This fall I had the pleasure of hosting two winners of the INFORMS Undergraduate Operations Research Prize in my Prescriptive Analytics course at Bucknell University (ANOP 370). These talks are now available online.
Kayla Cummings (2018 winner) talked about vaccine pricing: https://mediaspace.bucknell.edu/media/CDC+as+a+Strategic+Agent+in+Public+Sector+Vaccine+Pricing+-+Kayla+Cummings%2C+MIT%2C+09+30+2021/1_pmu631vu/185503823
Wes Gurnee (2020 winner) talked about fairness-optimized political districts: https://mediaspace.bucknell.edu/media/Fairmandering+Generating+Fairness+optimized+Political+Districts+-+Wes+Gurnee%2C+MIT%2C+11+11+2021/1_yu6gcqsm/185503823
Cummings has since extended her work to also analyze the pricing of COVID-19 vaccines.
I found very interesting when she mentioned that the hardest part of this extension was finding the right data, which took them 6 months but only shows up in the appendix of the paper.
You can find more comments on her talk in the thread that I posted on Twitter during her presentation:
Gurnee emphasized the benefits of decompositions in optimization and of leveraging symmetries in how districts are partitioned in order to make the algorithm scalable.
I also liked the comments about how challenging it is to sell ideas like this to organizations.
You can find more comments on his talk in the thread that I posted on Twitter during his presentation:
Both of them are currently pursuing their PhDs at the Massachusetts Institute of Technology’s Operations Research Center.
Whoever hire them when they finish will be very lucky: it was a pleasure to learn about their work when judging the award, and a blast to bring them to talk about it now!
Thiago Serra 