A fuel reloading analysis system for pebble bed reactor was developed by using a Monte Carlo code. The kinematic model was modified to improve the accuracy of the pebble velocity profile and to develop the model so that the diffusion coefficient is not changed by the geometry of the core. In addition, the point kernel method was employed to solve an equation derived in this study. Then, the analysis system for the pebble bed reactor was developed to accommodate the double heterogeneity, pebble velocity, and pebble refueling features using the MCNPX Monte Carlo code. The batch-tracking method was employed to simulate the movement of the pebbles and an automation system was written in the C programming language to implement it. The proposed analysis system can be utilized to verify new core analysis codes, deep-burn studies, various sensitivity studies, and other analysis tools available for the application of new fuel reloading strategies. It is noted that the proposed algorithm for the optimum fuel reloading pattern differs from other optimization methods using sensitivity analysis. In this algorithm, the reloading strategy, including the loading of fresh fuel and the reloading positions of the fresh and reloaded fuels, is determined by the interrelations of the criticality, the nuclear material inventories in the extracted fuel, and the power density. The devised algorithm was applied to the PBMR and NHDD-PBR200. The results show that the proposed algorithm can apply to satisfy the nuclear characteristics such as the criticality or power density since the pebble bed core has the characteristics that the fuels are reloaded every day.