Author: jay_sihm

As today was the last day of WEP, I was glad I could meet with and thank both Dr. Karmakar and Dr. Jiang for all their help. With Dr. Jiang, we explored further the mathematical side of optimization, as she explained linear and non-linear programming and multi-dimensional problems, along with showing me some of the data analytics done before optimization. Afterwards, I met with Dr. Karmakar as a final check-in on the whole experience. Although operations research and optimization were unfamiliar fields I was introduced to during WEP, I had a great time and am grateful to both Dr. Karmakar and Dr. Jiang for hosting and teaching me these past two weeks.

Linear programming example:

Today, Dr. Karmakar showed me various examples of data visualization to explain how interactive visualization helps in identifying patterns, discovering new trends, and understanding complex concepts. She started with a simple report on animal shelter employees, showing the variety of ways they can present information based on the client’s requests. Then, she brought up a more complex example on water bottle supply chain optimization and explained a couple of the response variables they consider, such as costs and KPI. She wrapped up by showing me some code used to create the data visualizations in SAS Studio. After asking a few questions about the data visualization process, Dr. Karmakar had me come up with some more questions on optimization for tomorrow.

Animal Shelter:

Supply Chain:

Code:

On day six, Dr. Karmakar was on campus, so she gave me a tour of her building. After introducing me to a few of her coworkers, she showed me the library, some workspaces, the cafeteria, and the outdoor area. Afterwards, I shared with her my literature review and, in preparation for the next day, she asked me to research a bit on data visualization and to create a Linkedin account.

Coworkers:

R Building:

On the fifth day, Dr. Karmakar had some conflicts and wasn’t feeling the best, so we decided to take a more relaxed day without meeting. I spent the day editing and finalizing my literature review, and by Dr. Karmakar’s recommendation, added a literature summary table to supplement my review.

Literature summary table:

On the fourth day of my WEP, I started with a meeting with Dr. Karmakar and Dr. Jiang on the studies I found on cell culture mediums to see if any would be viable as a demo sample dataset. After discussing what could be possible in their allotted time, Dr. Karmakar asked me to research how to organize, and then write a comprehensive literature review on my research throughout the week to present to her team.

Literature review:

Today, I researched studies in regenerative medicine, namely in cell culture media and bioreactors, to find a dataset not reliant on time series as Dr. Karmakar asked yesterday. As some unexpected meetings and conflicts came up, we didn’t have a video call as planned, but we kept up a correspondence via Teams on my progress and questions.

Some research:

   

Today, I met up with Dr. Karmakar in the morning and gave a summary of my work yesterday. After a couple of notes, she assigned my next task: finding a dataset from one of the biomanufacturing processes I found that had numerous predictor variables and recreatable data. In the afternoon, I had another meeting with Dr. Karmakar, this time with her colleague Dr. Wendy Jiang. Dr. Jiang introduced me to one of their department’s current projects, creating an experiment to increase penicillin concentration, and showcased her current dataset and graphs using arbitrary numbers. Then, as she has a conflict tomorrow morning, Dr. Karmakar asked me to look for a dataset in regenerative medicine not dependent on a time series for tomorrow.

Dr. Jiang’s dataset and graphs:

During the next two weeks, I will be following senior operations research specialist Dr. Nabaruna Karmakar at SAS. Additionally, SAS is also a part of the Data4good movement, which applies data analytics to humanitarian movements. I’m looking forward to the upcoming weeks to explore how a company like SAS uses data analytics for both internal business needs, along with broader social ones.