MACHINE LEARNING FOR PARAMETRIC COST ESTIMATION OF AXISYMMETRIC COMPONENTS

Abstract Machine learning (ML) is a well-established research topic in Industry 4.0 boosting its adoption. ML also used for manufacturing cost estimation during design. Such approaches are commonly to estimate the of mass-produced parts. Many consolidated historical data available training regression models. Unfortunately, very often, such database not available. The paper defines an approach parametric axisymmetric components. model derives from automatic software analytically estimating cost. With proper set simulations, tool can generate large amount training. presents steps developing using ML. based on CRoss Standard Process Data Mining method. proposed method was develop one (to total that considered raw material and cost). obtained Relative Error 23.52% ± 1.37%, coherent with E2516 − 11, Classification Cost Estimate System.