In recent decades the participation of private capital in infrastructure projects has increased. Considering the high level of risk of some of these projects, there may be a need for some sort of government support mechanisms that reduce these risks in order to continue to attract the private investor interest. One such alternative is the Limited Minimum Revenue Guarantee, where the investor is guaranteed a minimum revenue level for the project, limited to a preset ceiling, or CAP, which reduces the government’s risk. This work proposes a method to allow the government to define the optimal parameters for both the Minimum Revenue Guarantee and the CAP, so that the private investor is attracted to the project at the lowest cost and risk possible to the government. The interest of the private investor is measured by the level of risk involved, which is obtained from the probability distribution of the project´s NPV. The metric for the government risk is assumed to be the Omega performance measure, which is determined from the probability distribution of the guarantee under the real options approach. Due to the complexity of the problem, an analytic solution is not feasible and it is shown that the optimal parameters can be determined with numerical methods. As a secondary result, it is also shown that when the government analyses these guarantees in the context of a portfolio of projects, the risk to the government is reduced due to the diversification effect.