Tree Based Approaches for Predicting Concrete Carbonation Coefficient

Carbonation is one of the critical durability issues in reinforced concrete structures terms their structural integrity and safety may cause fatal deterioration corrosion steel reinforcement if ignored. Many researchers have performed a considerable number studies to predict carbonation structures. However, it still challenging depth or coefficient, as they depend on various factors. Therefore, creating model that can learn from available data using Data Driven Techniques (DDT) step forward this research field. This study provides new approaches coefficient through Model Tree (MT), Random Forest (RF) Multi-Gene Genetic Programming (MGGP) approaches. With 827 case studies, predicted models be seen function set conditioning factors, which are statistically significant explaining mechanism. The results obtained MT, RF MGGP were compared with those Multiple Linear Regression (MLR), Artificial Neural Networks (ANNs) (which previously developed). reveal perform better than previous models. Moreover, MT technique displays its output series equations, multiple trees form single equation, more user-friendly applicable practice.