The competitiveness of conventional press & sinter technology mainly depends on the ability to obtain tight tolerance on sintered products. In order to maintain this strategic advantage in spite of the rapid global market changes, a continuous improvement in the dimensional accuracy of the products has to be pursued. One of the major limits in the dimensional precision of sintered products regards the anisotropic dimensional change occurring on sintering. Despite this problem is well known, an effective design procedure accounting for the anisotropic behavior of dimensional variations is far to be reached. The main reasons concern the multiphysical mechanisms involved and the effect of material, geometry and process condition on the final results. This work aims at developing a design methodology accounting for the anisotropy of dimensional changes on sintering. This study has been performed considering both the fundamental principles and the industrial application, aiming at proposing: - a solid theory considering the mechanisms which determine the anisotropic dimensional changes; - a practical and effective design tool for the industrial application. The role of uniaxial compaction on the origin of anisotropic dimensional change was firstly investigated. AISI 316L ring shaped samples were compacted at different geometries, and four different particle sizes. During single action compaction, forces acting on the tooling and powder column, and related displacements, were recorded by the press in order to derive the compaction mechanics of the powder mixes. Further, the dimensions of the samples were measured before and after sintering. A linear trend was observed correlating the deviatoric stresses occurring in compaction to the anisotropic dimensional variations on sintering. This result offers a new perspective in the prediction of the anisotropic dimensional change, and could lay the foundation of a solid model. Aiming at developing an effective design tool to predict dimensional change on sintering, the analytical design procedure previously developed by the research group at the University of Trento was validated on real parts. A Club Project was promoted by EPMA, collecting the University of Trento and five qualified industrial partners. These companies provided five different real parts characterized by different materials and geometries, whose dimensions were measured before and after sintering. The comparison of the measured and the predicted sintered dimensions demonstrated that this design approach can be an effective tool for designers. Further work could implement the promising results obtained investigating the compaction mechanics in the design procedure, aiming at defining a powerful tool to design PM parts accounting for anisotropic dimensional changes.