The　new　product　development　(NPD)　problem　is　of　great　importance　for　company　profits,　even　its　survival.　Generally,　there　are　multiple　factors　to　assess　the　popularity　of　the　product.　The　main　challenge　of　the　problem　lays　in　that　the　modeling　and　decision　making　process　must　be　open　for　human　involvement　considering　these　factors　simultaneously.　In　another　word,　experts　and　decision　makers　must　be　able　to　understand　it　so　that　(1)　the　knowledge　and　experience　of　experts　and　decision　makers　can　be　integrated,　(2)　the　assessment　results　can　be　accepted,　and　(3)　certain　adjustments　on　the　initial　product　development　plan　can　be　made　as　a　feedback.　In　this　study,　the　Belief　Rule　Base　(BRB)　is　applied　to　solve　the　product　development　problem.　The　BRB　under　the　disjunctive　assumption　can　help　further　downsize　BRB　as　well　as　maintain　a　high　modeling　accuracy.　Moreover,　a　disjunctive　BRB　parameter　learning　model　and　an　optimization　algorithm　are　applied　as　well.　A　lemonade　product　case　is　studied　in　a　comparative　fashion　to　validate　the　efficiency　of　the　proposed　approach.