Effective software project planning is essential for accurate cost estimation, optimizing resources and ensuring project success. This study integrates Fuzzy Logic and Genetic Algorithms (GA) to improve prediction accuracy in software project cost estimation, addressing challenges like computational complexity, diverse project parameters, dynamic requirements handling, and integration into project management frameworks. Using the Desharnais, Maxwell, and Kitchenham datasets, the research employs Word2Vec embeddings for feature extraction and Recursive Feature Elimination (RFE) for optimal feature selection. The proposed model combines Fuzzy Logic for initial modelling, GA for parameter optimization, and RFE to select key features. Evaluation metrics, including Mean Absolute Error (MAE), R-squared (R²), and Root Mean Squared Error (RMSE), illustrate varying levels of accuracy. For the Desharnais dataset, the combination of fuzzy logic and Genetic Algorithm (GA) significantly improves the performance, reducing MAE from 0.621 to 0.419 and RMSE from 0.453 to 0.323. Similarly, on Maxwell, MAE decreases to 0.642 and RMSE to 0.521 from 0.946 and 0.767, while on Kitchenham, MAE improves to 0.304 and RMSE to 0.312 from 0.561 and 0.521 with fuzzy + GA. This study not only enhances software cost estimation but also suggests future improvements in model parameterization and real-time data integration for more precise project planning and management.