were considered to support; strategic, tactical or operational. This was done by analyzing whether they addressed mainly long-term goals, medium-term plans or day-to-day activities. Additionally, articles were classified based on their product specificity; whether a specific product or group of products was mentioned in the title or abstract as the subject of the study. Classifying studies based on product specificity allows more detailed insights and comparisons across different contexts, and more precise recommendations for practitioners.

Further analysis of the studies was conducted by seeking to answer the following questions:

Finally, general themes were outlined concerning the purpose of using the cost estimations and models.

Table 2 presents how the articles fall into the categories defined by decision-making level and application focus.

Collectively, studies without a clear application focus contribute to strategic planning by offering insights into technological trends, cost management, environmental sustainability, supply chain design and business model development. More specifically, these studies offer general cost models for AM (Baumers et al., 2016; Baumers and Holweg, 2019; Ding et al., 2021; Fera et al., 2017; Franchetti and Kress, 2017; Gitlin et al., 2018; Kadir et al., 2020; Lindemann and Koch, 2016; Minguella-Canela et al., 2019; Shehab et al., 2018), one focusing on printer electricity consumption (Ajay et al., 2016), or discuss the economics of AM on a general level (Abdulhameed et al., 2019; Caviggioli and Ughetto, 2019; Mellor et al., 2014; Niaki and Nonino, 2017; Sharma and Dixit, 2021b; Thiesse et al., 2015).

Another general theme is the impact of AM on overall management. This includes identifying the implications of AM for supply chains (Afshari et al., 2019; Bonnín Roca et al., 2019; Eyers and Potter, 2015; Faure and Li, 2020; Feldmann and Pumpe, 2017; Khajavi et al., 2018a; Ott et al., 2019; Pahwa and Starly, 2020; Rudolph and Emmelmann, 2017; Tasé Velázquez et al., 2020; Zanoni et al., 2019; Alogla et al., 2021), business models (Godina et al., 2020; Piller et al., 2015) and AM management (Khorram Niaki and Nonino, 2017). Additionally, the environmental impacts or sustainability of AM have been explored by several studies (Kellens et al., 2017; Kianian and Larsson, 2016; Ma et al., 2018; Niaki et al., 2019; Ribeiro et al., 2020; Yao and Huang, 2019; Yosofi et al., 2019).

Some studies focus on specific AM methods, e.g. forecasting the future development of SLM (Kopf and Lanza, 2016), build time estimation in SLS (Zhang and Bernard, 2014), FDM printer design (Sood and Pradhan, 2020), environmental impacts of SLM (Faludi et al., 2017), life cycle cost analysis for laser-based powder bed fusion (Nyamekye et al., 2020) and economic and environmental analysis of FDM, SLS and MJF (Tagliaferri et al., 2019).

Other topics include intellectual property rights issues in the 3D printing context (Depoorter, 2013), the use of AM in mechatronics learning (Kaneps and Gerina-Ancane, 2016) and using system engineering framework for AM (Togwe et al., 2018)

Literature reviews summarizing research findings have been conducted for such application areas as concrete structures (Khan et al., 2021), cleft care (Virani et al., 2022), surgery (Serrano et al., 2020), the aviation industry (Gisario et al., 2019), the automotive industry (Yi et al., 2019), chemical separations (Davis et al., 2021) and the defense sector (Busachi et al., 2017).

One general theme in research is determining the economic viability of AM in a selected application area, and some studies draw straightforward conclusions on this; WAAM is reported to be less economical than fabrication by traditional means at smallholder rice farms (Gu et al., 2020), conventional construction methods outperform AM in the production of geometrically simple concrete constructs (Pekuss and García de Soto, 2020), and powder bed fusion AM-based spare part supply is noted in many cases to not be cost competitive with traditional spare parts supply that is based on warehouses (Zhang et al., 2019a).

