This study intends to conduct a comparative test of efficiency of the 3 merged state-owned banks using 5 merged Simulation models which will be measured by the 2-stage Data Envelopment Analysis (2-stage DEA) method with Islamic banks and also conventional in Indonesia. This study uses a quantitative approach using the 2-Stages Data Envelopment Analysis (DEA) method. By using in the first stage this method can measure the efficiency score generated from the input and also the output that has been determined, among others, by inputs namely Savings, Administrative Costs, Personal Costs. The outputs used are Financing and Income. In the second stage the method used is Tobit Regression which will see whether there is an influence between the dependent variables determined on the efficiency score. The dependent variable consists of 2 types, namely macrobanks consisting of GDP, Inflation, and Interest Rate. While the other is the Microbank variable including NPF, FDR, and ROA. In this study, 5 combination simulations will also be examined, which will see which combination is the most efficient. From the results of the study, it was found that the efficiency of Islamic banking as a whole is still below the efficiency of conventional banking. Meanwhile, by using 5 simulations of the merger planned by the government, namely combining BRI Syariah, BNI Syariah, and Mandiri Syariah, it was found that the merger of the 3 banks is the most optimal option in terms of efficiency. In this study also found that the dependent variable that affects the efficiency score is the ROA variable. Meanwhile, for other variables, namely NPF, FDR, GDP, Inflation, and also Interest rate does not significantly affect the efficiency score.