A growing body of empirical literature has attempted to measure the efficiency of banking sector using Data Envelopment Analysis (DEA) by focusing on different aspects of banking services. However, standard DEA models often fail to sufficiently discriminate between efficiency scores of banks particularly with small sample size. Moreover, sometimes knowledge about different banking operations is available that needs to be incorporated in the evaluation method to assess their impact on the performance of banks. This research deals with the efficiency evaluation of banking sector through DEA based on additional information about multiple banking operations without which efficiency is generally overestimated. The main objective of this thesis is to develop a better informed DEA model that is capable of incorporating additional information about different bank specific characteristics by overcoming the problem of poor discrimination. For this purpose, the current study has proposed a novel methodological integration of DEA with production trade-offs in banking context and named it “DEATOB Framework”. This framework is universal in nature and can be applied to banking sectors of other countries. The study also aims to provide the empirical application of DEATOB Framework for which a sample of 29 commercial banks of Pakistan is selected. The results indicate that this framework evaluates banks on the basis of additional characteristics and provides better discrimination between good and bad performers as compared to the standard DEA model. The final objective is to extend the proposed framework to other banking models. For this purpose, the profitability model is chosen considering the profit maximization goal of banks and a separate PDEATOB Framework is developed. An empirical application of this framework is also provided to demonstrate its workability. This thesis also provides an insight on scale efficiency and relationship of efficiency with the banks size and ownership after application of the proposed frameworks.