Purpose -The purpose of this paper is to propose a new approach to optimize the TiO 2 concentration on a resistive-type humidity sensing mechanism (RHSM) based on artificial neural network. Design/methodology/approach -First is the modeling of the sensing mechanism. Using neuronal networks and Matlab environment to accurately express the output of the sensing mechanism, this model thus takes into account the parameter, non-linearity, hysteresis, temperature and frequency; furthermore, the TiO 2 concentration effect on the humidity sensing properties in the model is investigated. In a second step, the Matlab environment is used to create a database for an ideal model for the sensing mechanism, where the response of this ideal model is linear for any above parameters value. Findings -An analytical model for the sensing mechanism "SM" and the ideal model "IM" has been created. The bias matrix and the weights matrix were used to establish the SM model and the IM on performance simulation program with integrated circuit emphasis simulator, where the output of the first is identical to the RHSM output and the output of the last is the ideal response. Originality/value -The paper proposes an electrical circuit used to optimize the TiO 2 concentration of a resistive humidity sensing mechanism.