Akademik Veri Yönetim Sistemi
INTERNATIONAL JOURNAL OF INDUSTRIAL ENGINEERING-THEORY APPLICATIONS AND PRACTICE, cilt.28, sa.5, ss.477-489, 2021 (SCI-Expanded, Scopus)
This study has developed a machine learning-based framework to determine whether the planned hydroelectric power plants (HEPPs) are effective. First of all, the performance of HEPPs in Turkey was examined via DEA, and efficiency measurement was performed using the output-oriented BCC model. Then, classification models based on machine learning were established by using the obtained efficiency scores and input variables used in DEA. When identical comparisons were made using seven different classification models, REPTree was found to be the superior model. Finally, an interface based on the decision rules derived from RepTree was created to facilitate the use of the established model. With this interface, the HEPP's efficiency can be determined by the relevant inputs before a HEPP investment decision is made Thanks to this reasonable and intelligent framework, strategic decision support is provided to decision-makers in the field of energy.