Análise multifatorial e preditores para características de importância econômica de matrizes suínas em granja multiplicadora

• Autor(a) principal: Leal, Guilherme Brunno de Medeiros
• Data de Publicação: 2021
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UFG
• dARK ID: ark:/38995/0013000001cf3
• Texto Completo: http://repositorio.bc.ufg.br/tede/handle/tede/11714
• topic: Análise multivariada; Longevidade da porca; Granja multiplicadora; Multivariate analysis; Sow longevity; Breeding farm; CIENCIAS AGRARIAS::ZOOTECNIA::PRODUCAO ANIMAL

The present work aimed to investigate the relationship between the productive characteristics and to identify the factors that affect the number of weaned piglets (NLD) per sow/year, the accumulated productivity in kilograms (kg) of weaned piglets (LD) per sow/year and the productive longevity of swine breeders in order to support the selection process of a breeding farm. In chapter 2, factor analyzes were performed in order to understand the relationship between the characteristics and multiple linear regression to identify the main predictor variables for the variables NLD per matrix/year and accumulated productivity in kg of LD per matrix/year. In chapter 3, the females were evaluated as to their longevity, according to two distinct stayability criteria, STAY1 and STAY2, which consider as longevity, respectively, the females that presented the number of live born piglets (NLNV) and kilogram of piglets weaned to the fourth parturition above the herd average. It was identified by the main latent factors that 55% variance between breeders on the farm is explained by the characteristics of maternal ability, piglet viability, fertility and perinatal mortality. The best regression models for prediction explained approximately 66% of the variation both for the NLD traits and for the total weight of weaned piglets adjusted for 21 days (PTD21), evidencing the characteristics number of stillborn piglets (NLNAT), number of dead piglets at birth (NLMN), mean piglet birth weight (PMN) and number of piglets weaned (NLD) with common explanatory factors. It was not possible to obtain a good predictive model for the characteristic mean weight of weaned piglets adjusted to 21 days (PD21) in the present study. The analysis of females by NLNV and PD21 categories indicated that sows that present NLNV below the herd average do not have the potential to exhibit an adequate accumulated production under the evaluated conditions of the farm. The best predictive models for female longevity by STAY1 criteria include age at first mating (IPC) and NLNV and for STAY2 it included only IPC. The prediction model proposed for STAY1 was more efficient (78%) than for STAY2 (55%). Selection for STAY1 can be performed with the NLNV of the first calving, while for STAY2 should preferably be performed later, by the accumulated NLD of the first three calvings. Survival analysis showed that by STAY1, females would be discarded later (8th parturition), when compared to STAY2 (6th parturition).