Mixed model methods allow genetic merit to be compared directly for pigs measured in different environments or generations. They take into account genetic relationships among all individuals in the population, and allow continuous selection throughout the animal's life. The advantage is more accurate selection in larger populations. The risk is that selection decisions are decentralised to many different sites, requiring a high degree of supervision.

In 1986 a “group nucleus” using BLUP and AI was established for a pig dam line in order to improve both litter size and lean growth. Selection occurs at several different stages in the life cycle, and this study examines selection differentials during performance testing together with the loss of differential resulting from visual appraisal.