Designing a Recommendation Model Based on Tobit Regression, GANN-DEA and PSOGA to Evaluate Efficiency and Benchmark Efficient and Inefficient Units

The main purpose of this study is to design a privatized proposal model for Tavanir regional electricity distribution and transmission companies. This proposed model is based on Tobit regression, GANN-DEA and PSOGA to evaluate the efficiency and modeling of efficient and inefficient units. This three-step process is benefited a hybrid data envelopment analysis model with a neural network optimized by a genetic algorithm to evaluate the relative efficiency of 16 Tavanir regional electricity companies. To measure the effect of environmental variables on the average efficiency of companies, two-stage data envelopment analysis and Tobit regression were used. Finally, with a hybrid model of particle mass algorithm and genetic algorithm, we have modeled for efficient and inefficient units. The average of efficiency of regional electricity companies during the years 2012 to 2017 has increased from 0.8934 to 0.9147. And companies in regions 1, 2, 4, 5, 8, 12, 13 and 16 have always had the highest efficiency average (one). And the power companies in regions 10 and 11 with the average efficiency values of 0.7047 and 0.6025 had the lowest efficiency values.