Nonetheless, some studies do suggest that AM outperforms conventional manufacturing methods in specified application areas. Rapid-laser reactive sintering is reported to be more cost competitive than conventional furnace sintering in the production of protonic ceramic fuel cells (Mu et al., 2020), AM-based prototyping has been reported to lead to significant cost reduction (Khorram Niaki et al., 2019), and AM-based tooling is reported to be less accurate but cheaper than CNC machined tooling (Dippenaar and Schreve, 2013). Fused fabrication using recycled plastic is reported to be a cost-efficient method for the manufacturing of large sporting equipment products when compared to commercial equivalents (Byard et al., 2019), and Zhang et al. (2013) find printed open source optics equipment to be considerably less expensive than commercial equivalents. Petersen and Pearce (2017) assert that a desktop 3D printer is a profitable investment for average households, since the costs of manufacturing with a low-cost desktop printer using free 3D-models provided by online communities, are generally lower than the sales prices of their commercial equivalents. However, this kind of method-based estimation ignores such factors as reduced visual quality, reduced lifetime of the most common printing material (polylactic acid) and the relatively high time investment associated with desktop 3D-printers (Rintala, 2021).

These studies collectively highlight that AM can be cost-effective in specific applications like prototyping and tooling, but traditional methods often remain more economical for simpler constructs and spare parts supply.

Some of the comparative studies provide a straightforward conclusion on the financial viability of AM in case of specific products. In cases of metallic flat washers (Guarino et al., 2020), connecting rods (Srivastava et al., 2020) and microreactor plates (Raoufi et al., 2020), traditional manufacturing methods were shown to be more cost efficient than AM, and Schneider et al. (2019) report a case in which dental implant planning and placement that used AM was more costly than a conventional implant planning and placement process. However, lower costs compared with traditional manufacturing methods are reported in the cases of acetabular implants (Tack et al., 2021), component breeding blankets for a special type of pebble bed nuclear fusion reactor (Neuberger et al., 2020) and small batch production of polymer tools for injection molding (Kampker et al., 2020). In the case of the GE engine bracket (Laureijs et al., 2017), considering the part redesign for AM and the resulting fuel savings during brackets lifetime, the AM part has demonstrated to be less expensive than a forged one in many scenarios.

AM has been presented as a low-cost or cost-effective solution for manufacturing prosthetic hands (Cabibihan et al., 2021; Zuniga et al., 2015), nose bolus (Albantow et al., 2020), mandibular implant plates (Moiduddin et al., 2020), microfluidics for biomedical analysis (Yang et al., 2020), molds for cranioplasty implants (Low et al., 2019), corrective osteotomies (Inge et al., 2018), orthopedic implants (Van de Kleut et al., 2018), orbital implant templates (Callahan et al., 2017), thumb orthoses (Fernandez-Vicente et al., 2017), cell-stretching apparatuses (Toume et al., 2016), metallic control valve trims (Singh et al., 2020), solar collectors (Martín-Sómer et al., 2021), jewelry (Tu et al., 2020), flow models for research purposes (Mattio et al., 2017), scattering measurement instruments (Nadal-Serrano et al., 2017) and different types of sensors (Cennamo et al., 2021; Cho et al., 2017; Leigh et al., 2012). In multiple cases concerning 3D-printed models for medical planning or training (Ballard et al., 2020; Bettega et al., 2019; Eltes et al., 2020; Gillett et al., 2021; Nebor et al., 2021; Parotto et al., 2017; Richter et al., 2021; Ruedinger et al., 2019; Seifert et al., 2020; Shen et al., 2020; Smektała et al., 2016; Spaas and Lenssen, 2019; Wang et al., 2018; Watson, 2014; Yang et al., 2015), AM was presented as a low-cost solution. Some studies focus more on the achieved system-level improvement, for example, reported reduced operation time (Ballard et al., 2020; Yang et al., 2015), reduced risk of operative complications (Katkar et al., 2018) and educational value (Parotto et al., 2017; Seifert et al., 2020).

