The　results　of　carbon　footprint　assessment　of　products　depend　on　the　selection　of　data　types,　sources,　assessment　approaches.　The　purpose　of　this　study　is　to　develop　a　new　method　which　is　combined　DOI-Monte　Carlo　with　sensitivity　analysis　and　data　quality　analysis　method　for　carbon　footprint　assessment　of　products.　For　this　new　approach,　firstly,　the　primary　data　impacting　on　the　assessment　result　were　chosen　through　data　sensitivity　analysis;　then,　with　DOI-Monte　Carlo　analysis　the　uncertainty　of　the　primary　data　and　the　key　data　that　affect　the　evaluation　results　were　obtained.　As　a　result,　the　accuracy　of　carbon　footprint　assessment　could　be　improved　more　specific　by　optimizing　data　collection　scheme　according　to　the　above　data　analysis　method.　As　a　case　study,　the　developed　method　was　applied　to　the　carbon　footprint　assessment　in　the　pre-printing　stage　of　one　plastic　flexible　packaging　printing　company　in　China.　This　approach　can　be　used　for　carbon　footprint　assessment　of　many　products　by　improving　the　uncertainty　and　data　quality.