Diesel exhaust aftertreatment solutions using injection, such as urea-based SCR and lean NOx trap systems, effectively reduce the emission NOx level in various light vehicles, commercial vehicles, and industrial applications. The performance of the injector plays an important role in successfully utilizing this type of technology, and the CFD tool provides not only a time and cost-saving, but also a reliable solution for extensively design iterations for optimizing the injector internal nozzle flow design. Inspired by this fact, a virtual test methodology on injector dosing rate utilizing CFD was proposed for the design process of injector internal nozzle flows. For a low-pressure (less than 6 bar) injector application, the characteristic Reynolds number based on the diameter and mass flow rate of the inlet, return flow outlet, and nozzle exit of the injector might range from 2000 to 20000, therefore, employing a flow-physics based viscous model for building up a virtual test methodology is critical to properly capture the fluid dynamics of injector internal nozzle flow. In this study, a transition three-equation eddy-viscosity model was used to calculate the dosing rate for injectors that have different configuration features, and the computational results of the proposed virtual test methodology were validated with the test data measured in the Tenneco Injector Flow Lab. The results also demonstrated the virtual test methodology can accurately predict the fluid dynamics of boundary layer development and calculate the onset of the transition to cope with the transitional flow behavior. Several design iterations were studied using the validated virtual test methodology to investigate the impacts of injector key geometric parameters on the dosing rate.