Some studies refrain from general-level cost estimations and focus on the advantages achieved with AM, such as weight reduction in the case of robot parts (Junk et al., 2019) and gears (Kamps et al., 2018). Thomas et al. (2021) studied the process enhancement achieved with 3D-printed feed spacers in different membrane distillation configurations. Theodosiou et al. (2019) report that microbial fuel cells with 3D-printable materials such as terracotta and air dry Fimo outperformed those using conventional cation exchange membranes. Manoharan et al. (2019) compared two metal AM processes for manufacturing microscale plate reactors. In this case, AM enables high surface-area-to-volume ratio, and thereby reducing the size and cost of chemical reactors.

As a summary, it can be said that for strategic decision-making, the literature provides examples of products where AM is a cost-effective solution. However, except for a few examples of small special parts, most of these examples represent the medical sector.

Decisions related to the choice of a specific AM technology or its cost assessment can be considered tactical because they are related to implementation of AM strategy. Cost estimates and comparisons can be used to support equipment and service procurement decisions.

These studies focus on such topics as manufacturing method selection (Khaleeq uz Zaman et al., 2017), focusing on machine and materials (Palanisamy et al., 2020), AM infrastructure (Reiff et al., 2019) and integration of traditional manufacturing and AM (Manogharan et al., 2016, 2011; Nagulpelli et al., 2019). Some of these studies evaluate or model the costs of specific technologies, namely multi-jet-fusion (Šoškić et al., 2021), polyjet process (Rimašauskas et al., 2014), direct metal laser sintering (Baumers et al., 2012; Di and Yang, 2021), laser powder bed fusion (Schneck et al., 2020), SLS (Sharma and Dixit, 2019), SLM (Rickenbacher et al., 2013), laser-powder bed fusion (Piller et al., 2019), WAAM (Cunningham et al., 2017), cold spray (Stier, 2014), FFF (Urbanic and Saqib, 2019), contour crafting (Zhang et al., 2019b), mask image projection stereolithography (Yang and Li, 2018), inkjet-based 3D printing (Zhao et al., 2021), freeform injection molding (Chaudhuri et al., 2021) and metallizing structures fabricated with FFF (Romani et al., 2021). This type of study can provide a benchmark price, for example, for the purchasing of printing services, although this goal has been raised to the title level only in the case of laser sintering service providers (Baldinger and Duchi, 2013). Sharma and Dixit (Sharma and Dixit, 2021a) present a general cost comparison of SLS with injection molding.

Some of the studies focus on an economic analysis of a production process detail—such as lattice structures in metal powder bed fusion (Flores et al., 2020), the use of in-line nitrogen-helium blending in cold spray (MacDonald et al., 2019), recycled powder feedstocks in powder bed fusion (Barclift et al., 2016) and the risk of failure in laser sintering (Baumers and Holweg, 2016)—or focus on tradeoff analyses—such as strength to cost in FDM (Mostafa et al., 2018), complexity to energy consumption in electron beam melting (Baumers et al., 2017a, 2017b), printing parameters to cost in FDM (Singh et al., 2014) and cost to the shape memory properties and material solidification chemistry in 4D printing (Han et al., 2021).

Some of the studies refer to a group of technologies by printing material or printer type. These include cost-modeling for material extrusion processes (Kampker et al., 2019), a cost estimation for laser AM of stainless steel (Piili et al., 2015), a lifecycle economic analysis of open-source printers (Wittbrodt et al., 2013), cost modeling tools for design in metal AM (Barclift et al., 2017) and techno-economic modeling of 4D printing (Han et al., 2021). The rest of these studies discuss non-planar printing strategies, harmful particle emissions from thermoplastic AM (Sittichompoo et al., 2020), printing material (graphene) production (Dong et al., 2018), photovoltaic modules as power sources in mobile AM (Fateri et al., 2015) and a survey on AM device purchasing intentions in the metal casting industry (Lynch et al., 2017).

