The rapid growth of industrial infrastructure creates ecological issues such as climate change. Field indecisiveness affects agricultural yields due to improper measurement, field assessment, selection of sensors and deployment of sensors. The accurate prediction of changes in weather parameters, field assessment and soil parameters, has become an outstanding challenge for the agricultural IoT. To solve this problem, we propose a (t, n) sensor selection mechanism and a soil temperature, humidity, air- and water-quality measurement (THAM) index for node stipulation, based on a smart decision-making system for the agricultural domain that considers the temperature quotient, an NPK fertilizer regulatory model and the agronomy function. The (t, n) node stipulation index defines an optimal number of sensors to monitor the field. The temperature quotient considers soil temperature and moisture to assess the growth rate. The agronomy function, based on water pH level and SO2 concentration level in air, assesses the production yield rate of the field. This framework improves the prediction performance for detecting abnormal conditions by 75%, with a reduction in the creation of unimportant data and the resources loss rate. It increases the agricultural production yield compared to existing systems. © 2019, Springer Science+Business Media, LLC, part of Springer Nature.