Hi @rohit_SB,
You have to address the duplicates at the analysis level by using level-aware calculations. Take a look at use case #2 in this article.