In this paper, we investigate an adaptive energy management and power splitting system for a fuel cell hybrid electric vehicle. The battery pack is the main power source whereas the fuel cell is considered as a range extender that cannot sustain alone the vehicle traction power. In addition, the fuel cell contributes to reduce the battery pack degradation by limiting its depth-of-discharge (DoD). This energy management system is based on a two layer architecture in which the upper layer computes the anticipated end-of-trip DoD using online mass estimation. The lower layer is designed to split the driver power demand by minimizing a cost function which includes the hydrogen/electricity cost ratio. Therefore, the best trade-off between reducing battery pack degradation and using cost effective energy is provided. Furthermore, the system allows the fuel cell to operate at its maximum efficiency. Comparative study results indicate that using online mass estimation improves the overall fuel consumption efficiency whilst contributing at the same time to DoD reduction. (C) 2013 Elsevier B.V. All rights reserved.