Q. 125  A capable process is one that produces an output that meets customer specifications. A stable process is one that has controlled variation and operates within the control limits. Give examples of situations when a capable and stable process may suddenly be rendered incapable. 

 

Note for website visitors - Two questions are asked every week on this platform. One on Tuesday and the other on Friday.

• All questions so far can be seen here - https://www.benchmarksixsigma.com/forum/lean-six-sigma-business-excellence-questions/
• Please visit the forum home page at https://www.benchmarksixsigma.com/forum/ to respond to the latest question open till the next Tuesday/ Friday evening as per Indian Standard Time.  
• The best answer is always shown at the top among responses and the author finds honorable mention in our Business Excellence dictionary at https://www.benchmarksixsigma.com/forum/business-excellence-dictionary-glossary/ along with the related term.

 

Enumerated below are few pointers which I potentially think can lead to a situation where a capable and stable process
can suddenly render incapable

 

1. Absence of a Robust Control Management System

1. Post Analyze and Improve phase, the solutions implemented needs to have a robust Control Management System The deployed solutions needs to be monitored for its effectiveness and also project owner needs to see if there is any change in business environment that can potentially drive a change in input variables which in-turn can impact the focused out put metric.
2. Further, given today's dynamic business environment, failure in updation and hand over of actions to appropriate owners will result in rendering the process incapable.
3. Also, the mindset of driving actions can move from Process Dependency to People Dependency which can weaken a stable process

2. Compromised Quality Management Process

 

A Stable and Capable process with a defined Quality Management process, if not reviewed in the event of any
increased demand scenario can potentially lead to systematic failure of a stable and capable process.

With increased demand from client, there could be a potential rush to complete the job (say production) within
stipulated timelines. If there is a compromise in Quality Management System, we could land up delivering poor
quality products rending the process incapable. 

Manufacturing and PC industry has seen several recalls of their vehicles, batteries, motherboards etc., possibly
due to aforementioned reasons

 

3. Shifting Business Priorities - Siloed Views by Management

 

Is your balancing metric now deemed critical to take over as primary metric? 

Lets take an example where we have worked a process improvement project using DMAIC method to reduce the "Cost of
Dissatisfaction" which has resulted in significant cost savings while your Customer Experience metric is kept
under control.

The Management is now under pressure with the revised Customer Experience goals. There could be some systematic
compromises done on few of your action plans on cost savings project (which could still be under specified limits)
which gradually can take the focus out as it could be the same set of contact center agents who need to drive both
the actions. A conflicting and Siloed approach could render your cost saving project incapable.

 

4. Special Cases like Mergers & Acquisitions.

 

A M&A situation can result in People Motivational Issues, Organization Culture differences, Business priorities
etc., which in turn can result in impacting an existing stable / capable process.

 

An Interesting Read through

 

https://en.wikipedia.org/wiki/Space_Shuttle_Columbia_disaster     - Enclosed herewith is a Wikipedia artilce which I found during my internet reading.  A direct excerpt of key call out "... where Bipod ramp insulation had been observed falling off, in whole or in part, on four previous flights: STS-7 (1983), STS-32 (1990), STS-50 (1992) and most recently STS-112 (just two launches before STS-107). All affected shuttle missions completed successfully. NASA management came to refer to this phenomenon as "foam shedding". As with the O-ring erosion problems that ultimately doomed the Space Shuttle Challenger, NASA management became accustomed to these phenomena when no serious consequences resulted from these earlier episodes...."

What was successful in previous four launches with the same issue, proved fatal for Shuttle Columbia which
resulted in its failure.

A process is said to be capable if it meets specification limits which are decided by customer . On the other hand a stable process is one which is statistical controlled and meets control limits which are derived from process itself. Stability of process relates to change in variation and has control limits which are lower than specification limits. A process will be stable if gives consistent output. SPC charts are used to determine if the process is stable or not.For example X bar charts are used to monitor process average.

 

There is no direct link between process stability and process capability . by knowing that the process is capable tells us nothing about the process is stable or not and similarly by knowing only process is stable tells us nothing about the process is capable or not. there fore we should know both control as well as specification limits for any process.

 

There is only one link between these two is that process capability should be performed only after performing process stability. because if we dont know that process is stable or not and its not giving any consistent output , how can we find out process capability. Process capability represents the performance of the process over a period of stable operations

 

if there is only common cause of variation present in any process, then the output of that process forms a stable distribution over a period of time and can be predictable. A stable and capable process will be incapable and unstable if there is any special cause of variation  because it reflects unexpected results and output goes beyond specification and control values. Process will be unstable if any value goes beyond control limits and any special cause gives any output which is out of specification limits. Special Cause of variation is a type of variation that are not always acting in the process. The process may be unstable but can be capable inn case of common cause of variation. Because common cause of variation leads to many pattern in control charts to identify this , either the mean is shifted or have a repeatable distribution over a period of time. When we are plotting a process on control charts, process stability can be checked by several patterns like points are outside control limits, there is any upward or downward trends, points are only one side of central line, repeating cycles. Output is in specification limit but not stable as it varies over a period of time.

A process will not be capable if the specification limits becomes unrealistic, process is giving stable output within control  limits but not have realistic specification limits can be one of the reason that the process is not capable. and other reason for the same is any special cause of variation in any process which  is not acting before in the process.

 

the purpose of capability study is to find out whether a process is capable of meeting customer requirements and to take any action if its not meeting requirement.

