Rutgers Business School students leverage data modeling and machine learning to win case competition

Using lessons they learned in the classroom, the students spent 24 hours developing a solution that impressed the judges with its capabilities. They competed against 32 other teams to win the $5,000 top prize.

NEW BRUNSWICK, N.J., March 11, 2026 /PRNewswire/ -- A team of Rutgers Business School graduate students won first place in the University of South Carolina's Big Data Health Science Case Competition after designing a tool to help doctors determine treatment options for patients undergoing musculoskeletal care.

Bhargavi Varanasi, a medical doctor from India pursuing her a Master of Health Care Analytics joined Debanshu Poddar, Anjaney Srinivas and Vivek Chakraborty, classmates in the Master of Supply Chain Analytics Program, to form the Rutgers team. The students were advised by Rutgers Business School Professor David Dreyfus, director of the Health Care Analytics Program.

Find out more about the specialty master's programs offered at Rutgers Business School during a Graduate Open House on Saturday, March 28, from 9 a.m. until 1 p.m. in Newark. Register today.

"Building this in 24 hours was a blur of whiteboarding and debugging, but it pushed us to grow," the team members said in an email. "The true value was learning to translate 'black box' math into a transparent, trustworthy tool. Realizing there's a real person making a life-altering choice behind every data point made the sleepless night completely worth it."

An intense competition

The two-day, online competition attracted 32 teams from across the country. It required the students to data clean 100,000 pieces of patient information, deciding what was important and what was unnecessary. 

"Understanding the data was the main thing," Chakraborty said. Sorting through the data was time-consuming, and they were processing so much data, he said, that their computers repeatedly froze.

And that was just the beginning. They used their modeling knowledge to determine predictive values for different outcomes, including patient recovery, rehabilitation time and cost of surgery or care. They also had to demonstrate their programming, analytical, collaboration and communication skills to develop and explain their tool to the judges.

While Varanasi helped to identify the important data – what would be most relevant to determine a patient's treatment and recovery time – Srinivas went over the case again and again to ensure the team understood the problem and was approaching it correctly. Vivek did research to find data that they felt was missing, and Poddar crafted the presentation.

First among seven finalists

As they prepared for the final round, they had to do more than only refine the work they had presented previously. They were required to integrate more information about patient treatment preferences into their solution.

The Rutgers Business School students won the final round against teams from Carnegie Mellon, Dartmouth College, University of South Carolina, University of Iowa, and Florida State and Missouri State universities.

Students from the University of South Carolina won second place, and third place went to Missouri State University's team. The Rutgers team will share the first-place prize of $5,000.

"This victory reflects not only the team's technical excellence and analytic depth, but also their ability to translate data science into meaningful healthcare solutions," Dreyfus said after the team's win. "Their shared decision-making model demonstrates the very best of interdisciplinary collaboration between healthcare and supply chain analytics."

The competition reflected a growing real-world demand in health care for tools that will allow physicians to access quantitative data to support their care decisions and better explain to patients such things as the would benefit from surgery or physical therapy and how long their recovery will take. 

The team wasn't told why they won the competition, but Srinivas theorized that it had to do with the quality of their final tool. "We were quite robust with our solution," he said. 

SOURCE Rutgers Business School