You can read defect counts for subgroup samples from a DATA= data set specified in the PROC SHEWHART statement. Each process specified in the UCHART statement must be a SAS variable in the data set.