Part I of this five-part paper presents a robust and comprehensive computational methodology, developed and validated for physically realistic and economically feasible results in racing. The methodology was applied, in Parts II-V, for the flow predictions and optimization in the entire intake cowl, complete intake manifold and the intake and exhaust valve regions of a V8 racecar engine. The validations had been performed through “blind” experimental tests for four intake and three exhaust valve cases, revealing consistently predicted flow rates within an acceptable error band. Additional validation was obtained on the Talladega racetrack where recommendations based on CFD predictions of intake cowl modifications lead to a significant time reduction of 0.286 seconds per lap. An original method to obtain a detailed electronic description of complex geometry from actual prototype hardware was developed. New grid generation techniques were developed in order to obtain high quality and high density grids and reduce the time to mesh the classes of complex geometries encountered in racecar applications. Despite having over 11 million cells, all grid meshes proved to be very high quality, with maximum and average skewness of 0.76 and 0.28, respectively. An assessment of the RKE, RNG, Standard k-ɛ, and RSM turbulence models, used in conjunction with non-equilibrium wall functions, was performed for the medium-lift intake valve case. The results indicate that the RKE turbulence model yields the best overall compromise, as it is robust, requiring a moderate computational effort, relatively “easy” to use, and reasonably accurate. Due to its well-known high numerical stiffness, the RSM model was found to be not at all practical at this time. Effective solution management strategies were developed to rapidly attain “full convergence” for all the cases, while satisfying unusually strict convergence criteria. Overall, the comprehensive CFD methodology and the new geometry, grid and convergence procedures developed by the authors proved collectively to be an effective and economical design tool to make “go, no-go” type decisions by racecar teams.