Despite the massive use of Reclaimed Asphalt Pavement (RAP) in Hot Mix Asphalt (HMA) production, the chemo-physical phenomena that characterise the fabrication of these mixtures have not yet been completely explored. The detection and understanding of these mechanisms as well as the study of the heterogeneity that characterizes high RAP mix production are fundamental to improving the approach to recycling, because they represent the source of the mixture characteristics and performance. This thesis aims to provide a framework for the characterisation of several phenomena and mechanisms occurring during a new mix when RAP material is involved. The variability of binder rheology with different ageing levels was analysed to evaluate the RAP binder heterogeneity and its evolution with time. This stage provided important information for the development of new binder blending charts to understand how the aged and the virgin binder interact and to ascertain whether the blend is homogeneous. The artificially simulated behaviour, assuming the hypothesis of 100% blending, would have provided important information regarding the blend characteristics if the behaviour of the binder had been confirmed in the mixture evaluation. However, in the next stages other phenomena occurred such as RAP particle clustering and the presence of re-activated and non-reactivated binder. Thus, a new methodology was developed using a Clustering Index (ICG*) to detect and quantify the RAP clusters. The same methodology allowed the investigation of the mobilisation of RAP binder and highlighted the presence of different layers in the RAP binder film thickness such as an over-aged crust and softer bitumen under the crust that acts as glue facilitating the clustering. The framework that emerged after these investigations led to the rethinking of the mix design approach, especially when high RAP quantities are used. Therefore, a new methodology was developed for estimating the dosage of virgin bitumen required in asphalt mixtures with or without RAP, investigating the specific contributions of all the components of the mixtures (virgin aggregates, filler and RAP) and taking into account the RAP cluster phenomenon and its consequences during the mixture. Indeed, new revised surface area factors were computed thanks to the 3D images of several stones for the calculation of the specific surface area of the aggregates and the critical filler concentration allowed the volume of bitumen required for coating the filler to be obtained. Finally a RAP clustering prediction model was developed to readjust the grading curve after the RAP particle agglomerations occurring during the mix. The new theoretical formula, thanks to the separation of all the contributions, provides results valid for mixtures with and without RAP (the terms related to the RAP become zero if the RAP percentage introduced in the formula is zero). The verifications carried out in the laboratory gave positive and promising results and the methodology is considered successful in estimating the value of the optimal bitumen quantity for mixtures with and without RAP.