The conversion of organic waste into high-value bio-products via Black Soldier Fly Larvae (BSFL) bioconversion presents a promising avenue for sustainable waste management, circular bio-economy, and biofuel production. However, existing studies lack a structured methodology to systematically assess the impact of substrate macronutrient composition (proteins, carbohydrates, lipids, and moisture content) on larval growth. This research introduces a computational framework leveraging Model-Based Design of Experiments (MBDoE) within the PYOMO optimization platform to parameterize the BSFL growth model using optimization techniques such as sum of squared errors (SSE). By integrating MBDoE with the Gompertz growth model, we design optimized experiments that maximize information gain while minimizing experimental resources. The Fisher Information Matrix (FIM) is utilized to assess parameter identifiability, ensuring precise estimation of growth kinetics under varying feed conditions. Preliminary results demonstrate that after seven iterations, MBDoE successfully predicts BSFL growth parameters with high accuracy, optimizing substrate composition to enhance biomass yield. Compared to traditional empirical approaches such as black-box modeling, full-factorial, and fractional factorial designs, this methodology significantly improves experimental efficiency, accelerates model identification, and reduces parameter estimation uncertainty. Moreover, it enables the design of dynamic experiments while accounting for constraints and system bounds. This study advances computational modeling in biomass utilization by providing a scalable, data-driven methodology for predicting BSFL growth dynamics while showcasing the feasibility of insect-based waste valorization, bridging the gap between experimental biology and process systems engineering.
