Thesis (M.S., Mechanical Engineering)--University of Idaho, June 2015 | The primary objective of this study is to evaluate the accuracy of using turbulence models to predict entropy generation rates in bypass transitional flows under various pressure gradients in a spatially growing boundary layers over a flat plate. Entropy generation rates in such flows are evaluated using CFD methods. Various turbulence models are assessed by comparing results with DNS data and two recent CFD studies. 3D bypass transition simulations are also performed using LES and IDDES models. Results show better agreement with DNS than any previous RANS studies. The k-? 4eqn. transition model shows closest agreement to DNS data for 2D flow. The SGS models are highly dissipative and LES will require much finer grid within boundary layer region and freestream. These models fail to predict bypass-transition accurately due to lack accurate unsteady fluctuations and grid resolution. Certain limitations and issues are identified and recommendation are made for future studies.