AB Ülkelerinin Çevre ve Atık Yönetimi Performanslarının Değerlendirilmesi: Veri Zarflama Analizi ve Yapay Sinir Ağlarının Birlikte Uygulanması.

With the development of industry, the increase in human needs has led to an increase in production and consumption. The inadequacy of waste recycling in the production-consumption-waste cycle chain brings environmental, climatic, economic and many problems together. Within the framework of the 2015 Circular Economy Action Plan, the European Union has taken many decisions, including waste management. Improvement of recycling capacity and waste management for EU member and candidate countries is at the top of the targeted strategies. In this study, after the 2015 Circular Economy Action Plan, the waste management performances of EU countries for the years 2017, 2018 and 2019 were examined with the Artificial Neural Networks method based on Data Envelopment Analysis (DEA). DEA input variables; municipal waste generation per capita, import of waste for recycling, national expenditures for environmental protection, real GDP per capita, human development index, plastic packaging waste production per capita, output variable; taken as the recycling rate of municipal waste. Efficiency scores obtained with DEA were used as output variables in artificial neural networks, and predictions were made for the efficiency scores of countries in 2019 with artificial neural networks feed-forward networks. In the findings, it has been concluded that Slovenia, Latvia, Germany, Ireland, Luxembourg and Belgium are the countries that are active in all years and are frequently included in the reference clusters of other countries. It has been observed that the 2019 prediction values obtained from the data set trained with Feed Forward Artificial Neural Networks are very close to the real efficiency values. [ABSTRACT FROM AUTHOR]