The aim of this study was to characterize the morphology of the creole goat of the Volcanes de Colima physiographic sub-province, Mexico. Seventeen body measurements were taken from 371 goats older than two years and four racial indices and 17 functional indices were calculated. Data were analysed using Pearson correlation, principal component, and hierarchical cluster analysis. The rump angle, chest width and all measurements of the ears and udder showed the greatest variability. Also, the goats showed great morpho-structural harmony. Racial indices showed that goats are dolichocephalic, ellipometric, and have a convex rump, while functional indices indicated dual-purpose zootechnical aptitude. Four principal components explained 84.5% of the variation of body measurements. Rump angle, hearth girth, body length and ear l...

            The aim of this study was to characterize the morphology of the creole goat of the Volcanes de Colima physiographic sub-province, Mexico. Seventeen body measurements were taken from 371 goats older than two years and four racial indices and 17 functional indices were calculated. Data were analysed using Pearson correlation, principal component, and hierarchical cluster analysis. The rump angle, chest width and all measurements of the ears and udder showed the greatest variability. Also, the goats showed great morpho-structural harmony. Racial indices showed that goats are dolichocephalic, ellipometric, and have a convex rump, while functional indices indicated dual-purpose zootechnical aptitude. Four principal components explained 84.5% of the variation of body measurements. Rump angle, hearth girth, body length and ear l...

En esta tesis construiremos un algoritmo de descomposición asociado a un problema de optimización convexa separable con restricciones lineales, en particular lo aplicaremos a problemas de programación lineal. Este algoritmo aprovecha la estructura separable de la función objetivo del problema original considerando en cada iteración subproblemas de optimización para cada componente de la función objetivo, siendo estas de menor tamaño que el problema original e independientes entre sí, lo cual permite resolverlos de forma paralela, disminuyendo el costo computacional.