This work was devoted to optimizing the operating conditions of a low-energy consumption, high frequency piezoelectric ultrasonic reactor used for biodiesel production from waste cooking oil (WCO). The optimization was conducted on the basis of exergetic and technical constraints using a consolidated combination of soft computing techniques. Transesterification temperature, residence time, methanol/oil molar ratio were input variables, while functional exergy efficiency (FEE), universal exergy efficiency (UEE), normalized exergy destruction (NED), and conversion efficiency (CE) were considered as output parameters. Four independent ANFIS models were first developed to map the outputs as a function of the inputs. The developed ANFIS models were then introduced to a hybridized optimization paradigm obtained by consolidating the elitist non-dominated sorting genetic algorithm (NSGA-II) and linear interdependent fuzzy multi-objective optimization (ALIFMO) approaches. The optimization was set to maximize the FEE and UEE and to minimize the NED, while satisfying the ASTM standard on CE (i.e., biodiesel content > 96.5%). All the developed ANFIS models excellently estimated the outputs with an R-2 approximate to 1.0. The transesterification temperature of 60 degrees C, residence time of 10 min, and methanol/oil molar ratio of 6.20 were selected as the best operating conditions among the spectrum of the solutions suggested by the developed approach.