In recent decades, the conceptualisation of translation competence and its development has attracted significant attention from translation researchers. Existing literature on translation competence is characterised by the prevalence of multi-componential models of translation competence, with inadequate attention paid to the interplay between competence components in the translation process. Therefore, this doctoral research sets out to re-conceptualise translation competence from a problem-solving perspective so as to understand translation competence and its development in the translation process. By re-defining and re-modelling translation competence with inspiration from problem-solving studies, this research proposes a conceptualisation of translation competence and its development that accommodates the translation process and the learning process. In order to validate the relevance of the proposed theoretical framework, a longitudinal study was conducted among a small group of Chinese students from an MA translation programme, using the translation task-based interview as the main research instrument. Findings from the empirical study have demonstrated the relevance and strength of the theoretical framework as well as revealing individual and shared paths of translation competence development among the learners. This research enriches the current understanding of translation competence and its development. It introduces a fresh perspective for conceptualising translation competence, proposes an effective instrument for empirical competence research, and identifies possible directions for further research. It also has practical implications for translation pedagogy, offering theoretical and empirical support for some recent approaches and trends in translator education and training.