K-means clustering algorithm approach for the segmentation of organic banana producers according to production ratio

Abstract

In Peru, the exportation of organic bananas has increased considerably in the last ten years, the arrival of the “boom” of organic bananas is imminent, therefore it is necessary better management of small producers and associations for this organic resource, providing them with new technologies that help them to better organize, develop and monitor production. In the present study, a segmentation algorithm based on k-means has been developed, through which it was possible to group the weekly production of an organic banana packhouse, formulated on its production ratio. This project has been carried out to facilitate the monitoring of the production of each sector of the agricultural cooperative APBOSMAM (Asociación de Productores de Banano Orgánico Sector el Monte y Anexos Mallaritos), which provided production data. In the development of this study, two production ratios have been considered for the analysis: “Plant population per hectare ratio” and “Pre-bagged bunches ratio” by each sector during a week. From formed groups, it has been possible to identify the sectors with the lowest production and the possible factors causing this low contribution at the time of exporting the fruit.