D
D.Salman
Dear experts,
Is process behavior charts the only way to analyze deviations from targets?
Many thanks in advance
Is process behavior charts the only way to analyze deviations from targets?
Many thanks in advance
Are Process Behavior Charts the only way to Analyze Deviations from Targets?
Re: Is process behavior charts the only way to analyze deviations from targets?
Hello D. Salman,
Just to make sure; what is your definition of "Process Behavior Charts?"
Are you referring to regular SPC control charts for variables?
Stijloor.
Hello D. Salman,
Just to make sure; what is your definition of "Process Behavior Charts?"
Are you referring to regular SPC control charts for variables?
Stijloor.
Re: Is process behavior charts the only way to analyze deviations from targets?
I am a little wary of your use of the phrase "deviations from targets". Generally, when using SPC to try to understand deviations from targets we figure out the variability of the data itself - the deviations (standard deviation) away from the average of the data. That lets us know what the process has been capable of, and if stable, allows a prediction of future performance. Then we can look at the center line and the control limits, and compare that against any targets to see if the past performance has been hitting the target or not.
So I may have the following combinations occurring:
An unstable process from which I can make little prediction as to if I will hit the target in the future.
An unstable process with a trend towards the target (improving)
An unstable process with a trend away from the target (getting worse)
A stable process not capable of hitting the target
A stable process capable of (and predictably) hitting the target
You would need to categorize your results into one of these categories in order to understand "deviation from the target", and yes, you could use SPC to do this.
I am a little wary of your use of the phrase "deviations from targets". Generally, when using SPC to try to understand deviations from targets we figure out the variability of the data itself - the deviations (standard deviation) away from the average of the data. That lets us know what the process has been capable of, and if stable, allows a prediction of future performance. Then we can look at the center line and the control limits, and compare that against any targets to see if the past performance has been hitting the target or not.
So I may have the following combinations occurring:
An unstable process from which I can make little prediction as to if I will hit the target in the future.
An unstable process with a trend towards the target (improving)
An unstable process with a trend away from the target (getting worse)
A stable process not capable of hitting the target
A stable process capable of (and predictably) hitting the target
You would need to categorize your results into one of these categories in order to understand "deviation from the target", and yes, you could use SPC to do this.
Re: Is process behavior charts the only way to analyze deviations from targets?
actually a this point your question is very difficult to answer. What is your situation? what are you trying to understand? details and data if you've got it will get you much more useful answers quickly.
actually a this point your question is very difficult to answer. What is your situation? what are you trying to understand? details and data if you've got it will get you much more useful answers quickly.
D
D.Salman
Re: Is process behavior charts the only way to analyze deviations from targets?
Dear Mr. Steve,
Thanks for the information, but I couldn’t understand what you mean of an unstable process from which I can make little prediction as to if I will hit the target in the future.
Can I have an example from the service industry?
Many thanks in advance
Dear Mr. Steve,
Thanks for the information, but I couldn’t understand what you mean of an unstable process from which I can make little prediction as to if I will hit the target in the future.
Can I have an example from the service industry?
Many thanks in advance
D
D.Salman
Re: Is process behavior charts the only way to analyze deviations from targets?
Dear Stijloor,
what I meant is control charts.
Dear Stijloor,
what I meant is control charts.
Re: Is process behavior charts the only way to analyze deviations from targets?
An example in the service industry might be customer wait times for the service (be it telephone hold times, cash register queue times, emergency room waiting times). If the control chart of waiting times is "out of control" - points frequently above and below the control limits, runs of 7 in a row the same side of the average, with no discernable predictable patterns, there is no way to provide the customer an estimate of how long they will have to wait. Nor is there a good way to staff the process, just need to keep a bunch of extra people on hand (spending much of their time idle) to open more registers, phone lines, etc. when the random need arises.
Fortunately, many waiting processes have some predictability to them. Such as there will be a lunch-time rush at a certain time of day. If I can formulate a prediction of waiting times (and more importantly, a prediction of numbers of customers and how long their service will take), then I can increase staffing during lunch-time rushes, or offer customers benefits to use the service at off-peak hours.
An example in the service industry might be customer wait times for the service (be it telephone hold times, cash register queue times, emergency room waiting times). If the control chart of waiting times is "out of control" - points frequently above and below the control limits, runs of 7 in a row the same side of the average, with no discernable predictable patterns, there is no way to provide the customer an estimate of how long they will have to wait. Nor is there a good way to staff the process, just need to keep a bunch of extra people on hand (spending much of their time idle) to open more registers, phone lines, etc. when the random need arises.
Fortunately, many waiting processes have some predictability to them. Such as there will be a lunch-time rush at a certain time of day. If I can formulate a prediction of waiting times (and more importantly, a prediction of numbers of customers and how long their service will take), then I can increase staffing during lunch-time rushes, or offer customers benefits to use the service at off-peak hours.