On the environmental performance analysis: A combined fuzzy data envelopment analysis and artificial intelligence algorithms

Greenhouse gases (GHG) remain in the atmosphere for a very long-time causing alarmingly fast warming worldwide (global warming); especially Carbon dioxide (CO₂) emissions have become a worldwide concern because of their harmful effects on the climate, and they are considered as an undesirable product of a lot of production systems. Various models dealing with undesirable outputs for measuring environmental efficiency have been employed to control greenhouse gas emissions via forecasting and/or optimizing their emissions. In this regard, this study proposes a novel modified Fuzzy Undesirable Non-discretionary DEA (FUNDEA) model to Measure environmental efficiency, and combine it with some novel artificial intelligence algorithms (Artificial Neural Network (ANN), Gene Expression Programming (GEP) and Artificial Immune System (AIS)) in order to predict optimal values of inefficient Decision-Making Units (DMUs) for being more efficient and mitigating their Co₂ emissions in the uncertain environment for the first time herein. The model is applied to a dataset of 24 Iranian forest management units. Although our findings show that 17 DMUs are inefficient with a weak efficiency dispersion, these inefficient DMUs could improve their efficiency border by following the combined approaches (FUNDEA-ANN, FUNDEA-GEP and FUNDEA-AIS). As a consequence, the applied FUNDEA- artificial intelligent approaches are performed very well in predicting the optimal values of CO₂ emissions and, hence increasing the total environmental efficiency.