Texas A&M teams place first, second in inaugural Dow Big Data Challenge

April 26 marked the culmination of Texas A&M University’s INFORMS Student Chapter inaugural Dow Big Data Challenge. 

Supported by The Dow Chemical Co., the Dow Big Data Challenge pitted a total of 27 teams against each other in a three-week race to creatively dissect and analyze two years of real-world Dow data.

Entries for the inaugural Dow Big Data Challenge consisted of teams of students from the Department of Industrial and Systems Engineering  at Texas A&M University and from the University at Buffalo. The challenge brought leadership from both Texas A&M and Buffalo INFORMS Student Chapters together to coordinate all aspects of the competition from entry to award. 

The 2013 Dow Big Data Challenge winners are:

  • First place: Texas A&M University's Shilan Jin, Yanjun Qian and Su Zhao 
  • Second place: Texas A&M University's Gokhan Memisoglu, Abhilasha Katariya and Samyukta Sethuraman
  • Third place: University at Buffalo's Georgia Cruz, Michael Moskal, Geoff Gross and John Coles

In total, 67 graduate students and 21 undergraduates entered the competition to assess rail car demand and transportation data for Dow’s North American rail car fleet. Students were encouraged to use their knowledge of optimization, data analytics, computer science, simulation, statistics and experimental design to uniquely identify ways to improve Dow’s rail fleet management.

Of the 27 original teams, nine submitted final reports and presentations to faculty judges from both universities; the judges then selected five team finalists. Finalists then presented directly to three Dow representatives on the Texas A&M campus, with the University at Buffalo attending and presenting live via video conference.  Prizes were awarded to the first, second and third place teams whose creativity and innovation in their data analysis best captured Dow’s interest and attention.

As the inaugural Dow Big Data Challenge Sponsor, Dow contributed financial support for prizes awarded to the top three teams in addition to supplying the challenge’s data. 

“We are extremely excited to partner in the challenge and were very impressed with the submissions received,” said Dow’s Stew Serpas, who worked closely with Dow's Kevin Hunter and Doug Sturgis throughout the inaugural challenge. “This type of innovative, integrated collaboration with academia adds tremendous value not just to the rail fleet management team, but to Dow as a whole."

The Texas A&M University organization team was lead by current INFORMS Student Chapter president Michelle Alvarado, with  support from president elect Gokhan Memisoglu and vice president of communications Shachi Pande. The student chapter also received help from faculty mentor Dr. Justin Yates.

John Coles, challenge participant and former University at Buffalo INFORMS Student Chapter president, said, “This has been an incredible opportunity for our students to work on a challenging and realistic problem. It is great practice for us as we prepare for careers in a data-rich world.” 

The University at Buffalo organization team for the Dow Big Data Challenge included Coles, current student chapter president Geoff Gross, president-elect Michael Moskal and faculty adviser Jun Zhuang.

The Dow Big Data Challenge posed many unique communication challenges as students from both universities collaborated to make this competition a reality. Online conference calls, video chats, and presentations were a regular part of the endeavor. Texas A&M’s information technology team (Dennis Allen, Mark Hopcus and Mark Henry) played a critical role in overcoming communication barriers and ensuring that the best technology was available during student presentations.

Plans for the 2014 Dow Big Data Challenge are already in discussion with all parties set to reprise their roles in year two.  The Texas A&M University and University at Buffalo INFORMS Student Chapters anticipate a stronger second year and will use their first-year experiences to help pave the way for other student chapters to either participate in the Dow Big Data Challenge or to develop their own. 

Yates said, “This is not a one-time challenge and our goal in this has always been to increase the uniqueness and diversity of our students’ experiences outside of the classroom. Our hope is that our template can be used to increase the availability of these opportunities to students nationally and that we show how the creativity and hard work of our student leaders transforms and distinguishes their organizations.”