Milk modeling with PLS-SEM

Using data from 38 Awassi ewes across three lactation stages, we applied Partial Least Squares Structural Equation Modeling to integrate latent milk quality traits—freshness, microbiological integrity, chemical composition, flavor, and color—with yield. Our framework reveals time-dependent shifts in the drivers of milk production, highlighting the trade-offs between quality attributes and volume. These insights enable precision management of udder health, nutrition, and processing strategies to optimize economic returns and product quality, supporting sustainable sheep-dairy systems by balancing yield efficiency with high-quality standards that meet environmental and social demands.