Geothermal techno-economic models currently in widespread use do not provide the means to jointly account for parameter uncertainties, dynamic operational strategies, and power-plant design flexibility in an integrated analysis. For available academic and government- provided tools, geothermal power generation cost estimates typically start with single-value inputs, although support for user-specified distributions capturing uncertainty in parameter values is becoming more commonplace. The missing piece in determining project value is allowing for flexible responses to uncertainties, where early architectural choices enable future conditions-based design modifications, and rules simulate field-management decisions made during the lifetime of a plant. This pap er proposes a different template for estimating power-project value that incorporates design flexibility. First, the static model is defined with deterministic parameter inputs. Significant uncertainties like the initial subsurface conditions, variation in the local environment over time, and broader risks like disruptions to the electricity market from national electrification, are evaluated through a sensitivity analysis. The most sensitive features are assigned probability density functions, each sampled in repeated model runs to form a M onte Carlo solution ensemble. This base model is then enhanced with decision rules for executing design flexibilities. M ulti-dimensional analysis of the final results provides decision-makers with insights into the optimal choice of facility design, construction timeline, and strategy among those tested that best mitigates the risk of poor economic outcomes for the geothermal investment. This study applies the proposed modelingapproach to a hypothetical Enhanced Geothermal System (EGS) expansion of an existing plant in New M exico. The modeled concept uses modular power-plant units targeting a shallow reservoir, offset from the hydrothermal system currently utilized for producing electricity. Each module comprises a single injector-producer pair connected to a binary cycle generator based on a present -day commercial system analog. The initial cost model provides a static assessment of capital expenses, operating and maintenance costs, and income from power sales to determine the Net Present Value (NPV) over the useful life of the plant. After supplementingkey model parameters with probability distributions, the model uses multiple decision rules to adjust the plant design as operatingconditions change over time. These rules are implemented in succession, defining scenarios with results ensembles compared using summary metrics, histograms, and target curves. Insights from the scenarios are enhanced by optimizing decision-rule threshold criteria, thereby characterizing a field-management strategy that maximizes upside potential without increasing downside risk.