The objective of this work was to evaluate the precision of Artificial Neural Networks (ANNs) to estimate zootechnical indexes, based on thermal and physiological variables of pregnant sows. This study was carried out from January to April 2005, in a swine industrial production farm in the gestation section with 27 primiparous gilts, allocated in individual pens and after on farrowing pens where it was quantified animal production indexes of piglets from the study. Therefore, an ANN backpropagation was implemented, with one input layer, one hidden layer, and one output layer with tangent sigmoidal transference functions. Air temperature and respiratory frequency were considered as input variables and weight of piglet at birth and the number of mummified piglets as output variables. The trained ANN presented a great generalization power, which enabled the prediction of the answer-variables. Characterization of the environment of gestation and maternity was appropriated if compared to the real data, with few under or overestimated tendencies of some values. The use of this specialist system to predict zootechnical indexes is viable because the system shows a good performance for this use.
animal production precision; specialist systems; swine production
