FIRAS HASHIM KAMAR, FARRAH E. NIAMAT, AYAD A.H. FAISAL, AHMED A. MOHAMMED, AURELIA CRISTINA NECHIFOR,GHEORGHE NECHIFOR USE OF ARTIFICIAL NEURAL NETWORK FOR MODELING AND PREDICTION OF REACTIVE RED DYE REMOVAL FROM WASTEWATER USING BANANA PEELS BIO-SORBENT Bio-sorption of red dye from aqueous solutions onto banana peels was investigated. Effects of initial pH, bio-sorbent dose, initial concentration, contact time, and temperature were studied and they found of 3, 0.4 g/100 mL, 50 mg/L, 100 min and 298 K respectively with removal efficiency of 93.44%. Artificial neural network was used for prediction of adsorption efficiency and its outputs showed a better fit than other traditional isotherm models. The negative values of DG° and DH° indicate that the bio-sorption of red dye was favored and exothermic. The sensitivity analysis signified that the pH was the most influential variable.