METSK-HD-Angeleno: How to predict fruit quality using Multiobjective Evolutionary learning of TSK systems

The United Nations Food and Agriculture Organization (FAO) ranks Spain 13th in the world in plum and sloes production and 4th in the member states of the European Union. Cultivation of this fruit is clearly important in our country, and is even more so in Extremadura, a Mediterranean region in south-western Spain, which focus their economic activity on the primary sector. A plum production must be differentiated by its quality, but the quality of the fruit is traditionally perceived by the experience of farmers and technicians, based solely on their visual perception. This traditional decision-making process sometimes leads to errors in determining the optimum date for harvesting.Among the quality parameters used by the food industry are the soluble solids content and/or the firmness. These parameters will measure the fruit quality, allowing to obtain the best quality of the fruit when the optimal values are reached. The parameters must be calculated using destructive techniques and sophisticated laboratory equipment. In this work we present a new method to predict the soluble solids content or the firmness of a fruit, by means of software techniques that do not require a destruction fruit process and expensive laboratory equipment. The results presented in this work allow us to affirm that it is possible to provide farmers and agricultural technicians with software tools that help them make the right decision regarding the ripening of their fruit, in order to obtain the highest quality products and be more competitive in the sector.