REVISTA DYNA NEW TECHNOLOGIES

Key words:

Lógica Difusa, Sistema de Inferencia, Gallinas de Postura, Estrés, Clasificación, Fuzzy Logic, Inference System, Laying Hens, Stress, Classification.

Article type:

ARTICULO DE INVESTIGACION / RESEARCH ARTICLE

Section:

RESEARCH ARTICLES

In this work a fuzzy inference system design to estimate if laying hens present some level of stress without an expert intervention, as a poultry veterinarian, is presented. Additionally, in small farms usually hens diagnosed as stressed are isolated during some days until they are recovered. Moreover, isolated hens are diary examined by the expert to diagnose if the stress has disappeared. Here, it is important to point out that experts usually are unable to estimate the number of days that the hen will need to be kept in isolation until they are recovered. Therefore, as an additional advantage of the proposed FIS is that it can estimate this parameter. In this way it is shown that with the proposed FIS it was possible to determine with an accuracy of 98.62% the stress level and of 90.41% the number of days that the hen will require in isolation. Furthermore, the confusion matrices for these two variables shown that in most of the cases the incorrect estimations have an error of ±1 stress level and ±1 days of isolation. Finally, it is shown that it is feasible to implement this FIS because of the inferences are performed with variables that can be easily measured or identified by non-experts in the hen stress diagnostic. Consequently, it can help to minimize the frequency of the expert intervention in the process.

Keywords: Fuzzy Logic, Inference System, Laying Hens, Stress, Classification.