Better With UX: My Mechanical Engineering Capstone Project
Long before I decided to pursue a career in UX, before I even really knew what a UX designer does or how user experience design works, I obtained my bachelors from the University of Washington in Mechanical Engineering. I was drawn to MechE because i always excelled at math and science, physics and materials science in particular, and the field offers a lot of opportunities for designing real world objects.
The University of Washington is a public research university, and my final year of undergrad I led a team of 4 classmates in investigating the possibility of using 3D printers to make prosthetic limbs in a capstone project that we worked on during our last semester of college. The purpose of the project was to make prosthetics more accessible and affordable for those who required them. As the team lead I took on a project management role, keeping us on schedule and delegating tasks, and I also took on the role of materials science expert.
There was a lot to take into consideration when it came to the idea of 3D printing prosthetics. As mechanical engineers our main concern was the strength of the materials, which was determined by the 3D printable plastic we chose, the capabilities and print weave patterns of the 3D printer, and any additional materials we could use to increase the strength of the prosthetic. But the strength requirements would be based on the user and how they used the product.
Who are we designing for? Are we designing for long term durability or easy customization? Does the user have access to a 3D printer, or is this something they send away for? All of these factors would help us determine the material properties and the printer capabilities necessary.
The first thing we did wrong, we made a lot of assumptions. Looking back on the project now through a user experience lens, I see how we could have saved time and effort by conducting a thorough UX research study and narrowing down the scope of our project.
One familiar UX tool we did use was creating user personas, to inspire empathy and help us stay on track with the project. We created 3 to be exact.
Hailey, the 5 year old that lost her foot in a tragic accident. The main concern of her and her caretakers is that she will continue to need new prosthetics as she physically grows, which is costly.
Ryan, who is under the poverty level and does not regularly have access to health care. He lost his leg up to the knee as a result of poor health care. His main concern is his inability to afford a traditionally made prosthetic.
Jon, who is in the armed forces. He lost his arm to his elbow in an on the job accident. He doesn’t share the concerns of the other personas, because he has reliable and affordable access to healthcare.
These stories were created for the most part off the top of my head, and I probably got most of my inspiration from movies. While we had done some basic quantitative research online determining high overview level statistics about the number of amputees in the US, we didn’t determine the demographics amongst amputees, their lifestyles or their access to healthcare and prosthetics. We did not do enough investigation into their needs.
More egregious, we didn’t even consider a qualitative approach to our research phase. Ideally after conducting a more thorough quantitative study, we would have honed in on one subset of the amputee population, and conducted interviews with several members of that population to determine their goals and pain points. This would have greatly narrowed the scope of our project and insured that what we were working on was actually addressing the needs of the users.
The second issue is that we created three different personas that require three different products. Assuming that the made up needs for these personas are anywhere close to those of the real users, Hailey needs a continual supply of prosthetics that are all but disposable and do not need to stand up to high stress over time. Ryan needs a durable prosthetic that will last him as long as possible but is still affordable. And while Jon may enjoy being able to customize and print his own prosthetic, and really make it something that suits his personality and aesthetic, he has access to traditional prosthetics so it is less of a need and more of a luxury.
We approached the problem as engineers, trying to find a material/printer combination that would fit the needs of all three users. Easy, quick and inexpensive to produce, with a material strong enough to withstand the forces applied when used as a major lower limb prosthetic. The choice of printer wasn’t much of a choice, in order to complete our testing we had to use the equipment available to us on campus. For our plastic we decided to use PETG (glycolized polyester) due to its ease of printing and strength properties.
We created different samples of composites, consisting of different ratios of PETG combined with a carbon fiber weave applied using an epoxy resin. These samples were tested for tensile and flexural strength. Ultimately we failed to get consistent results based on many factors, including the 3D printing equipment available to us and inconsistencies in our curing process that adhered the carbon fiber to the plastic.
We had fallen into the classic trap well known in the UX field. We were trying to design for the needs of too many, trying to cover all our bases. Had we decided to address the needs of one of the personas, rather than all 3, it is probable that we could have better chosen materials that fit the criteria and made more progress on our overall project.
If I could do it over, I would have started the project collecting quantitative data, finding statistics on amputees and their access to healthcare here in the US. Based on these findings I would have narrowed down the users to one specific persona, with specific needs. I would have interviewed amputees that fit this criteria, removing the assumptions and biases my team and I unknowingly had in an attempt to actually address what the user needs are. From there we would have had a much more manageable scope for the strength requirements of our materials, and we could have made a bigger impact with our projects findings.
