A mathematical strategy for predicting farmers’ soybean seed variety demand took home the grand prize in the second annual Syngenta Crop Challenge in Analytics.
The BioSense Institute in Serbia team, which included Oskar Marko, Sanja Brdar, Marko Panić, Isidora Šašić, Milivoje Knežević, Danica Despotović, Vladmir Crnojević, and Zorana Djindjica was awarded a $5,000 prize for its entry, “Portfolio Optimization for Seed Selection in Diverse Weather Scenarios.”
“The overall quality of submissions was at exactly the level of analytical and mathematical thinking we are looking to bring to the agriculture space,” said Joseph Byrum, Ph.D., MBA, PMP, senior R&D strategic marketing executive with Syngenta and Syngenta lead for the Crop Challenge committee. “Very little separated all the finalist submissions—but there was excellent clarity in the logic of the BioSense team’s submission. They put a great deal of thought and contemplation into a very complex problem, and then solved it systematically.”
“It is an exceptional honor to win first prize—all the teams were exceptionally good and all had different approaches,” said Oskar Marko, team lead for the winning submission. “We feel very fortunate that our approach proved to be the most effective.”
Marko’s team from the BioSense Institute, a multi-disciplinary research organization affiliated with the University of Novi Sad in Serbia, received fourth place in last year’s Crop Challenge in Analytics. After the competition, the team used material it developed in that Challenge submission to help its university receive agriculture research funding from Horizon 2020, a European Commission-led program that fosters research and innovation in the sciences.
This year’s Challenge tasked participants to use data analytics to predict which soybean seed variety—or assortment of varieties—is most likely to be chosen by farmers within a specific growing region. Since seed variety selection is one of the most important decisions farmers make each season and no two seasons are alike, data driven models are increasingly being deployed in making seed decisions.
The finalists made their presentations on April 3, 2017, at the INFORMS Business Analytics Conference in Las Vegas, Nevada. Programs were evaluated based on the rigor and validity of the process used to determine optimal seed varieties, the quality of the proposed solution, and the finalists’ ability to clearly articulate their solution and its methodology.
The runner up submission, “A Decision Making Approach for Soy Seed Variety Selection via Hedging Against Weather Risk,” authored by Zhongshun Shi, Yu Zhao, Xi Zhang, and Leyuan Shi from Peking University, China, received a $2,500 prize; and the third place entry, “A Hierarchical-Ensemble of Machineries to Optimize the Choice of Soybean Varieties,” authored by Durai Sundaramoorthi, Lingxiu Dong, Iva Rashkova, and Piruthiviraj Sivaraj from Washington University in St. Louis, Missouri, received a $1,000 prize.
Details regarding the 2018 Crop Challenge will be announced next month, with submissions due January 2018.
Moving Agriculture Forward
The AGDAILY Digest is the information superhighway for your country road.