The current trends in the automotive industry toward electric vehicles are creating increasing interest on methods to quantify and predict the noise, vibration and harshness (NVH) behavior caused by road noise, as well as secondary noise sources. In this regard, component-based TPA (Transfer Path Analysis) has been recently explored. It is a powerful methodology allowing for a virtual prototype vehicle NVH prediction starting from independent sources and component models. The aim of this paper is to analyze the potential of the component-based TPA in the context of road noise. This methodology, on one hand, allows the tire manufacturers to characterize their tires on a tire test-rig independently of any vehicle and, on the other hand, enables the automotive OEMs to perform the prediction of the full system vehicle behavior starting from the test-rig measurements. However, some challenges are still hindering the application of the methodology in this context. In this regard, all the component-based TPA methodology steps are investigated in this work. A slick tire, selected for this analysis, is characterized on a test-rig by a set of blocked forces, which only depend on the source. The uncoupled tire and vehicle are experimentally characterized through the measurement of their frequency-response functions (FRFs) under a static load condition. Frequency Based Substructuring (FBS) is applied to these substructures in order to synthesize the coupled vehicle FRFs. Both rear tires are coupled to the vehicle using this approach. Finally, the vehicle road noise is predicted by propagating one of the tires’ blocked forces through the synthesized full vehicle FRFs in a modular approach. The method is validated by comparison with direct vehicle measurements. Transferability of blocked forces is also assessed by direct comparison between blocked forces estimated on vehicle and test-rig. Importantly, the rolling effect is identified and considered in a direct propagation process, analyzing its implications in the vehicle noise prediction. This paper also intends to serve as a guideline to industrialize the experimental and processing steps.