Precise modeling of asset price trajectories in turbulent financial markets, especially cryptocurrencies, has always faced challenges such as complex temporal dependencies and nonlinear behavior. Classical models, such as Brownian motion, have limitations in simulating features like long memory and mean reversion. In this study, by combining the fractional Brownian motion process and the Vasicek model, and utilizing a genetic algorithm for optimizing the combination weights, a novel framework for asset price path simulation is presented. In this approach, the aforementioned models are implemented separately based on historical data, and their results are combined using optimal weighting coefficients. Then, by defining an objective function based on minimizing the prediction error, the optimal weights are calculated using the genetic algorithm. The findings, based on Bitcoin data, indicate that the proposed hybrid model demonstrates greater accuracy compared to individual models and offers acceptable performance in forecasting price fluctuations. This framework can serve as an efficient tool in financial decision-making, risk management, and the design of investment strategies in highly volatile markets.