In Public Bicycle System (PBS), bicycle redistribution is one of the most important issues for efficient and effective operations. Although, bicycle redistribution strategy was often noted, it was rarely studied in a through manner. The major purpose of this study is to find the number of bicycles to redistribute them on full or empty stations and to identify the station visiting sequence of redistribution vehicle to minimize travel time and dissatisfaction incurred from users. The schedule of bicycle redistribution vehicle was obtained by three steps with several assumptions. The first step is to find redistribution timing to prevent full or empty stations. The second is to set up time windows to redistribute bicycle to prevent full or empty states. Finally, the last is to schedule the bicycle redistribution vehicle as the dynamic Bicycle Redistribution Vehicle Routing Problem (BRVRP). One hypothetical PBS is conceived to examine the possibility to apply the dynamic BRVRP suggested in this study to either existing or proposed PBS. The example system consists of 20 stations, 1 redistribution vehicle, and 1 center, the condition of which is more or less similar to existing system. The simplex dual algorithm programmed with Xpress-Mosel system provided proper solutions which describe the visiting station sequence of redistribution vehicle while meeting the needed bicycles to be moved and the capacities of stations. The methodology set forth in this study may help design the PBS and bicycle redistribution strategies to be used in reality. The tentative conclusions resulted from a hypothetical PBS, however, await further refinements and correction. Future research such as applications to existing PBS, comprehensive integrated modeling, and more precise time-dependent variables can also be considered to be of better use.