Decision-making strategies should be robust and adaptive to consider two key characteristics of climate change effectively: non-stationarity and large uncertainty. A robust strategy is satisfactory over a wide range of uncertainties, while an adaptive strategy updates itself continually with its previous performance and new information. This chapter introduces two well-known decision-making methodologies: robust optimization for the robust perspective and real option analysis for the adaptive perspective. The concept of robust optimization is to save infeasible scenarios that violate the constraints being considered by incorporating a penalty function, which is a function of the violation degree for each scenario. As a case study of robust optimization, a robust stochastic dynamic programming model was applied to the monthly reservoir operations of a single dam. A large number of simulations that tested various changes in average and standard deviation values revealed trade-off relationships between different metrics, where the robust stochastic dynamic programming model performed better in terms of frequency and magnitude under extreme conditions. As the second case study, the traditional benefit–cost analysis was expanded into the real option analysis model by incorporating options and scenarios into its evolutionary procedure. The real option analysis model was applied to evaluate a water infrastructure planning project for drought mitigation in Korea. It was demonstrated that the tested project would not be economically feasible when it is constructed, but could become feasible if the probability of severe drought increases by 20%.