The Impact of Additive Manufacturing on Supply Chains and Business Models: Qualitative Analyses of Supply Chain Design, Governance Structure, and Business Model Change

Recent global crises like the COVID-19 pandemic challenge traditional global supply chains (SCs). Their disaggregated, “fine-sliced” character comes with a high risk of disruption, and current supply bottlenecks (e.g., the chip shortage in the automotive industry) demonstrate that there is often no quick fix. Firms are increasingly under pressure to react and (re-)design their SCs to increase their resilience. Additive manufacturing (AM) technologies are acclaimed for their potential to foster the shift from global SCs to shorter, decentralized, and more resilient SCs. The key feature of AM technologies lies in their inherently digital and flexible nature. Their specific characteristics are envisioned to enable location-independent manufacturing close to or even at the point of demand and lead to a commoditization of manufacturing infrastructure for flexible outsourcing to local partners. Moreover, AM technologies are expected to revolutionize the way firms do business and put traditional business models at stake. This doctoral thesis is motivated by the outlined potential of AM and the resulting impact on firms’ supply chain design (SCD) and business model choices. The extant literature raises high expectations for AM. However, concrete and real-world insights from specific application domains are still scarce. This thesis seeks to fill the gap between high-level literature-based visions and currently emerging realistic business models and SCDs for AM. Thereby, AM is understood as a potential intervention emanating from outside firms and requiring them to react by realigning their business models and SC structures to maintain a fit. This thesis aims to build an in-depth understanding of these mechanisms and, hence, of the inner causal processes involved in the AM SCD and business model choices. This concentration on the rationales and underlying behavioral patterns is formalized with primarily exploratory (how and why) research questions that are addressed with qualitative research methodologies, mainly case study research and grounded theory. These methodological practices are applied in the industrial AM context, entailing an embedding of this thesis in challenging industries where AM applications have already started to create value (i.e., in the aerospace, rail, automotive, and machinery and equipment industries). The selected research approaches are mostly inductive and, hence, strongly driven by the data collected from this context (e.g., in interviews, by reviewing documents, and by analyzing websites). Additionally, this thesis relies on grand theories, namely transaction cost economics, the resource-based view, and configuration theory, to discuss the findings in their light and to interpret and distill nuances of these theories for their application in the industrial AM context. This thesis is cumulative, consisting of four studies that form its main body. These studies are organized in two parts, part A and part B, since two domains of strategic decisions are targeted jointly, the business model development (part A) and AM SCD choice (part B) for industrial AM. Different perspectives are associated with the two parts. Logistics service providers (LSPs) are in a critical position to develop AM business models. Based on the expected shift to decentralized, shorter SCs, the traditional business models of LSPs are at risk, and their inherent customer orientation puts them under pressure to adjust to their customers’ needs in AM. In part A, study A.1 applies a process-based perspective to build a broad understanding of how LSPs currently respond to AM and consumer-oriented polymer 3D printing with specific AM activities. It proposes six profiles of how LSPs leverage AM, both as users for their in-house operations and as developers of AM-specific services for external customers. A key finding is that the initiated AM activities are oftentimes strongly based on LSPs’ traditional resources. Only a few LSPs are found whose AM activities are detached from their traditional business models to focus on digital platform-based services for AM. In contrast to the process-based perspective and focus on business model dynamics in study A.1, study A.2 takes an output perspective to propose six generic business model configurations for industrial AM. Each configuration emerges from the perspective of LSPs and is reflected by their potential partners/competitors and industrial customers. Study A.2 explores how the six generic configurations fit specific types of LSPs and how they are embedded in a literature-based service SC for industrial AM. In combination, studies A.1 and A.2 provide a comprehensive understanding of how LSPs are currently reacting to AM and an empirically grounded perspective on “finished” AM business models to evaluate and refine literature-based visions. Part B of this thesis is devoted to the mechanism of (re-)designing SCs for AM, which is investigated from the perspective of focal manufacturing firms based on their dominant position in SCs. Two dimensions are used to characterize AM SCDs, their horizontal scope (geographic dispersion) and vertical scope (governance structure). The combination of both dimensions is ideally suited to capture the literature-based vision of shorter, decentralized AM SCs (horizontal scope) with eased outsourcing to local partners (vertical scope). Study B.1 takes a firm-centric perspective to develop an in-depth understanding for AM make-or-buy decisions of manufacturing firms, the outcomes of which determine the SC governance structure. This study elaborates how the specific (digital and emerging) traits of industrial AM technologies modify arguments of grand theories that explain make-or-buy decisions in the “analog” age. In comparison, study B.2 shifts from a firm-centric to a network perspective to rely on both dimensions for investigating cohesive AM SCD configurations. More specifically, study B.2 explores four polar AM SCD configurations and reveals manufacturing firms’ rationales for selecting them. Thereby, it builds an understanding for why manufacturing firms currently have valid reasons to implement industrial AM in-house or distributed in a secure, firm-owned network. As a result, combining both studies provides an understanding of why manufacturing firms currently select specific governance structures for industrial AM and opt for SCDs that differ from the literature-based vision of decentralized, outsourced AM. Overall, this thesis positions itself as theory-oriented research that also aims at supporting managers of manufacturing firms and LSPs in making informed decisions when implementing AM in their SCs and developing AM-based business models. The three studies A.1, A.2, and B.2 contribute to initial theory building on how and why specific AM business models and SCDs emerge. With their focus on developing an understanding for the causal processes (how and why) and by assuming a process-based and output perspective, they can draw a line from firms’ current reactions to sound reflections on future-oriented, high-level expectations for AM. As a result, the studies significantly enrich and refine the current body of knowledge in the AM business model literature on LSPs and the operations and supply chain management literature on AM SCDs, focusing on their geographic dispersion and governance structure. This thesis further contributes with its context-specificity to building domain knowledge for industrial AM, which can serve as one “puzzle piece” for theorizing on how AM and other digitally dominated (manufacturing) technologies will shape the era of digital business models and SCs. In particular, study B.1 stands out by its focus on theory elaboration and the objective of developing contextual middle-range theory. It reveals that emerging digital AM is a setting where the argumentation of grand theories provides contradicting guidance on whether to develop AM in-house or outsource the manufacturing process. Such findings for industrial AM raise multiple opportunities for future research, among them are the comparison with other industry contexts with similar characteristics and the operationalization of the propositions developed in this thesis in follow-up quantitative decision-support models.