The　traditional　MPCA　model　takes　the　entire　batch　data　as　a　single　object,　and　it　is　difficult　to　reveal　the　changes　of　process　correlation　from　stage　to　stage.　Considering　that　multiple　phases　with　transitions　from　phase　to　phase　are　important　characteristics　of　many　batch　processes,　it　is　desirable　to　develop　stage-based　models.　However,　some　stage-based　monitoring　methods　may　occur　false　alarm　and　missing　alarm　at　the　beginning　and　end　of　each　stage,　because　the　hard-partition　and　misclassification　problems.　To　overcome　the　above　matters　flexibly,　a　novel　multiple　PCA　batch　monitoring　approach　using　fuzzy　clustering　soft-transition　is　proposed.　It　reduces　the　false　alarm　and　missing　alarm　for　batch　process　in　on-line　monitoring　due　to　batch　variation.　The　proposed　method　is　applied　to　detect　and　identify　faults　in　the　well-known　simulation　benchmark　of　fed-batch　penicillin　production.　The　simulation　results　demonstrate　the　effectiveness　and　feasibility　of　the　proposed　method,　which　detects　various　faults　more　promptly　with　desirable　reliability.　©　2010　IEEE.