Predicting the Best Green Growth Performance With Integrated Intuitionistic Fuzzy and Metaheuristic Algorithms

Although various studies monitor and evaluate green growth performance, no heuristic-based hybrid studies test the reflection of green growth indicators. This study aims to identify the best green growth performance by using intuitionistic fuzzy and metaheuristic algorithms over green growth indicators, contributing to filling the gap in the field and providing inferences about green growth to decision-makers. For this purpose, the green growth levels of 31 selected countries were analyzed based on environmental and resource efficiency indicators, and the best result was predicted. The study was conducted in two phases. In the first phase, countries were ranked according to their green growth performance using intuitionistic fuzzy methods. In the second phase, the best performance value was estimated at different population levels using metaheuristic algorithms. The research results show that the renewable electricity generation variable is the most important criterion. Ireland (1.00), Switzerland (0.98), and Costa Rica (0.95) are the countries with the best green growth performance, respectively. In green growth performance estimation, the ANFIS-TLBO model (R² = 0.908; MAE = 0.196; SMAPE = 0.689; MSE = 0.050; RMSE = 0.223; MBE = 0.188) demonstrated the closest estimation accuracy to the real values. In this study, for the first time, a hybrid model combining the intuitionistic fuzzy method and a metaheuristic algorithm was tested and proposed for green growth performance assessment. With this originality, it is expected that the results of this article will make a significant contribution to the literature gap and serve as a guide for policymakers and researchers.