Software Process Models…..

Six Sigma – iSixSigma Forums Old Forums General Software Process Models…..

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
  • #36632


    Hi all,
    Can any one help me clear my doubts regarding the following question?1.What are the objectives of Process Models?Can we find out specific ways to to measure the effectiveness of process models? If yes, how?2. In a given scenario, how can we select one process model over other? Means what are the advantages and disadvantages of different process models? And how can we select the most appropriate model for a given situation? Thanks n Regards,


    Bob J

    You may want to post your question on the Software/IT forum…  You may get more responses to contrast with my opinion below…
    Software models are designed to structure the development of software.  Typically their effectiveness is primarily measured by how quickly they can facilitate the development of software that is fully functional and (relatively) defect free.
    A lot of the decisions driving the selection of the appropriate process model stem from the type of product you are trying to develop.  A good general reference detailing many of the common models and associated advantages/disadvantages is “Metrics and Models in Software Quality Engineering” by Stephen H Kan. 
    Hope this helps….
    Best Regards,
    Bob J


    John J. McDonough

    As Bob J points out, in the software development world, process models are most often used during development.  However, just like in any other field, process models can help in a variety of activities, and in the Six Sigma context, we are typically talking of process improvement.
    It sounds as if you are looking for a modeling methodology to choose. However, I submit that the choice of any particular diagramming convention is not nearly as important as the discipline of documenting the model in the first place.
    More signficant is the choice of WHAT to model. In the software environment, we could model the functions, we could model the data, we could model the actual operation of the software, we could model the network interactions, or we could model the business process that the software supports.  Any of these could give us insights that we might not otherwise have, and the choice of what to model will significantly influence what we learn from the model, probably far more than the choice of particular symbology.
    If you are looking at new development, you may well want to do each of these models.  For process improvement, yoy may well get away with only one, depending on what you are trying to improve.
    Don’t let yourself get overly fixated on the choice of tool.  Look at the problem, decide what you need to learn, then decide what kind of model to generate.

Viewing 3 posts - 1 through 3 (of 3 total)

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