The Seine catchment (located in the North of France) includes a system of four large reservoirs to protect against floods and water shortages multiple locations downstream including the Paris region. For each downstream key location, the reservoirs try to keep the streamflow between minimum and maximum thresholds, which are all objectives to be achieved. We propose here a framework assessing the risk of failing to meet each of these objectives under specific hydroclimatic conditions considering the current filling state of each reservoir and the calendar day. First, a semi-distributed hydrological model of the basin calibrated on influenced flows over the period 1959-2019 is used to simulate naturalised flows for a given climatic daily time series. Then, for each objective, a dynamic programming optimisation provides optimal filling curves that are statistically analysed to produce yearly probabilistic upper (for floods) or lower (for droughts) filling curves. These results are highlighted by an interactive graphical user interface based on an R Shiny application showing the risk probability of future failure of each objective depending on the calendar day and the storage state of each reservoir. This assessment is conducted with historical climatic observations as well as a set of the recent French DRIAS 2020 climatic projections for different future periods: 2021-2050, 2041-2070 and 2071-2100.