The BDI model, as a practical reasoning architecture aims at making decisions about what to do based on cognitives notions as beliefs, desires and intentions. However, during the decision making process, BDI agents also have to make background decisions like choosing what intention to achieve next from a set of possibly con icting desires; which plan to execute from among the plans that satisfy a given intention; and whether is necessary or not to reconsider current intentions. With this aim, in this work, we present an abstract framework which integrates a Possibilistic Defeasible Logic Programming approach to decision making in the inner decision processes within BDI agents.XIV Workshop agentes y sistemas inteligentes.Red de Universidades con Carreras en Informática (RedUNCI)