Investigating the effects of superplasticizer and recycled plastics on the compressive strength of cementitious composites using neural networks

The potential application of neural network (NN) models to estimate the compressive strength ([Formula: see text]) of cementitious composites under a variety of experimental settings and cement mixes was investigated. The data were extensively collected from previous literature, and the bootstrap resampling tests were applied to estimate the statistics of the parameter correlations. We find that the NN model that involves the coarse and fine natural aggregates ([Formula: see text] and [Formula: see text]), superplasticizer ([Formula: see text]) and recycled plastics ([Formula: see text]) as the features can accurately predict the [Formula: see text] (R(2) ∼ 0.9), without the need to specify the type of [Formula: see text] and the structure of [Formula: see text] in advance. The developed NN model holds promise for revealing the global dependency of [Formula: see text] on these parameters. It suggested that increasing 100 kg/m(3) of [Formula: see text] could increase [Formula: see text] by ∼4 MPa, but the usage of [Formula: see text] more than 700 kg/m(3) could negatively affect [Formula: see text]. How the [Formula: see text] varying with [Formula: see text] is apparently nonlinear. Within the optimum limit, adding 1 kg/m(3) of [Formula: see text] could enhance the [Formula: see text] by ∼2 MPa. Contrarily, additional 1 kg/m(3) of [Formula: see text] results in a decrease of ∼0.2 MPa of [Formula: see text]. The mixture-type independent models developed here would broaden our understanding of the global influential-sensitivity among these variables and help save cost and time in the industrial applications.