We introduce and analyze an innovative transportation system called Flexible Mobility on Demand (FMOD) which provides a menu of optimized travel options in real-time. This menu consists of different modes with varying level of service including mini-bus, shared-taxi and taxi. In order to analyze FMOD, a modeling framework is developed that integrates scheduling, routing, assortment optimization and choice modeling methodologies. Simulation experiments are conducted for Hino city in Tokyo with different number of vehicles in the fleet.