STRENGTH PREDICTION OF RC INTERIOR BEAM-COLUMN JOINT AND CONTRIBUTION ANALYSIS OF DESIGN FACTORS USING MACHINE LEARNING

This paper reports a machine learning model to predict moment capacity of interior column-beam joints learning from a database of previous experimental results. Contribution analysis of design factors with the model reveals the effects not previously noted, such as that even though they are equivalent in terms of increasing the tensile force of the reinforcing bars, increasing the cross-sectional area and increasing the strength have different effects and that there is a limit to the effect of increasing the strength.