Novel Evolutionary-Optimized Neural Network for Predicting Fresh Concrete Slump

Accurate prediction of fresh concrete slumps is a complex non-linear problem that depends on several parameters including time, temperature, and shear history. It also affected by the mixture design various ingredients. This study investigates efficiency three novel integrative approaches for predicting this parameter. To end, vortex search algorithm (VSA), multi-verse optimizer (MVO), shuffled evolution (SCE) are used to optimize configuration multi-layer perceptron (MLP) neural network. The optimal complexity each model was appraised via sensitivity analysis. Various statistical metrics revealed accuracy MLP increased after coupling it with above metaheuristic algorithms. Based obtained results, error decreased up 17%, 10%, 33% applying VSA, MVO, SCE, respectively. Moreover, SCE emerged as fastest optimizer. Accordingly, explicit formulation SCE-MLP introduced capable practical estimation slump, which can assist in project planning management.