1Department of Computing Technologies, SRM Institute of Science & Technology, Chennai, Tamil Nadu, India
2Department of Chemistry, Indian Institute of Technology Bombay, Mumbai, Maharashtra, India
*Corresponding author: up0625@srmist.edu.in
Adoption of the use of artificial intelligence (AI) in agricultural disease control has proven to be efficient in terms of technical improvement but economic justification has proved to be a foundation pillar in implementation strategies. This is a review article, which summarizes the existing approaches to measuring the financial changes brought about by AI-based crop disease detection and mitigation systems. The authors focus on the schemes of yield losses estimation, methods to value economic types of returns, and modelling, methods of return on investment (ROI) evaluation in varying agricultural application situations. By systematically reviewing the current body of such literature and case studies, this review confirms in part that AI- based early disease detection can yield losses of 15-40% of the end yield dependent on type of crop and time of intervention with the same investment returning on investment in the range of 150, 400% on a 3-year deployment period. But even there, there are major discrepancies in methodology in the area of impact assessment, especially in setting up of baseline, modelling attribution and long-term measures of sustainability. The current paper suggests the system of integrating the disease incidence modelling with the market-responsive economic valuation with the incorporation of AI model performance measures and the measures of the regional variability. Important results suggest that the financial feasibility is highly sensitive to other factors such as size of the fields, diseases and predominance, prices of crops in the market and the availability of infrastructure. The review reveals severe gaps in the research on applicability of smallholder applications, climate adaptation Case scenarios, and policy mechanisms used to enhance faster economic implementation of AI in crop protection. The suggested framework not only provides a methodology that a researcher and practitioner can reproduce to assess AI interventions, but it also emphasizes the one that requires standard measures of economic impacts in agricultural AI studies.
Artificial intelligence, Crop disease detection, Economic impact assessment, Yield loss modeling, Return on investment, Precision agriculture, Disease mitigation, Agricultural economics, AI adoption, Financial sustainability
