Depletion of conventional oil fields and increasing requirements for the economic justification of investment projects enhance the significance of developing hard-to-recover reserves, including heavy oil, which is characterized by technological complexity, high energy intensity, and multifactor uncertainty. This study proposes a balanced risk assessment framework that integrates Monte Carlo simulation with a Mamdani-type fuzzy inference system. Quantitative project parameters are modeled as random variables, while poorly formalized geological, technical, operational, environmental, social, and political-regulatory risks are aggregated into an integral fuzzy risk index. The proposed framework is tested using open data for a heavy oil project and four integration variants: a baseline Monte Carlo model without fuzzy risk adjustment, a model with risk-adjusted mean values, a model with risk-adjusted standard deviations, and a combined model. The fuzzy inference system produced a high integral risk index, Rhtr = 0.782. In the baseline model, expected NPV was −1.79 billion USD and the probability of negative NPV was 66.52%. When the integral risk index was incorporated through shifts in expected values, expected NPV decreased to approximately −6.76 billion USD and the probability of negative NPV increased to 89.90%. The combined model produced a similar central shift, with expected NPV of −6.75 billion USD and P(NPV < 0) of 89.44%, while also reflecting increased distributional uncertainty. The results show that the main economic effect of integral risk is associated with systematic changes in expected CAPEX, OPEX, production rate, and decline rate, whereas variance-only adjustment has limited influence on central project metrics. The proposed approach provides a transparent tool for linking qualitative risk assessment with probabilistic economic evaluation and supports more conservative investment decision-making in heavy oil development projects.