The increasing integration of renewable energy sources (RES) and the transition towards a decarbonized energy sector present significant challenges, particularly in demand-side management. Thermal energy storage (TES) systems offer a cost-effective solution for enhancing energy flexibility in building heating systems. However, improper sizing and operation of TES systems can lead to increased investment costs and energy losses. To bridge this gap, this study proposes a novel, optimization-based framework for the systematic sizing and operation of TES systems. The methodology encompasses two key components: (1) an innovative TES sizing framework that integrates system modelling and optimization-based sizing leveraging historical thermal load data; (2) validation and performance evaluation of the sizing outputs through building energy simulations across three diverse building types and climatic conditions. Key findings demonstrate the framework's ability to adapt to various scenarios, achieving operational cost reductions of up to 35 % and significantly enhancing the energy flexibility in terms of flexibility factor by up to 1.03. Furthermore, the proposed framework is shown to effectively optimize TES capacities to unique building load patterns. These results highlight the framework's potential as a robust tool for optimizing TES in buildings, contributing to flexible and cost-efficient energy systems.