Cost estimations have been presented for spare parts in the process industry (Cardeal et al., 2021), spare parts in agriculture (Łukaszewski et al., 2021), 3D-printed material with embedded sensors (Mekid et al., 2015), automotive parts (Bubna et al., 2016), complex-shaped products (Ahmad et al., 2015), gas turbine components (superalloys) (Moor et al., 2010) and 3D-printed food (Dabbene et al., 2018). Yi et al. (2019) identified in a research project that cost estimation is one of the key issues for application of AM technologies in the German vehicle industry in the long term. Gatto et al. (2015) report an educational project that included AM-enabled prototyping and cost analysis. Ju et al. (2018) present a preliminary analysis to enable cost estimation in the construction context. Abdalla et al. (2021) focus on eco-efficiency of AM in the construction context. Some studies present real-time costing tools to be used during the design of buildings (Bañón and Raspall, 2021) or consumer products (Dinda et al., 2017).

Some of the studies focus on cost comparisons in different settings, e.g. manufacturing 26 design solutions for COVID-19 with AM versus injection molding (Prabhu et al., 2021), tooling manufacturing for composite parts production with FDM versus liquid crystal display (Cicala et al., 2021), additively constructed fiber-reinforced concrete versus conventional concrete masonry and cast-in-place construction (Kreiger et al., 2019), SLS versus injection molding in small batches of plastic products (Brajlih et al., 2016), AM versus conventional production for geothermal tools (Price et al., 2021), factory versus on-site printing of geopolymer construction products (Munir and Kärki, 2021), AM-enabled remanufacturing versus new product manufacturing (Xu and Feng, 2014), metal parts production with SLM versus high-pressure die casting and 5-axis machining (Atzeni et al., 2014), traditional high-pressure die casting versus direct metal laser sintering in small/medium batch metal parts production (Atzeni and Salmi, 2012) and traditional injection molding versus rapid manufacturing combined with redesign in medium volume production of plastic parts (Atzeni et al., 2010). Although AM is typically presented as a comparable production method, these studies do not draw simple conclusions about the superiority of any technology in the selected application area. Instead, for example, Facchini et al. (2019) conclude that the economics of WAAM for aeroengine components depends on the batch size.

Some studies estimate or analyze the cost of AM manufacturing for products such as gas burner heads (Cicconi et al., 2021), venturi nozzles (Fritz et al., 2013), aeroengine components (Thomsen et al., 2017), automotive exterior parts (Wiese et al., 2021), optomechanical cage systems (Winters and Shepler, 2018), small scale wind turbines (Davis and Madani, 2018), wafers (Chen and Tsai, 2018), hydraulic manifolds (Rolinck et al., 2021), hip endo-prostheses (Mihaela-Elena et al., 2017), meniscus transplants (Zhang and Wang, 2017), custom ventilation masks (Willox et al., 2020), biomedical implants (Emelogu et al., 2016), medical device prototypes (Ulmeanu et al., 2018), removable dentures (Srinivasan et al., 2021) and foot orthosis (Jumani et al., 2016). Lawand et al. (2020) present a life cycle cost model that considers the use of AM for repair and manufacturing and demonstrated its use with an aeroengine component. Westerweel et al. (2018) present a life cycle cost model that is demonstrated with two equipment components. Flores Ituarte et al. (2018) performed a case study of a laser-sintered automotive component and present a cost-comparison model based on that. A literature review by Le Néel et al. (2018) discusses the economic implications of using binder jetting (3D printing) and selective laser sintering for sand mold fabrication.

Some of the studies focus on cost comparisons in different settings, e.g. injection molding versus FDM for mobile case covers (Minetola and Eyers, 2017), a prototype with polyjet process versus FDM (Maurya et al., 2019), orthopedic scoliosis braces with FDM versus thermoforming (Redaelli et al., 2020), distributed versus centralized manufacturing of medical parts (Verboeket et al., 2021), cranial implants manufactured with or without supports (Moiduddin et al., 2021) and hybrid manufactured casting moldings with different mechanical properties and dimensional accuracy (Singh, 2012).

