Current aircraft conceptual design practices result in the selection of a single (hopefully) Pareto optimal design to be carried forward into preliminary design. This paradigm is based on the assumption that carrying a significant number of concepts forward is too costly and thus early down-selection between competing concepts is necessary. However, this approach requires that key architectural design decisions which drive performance and market success are fixed very early in the design process, sometimes years before the aircraft actually goes to market. In the presence of uncertainty, if the design performance is examined for individual scenarios as opposed to measuring performance of the design with aggregate statistics, the author finds that the single concept approach can lead to less than desirable design outcomes. This thesis proposes an alternate conceptual design paradigm which leverages principles from economics (specifically the Nobel prize-winning modern portfolio theory) to improve design outcomes by intelligently selecting a small well diversified portfolio of concepts to carry forward through preliminary design, thus reducing the risk from external events that are outside of the engineer’s control. This alternate paradigm is expected to result in an increase in the overall profit by increasing the probability that the final design matches market needs at the time it goes to market. This thesis presents a portfolio based design approach, which leverages dynamic programming to enable a stochastic optimization of alternative portfolios of concepts. This optimization returns an optimized portfolio of concepts which are iteratively pruned to improve design outcomes in the presence of scenario-driven uncertainties. While dynamic programming is identified as a means for doing a stochastic portfolio optimization, dynamic programming is an analytical optimization process which suffers heavily from the curse of dimensionality. As a result, a new hybrid stochastic optimization process called the Evolutionary Cooperative Optimization with Simultaneous Independent Sub-optimization (ECOSIS) has been introduced. The ECOSIS algorithm leverages a co-evolutionary algorithm to optimize a multifaceted problem under uncertainty. ECOSIS allows for a stochastic portfolio optimization including the desired benefit-to-cost tradeoff for a well-diversified portfolio at the size and scope required for use in design problems. To demonstrate the applicability and value of a portfolio based design approach, an example application of the approach to the selection of a new 300 passenger aircraft is presented.