iSixSigma

Cycle Time Targets

Six Sigma – iSixSigma Forums Old Forums General Cycle Time Targets

Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
    Posts
  • #38273

    Bambace
    Participant

    Ok now I really the experts assistance.
    I have a process with multiple gates in the process.   Let say that I have 4 major steps that equate to my wing to wing cycle time.    I have calculated my span and median but I am being asked to determine what the target is.    At what point does the transaction become a defect?   Based on input from process experts they feel a perfect transaction takes around 10 days. If I was to throw out the number of 10 days being my target and anything over 10 days is a defect then the question senior management will ask is ” Well how did you come up with 10 days as your target?”
    How would I use data to determine my target?   How do I use data to determine when something becomes a defect as opposed to work-in-progress
    Hopefully someone can help!!!!!!!!!

    0
    #114375

    Jessica,
    You’ve identified a common problem that is also found in manufacturing. Consider a tolerance of 50 microns +/- 5 microns – is a part with a 52 microns thickness really of superior quality to a part of 49 microns thickness? Of course not; it is sheer nonesense. Therefore, the part with a 51 micron thick film is not necessarily a defect: So how does one solve this problem?
    The first step is to find out how to design a single tolerance. I am currently writing an article on this subject which I’ll publish in the forum.
    The second step is to consider the ‘capacity’ of your process system, because a target must be set by the capacity.
    To my mind there are two ways to work out capacity, the first is to use queuing theory – a Monti Carlo modelling system such as Simul8. The other way is to use Ikko Nagashi, or single flow, where the tack time corresponds to a theoretical standard time for each step.
    Good luck,
    Andy

    0
    #114376

    Jessica,
    (Errata – sorry I was in a hurry!)
    You’ve identified a common problem that is also found in manufacturing. Consider a tolerance of 50 microns +/- 5 microns – is a part with a 56 microns thickness really of superior quality to a part of 54 microns thickness? Of course not; it is sheer nonesense. Therefore, the part with a 51 micron thick film is not necessarily a defect: So how does one solve this problem?
    The first step is to find out how to design a single tolerance. I am currently writing an article on this subject which I’ll publish in the forum.
    The second step is to consider the ‘capacity’ of your process system, because a target must be set by the capacity.
    To my mind there are two ways to work out capacity, the first is to use queuing theory – a Monti Carlo modelling system such as Simul8. The other way is to use Ikko Nagashi, or single flow, where the tack time corresponds to a theoretical standard time for each step.
    Good luck,
    Andy

    0
    #114377

    billybob
    Participant

    Hello folks,
    Do you have any historical data as to how long it took to process a transaction under your current system? How does you company employ manpower without knowing how much work has to be done in the current state?
    I wouldn’t trust any expert who says its takes “around 10 days”  go get the data yourself. 
    I wonder if the possum will see his shadow in the moonlight tonight? Will it be 3 more weeks until my next bath or not?
    Later,
    Billybob
     

    0
    #114378

    Anonymous
    Participant

    Make an x-bar & R chart based on current times, and be sure to collect continuous data for this (i.e. # of days, not just good/bad).  This will give you a sample mean & range and let you look for stability.  This will at least show you how the process behaves today.
    To decide on what the mean “should be”, can a value be derived for different times?  For example, how much $ savings can be realized by completing the task in 9 days vs. 10? 8 vs. 9?

    0
    #114388

    Shrestha
    Participant

    Try using the Value Stram Mapping approach from Lean – it is much simpler, direct and quick.
    Niraj 

    0
    #114390

    Peppe
    Participant

    Was this process designed before start or it is free running ? Experts you refer to are worker with experience or designer of the process ? A target, usually, must be defined before to start to do a process by your customer or benckmarking, as indicated in other answers,  following adjustments are improvements.
    Question of Senior management is correct.
    Rgs, Peppe

    0
Viewing 7 posts - 1 through 7 (of 7 total)

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