PARAMETRIC COST MODELLING OF COMPONENTS FOR TURBOMACHINES: PRELIMINARY STUDY

Abstract The ever-increasing competitiveness, due to the market globalisation, has forced industries modify their design and production strategies. Hence, it is crucial estimate optimise costs as early possible since any following changes will negatively impact redesign effort lead time. This paper aims compare different parametric cost estimation methods that can be used for analysing mechanical components. current work presents a methodology which uses non-historical data database population. settled using should obtained from analytical models implemented in software. Then, compares modelling techniques (artificial neural networks, deep learning, random forest linear regression) define best one industrial Such have been tested on 9 axial compressor discs, dimensions. by considering other materials batch sizes, was reach training dataset of 90 records. From analysis carried out this work, conclude machine learning are valid alternative traditional regression ones.