Universidad Pablo de Olavide, de Sevilla


UPO and ec2ce apply artificial intelligence to predict the evolution of the olive fly plague

This predictive model allows an analysis with four weeks in advance, avoiding the explosion of the pest and minimizing the use of pesticides

Olive tree fields

Olive tree fields

Researchers at the Pablo de Olavide University, with the coordination of Professor Francisco Martínez Álvarez, participate in a project by ec2ce which aims to develop a predictive model to control a pest of olives (Bactrocera oleae or fruit fly). This study allows anticipating decisions which favors a controlled application of phytosanitary products and it improves the sustainability of the olive wood.

The project, ‘IA2GIP: Applied Artificial Intelligence for integrated management of plagues’ has been financed by the Ministry of Economy, Industry and Competitiveness, within the State Plan for Scientific and Technical Research and Innovation 2013-2016. Also others entities participate such as Novadrone, Andalusian Institute of Technology (IAT) and the University of Seville.

The main innovation of the project is the integration of mathematical modeling, artificial intelligence and sensors to provide recommendations to farmers. It requires the development and combination of various technologies and the use of deep learning, genetic algorithms, neural networks, fuzzy logic and techniques of optimization which allow the use of huge amounts of information in real time. This study combines the use of large data volumes (big data) with artificial intelligence systems to create a tool which supports the decisions about pest control systems. This investigation will be based on public and private data, and field data taken in a traditional way and others taken by unmanned air systems.

Francisco Martínez Álvarez

Francisco Martínez Álvarez

Olive fly is the main plague affecting olive oil producers, generating losses of up to 80% of the value of the harvest, due to production decrease and oil quality. This insect attacks directly the fruit and the pest evolves in an explosive way making it difficult to control and assuming very important costs for producers.

The use of predictive models allows to determinate the explosion of the affection of the fly and it would make it possible to apply control systems in time. This tool will be easy to use by farmers and they will be able to improve the production, increasing the quality of oil and reducing the use of pesticides, protecting the environment and practicing sustainable agriculture.

 


Archivado en: , , , , , ,