Jump to content

Difficulty in measuring accuracy...

Recommended Posts

Hi Team,


Need your advice.... how we can measure Accuracy of the SLA... sounds very simple...


There is an Client SLA where team has to maintain 99% accuracy for each cases...


Team receives different cases from client where each cases is different from the other, CTQ wise certain CTQs are similar to each cases and some are different. So we have different opportunities for each cases....


In this situation we cannot have constant value for each cases as denominator to measure accuracy..


Below mentioned are the few samples taken from the monthly volume data... where we identified the actual opportunities from each cases....


Day        Volume/Cases  Total Opportunities

Day 1     123         98

Day 2     143         22

Day 3     109         62

Day 4     112         29

Day 5     141         40

Day 6     121         62

Day 7     132         63

Day 8     145         81

Day 9     138         88

Day 10   104         22

Day 11   144         81

Day 12   115         38

Day 13   135         21

Day 14   138         65

Day 15   123         95


On a monthly basis the volume or cases used to be in an average of 1456 cases.... how we can measure its accuracy.... (client target 99%)


Challenging part is to identify opportunities count for each cases as it is one of the time consuming activities....how we can resolve such constraints....


Hope you understood the concern....

Link to post
Share on other sites
  • 2 weeks later...
  • 2 years later...

Dear Rajiv

Hope this helps if iam not late.


When we visit "Subway" for a roll. We have these chefs work on each roll different depending on the request. Well the CTQs are similar in all of their delivery, yet the taste varies depending on the activities while making the roll. If he adds too many vegetables, or sausages or the meat or underheat the bread - the accuracy with the taste falters.

I find your challenge similar to a typical manager's worry on ensuring the right taste of each roll as per their brand( in your case the SLAs )


What I can think of is if we break the activities of these cases as they have similar CTQs into segments( sequential or parallel )/ work blocks and check for the accuracy ( based on the criteria set for each segment) on a random basis and then extrapolate the entire work - we might have an answer.


I can also recommend the structure followed by the famous " Dabbawalas" of Mumbai. They break it down into stages, track it and ensure the delivery to the right candidate.

Better watch out for Rolled throughput yield.


Share your thoughts and good luck.



Nagraj Bhat


Link to post
Share on other sites
  • 10 months later...
On 6/13/2014 at 1:44 PM, Rajiv. said:

Team.... can you please suggest.... how can I resolve above process constraints......


Hi Rajiv, 


Opportunities need to be counted each time if you wish to use the DPMO method. This will lead to Total Defects out of Total Opportunities over a period of time being utilised for accuracy assessment. 


If opportunity count is time-consuming and you want to use a method of counting defectives (instead of defects), you have an option. Using this approach, you will lose the level of detail that is needed for analysis and improvement but can give you a reasonable idea of performance. 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Who's Online (See full list)

    There are no registered users currently online

  • Forum Statistics

    • Total Topics
    • Total Posts
  • Member Statistics

    • Total Members
    • Most Online

    Newest Member
    Mahtab Khan
  • Create New...