Validating algorithms designed for fertilization control in rice farming system

Abstract Rice yields in Rwanda seldom reach values expected for irrigated production, part due to poor fertilization management. To adapt well-known best practices local conditions, two fuzzy algorithms were designed conditions of rice production Rwanda. Fuzzy have a linguistic rule base 183 IF THEN statements linking measurable field crop yield. Interviews with government agricultural experts and published knowledge site used create the which incorporates known nutrient requirements different growth stages rice. are intended use an Internet Things (IoT) system. validate algorithms, historical weather data drive yield simulations plots during first growing season 2020 at sites northeast Predicted compared measured scenarios irrigation levels amounts without full connectivity. Insights provided on potential these methods improve management increased low-income countries discussed.