IoFarm: A novel decision support system to reduce fertilizer expenditures at the farm level

Farmers and consultants make the best use of existing resources to increase profitability, and this begins at the operating resources level. Existing literature primarily deals with the optimal intensity of fertilization. A multitude of different fertilizers, with fluctuating price ratios, however, promises an additional economic optimization potential that can be achieved by cleverly combining these fertilizers to form a least-cost combination. This potential is extended if individual costs of fertilizer application are considered in parallel. The objective of the IoFarm decision support system is to take up this overall complex and to determine individual fertilizer strategies for farms, which also determines the economic optimization potential. In terms of methodology, the solution to this problem is based on mixed integer nonlinear programming (MINLP). The restrictions and parameters used for this were taken from the literature and where necessary were derived in a simplified form in order to keep usability a practical tool. The subsequent evaluation of the economic performance was carried out by conducting an experiment. The participants were asked to define a fertilizer strategy for a simplified farm with three fields and three crops over three years. Despite the considerable amount of time it took the testers to conduct this investigation, IoFarm performed 19% better in terms of costs. Our results show surprisingly clearly the complexity of this decision-making process and the previously unused monetary and time optimization potential behind it. The agronomic performance of IoFarm has already been confirmed in another study.