For managing the supply-demand imbalance in free-floating bike sharing systems, we propose dynamic hubbing (i.e., dynamically determining geofencing areas) and hybrid rebalancing (combining userbased and operator-based strategies) and solve the problem with a novel multi-objective simulation optimization approach. Given historical usage data and real-time bike GPS location information, the basic concept is that dynamic geofenced areas (hubs) are determined to encourage users to return bikes to desired areas towards the end of the day through a user incentive program. Then, for the remaining imbalanced bikes, an operator-based rebalancing operation will be scheduled. The proposed modeling and optimization solution algorithm determines the number of hubs, their locations, the start time for initiating the user incentive program, and the amount of incentive by considering two conflicting objectives, i.e., level of service and rebalancing cost. In this study, for free-floating bike sharing, the level of service is represented by the walking distance of users for locating a usable bike, and the rebalancing cost is weighted incentive credits plus operator-based rebalancing cost. We implemented the proposed method on the Share-A-Bull free-floating bike sharing system at the University of South Florida. Results show that a hybrid rebalancing and dynamic hubbing strategy can significantly reduce the total rebalancing cost and improve the level of service. Moreover, taking the advantage of crowdsourcing (or job-sharing) reduces negative impacts—energy consumption and green house gas emissions—of the operation of rebalancing vehicles and makes bike sharing a more promising environmentally-friendly sharing transportation mode.