Syngenta, a Switzerland-based Agribusiness company, is a leading developer of crop varieties and seeds. Syngenta’s R&D center is working on methods and processes to increase the quality and quantity of soybean crops. It is estimated that the global population will increase to 9.6 billion by 2050, and crop production needs to rise to meet its needs. Syngenta’s soybean R&D team uses advanced analytics and breeding principles to develop a plant variety that yields higher quantities. The analytics group has developed a step-by-step procedure to produce the finest varieties of soybean. Decisions related to cost, time, and success in developing better-quality crops are taken during the first three stages of this process. Decisions on commercializing the produced variety of soybean are taken in the last two stages. Syngenta developed four tools based on simulation models to help develop a superior soybean variety: • The Trait Introgression Tool (TI) This tool uses discrete-event and Monte Carlo simulation models to model the biology of mating two varieties to create progeny as well as the success of a specific traits’ transfer to the next generation of the variety. The Monte Carlo simulation allows the planning of each step and examining the effects of the plan based on cumulative cost and time. If the cost, time, and probability of transferring the planned traits is unacceptable, the project lead can change one or more parameters, rerun the simulation, and view the results on a dashboard. Thus, the TI tool models genetically distinct varieties of soybean and helps project leads in selecting plant varieties for self-pollination and cross-pollination. • The Breeding Project Lead Tool (BPL) The BPL tool is also based on a discrete-event simulation platform. The user defines not only the genetic composition and relative maturity of each parent variety but also the conditions related to actually running breeding experiments: trail locations, government restrictions, etc. Its results help in defining the time, cost, and success of developing a particular commercial variety. This helps in determining the optimal use of available facilities, fields, and plant materials. • Yield Trial Design (YTD) Optimizer The YTD optimizer integrates an Excel spreadsheet with the simulation model to permit input of a large number of plant varieties and develop results for each one. It evaluates the solution designs through a large random sample of feasible solutions. The model takes historical data to fit the probability distributions needed for the simulation model. A simulation run generates a number of varieties with a mean and variance. The varieties with a higher mean and lower variance are selected as superior varieties for the next trial stage. Benefits The TI tool and YTD optimizer have helped develop and simulate experimental designs prior to initiating any actual field trials. The TI tool simulates various plant varieties and estimates the cost and time for developing each design. The BPL tool improves decision-making by allowing a project lead to evaluate the results of altering any decision in the development of a plant variety. This shows Syngenta the highest probability of developing a new variety within the cost and time constraints, and it has been estimated that these tools saved $287 million across various seeds development efforts from 2012 to 2016.
Questions for Discussion
1. Describe the operation research tools developed by Syngenta to improve the quality and quantity of soybean production.
2. How do the Trait Ingression and Breeding Project Lead tools use discrete-event simulation to select the best plant variety?
3. List some of the benefits to Sygenta of using simulations in OR tools.