Question: Differentiate between a stable process and a capable process. Is Process Stability supposed to be a pre-requisite for all type of processes? Explain with appropriate examples.
Process stability: means consistently producing the output in the process over time. If a process is consistent over time & distribution of the data is within the control limits. then we call this process as “Stable within control”. If there is a shift from the mean but the data points fall into the control limits set is called “Stable – out of control”. This requires attention since there is a common cause variation.
Process Capability– means a measure of the process and how capable the process is in maintaining the customer’s needs / expectations. It explains us how good or how bad the output is. Process capability is measured and represented in “Cp, Cpk, Pp, Ppk”.
If the process is stable but not capable, the customers will still be satisfied but not pleased/delighted.
Eg. A customer requires pant size of 40 for his usage. Also he says anything that fits him between 38 & 42 would be okay. In this case, the store consistently supplies him the size of 39. The customer is happy but not delighted. When the supply is accurate the customer would be delighted. Here the process is stable but not capable in meeting the customer’s expectation. You should always concentrate on a target to delight the customer and not on the range though given by them.
Process stability | Process Capability | |
Definition | Predictable output consistently. | Meet customer's expectation at all times.(100%) It compares the process performance agaings tehe specifications given by client. Capability has nothing to do with stability. Both has inherent relationship. |
Central tendancy measures | Constant mean and constant variance is required to say a process is stable. | Range and target is provided by the client. |
Limits | Control limits(UCL & LCL ) are in place | Spec limits (USL & LSL) are in place. |
Usage of variations | Constant random variations / controlled variation is exhibited by a stable process | Controlled and uncontrolled variations can be seen in the process. |
Grphical tool used | Control chart / Scatter plot | Histogram |
Interpretation | A random fluctuation around the constant mean over a period of time is said to be the stable process. When the pattern is seen and variation is uncontrolled, though it falls within the control limits, the process is not stable. It is always important that we require controlled variation, and constant mean. | if the histogram falls within the specification limits, the process is capable. If the variation is too high when mean is shifted below the target, then the process is incapable. |
Distribution | Constant distribution is required to say the process is stable. | Normal distribution is required to say process is capable. |
Order of stability / Capability | Stability comes first | only when stability and normality of the process is tested, a process capability is tested. |
Measurement index | Root cause for variation would be identified. | Cp, Cpk, Pp and Ppk is calculated. |
Usage of Cp & Cpk, Pp & ppk
Use Cp, Cpk for samples and Pp & Ppk for population to arrive at the capability index.
Definitions
Cp= Process Capability. A measure for a centered process
Cpk= Process Capability Index. Adjustment ofCpfor the off centered process.
Pp= Process Performance. A measure of process performance for the centered process.
Ppk= Process Performance Index. Adjustment ofPpfor the off centered process.
Cp – Capability index will tell you how fit the data are into the USL & LSL. Also make sure the process is well centered around the average.
If the process is centered – then Cp is equal to Cpk. If not centered, then Cp > Cpk.
When the Cp & Cpk is > or = 1 the nthe process is capable.
Cp & Cpk comparison to sigma
Cp and Cpk of:
· 1.00 is equal to 3 Sigma
· 1.33 is equal to 4 Sigma
· 1.67 is equal to 5 Sigma
· 2.00 is equal to 6 Sigma
. Is Process Stability supposed to be a pre-requisite for all type of processes?
Deming has quoted” Only when the process is stable, the process is capable of producing output”. This means the capability cant be checked before studying the stability and normality. The order is like this… Study stability, normality and then capability of any given process.
A stable process is always a prerequisite for all processes to meet customer’s expectation or calculating Process capability, because a process can’t be capable if the process is out of control.
In a DMAIC process , a BB should always check
1. Cehck if the process is stable or not simply by using the control charts.
2. Check if the data is normal or non normal to calculate the capability.
If the process is not stable , then start focusing into the root cause. Try to eliminate the roort causes identified from the control chart, make the process stable. Make sure the data is normal. Then calculate the capability. Remember if the process is out of control, then no use in calculating capability. Because capability depends on the data where the process happened when the data is collected.
Always remember root cause identified is eliminated ,but not to improve the process, to get the process where it belongs to. Dr. Deming used to use a very simple analogy in his seminars: “If this building catches on fire, we must put out the fire. Putting out the fire does not improve the building. All it does is get the building back to where it should have been all along – no fires!”
For Eg. When we apply for the credit card, the agent tells us that we would receive the kit within 7 working days. But here customer expects a shorter delivery. Though there are aware of the process, they expect the shortest. Anything that exceeds the time period mentioned by the agent, will annoy the customer. There the process is said to be unstable and the variation would be high. In such cases, the customer makes many phone calls to the agent ./ bank to check the status. This involves cost, time and human intervention. Also, this makes the business weaker sinceword of mouth is the best method to increase sales pitch in the bank. It ruins the sales of the company.
This variation is to be identified and eliminated inorder to satisfy the customer, and put back the process where it had been / where it should be to improve this process.
Conclusion:
Then process has to be stable , the data is to be normal and then the process is calculated whether it is capable to meet the customer needs.
Any process which is unstable can’t be capable to meet the customer’s expectation.
In simple words thus, stability and capability need to be treated hand in hand in terms of interpretations, but at all times, the word stable needs to come before saying the word capable.
thanks
Kavitha
 - Ippei Blog)