One of the problems of the evaluation for Real Options is the need to have complete markets so that non arbitrage methods can be used for its solution. When that is not the case, or when the determination of a dynamic portfolio of market securities that replicate the stochastic characteristics of the project is not feasible for any reason, the alternative is to use an exogenous and arbitrary discount rate. Another problem is the inclusion of two or more uncertainty sources in the mathematical modeling of the project, which brings a certain degree of complexity to the problem, especially when those uncertainties involve private risk, not correlated with the market. This work synthesizes some Real Options Theory concepts developed by several authors with Decision Analysis tools to propose a method for evaluation of projects in incomplete markets by dynamic programming using an innovative algorithm to model the project`s stochastic process with a binomial lattice and decision tree. The method is computationally intense, but simpler and more intuitive than that the traditional methods of Real Options, allowing for a greater flexibility in the modeling of the problem. This methodology is applied to the problem of the valuation a highway concession in Brazil with managerial flexibility in incomplete markets and political risk.