Using the Process Sigma Calculator given here..

Six Sigma – iSixSigma Forums Old Forums General Using the Process Sigma Calculator given here..

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
  • #65828


    kalyani, as far as my knowledge goes,

    if you are referring a LOC as a line containing
    various ‘programming codes’ then you
    have to evaluate ‘in what all ways you can commit a
    mistake in that particular line’
    this will be called as opportunities in that line,
    since you are talking about software
    all the LOC will be having varying number of
    opportunities to go wrong, sum of all
    the opportunities will be called as TOP or total

    Hope this helps
    You can also download sigma calculator, in, that is
    more eloborate in my purview.



    I refer to my reply to your earlier posting on DPMO use in software.

    Be careful in using line of code as the unit for opportunity for defect – this is generally too low a level and too distant to the customer. A better unit is ‘program’ or ‘function’ and opportunities are such things as ‘technical failure’, ‘inappropriateness’, ‘unexpected outcome’ or ‘incomplete performance’

    If you do use LOC as the unit, then the opportunities will range from 1 (which is simple and many believe the only true answer) to something like 2 or 3 – write a line when it is not needed (waste), write a wrong line (failure), write something that is not accepted by the user (scrap).

    To get technical – if you use a compiled language then ‘error in syntax’ is not really an opportunity as this will be detected by the compiler (although you are still wasting time and effort in writing the wrong stuff). However if you use an interpreted language, then ‘error in syntax’ is yet another opportunity, as it can get through to the end user. The aim is to arrive at something that reasonably reflects the level of defect compared to opportunities, and there is never a right or wrong answer for this.

    For example, if you are writing in c++, and you count EVERY line of code written – FIRST TIME – then opportunities per line could be:
    * failed to compile (two syntax errors per line count as two defects)
    * failed to perform in module (every distinct line-edit counts as a defect)
    * failed to perform for user specification (every test failure and line-edit counts as a defect)
    * unwanted code (every line removed from the final code counts as a defect)
    * non-executed code (every line never used counts as a defect)

    Say I write 100 lines of code. I get three syntax errors and make three edit changes. The code goes for testing and we find two faults, requiring five edit changes. The code goes for user testing and the user changes their mind, requiring 30 lines to be re-written. One more syntax error, two more testing errors and the user had a further 5 changes needing 10 edits, and we scrap 15 lines of unwanted code. In use it is discovered that another 3 lines are never called.

    100 units, 5 opportunities per unit

    3+5+30+1+2+10+15+3 defects = 69

    DPMO = 1000000 x 69 / 500 = 138,000 or 2.6 sigma

    Of course, those 100 lines of c++ could be just one line of SQL, or you could have re-used 100 lines from a standard l




    I was trying to use the “Calculate your Process Sigma” link on this site. The advanced mode has Two text boxes : Total Units and Opportunities per Unit,. I want to understand the meaning of these two terms. If I am developing software and my unit is Line of Code (LOC), then what is the “Opportunities per Unit” ?

    Thanks in Advance




    Wrong! Opportunities are not based on the number of mistakes that can be made. They are based on th number of value added things that must go right.

    Your unit is the piece of software, younwould get one opportunity for each line of code.

Viewing 4 posts - 1 through 4 (of 4 total)

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