Hydro power producers are faced with the task of releasing water from the reservoirs in the right time. To do this there are tools using stochastic optimization that aims at maximizing the income of that producer. The existing methods have a high computing time and grow exponentially with the size of the problem. A new approach that uses linear decision rules is investigated in this thesis to see if it is possible to maintain the same quality of the solutions and in the same time decrease run times. With this method the hydro power producer receives policies as an affine function of the realization of the uncertainty variables in inflow and price. This thesis presents a deterministic model and then converts it into an linear decision rules, LDR, model. It also presents a way to model the uncertainty in both inflow to the reservoir and the spot price. The result is that the LDR approach generates reasonable policies with low run times but loses a lot of optimality compared to solutions that are used today. Therefore it is concluded that this approach needs further development before commercial use. The work described in this thesis has been done in cooperation with three master students at NTNU. The approach of using linear decision rules are the same in the two projects but the differences are the models evaluated.