Abstract
<jats:p>The sustainable production of alternative protein sources requires interdisciplinary approaches integrating biological experimentation, mathematical modeling, and computational optimization. This study presents a computational–experimental framework to optimize diets for producing protein-rich meal from the red Californian earthworm [Eisenia foetida], maximizing protein yield under population and budget constraints. Four diets based on cattle manure and regional organic residues were evaluated through population dynamics, biomass growth, feed conversion efficiency, and protein content determined by the Kjeldahl method. Significant differences were observed [p < 0.05], with a combined diet achieving the highest protein content [49.75 ± 1.2%]. A logistic growth computational model coupled with linear and multi-objective optimization was developed using experimental data, achieving prediction errors below 5%. The optimal strategy consisted of a mixed diet producing up to 0.368 kg of protein without exceeding population or budget limits. Optical micrography confirmed structural differences associated with diet composition, supporting sustainable protein production and circular economy applications.</jats:p>