This　paper　proposed　a　product　cost　estimation　method　based　on　principal　component　analysis　(PCA)　and　artificial　neural　network　(ANN)　for　generalized　modular　design　of　machine　tool.　In　the　first　stage,　PCA　was　applied　to　identify　the　principal　components　of　product　modular　features,　which　was　conducted　by　analyzing　the　product　cost　components　and　their　influencing　factors　driven　by　features　of　modules　firstly,　and　then　by　calculating　the　eigenvalue　and　eigenvector　of　correlation　coefficient　matrix　to　reduce　the　dimension　of　the　data,　later　by　defining　the　first　few　principal　components　which　contain　most　of　the　feature　variables.　In　the　second　stage,　the　mapping　from　the　restructured　product　modular　feature　to　the　product　cost　was　established　by　general　regression　neural　network　(GRNN).　At　last,　the　simulation　results　demonstrate　that　the　proposed　algorithm　is　effective　and　speedy.　©　Springer-Verlag　Berlin　Heidelberg　2012.