A new method, called modified Gompertz model, was proposed to interpret solid state decomposition. To verify the validity of this new model, the thermal degradation of polypropylene (PP) was performed and the results of curves fitting of thermogravimetry (TG) and derivative thermogravimetrical (DTG) from the Gompertz model were compared with those obtained from the traditional general model. The Arrhenius kinetic parameters used in the general model were derived from three traditional methods (i.e., Friedman, Freeman-Carroll, and Ozawa methods). For the Arrhenius parameters calculation, the heating temperature range was set from 298 to 1073 K and three different heating rates of 5, 10, and 15 K/ min were selected. The sum of square errors (SSE) and the coefficient of determination (r(2)) were calculated for all simulated data obtained from the modified Gompertz model and the general model. Statistical results show that all SSE values obtained from the modified Gompertz model are much smaller than those from the general model, and every r(2) value of conversion of modified Gompertz model is greater than 0.987. The writers found that the TG and DTG curves simulated by the modified Gompertz model have a better fit to the experimental data than those curves reproduced from the general model with values of Arrhenius parameters obtained from the traditional techniques. These results suggested that the modified Gompertz model can be considered as a simple way to describe the thermal decomposition reaction behavior. (C) 2012 American Society of Civil Engineers.