Orateur
Description
The cheese-making process although standardized faces variability, impacting performance indicators, i.e. product quality reproducibility and economic result. To enhance performance indicators, addressing all variability sources is essential. Despite vast data collections all along the cheese-making process, the potential for its exploitation remains largely underutilized. Integrating artificial intelligence, particularly machine learning (ML), offers new perspectives through data-driven, multi-objective optimization (MOO) approach. In this study MOO was used to enhance the overall performance of a cheese manufacturing plant. ML models elucidate variability in each performance indicator, establishing complex parameter-indicator relationships. MOO addresses conflicting performance objectives simultaneously and enables compromises between indicators, proposing viable solutions. In summary, the use MOO with ML to maximize overall performance represents a breakthrough for dairy industries.
