The aim of this study is to evaluate optimum process conditions for water soluble polysaccharides (WSP) of plant waste materials such as hazelnut skin by using response surface methodologies (RSM). Polysaccharides from hazelnut skin were extracted by means of probe type sonication system using distilled water as a solvent with several sonication time, temperature and amplitude. Response surface models were applied by both Box-Behnken Design (BBD) and Central-Composite Design (CCD) for determining optimal condition which satisfies both maximizing polysaccharide yield and minimizing applied energy for sonication. Results showed that both BBD and CCD models represented the system with high regression coefficient (R-2>0.95) and optimum conditions are 16.15min, 45C, 50.29% amplitude with 0.733 desirability, 15.97min, 45C, 47.27% amplitude with 0.712 desirability, respectively, for designs. Results of given conditions are 1.74% yield and 78.75 kJ consumption for BBD and 1.64% yield and 71.68 kJ for CCD. These results were validated by comparing experimental results at 15min, 45C, 50% amplitude. As a consequence, it was observed that selected statistical models represent the system with a high accuracy and associating scores of each optimal condition is close enough to experimental results.