The aim of this work is predict the influence of containing nanofillers in PVA/PVAc (Poly (vinyl alcohol)/Poly (vinyl acetate)) blends on Young’s modulus of a composite film by response surface methodology (RSM). PVA and PVAc composite films prepared by mixing of water systems: PVA solution and PVAc emulsion with subsequent drying at room temperature. They show wide range of strength-deformation characteristic, which can be controlled by component ratio. Effect of small amounts (1 - 10 wt %) of nanoparticles: monmorillonite clay (MMT) and microcrystalline cellulose (MCC) has also had an influence on structural and strength deformation characteristics of PVA, PVAc and PVA/PVAc blends. The PVA/PVAc (Poly (vinyl alcohol)/Poly (vinyl acetate)) blend with nanofillers studied contained 10% of dioctyl phthalate softener. The solution of the predicted properties is divided into the following stages: choice of control parameters and establishment of the domain of search, elaboration of plans of experiment for the chosen number of reference points, execution of the experiments (physical experiment or computer simulation), determination of simple mathematical models from the experimental data, design of composite on the basis of the discovered mathematical models and at the end verification experiments at the point of optimal solution. In each of these stages it is possible to solve a problem by various methods. So, there are many ways to obtain the plans of experiment. Instead of Experimental Factorial Design the parametric data of experiments was used for optimization of Young’s modulus. The tensile tests were carried out by a Zwick/Roell universal testing, according to EN ISO 527. The method of polynomial approximation of the table data has been at the basis of creating RESINT program, which was used in this research. This method is based on selection of simple functions using least squares. [1]. Simple mathematical models obtained from data of experiments are used as objective functions and constraints in the optimal design problem [2]. In the present investigation penalty function method with random search is used [3]. The last stage is verification experiments in the point of optimal solution.