Abstract: For the dispersed particulate fuel, the effect of particle randomness and double heterogeneity impedes the solution with the existing transport scheme. The Sanchez-Pomraning method improved the conventional method of characteristic (MOC)with the renewal equation to solve the dispersed particulate fuel problem in transport calculation. A 2D cylindrical transport problem for dispersed particulate fuel were calculated. The results show that code can ensure a good calculation accuracy under the conditions of different packing fractions, different particle sizes and multi-particles co-existing in the stochastic medium. Most biases of the eigenvalue (effective multiplication factor) are lower than 100 pcm. The highest bias occurs in the problem of QUADRISO burnable poison particle filled, and gets to 163 pcm. The code based on Sanchez-MOC method can provide believable results for the research about the novel fuel.