The response surface methodology (RSM) is a collection of mathematical techniques useful for developing, improving and optimizing process. In this study, RSM technique was applied to evaluate the effect of the components proportion on the mechanical properties of low density polyethylene (LDPE)/ poly (3-hydroxy-butyrate) (PHB) blends filled with castor oil cake (CC). The blends were prepared by melt mixing in a twin screw extruder. Low density polyethylene, poly (3-hydroxy-butyrate) and castor oil pressed cake were represented by the input variables designated as LDPE, PHB and CC, respectively. As it was desirable to consider the largest LDPE content in the ternary system, the components of the mixture were subjected to the following constraints: 0.7 ¡Ü LDPE ¡Ü 1.0, 0¡Ü PHB¡Ü0.3 e 0 ¡Ü CC ¡Ü0.3. The mechanical properties of the different mixtures were determined by conventional ASTM tests and were evaluated through analysis of variance performed by the Minitab software. Some polynomial equations were tested in order to describe the mechanical behavior of the samples. The quadratic model in pseudo components was selected for describing the tensile behavior because it was the most efficient from a statistical point of view (p-value ¡Ü 0.05; coefficient of determination (r2) close to 1 and variation inflation factor (VIF) values < 5). The results showed that the LDPE Young¡¯s modulus increases but the other tensile properties and impact resistance deteriorate with the addition of PHB or CC. The tensile strength values of binary mixtures of LDPE lie in the range from 8.9 to 10 MPa. As some commercial grades of LDPE have mechanical strength in this range, it may be inferred that the addition of a certain amount of PHB or CC to LDPE may be considered as a possibility for obtaining LDPE based materials with increased susceptibility to biodegradation. The cubic model in pseudo components was selected for describe the flexural strength of the samples because it was the most adequate