Do you have only one value for Current_ROL_Week or does the value change based on dimension? If there are multiple values, I think you will have to use Aggr() function with if statement instead of using set analysis to get this done. Would you be able to provide few rows of data with the expected output to help us understand what you are trying to do?
With strings are no "><" comparings possible - only "=" or "like". This meant in your case that you would need to create a list of values for this data-set comparing within the set analysis, maybe something like this:
concat(distinct if(num(keepchar(Current_ROL_Week, '0123456789')) <=
num(keepchar(rol_week, '0123456789')), rol_week), chr(39) &', ' & chr(39)) & chr(39)
Better and a lot easier than this would be to double your week-field within the datamodel with pure numeric content like:
1,2,3,4..... And this is a general recommendation to use only pure numeric fields (without any formattings) for calculating or comparing.