Victor Verboeket

Additive Manufacturing (AM) – also known as 3D-printing – is one of the digital technologies associated with the 4th industrial revolution. The objective of this dissertation is to investigate the potential of AM for designing vastly improved future Supply Chains (SCs). A lack of knowledge about the effects of AM on SC design is considered a gap in the literature. To bridge this gap, this dissertation includes theoretical and empirical research. Following the introduction in Chapter 1, the main body of this dissertation consists of Chapters 2 to 5, each representing one academic article. The final Chapter 6 uses the Design Science Research (DSR) strategy, including the CIMO-logic, to integrate theoretical and empirical research. The CIMO-logic explains how interventions (I), in a certain context (C), lead to outcomes (O), through specific mechanisms (M).

The Chapter 2 article uses a systematic literature review. The results show that the main advantages of AM include the possibility to manufacture complex shapes, thus allowing for direct integration of functions into the product, or to manufacture lightweight products. Moreover, since AM does not require product-specific machines and tools, and digital files easily bypass physical boundaries, in-situ production may be allowed. These characteristics shorten and simplify SCs. Using AM in the SC may reduce cost and assets, and may improve order lead-time, but mainly for small products and short series. This article contributes to the academic knowledge base by developing 18 propositions, a conceptual model, a research agenda, and a roadmap illustrating the bottlenecks, serving as thresholds that prevent the full roll-out of AM.

The Chapter 3 article uses literature synthesis and DSR’s CIMO-logic. This article contributes to the academic knowledge base by constructing six theoretical AM SC ‘mechanisms’ that can be used to design better performing SCs by using AM.

The Chapter 4 article combines case study research and scenario analysis to investigate a change from centralized to distributed AM production of surgical guides used in University Medical Centers. The results show that – when changing from centralized to distributed AM – the SC lead-time improves, but at significant cost increase, since investments in machines and facilities at each of the distributed locations are required.

High AM machine costs, post-processing issues, and low manufacturing speed are bottlenecks that need to be eliminated before distributed AM becomes competitive with centralized AM. This article contributes to the academic knowledge base by increasing the understanding of the relevant factors in deciding upon centralized or distributed AM production of personalized medical care products SCs.

The Chapter 5 article combines case study research and scenario analysis to investigate implementing AM in a traditional-manufacturing orthopedic shoe SC. The results show that using AM to replace traditional-manufacturing of tools and components deteriorates SC cost and lead-time performance. However, when AM of complete final products replaces traditional-manufacturing of components, lead-time significantly improves at competitive SC costs. Yet, currently, the quality of AM produced shoes is insufficient, AM production costs are too high, and traditional craftsmanship knowledge needs to be integrated into the digital design. This article contributes to the academic knowledge base by pinpointing the required conditions for AM to significantly improve SC performance in personalized medical care.

Chapter 6 uses DSR to bridge theory and practice and compares the empirical CIMO reasoning chains derived from Chapters 4 and 5, with the theoretical mechanisms synthesized from theory in Chapter 3. This chapter contributes to the academic knowledge base by (partly) confirming the existence of four of the six theoretical mechanisms. The empirical mechanisms are valid for other non-inventory holding contexts but can be extended with the inventory-related elements. These mechanisms can be used to design significantly better performing SCs using AM.

The main conclusion of this dissertation is that AM offers great opportunities for designing significantly improved SCs, but bottlenecks, serving as thresholds prevent the full roll-out of AM. Besides, merely using AM to replace elements of the traditional SC can deteriorate SC performance. Therefore, to fully exploit AM’s specific advantages, a complete redesign of the existing SC is required. The mechanisms developed in this dissertation, may serve as a tool for SC (re)design, which may ultimately lead to radical change of the existing SC structures.