Artificial intelligence tools for modeling and optimizing solid-state fermentation processes

Abstract

Solid-state fermentation (SSF) is a process used to produce enzymes and secondary metabolites; however, its low efficiency limits its application, as it does not cost-effectively meet market demand. This article proposes modeling the operation and determining the optimal parameters of SSF processes through the application of artificial intelligence systems. To this end, programming algorithms were designed in MATLAB software to implement an artificial neural network (ANN), a genetic algorithm (GA), particle swarm optimization (PSO), and the artificial bee colony (ABC) algorithm. To verify the proposed method, the production of proteases used in the cheese industry was modeled and optimized. The results show that the optimal process parameters were correctly identified, the modeling precision and accuracy were increased (R² > 0.90), and the process resources required can be reduced. These findings suggest that the use of artificial intelligence systems in SSF processes is an effective tool to maximize their production.