Improving Red Cross EMS Operations in Tijuana, Mexico
A long-term project towards improving emergency medical services. UC San Diego - Global TIES Program. I started out as another student in a lab class but there were aspirations from the Red Cross in Tijuana (RCT) and the professor that were not yet attained. With just 13 ambulances servicing 2-3 million people in Tijuana at the time, one route towards improving the state of emergency medical services (EMS) was to modernize the technological infrastructure in the Red Cross. One past work determined an optimal set of ambulance standby stations that would maximize coverage in the city, which became partially implemented. In my first quarter, I prototyped and demonstrated live location tracking and historical pathing accumulation that would help keep track of the very limited number of ambulances in service. In the following two quarters, I gradually led more of the project starting with the mobile app subteam and peaking at managing the overall project and keeping in touch with 4-5 subteam leaders. We improved the RCT's internal data collection from traditional, manual methods like pen and paper to using computers and wrote web applications that allowed them to take advantage of expanding cell data availability (see [5]).
After my first 3 quarters, I attempted to tackle a larger problem that researchers have been working on: dynamically optimizing resource allocation (including but not limited to ambulance positions) to maintain city-wide coverage while accounting for additional contexts such as time of day and severity of emergency cases. Simulating a dynamic and chaotic world to produce estimations and outcomes is challenging. We showed that a reduction in ambulances (for example, suppose many ambulances happen to become busy at the same time because several emergencies occurred simultaneously) can drastically reduce coverage in Tijuana but it is possible to react accordingly to reallocate resources to reduce loss of coverage while maintaining a required level of quality of service (such as ambulance response times). However, in all simulations, a widespread disaster at almost any scale will plummet the ability to address emergency cases. This is due to the extremely constrained number of ambulances. These simulated findings can demonstrate a strong need to allocate more EMS resources for the RCT.
[5] Timothy Lam, Hans Yuan, and Maurício C de Oliveira. Low-Cost Open-Source Solution to Optimize Emergency Medical Services in Developing Communities by Tracking, Dispatching, and Simulating. In 2019 IEEE Global Humanitarian Technology Conference, Seattle, USA, October 2019.
[5a] Link to Paper
After I graduated, I remained involved for some time. We began to think about modernizing the Red Cross's operations at scale. We participated in innovation contests for funding and networking, wrote proposals and conducted market research to show why our idea is executing at the right time, why competing solutions were not as adoptable, and why ambulance dispatchers may use our solution.
[7] Big Ideas Competition - UC Berkeley - Global Health 1st Place
[8] Hans Yuan, Timothy Lam, Jose Rodriguez, and Maurício de Oliveira. "Revolutionizing EMS in Developing Communities" Clinton Global Initiative University (CGI U) Exchange, Chicago, USA, October 2018.
Student Collaboration in Computer Science Pedagogy
On collaboration, especially pair programming, as a pedagogy in university introduction computer science. UC San Diego - CSE Department.
[9] Hans Yuan and Paul Cao. 2019. Hybrid Pair Programming - A Promising Alternative to Standard Pair Programming. In Proceedings of the 50th ACM Technical Symposium on Computer Science Education (SIGCSE '19). ACM, New York, NY, USA, 1046-1052.
[10] Hans Yuan and Paul Cao, Collaborative Assessments in Computer Science Education: A Survey, in Tsinghua Science and Technology, vol. 24, no. 4, pp. 435-445, Aug. 2019.
Orchestra Club
When I was a freshman in high school I heard that the band director attempted to start an orchestra some years before, but couldn't or didn't. I had an idea to start the orchestra as a club instead of a class first. That summer, I approached the band director with my idea. I entered his office and introduced the past attempt and my proposal for how we can solve that problem in a more gradual way. He cut me off, gave me his business card, and said we'd start in the Fall. I walked out of his office stunned, wondering what just happened and whether I had just succeeded. In my sophomore year, we started a club which stabilized at 7 people. I tutored string techniques by relaying them from my own personal violin lessons. I played as the lone violinist in the school's musical play (and not that well). I joined the band student leaders' on their camping trip to leadership academies, joking that I was the new head marching violin. In my junior year, we continued the club which doubled in size to 15. We added one or two other string players in the musical.
Towards the end of my junior year, we convinced the school administration to use the district's definition of "String Orchestra P" that was recently created at Rosemead High for our new class. In my senior year, I took "String Orchestra P" to satisfy my own Fine Arts requirement in a class with more than 20 students, old and new. In all 3 years I was present, we fundraised by preparing food and selling at the school's regular food fair. The funds we raised went back into buying more material as we continued to fundraise more. We were later on able to buy new sheet music and eventually a double bass. Orchestra lives on:
AP Physics C
When I was an incoming senior in high school, I finished AP Physics B. My teacher had talked about wanting to start the calculus version of the class. However, the summer before the school year started, whether the class would be made was thrown into doubt. On campus, I texted her 5-6 SMS lengths of text messages saying why we should have this class, how it would be beneficial. I did not get any replies back until a few minutes later when I bumped into her, she seemed to be on her way here anyways. She said she could barely understand what I was saying because all the messages came out of order but she got the gist of it and was on her way to administration to petition the class.
We were very fortunate that she wanted to embark on this journey too. In AP Physics B, she set very high, lofty goals and succeeded in meeting those rigorous timelines in time for the AP exam.
Above all, she was a mother figure to every single one of her students, regardless of how well her students were doing. She met each person where they were and lifted them up from there.
We started a petition to collect signatures from students who want to take the class and had about 15 students enroll. We took the AP exam at the end of the year and had some students with only Physics B make an attempt at the Physics C exam.
The first time I encountered her, she was live demonstrating to her students outdoors how to make ice cream from scratch with liquid nitrogen. Those who can show the connection of science and maths to the real world while having fun are gifted and talented. Her early departure continues to be felt.
Other
2022
Computing in Julia Pluto notebooks by MIT and 3blue1brown.
2021
Unreal Engine, Blender.
Bioinformatics Algorithms. Active learning with paper textbook and online lectures, implementing algorithms in Julia, Python, or C++ using Rosalind. On Chapter 2.
Forecasting in Economics and Business. Time series, statistics. Reimplementing techniques in R and verifying with R packages. Finished Chapter 3.
Game Theory Online. Online lectures, simple problem sets. Finished Week 1.
Surface-level introductory machine learning course in Octave (Matlab). Completed.