5 Replies Latest reply: Mar 13, 2018 1:32 AM by Prashant Sangle RSS

    I need some help.

    Mario Francisco Celleri

      I have information about the productive process of our business.

      Each table keeps specific information of the productive process that corresponds to it; for example, in table1 stores information about process1, table2 about process2, etc.

      Table 1
      roll number, date in which it went through the process, cod_machine, cod_employee, weight

      Table2
      roll number, date in which it went through the process, cod_machine, cod_employee, weight

      Table3
      roll number, date in which it went through the process, cod_machine, cod_employee, weight


      Table4 (Machine)

      cod_machine, machine, etc


      Table5 (Employee)

      cod_employee, employee name, etc


      I have created a dimensions of, machine, employee name, etc.


      It could generate a single table that contains the information of all the processes, or it is better to have it separately.

      How can I get for each process how many rolls (products) were processed?

      How can I get for each machine how many rolls (products) were processed?
      How can I get for each employee how many rolls (products) were processed?.


      And others....


        • Re: I need some help.
          Stefan Wühl

          All your process tables seem to show the same structure, right?

           

          I would concatenate all process tables into one fact table then (either using CONCATENATE LOAD prefix or this should be done automatically by Qlik when loading your tables. Add a new field that identifies the process:

           

          Processes:

          LOAD *,

                    'Process 1' as process

          FROM Table1;

           

          CONCATENATE (Processes)

          LOAD *,

                    'Process 2' as process

          FROM Table2;

           

          etc.

          • Re: I need some help.
            Prashant Sangle

            concatenate 1st 3 table with flag field. and other 2 table keep on association.

             

            try like

             

            Process:

            Load *,'Process 1' as ProcessName from Table1;

            concatenate

            Load *,'Process 2' as ProcessName from Table2;

            concatenate

            Load *,'Process 3' as ProcessName from Table3;

             

            MachineDetails:

            Load * from Table4;

             

            EmpDetails:

            Load * from Table5;

             

            In front end

             

            take table

            1:

            dimension: ProcessName

            expression: count(Products)

             

            2:

            dimension: Machine

            expression: count(Products)

             

            3:

            dimension: EmployeeName

            expression: count(Products)

             

            Regards,

            • Re: I need some help.
              Mario Francisco Celleri

              Thank you for your answers.

              Each table has different information, however the fields that I indicated in my example are common in each table; of course it´s with different names.

              Does the concatenate of the tables work if they have different information?, a join works?

              What happens for example if in a table I have two fields that refer to the machine and also to the employee ?, in that case how does qlik sense understand which of the fields should be considered?


              For example in the process table2 (this refers to the quality control), I have a field that tells me in which machine i was dried the cloth, and in another field it tells me in which machine the quality was checked.

              When you want to filter by machine that dried the fabric or in which machine the quality was checked; How would he do it?

                • Re: I need some help.
                  Stefan Wühl

                  You could use something like a canonical dimension:

                  Canonical Date

                   

                  Basically you load your records twice, concatenating the two fields into one. Either do this in your fact table (duplicating the records) or create a separate table to link the canonical dimension to your facts.

                   

                  Latter is probably better if your fact table is containing a lot other fact fields.

                   

                  HIC is demonstrating it for a canonical date dimension, but you can do the same for machines.

                  Just tag the machines in the canonical machine table with the usage (create a new field to distinguish 'washing', 'drying' etc.)

                  If you have a lot of these fields you need to concatenate in one of your tables, you can also use

                  The Crosstable Load

                  • Re: I need some help.
                    Prashant Sangle

                    for more understanding can you explain with sample data???

                     

                    Regards,