Aimed　at　solving　the　forecasting　problems　in　Collaborative　Planning,　Forecasting　&　Replenishment　(CPFR).　this　paper　uses　grey　model　and　time　series　model　to　predict　the　demand　of　manufacturer　and　retailer　separately.　Combining　the　results　of　single　forecasting　based　on　induced　ordered　weighted　averaging　(IOWA)　which　can　empower　the　estimation　value　after　sorting　by　prediction　accuracy　at　each　point　of　time.　This　paper　sets　up　a　collaborative　forecasting　model　and　obtains　the　collaborative　forecasting　value.　The　practical　application　results　show　that　the　collaborative　forecasting　model　can　improve　the　precision　of　demand　prediction.　Moreover,　this　model　can　help　both　manufacturer　and　retailer　cooperate　with　each　other　more　closely　and　share　the　information,　thus　it　can　reduce　the　inventory　of　supply　chain　and　enhance　efficiency.