Open Access
Savonen, Benjamin Lyle
Graduate Program:
Mechanical Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
April 16, 2019
Committee Members:
  • John Gershenson, Dissertation Advisor
  • John Gershenson, Committee Chair
  • Timothy W. Simpson, Committee Member
  • Nicholas Alexander Meisel, Committee Member
  • Jason Andrew Acimovic, Outside Member
  • 3D printing
  • Humanitarian Engineering
  • International Development
  • Low-Resource Contexts
3D printing has captured the attention of those working in international development and humanitarian efforts. The increased usability and lower costs of 3D printing increasingly democratize the technology for all, and those working in low-resource contexts may be able to especially benefit from the ability to locally manufacture products with 3D printing (LM3D). For these users, locally manufacturing products could result in lower costs, shorter delivery times, and the ability to customize products to unique settings. Implementations of LM3D in low-resource contexts, however, are still relatively few, and the field as a whole is just beginning to emerge. While relatively flexible, the 3D printing technologies enabling these implementations are still not without issues, and nearly all of them are not being designed with low-resource users in mind. The technology’s limitations restrict instances in which LM3D can be reliably applied in the field. The issue is that those developing 3D printing are generally unfamiliar with humanitarian or development work, and those working in these low-resource contexts are rarely experts on the technology. This research bridges the gaps in these understandings by painting a better picture of what constitutes an appropriate use of LM3D in low-resource contexts. Inspired by my own experiences working with 3D printing in the Kenyan healthcare system and the documented experiences of others looking to pioneer 3D printing in low-resource contexts, I set out to determine what factors should be considered when evaluating an opportunity to utilize LM3D in a low-resource setting. I hypothesize that these factors should be based on principles established in the existing literature on designing products for additive manufacturing and designing products for low-resource contexts. After identifying from the literature what considerations are relevant in deciding whether 3D printing is an appropriate intervention, I further hypothesize that these considerations can be assembled into a method that can enable any user to evaluate products for LM3D. To be robust enough for use in low-resource settings, LM3D must be able to be applied to a wide variety of products and scenarios with relative ease and consistent success. For this to be the case, it is important that LM3D is applied to products for which both the 3D printing process and the scenario of its use make sense. Based on the medical idea of triage, I created a set of algorithms across a wide range of considerations to screen out products and situations encountered in low-resource contexts that would be an inappropriate use of LM3D. With the intent of aiding those in the field to better understand these considerations of LM3D, the triage is designed to answer the questions of whether a product needed in a particular scenario can and should be manufactured with the available 3D printing technology. Using case studies that describe particular product needs that could arise in low-resource contexts, I tested the triage first to see if it could predict or agree with experts’ verdict for each product. I found the triage to be correct, but generally more conservative than experts who have experience with 3D printing in lower-resource contexts in identifying potential issues. I next tested the triage with novices without such experiences to see if, using the triage, their outcomes would agree with the experts’ recommended course of action for each. Once again, the triage was found to be helpful and correct, but conservative in its recommendation on whether to proceed with a design. The triage was found to be particularly helpful in guiding the novice users to the experts’ opinion in the instances where the decisions are more difficult. The triage method outlined in this dissertation, and the LM3D considerations that it is based upon, present many opportunities for future work, with the more complete development of each of the triage algorithms representing potential research projects in of themselves. This work, as was intended, is not a conclusion to the discussion of how LM3D can be applied to low-resource contexts, but rather represents a merging of two fields (additive manufacturing and design for low-resource contexts) and provides a framework for which future research at this junction should be built off of.