Disassembly modeling and planning is meaningful and instructive for remanufacturing and recycling of disused products. However, an actual disassembly process of products has uncertainties since a variety of unpredictable factors. To handle such uncertainty, this work presents a novel disassembly modeling and planning approach using an extended stochastic Petri nets. An extended stochastic disassembly Petri nets (ESDPN) model is defined to provide basis for disassembly planning. Moreover, based on it, some typical probability evaluation models is built accordion to different disassembly criterion. In addition, a hybrid intelligent integrating stochastic simulation and neural network (NN) is proposed to solve the proposed models, respectively. Some numerical examples are given to illustrate the proposed models and the effectiveness of proposed algorithms.