A data‐driven, machine learning framework for optimal pest management in cotton

Abstract

Citation: Meisner, M. H., J. A. Rosenheim, and I. Tagkopoulos. 2016. A data-driven, machine learning framework for optimal pest management in cotton. Ecosphere 7(3):e01263. 10.1002/ecs2.1263Abstract. Despite the significant effects of agricultural pest management on crop yield, profit, environmental quality, and sustainability, farmers oftentimes lack data-driven decision support to help optimize pest management strategies. To address this need, we curated a comprehensive data set that consists of pest, pest management, and yield information from 1498 commercial cotton crops in California's San Joaquin Valley between 1997 and 2008. Using this data set, we built a Markov decision process model to identify the optimal management policy of a key cotton pest, Lygus hesperus, that balances the tradeoff between yield loss and the cost of pesticide applications. Our results show that pesticide applications targeting L. hesperus are only economically optimal during the first 2 weeks of June, and pesticide applications were associated with increased risk of an unprofitable harvest. About 46% of the observations in our data set involved at least one pesticide application outside of this optimal window, demonstrating the need for a data-driven approach to crop management. Sensitivity analyses on parameter perturbations and reduced data set sizes suggest that our methodology provides a robust policy-making tool, even in noisy data sets.

Review of Artificial Intelligence Applied in Decision-Making Processes in Agricultural Public Policy

“…Bayesian networks have been applied to determine water use and the development of sustainable agriculture, as presented in the research by [126], carried out in the Apulia region of southern Italy. In the San Joaquin Valley in California, United States, a study was carried out to generate an optimal management policy for pest management in cotton crops, which balances the loss of yield and the cost of pesticide applications [127]. This application employs a Markov decision process model.…”

A Triangular Similarity Measure for Case Retrieval in CBR and Its Application to an Agricultural Decision Support System

“…The occurrence of pests has a close relationship with several factors, such as crop status, soil conditions, and meteorological information [58]. When managing pest problems, it is necessary to take pest, crop, and environment data into account.…”

Volatile Organic Compound-Based Predictive Modeling of Smoke Taint in Wine

“…Data-driven approaches, especially machine learning, can accelerate the discovery of the VOCs related to smoke taint and the predictive modeling of the smoke taint index. In recent years, machine learning algorithms as a part of the Artificial Intelligence (AI) have increasingly been applied in food science and agriculture for a sustainable food system, including predicting micronutrients, − creating food ontologies and knowledge bases, precision agriculture, and crop and animal management . Although VOCs in smoke-affected grapes and wine have been reported, the levels contributing to the smoke taint effect of VOCs have been evaluated, and a few studies that model the smoke flavor based on chemical composition have been published recently, the number of studies focusing on data-driven approaches, especially predictive modeling of smoke taint based on VOC concentrations, are still limited.…”