Multi-objective Optimization using ANN and Quadratic Model

This dataset contains CCD result data for pullulan extraction using the solvent method and codes for developing linear regression and artificial neural network (ANN) models to optimize the extraction process (i.e., solvent-to-broth ratio, pH, and incubation time). It includes scripts for performing sensitivity analysis, desirability testing, and multi-objective optimization for three responses (i.e. pullulan recovery, sucrose equivalent, and protein impurities). The package also provides code to generate response surface methodology (RSM), an ANN residual plot to evaluate model accuracy, and Pareto front analysis to identify optimal trade-offs between competing objectives.