This work is part of an ongoing research effort in which parameter estimation techniques have been utilized in recovering the kinetic parameters of a given reacting model using measurements collected from TGA devices. The goal of this research work is to develop a useful and universal estimation procedure to determine simultaneously the kinetic parameters involved in the chemical modelling under study such as combustion, pyrolysis, waste stabilization, etc. The present parameter estimation problem is solved with the Levenberg-Marquardt method of minimization of the least square norm representing the square difference between the measured mass variations during cardboard pyrolysIS nd the mass responses obtained with numerical solution of the model. An analysis of the linear dependency of the pyrolysis parameters needed to design a robust estimation tool is presented. In order to perform thIS nalysis, the sensitivity coefficients and the sensitivity matrix determinant were examined. Experimental data obtained with TGA during the pyrolysis process of cardboard are analysed in this paper and the unknown parameters involved in the kinetic modelling are estimated.