Applying a combined method to evaluate the relative efficiency and rank the Insurance companies

Data Envelopment Analysis (DEA) have been used as an efficient tool for evaluating and benchmarking from years ago. A weakness of such approach is that the efficiency of units depends on the number of units under evaluation and the number of input and output variables. Therefore, while the number of variables is more, the analysis carried out has less power to distinguish between units. Hence, the number of variables should be decreased in this case to be utilized in DEA model. There are several approaches to reduce the dimensions of data that between them, Principle Component Analysis method (PCA) is a powerful approach for reducing the indices. It also can be placed in data envelopment analysis models area and using it, the indices with minimum volume and maximum data can be considered. But since we know in today’s world there are many uncertainties and on the other hand, it is not possible to specify the exact value of the indices in many organizations due to various constraints, so, it is better to consider the indices in interval. In this research, PCA methodology has been used in order to decrease the dimensions in interval data and finally, the upper and lower bounds of efficiency of the remaining units would be calculated, using the interval data envelopment analysis. Eventually, we will rank the decision making units, using the interval numbers ranking technique.