Against　the　background　of　new-generation　geometrical　product　specifications　(GPS),　the　respresentaion　of　skin　model　by　using　discrete　mathematics　is　focused　in　this　paper.　Considering　geometric　constraint　requirements　and　tolerance　specification,　to　achieve　skin　model　simulation　with　random　deviations,　Markov　chain　Monte　carlo　(MCMC)　simulation　techniques　and　a　global-modeling　approach　based　on　principal　component　analysis　(PCA)　are　investigated.　Statistical　shape　analysis　(SSA)　model　and　a　local-modeling　approach　based　on　augmented　Darboux　frame　are　utilized　here　for　the　simulation　of　the　skin　model　with　systematic　deviations　which　are　produced　in　the　manufacturing　process.　The　concept　of　the　mean　skin　model　and　its　robust　statistics　are　also　introduced.　In　the　end,　a　case　study　based　on　a　part　is　applied　here　and　simulation　results　of　its　skin　model　validate　the　correctness　and　applicability　of　the　proposed　method.　©　2016,　Science　Press.　All　right　reserved.