The properties of SFRT (Short Fiber Reinforced Thermoplastic) are highly dependent on the fiber orientation distribution (FOD) and fiber length distribution (FLD). Microstructural models for predicting composite strength and modulus on the basis of the mechanical properties of constituent materials have been developed, and these are generally consist of formulae taking in to account for both fiber length distribution and fiber orientation distribution. The purpose of this research work is to determine FLD and FOD by image analysis. The result obtained from this technique is then used in the theoretical models for predicting mechanical properties. In this investigation one composite was used, consist of Poly (vinyl chloride) as matrix phase and nylon fiber as reinforcing phases. Polarized microscope was used to obtain pictures (photo micrograph) of different point in the composite. The pictures obtained from polarized microscopy were changed to binary picture for image processing analysis. Tracing method was applied to prepare picture for obtaining data related to orientation and fiber length. MATLAB software was applied to measure the orientation and length of fiber. A developed algorithmic program was used to measure the orientation and length of fibers. Input modeling was used to assess orientation and length distribution of the fibers. The distribution obtained has a good compatibility with normal distribution (FOD) and Weibull distribution (FLD). The probability density function of this distribution can be used to predict Tensile Strength in the appropriated models. The Modified Rule of Mixture was used to prediction of Tensile Strength, in this model two parameter related with FOD and FLD were used and the experimental results have good approximation with results obtained from modeling