Smart agriculture and digital twins: Applications and challenges in a vision of sustainability

• Published in: European journal of agronomy 2023-05, Vol.146, p.126809, Article 126809
• Main Authors: Cesco, Stefano, Sambo, Paolo, Borin, Maurizio, Basso, Bruno, Orzes, Guido, Mazzetto, Fabrizio
• Format: Article
• Language: English
• Subjects: Agricultural Intelligence; Decision Agriculture; In-silico simulation; Variable Rate N Fertilizer; Yield Stability
• ISSN: 1161-0301; 1873-7331
• DOI: 10.1016/j.eja.2023.126809

Smart agriculture – i.e., the increasing use of information technologies, sensors, autonomous vehicles, data analytics, predictive modelling, and other digital technologies related to agricultural activities – has been strongly argued for as a means to significantly contribute to increased food security, reduced water consumption, reduced fertilizer and pesticide input, and increased farm profitability. Despite this, the adoption rate of smart agricultural technologies is still low and varies significantly according to the specific technology and the geographical area considered. The goals of this paper are to: (1) propose a conceptual framework for smart agriculture and digital twins, which takes into account the needs and characteristics of the farms; (2) present the application of the proposed conceptual framework as a case study; and (3) shed light on the challenges of and the future perspectives on smart agriculture. We first propose a framework for the design of farm information systems consisting of four key phases (i.e., data collection, data processing, data analysis and evaluation, and information use) based on the infological approach. We then apply the framework to present and discuss a field application of smart agriculture and digital twins on crop nitrogen (N) fertilization. The case study, along with the cited literature, highlights the need to specify the optimal N fertilizer input as well as defining the spatial variability of the land area, the soil characteristics and crop yield, and the integration of these with temporal variability. Finally, we discuss challenges and future perspectives, with particular focus on geographical areas characterized by small average farm size. We argue that, thanks to digital twins, the wide set of data collected can enable predictive (and stability) analyses that if implemented can benefit the farmer and the environmental, social, and economic sustainability of the agricultural system. •We propose a conceptual framework for smart agriculture and digital twin.•We apply the proposed conceptual framework to a case study of N fertilization management.•We shed light on challenges and future perspectives on smart agriculture.