Comparison of NDT Data Fusion for Concrete Strength using Decision Tree and Artificial Neural Network

Fusion of Non-Destructive Test (NDT) data results in more accurate estimation concrete strength when compared to any single NDT data. Estimation from assumes importance for health assessment and evaluation existing buildings, particularly those near the end their design life. Application machine learning tools response surface method has found popularity recent years this purpose. In study, universally popular Artificial Neural Network (ANN) relatively un-explored Decision Tree (DT) are applied estimate rebound number ultrasonic pulse velocity collected literature, combined forms. A ranking system based on ratios multiple performance measures was demonstrated cases where different models adjudged better considering measures. From results, it concluded that fusion resulted accuracy, both ANN DT. Comparing selected as well ranks two tools, were perform DT models. The narrow range metrics obtained three divisions (into modelling sets) all imparted confidence robustness approach model development adopted study.