Syngenta Crop Challenge finalists to use analytics for farm seed selection

The Syngenta Crop Challenge finalists must use advanced analytics for developing a model for farmers to decide which seed varieties to use next planting season.
The Syngenta Crop Challenge finalists must use advanced analytics for developing a model for farmers to decide which seed varieties to use next planting season.

The Syngenta Crop Challenge finalists, who will use advanced analytics for developing a model for farmers to decide which seed varieties to use next planting season, were selected at the end of February by the Institute for Operations Research and the Management Sciences (INFORMS).

“Knowing the world is grappling for new ideas to help alleviate hunger challenges, this competition focuses specifically on using analytics to address that issue” Joseph Byrum, Syngenta head of soybean seeds product development and lead for the Syngenta Crop Challenge committee, said. “Syngenta is excited to see the finalists’ presentations and learn how the teams propose making crops more efficient for farmers across the U.S. and the world. It’s a great opportunity to illustrate the value that analytics brings to increasing efficiency and productivity,”

The finalists all authored models based on different subjects including balancing weather risk and crop yield for soybean variety selection by Bhupesh Shetty, Ling Tong and Samuel Burer; soy variety selection to maximize yield and minimize risk based on neural network prediction and portfolio theory, authored by Yu Zhao, Jingsi Huang and Ming Qin; the best soybean varieties for hedging risk of weather uncertainties - a deep learning and heuristic optimization approach, authored by Mark Rees, Yidong Peng, Jaremy Babila, Mike Lyons, Lily Huang, Yinghan Song, Chun-Yang Wei and Susan Arnot; soybean varieties portfolio optimization based on yield prediction using weighted histograms by Oskar Marko, Sanja Brdar, Marko Panic and Predrag Lugonja; the model for a decision assist tool to choose the best seed variety for best yield on known soil and uncertain future weather conditions by Nataraju Vusirikala, Mehul Bansal and Prathap Siva; and hierarchy modeling of soybean variety yield and decision making for future planting plan by Xiaocheng Li, Hyauyang Zhong, David Lobell and Stefano Ermon.

Each group of finalists will make a presentation during the INFORMS Analytics Conference in Orlando on April 12.