Performance Management in Agtechs: a predictive approach based on machine learning and grey system theory

• Grant number: 24/03042-0
• Support Opportunities: Scholarships in Brazil - Master
• Start date: September 01, 2024
• Status: Discontinued
• Field of knowledge: Engineering - Production Engineering - Production Management
• Principal Investigator: Lucas Gabriel Zanon
• Grantee: Elvio José Sêneda Filho
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Associated scholarship(s): 25/02843-2 - MACHINE LEARNING IN AGRICULTURE: A FARMER-CENTERED DECISION SUPPORT MODEL BASED ON EXPLAINABLE AI (XAI)., BE.EP.MS

Abstract

The increase of global population has as a consequence a higher demand on food, which stimulates the search for more efficient and sustainable solutions in the agricultural sector. In this context, the term Agtech indicates a set of solutions of digital technologies applied in agriculture. Also, it refers to agribusiness sector startups based on 4.0 technologies, such as Internet of Things (IoT). Considering performance management as a crucial topic to ensure operational efficiency in the agricultural sector, this research project aims developing a predictive model based on the association between machine learning (ML) and grey system theory (GST) approaches. The final objective is to support agricultural performance management in Agtechs through monitoring of productivity and risks. As the databases collected through IoT may present incompleteness, the grey system theory is intended to deal with it. This application is not found in the literature and may contribute to the development of the theme in academic and business environments. The project is structured in four stages: literature review; theoretical model development; computational model development and application in a partner Agtech; results analysis. It is expected supporting technological agricultural operations, encouraging the incorporation of ML approaches in performance management process, mainly in incompleteness databases scenarios. Thus, it is also expected encouraging more efficient agricultural practices to address the global question about food production.