Data Envelopment Analysis (DEA) is a non-parametric linear programming-based technique used for evaluating the relative efficiency of homogenous operating entities on the basis of empirical data on their inputs and outputs. It is suitable in cases where other approaches do not provide satisfactory results. In the first chapter, we introduced the basic DEA model, the CCR model, and through the dual problem we discovered sources of inefficiency that were not detected through the primal problem, input and output slacks. DEA identified efficient bank branches as benchmark members and inefficient bank branches. Sources and amounts of relative inefficiency, which were identified in each input and output, establish guidelines for needed improvements. In the second chapter we introduced ”two-stage” DEA, model in which, except inputs \(x_i\) and outputs \(y_j\), we use intermediate measures \(z_d\) as well. For every DMU, the first stage uses inputs to generate outputs which become the inputs to the second stage. The first stage outputs are therefore called intermediate measures. The second stage then uses these intermediate measures to produce outputs. The efficiency of the whole process can be decomposed into the product of the efficiencies of the two sub-processes. In the example with 24 non-life insurance companies we compared efficiency of the relational model and the independent model. In the independent model efficiencies are calculated independently by using basic CCR model from the first chapter. The results show that the independent model may produce unusual results for several companies while the relational model always produces meaningful and reliable results for all companies.