Covering one-third of the world's land, forests provide many essential resources and services on which more than 25% of the world's population directly depend for their livelihoods. Moreover, because of their importance in the carbon and water cycle and in the production of renewable materials, forests are one of the main means of mitigating the consequences of current and future global climate change. The need to balance ecological, economic and social roles to achieve sustainable management of forest resources is therefore evident from the local to the global scale. Accurate and up-to-date information on forests dynamics is essential to design and implement sustainable forest management plans. In particular, the development of customized management scenarios based on forest stand characteristics require accurate growth and yield estimates that can only be derived from permanent forest inventory data. This thesis focuses on the relevance of forest modelling from permanent forest inventory data to improve knowledge on current forest resource stocks and dynamics and help design sustainable forest management policies. The study concerns the forests of Wallonia, the southern Region of Belgium. Wallonia forests represent an interesting case study as they are intensively managed, highly fragmented, heterogeneous in structure and composition and undergoing fast change to adapt to new environmental and economic conditions. The first part of this thesis concerns the research for an appropriate modelling methodology to assess the level of productivity and species-station suitability of forest plantations. We developed an easily replicable methodology relying on stem analysis data to fit robust site index (SI) models which make it possible to estimate the level of productivity of plantations according to their top-height and age. New SI models were calibrated for the three main plantation species in Wallonia: Norway spruce, douglas-fir and Larches. These new models make it possible to account for the significant changes ...