Operational decision-making is about resource efficiency in day-to-day decisions. Concerning generic level studies (studies without a clear application focus), three themes clearly stand out, namely:

1. resource optimization, in particular topology optimization (Lei et al., 2016; Liu et al., 2019; López et al., 2018); optimization of packing (Araújo et al., 2019, 2020; Baumers et al., 2017a, 2017b), that is, the layout of objects printed at the same time (also referred to as kitting) (Khajavi et al., 2018b); optimization of parts consolidation (Nie et al., 2020); build direction (Di Angelo et al., 2020); energy usage (Ajay et al., 2017); and sustainability (Verma and Rai, 2017).

2. cost-effective design or DfAM (Kadkhoda-Ahmadi et al., 2019; Lalley et al., 2016; Newell et al., 2019; Ningrong et al., 2015; Prabhu et al., 2019; Pradel et al., 2018; Thompson et al., 2016; Turk et al., 2017), in particular parts consolidation (Stevenson et al., 2020).

3. cost estimation tools (Armstrong et al., 2017; Baldinger et al., 2016; Baumers et al., 2013; Cameron and Gordon, 2017; Chan et al., 2018; Evans et al., 2014, 2015; Fox et al., 2018; Mahadik and Masel, 2018; Shakirov et al., 2020; Yao et al., 2016; Yi et al., 2021).

Also several studies with an application focus present approaches for different resource optimization problems, such as infill pattern selection (considering mechanical properties, part costs and production time) (Baich and Manogharan, 2020), deciding between centralized and distributed manufacturing of spare parts (Li et al., 2019), topology optimization for metal AM (Ulu et al., 2019), capacity planning including build volume packing and the anticipation of build failure in medical devices domain (Baumers et al., 2018) and process-planning for hybrid remanufacturing (additive-subtractive) (Zheng et al., 2020), identifying the impact of travel speed and wire feed speed to production costs in WAAM of alloys (Yehorov et al., 2019), Colosimo et al. (2020) present in situ monitoring tools for metal AM. Oyesola et al. (2019) present a decision tool for the implementation of a laser AM system. Chen et al. (2018) present a computational solution for using prefabricated general building blocks as the cores of the manufactured items cost-efficiently with AM. Munguia et al. (2007) present strategies for estimating the feasibility of rapid manufacturing, taking into account the design. Song and Telenko (2019) illuminate key factors affecting desktop FDM printing failure rates in an open studio environment.

Articles rarely examine operational-level decision-making from the perspective of a specific product, but a few present approaches for AM supply chain optimization using case examples of biomedical implants (Chowdhury et al., 2020; Emelogu et al., 2019).

This study highlights that the existing literature provides valuable insights to support decision-making concerning AM at strategic, tactical and operational levels. At the strategic level, literature offers general cost models, comparisons of AM production costs with the costs of traditional manufacturing methods, insight in implications of AM for general management and estimations and case examples on economic viability in specific applications. Tactical-level decisions are informed by technology-specific cost models and estimations and comparisons of specific technologies in different application areas and products. Operational decision-making benefits from studies providing practical methodologies and tools that enhance day-to-day efficiency in AM processes, including tools for cost estimation, design of products and print batches, as well as optimizing printing process details.

Studies that report the superiority of AM over other manufacturing methods in end product cases emphasize the system-level perspective; the total benefit of an additively manufactured product is realized during its use and typically has required new design. AM is strongly linked to digital design, and the benefits of these two cannot be completely separated. This is particularly evident in health-care case examples, which have been widely reported. The results of this study imply that unless there is a need for customization or optimization of geometry, AM is unlikely to be a cost-effective method for end product manufacturing due to high investment and material costs, despite advancements in materials and methods.

The primary limitations of this study pertain to the research methodology. Assessing the economics of manufacturing is a practical topic, and it appears to have been reported with a bias in scientific literature; health-care applications are emphasized while industrial applications are less frequently reported. Consequently, this study does not provide a comprehensive view of industrial end products for which AM is economically viable. Nevertheless, the study offers an overview of the relevant research literature on the topic.

Future research could focus on providing case studies related to tooling in industrial contexts, as few examples were found in this literature review despite the known prevalence of this use case. Investigating this area could provide valuable insights and promote best practices that enhance the resilience of manufacturing companies.