Efficiency Measurement for Non-homogeneous Units in a Fuzzy Environment

Abstract—Data envelopment analysis (DEA) assumes homogeneity among the decision-making units (DMUs) for measuring the relative efficiency of DMUs. In actual practice the efficiency scores generated by DEA may reflect the underlying differences in environments rather than true operational efficiency when DMUs do not operate under similar conditions. This work considers measuring the relative efficiency of non-homogeneous DMUs in a fuzzy environment. A two-stage analysis incorporating imprecise environmental factors into DEA model for measuring the relative efficiency of non-homogeneous units is proposed. A fuzzy regression analysis is employed for adjusting the data of DMUs to account for variations in the fuzzy causes of non-homogeneity. DEA with fuzzy residual errors is then performed to construct a scalar measure of efficiency for all DMUs. An example of the performance assessment of municipalities' solid waste recycling and disposal channels is included. This is the first attempt to estimate the relative efficiency for non-homogeneous DMUs in a fuzzy environment.