Report on INFORMS O.R. & Analytics Student Team Competition, Las Vegas, 1-5 April 2017

About the organizer

The Institute for Operations Research and the Management Sciences (INFORMS) is the largest society in the world for professionals in the field of operations research (O.R.), management science, and analytics. The 2017 conference, focusing on analytics and O.R., will bring together nearly 1,000 leading analytics professionals and industry experts to share ideas, network and learn about a range of current topics.

 

About the competition and conference

This international competition was sponsored by Syngenta, a global biotechnology company produces agrochemicals and seeds. We were helping Syngenta to decide among thousands of soybean seed varieties, which are the good ones to commercialize. Data we have is the historical experiment results of seeds, and historical sales. We first transformed the business problem into an analytics one, conducted some feature engineering based on the understanding of data and context, and built machine learning model to predict sales of future varieties. 

We passed the first round of selection and became one of the eight finalists. We were invited to present at a conference, the 2017 INFORMS Conference on Business Analytics & Operation Research in Las Vegas. Judging panels and other participants of the conference listened to our presentation and asked several questions. We were then accessed based on performance at the presentation and quality of report write-up. Our team won the second price and were awarded during a lunch ceremony open to the conference participants.

 

What we learnt

First of all, we gained plenty analytics-related knowledge and experience, which will be definitely useful during future career. We also found out how useful it is to make a few jokes during presentation, especially to U.S. audience. 

 

From participants around the globe, we realized there can be so many different approaches to the same problem. Based on school project experiences, we expected that different team may differ in terms of variable choices, feature engineering and statistical model selection. Apart from that, there are much more areas where different teams take distinct approaches. For example, the task is to predict sales volume for each soya bean variety. Some of the teams set accuracy as the absolute priority; while we set our goal as achieving a balance between prediction accuracy and model interpretability.

Moreover, it is great to know that analytics is being acknowledged and applied in more and more areas, including more traditional industries like agriculture. Seeing people's passion in the competition and the conference, I have more confidence in the potential of analytics, especially applied in different aspects of operational research.

© Copyright 2001-2012 National University of Singapore. All Rights Reserved.

This page is dynamically generated by the Drupal CMS.