In electrospinning and nanofibers production, an electric field is generated between an oppositely charged polymer fluid and a collection screen. The nanofibers properties depends on several parameter such as applied electric field, the polymer concentrations, the rate of injection and nozzle-collector distance. In this work, the polynomial regression, neural network modeling using back propagation technique and Fuzzy logic method are used to establish a relationship between polyacrylonitrile nanofiber diameter and electrospinning parameters such as the polymer concentrations, nozzle-col lector distance and applied voltage. Obtained results indicate that the performance of neural network is better than fuzzy logic and polynomial regression methods.