Center for Digital Agriculture

Project Team

  • Dr. Luis Rodriguez, Associate Professor, Dept. of Agricultural and Biological Engineering
  • Dr. Richard Sowers, Professor, Dept. of Industrial and Enterprise Systems Engineering
  • Jay Solomon, Environmental and Energy Stewardship Educator, University of Illinois Extension

Abstract

Farm operations are inherently vexed by significant amounts of risk. Consider commodity grain operations in the Midwest which are saddled with large amounts of financial leveraging in increasingly uncertain environments. These factors can readily drive decision makers to sub-optimal event-based decisions, which seek to manage risk, but suffer not only increased costs and lost production, but also significant direct and indirect waste generation throughout the value chain. Previous works, considering individual farms, demonstrate that there are significant differences between what is possible and what operations are likely to be occurring in practice (Liao, 2017; Lin et al., 2019). This becomes particularly worrisome if we consider diverse farm systems where assets are managed across multiple distributed parcels, with combinations of manual labor and mechanization. Our long-term goal is to develop systematic approaches for event-based decision making and asset management applicable to an array of crops and farming systems. Our current hypothesis is that by improving event-based farm management we can improve productivity by 5% on individual grain farms typical of the Midwest. Our corollary hypothesis is that these effects compound as farm managers deal with increasingly distributed systems. Further, we would assert that analogous approaches will be applicable to other agricultural systems (e.g. fruits and vegetables), where gains are likely to be greater because manual labor prevails and current loss rates are even more dire. Our project has several key objectives: within the first year, we seek to develop and transfer previous work in this area into a computational system which leverages available data to facilitate agile event-based decision making. With this foundation, we would then seek to validate predictions and expand our logic to consider diverse farm systems including both mechanized assets, manual labor forces, and networks of distributed farm parcels across regions. Future studies could also be expanded to consider the impacts on related value chains spanning food, energy, and water; minimization of food related wastes; cold-chain management; and the promotion of food and agricultural sustainability.