iSixSigma

Gage RR

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • #38419

    Nagaraj Kini
    Participant

    I am working in an IT industry where we provide the support in problem resolustion of the running client applications We are doing a project to improve the SLA performance in this support project. The system is such that whenever client faces a problem he logs the problem into another software with the details of the problem & by what date he needs the solution. Also as soon as client select the problem type, a predefined SLA violation date will be displayed. Our team work on those problem & try to solve the problem with in the date needed by the client as well as the predefined SLA violation date. We generate a weekly report from this application, to understand the effeciency of the process & the success rate. My question is whether we need to carry out R&R for this? As the process is digitized & no manual interaction is there , my understanding is that there is no need to carry out the repeatability & reproducibility study. Please clarify

    0
    #114933

    Anonymous
    Guest

    Nagaraj,
    It is well known that the NHS fiddles patient waiting time measurements. The situation is avoidable if only measurements were ‘calibrated’ and audited on a regular basis.
    Now if the measurement of ‘waiting time’ had been treated as a metrology system, a member of the quality department would have seen from scatter diagrams of a hospital reported waiting time versus the family doctor’s recording of the waiting time the measured variance was extremely high, as much as one year, and many patients could have avoided being put on and off waiting lists and spending hours in ambulances parked in the hospital parking lots.
    I hope this will give you some fresh ideas …
    Regards,
    Andy

    0
    #114934

    Nagaraj Kini
    Participant

    Andy,
    I agree with you on patient waiting time measurements in hospitals that we need to ‘calibrate’ and audit as the data collection system or data recoridng system is a manual process. In my case, the date & time are capured by the application automatically, i.e. once the problem is logged,application takes the date & time as the system date.Our team work on those problems & problem resolution activity is carried out. Once the client says his problem has been resolved, our team here close the log & application takes the date & time from the cuurent system date. Also we send the problem log report to client & he agrees to the time taken for resolving the problem. Do you feel, still R&R study is required & if yes, how can we carry out that?

    0
    #114935

    Anonymous
    Guest

    Nagaraj,
    I would still ask your customer directly whether or not they agree with the ‘recorded’ downtime – especially when the delivery of a ‘solution’ doesn’t necessarily always fix the problem,  in which case a ‘new case’ might be recorded!
    Automatic test systems also pose problems for Gage R&R. I don’t particularly like the test, but it does provide a quick ‘sanity’ check.
    I believe any characteristation of a metroloy system should cover the entire measurement range, and not just where the process operates, and ideally it should relate to a ‘national standard’ or avmanufactured standard does not exist – for example photo resist dimensions.
    The there is the issue of sensitivity and linearity that Gage R&R does not address.
    Of course, the reason for the interest in the measurement variation is to ensure that when you make a decision it is correct. If it is not correct, then some corrective action is required.
    My advice would be to calibrate your ‘system’ with known standards and don’t assume anything!
    Regards,
    Andy
     

    0
    #114971

    Nagaraj Kini
    Participant

    Thanks Andy.
     

    0
Viewing 5 posts - 1 through 5 (of 5 total)

The forum ‘General’ is closed to new topics and replies.