Why this visual is showing extra rows

[Anonymous]

I have a table 'Sales' (3935 rows) with columns: 'SalesOrder','SalesOrderLine','ProductionOrderKey'. ProductionOrderKey is unique.

I have another table 'Production' (8324 rows) with columns: 'ProductionOrderKey','ProductionOrder'. Both columns are unique.

The last table is 'Operations' (69,159 rows) with columns: 'ProductionOrder','Operation','TimeTaken'. None of the columns are unique. TimeTaken is a numeric column.

The relationships between tables are as follows:

'Sales[ProductionOrderKey]'<-'Production[ProductionOrderKey]'

and 'Production[ProductionOrder]'->'Operations[ProductionOrder]'.

I have created a measure 'SumTimeTaken' = SUM('Operations'[TimeTaken])

I am creating a table visual with Sales[SalesOrder], Sales[SalesOrderLine], Production[ProductionOrder], SumTimeTaken.

However the issue is, when I add SumTimeTaken measure to visual, many duplicate rows get created for SalesOrder. I expect that for one ProductionOrder the TimeTaken is summed from Operations table. Attached below is the result I am getting.

How can I solve this? 

[amitchandak]

@Anonymous , I doubt there is a join issue between Table Operations and the table of unsummarized columns.

Ideally sales Order number and Line position should come from common dimension table

[Anonymous]

Yes, they SalesOrder and SalesOrderLine come from 'Sales', which is a dimension table