LAUSR

Compressive strength is an important and pertinent property for the characterization of concrete. It is known that compressive strength of hardened concrete depends on the efficiency factor (K) of silica fume and its percentage replacement (%SF). Consequently, the ultrasonic pulse velocity (UPV) is also an important factor for predicting compressive strength. The article examines the capability of relevance vector machine (RVM) and emotional neural network (ENN) for determination of compressive strength using the efficiency factor, percentage silica fume replacement and UPV as an input parameter. A comparative study has been carried out for the developed model using RVM and ENN. The analysis confirms that both RVM and ENN are a robust model for the prediction of compressive strength of hardened concrete.