Followings are steps to conduct capability study of any process:

1. Select a key quality characteristic of a specific process for study.
2. Confirm the measurement system which is used to collect the process data.
3. gather the data , collect data from the process or we can track data on control charts.
4. verify process stability and ensure the process is in statistical control . if the process is out of control identify and eliminate the cause.
5. verify that the data that was collected is normally distributed which can be find out by Histogram, Anderson- darling Test etc.
6. get the process specification which can be one sided or two sided.
7. Determine the process capability indices and interpret them.
8. Update the process control plan.

Capable process example is airline service, which is designed to meet customer satisfaction. 

Stable process example is aircraft operation, which has minimal variation from standard processes. 

 

Both these process may become incapable, if a passenger does not reach his/her destination on time. 

The brilliant example is government changing the automobiles norms from BS4 to BS6 from April 2020. All automobiles manufacturers will have to redraw recertify and revalidate their stable and capable process meeting the current norms of BS4.

 

2nd example Government setting new requirements for pre-paid smart energy metering and more than 300 million energy meters to be changed all over India in a period of 5 years or so. Almost all energy meter manufacturers redrawing revalidating and getting their products approved for safety norms for pre-paid and remote cut off features. Again the stable and capable product not having prepaid and remote cut off feature will be rendered useless.

 

 

This answer to the question is considered the manufacturing process environment, variable and normally distributed data

No two products or characteristics are alike, because any process contains many sources of variability. Differences are always presents in the products or characteristics;

What is process?

            A process is a set of activities that converts an input into desired output. A set of inputs are Men, Materials, Machines, Methods, Measurements and Environment. Though there is a control in input parameters (4M + 1E), the actual performance of the process can be measured through the process output. When we measure the output of the process and we can clearly see that there is a part to part variation. So variability is an inherent properties of every process and we accept the variability within specification.

            Some sources of variation in the process cause short-term, piece to piece variations e.g backlash in the machine, Spindle run out and its fixturing or the accuracy of journal entry. The other sources of variation cause changes in the output only over a period of time. e.g tool wear, temperature increases in the process.

            The measured individual values form a pattern that can be described as a distribution. The distribution shall reveal the process characteristics such as;

·         Location ( Central tendency)

·         Spread ( Variation)

·         Shape

However, process control aims to maintain the location to a target value with minimal variability

A goal of a process control system is to make predictions about the current and future state of the process. A process said to be capable when the output of the process meets customer / drawing specifications. And, a process said to be stable / statistical control when the output of the process has less variations and produces the predictable outputs over a period of time within control limits

Causes of Stable Process Variation:

            In any production process, regardless of how well its designed or carefully maintained it is, a certain amount of variation will be existed in the process. When inherent and natural causes present in the process, then it often called the process is in stable or control. This causes called as common causes or chance causes.

Some other kind of variations may occasionally be present in the output of the process and it changes the process to “out of control”. For examples, improperly adjusted toolings, operator errors or defective input material etc., such variability is generally large when compared to the chance causes, and it usually an unacceptable level of process output. This causes often called “assignable causes or special causes”

A process that is operating in the presence of assignable causes is said to be an “out of control process”

Let us take µ0 and s0 is a stable process mean and standard Deviation, time at t1 an assignable causes occur, the process mean will shift to µ1> µ0. At time t2 another assignable cause occurs, resulting in µ = µ0, the process standard deviation will shift to s1>s0. When other assignable causes occur, the process mean and sigma will be in out of control limits.

Assignable causes which shifts the process mean and /or increase the process standard deviation.A Control Chart is a tool for monitoring process stability or process control relatively for longer periods of time.

Assuming, there is no probability error (type I and type II) of the control chart, eventually the right control chart deployed to study the variability and stability of the process, the following are the few causes to make the process unstable but not limited to;

·         Control limit have been miscalculated  or misplotted

·         The measurement system has changed( e.g different appraiser or instrument)

·         The measurement system lacks appropriate discrimination

·         Mixed lot of input materials

·         Tool change / Fixture change  / Set up

·         Wrong loading

·         Voltage fluctuations

·         Over control or Tampering ( take action when action is not necessary for the process)

The change events to be recorded in the control charts and if any outlier due to this changes and it can be omitted, provided other data points are stable.

 To interpret the subgroup ranges or subgroup averages, the R Chart to be analyzed first because it shows piece to piece variability.

 For any process, the process must first be brought into stable (statistical control) by detecting and acting upon special causes of variation. Then its performance is predictable, Process capability can be assessed.

In order to test the special causes, I-MR chart to be made and following interpretations will help us understand the presence of special causes;

Test 1 : One point more than 3s from the center line ( mean)

Test 2 : Nine points in a row on the same side of the center line

Test 3 : Six points in a row, all ascending or descending

Test 4 : 14 points in a row, alternating up and down

Test 5 : Two out of three points more than 2s from the center line ( same side)

Test 6 : Four out of five points more than 1s of center line ( same side)

Test 7 : 15 points in a row within 1s of center line ( either side)

Test 8 : Eight points in row more than 1s from center line ( either side)

 Causes of incapable Process:

            Process capability is often termed “output of the process meets customer specification”. When discussing process capability, two contrasting concepts to be considered;

·         Process Capability

·         Process Performance

Cp, Cpk, Pp, Ppk are the indices which used to measure the process capability in short term and long term respectively

             The output of a stable process can be described by its statistical distribution. The statistics are estimates of distribution location (center) and spread relative to the customer requirements. A shift in process location, an increase in process spread or combination of these factors may produce the parts outside the specification limits. Process capability study is an activity to evaluate the processes and machines are capable to produce the parts as per the customer specification.

The following are the possible causes could make the process incapable not limited to ;

1.     Special cause variations  in the process

2.     Limitation of machines / process

3.     Part to part  variation and center shift  (Spread & Location)

4.     Poor Fixture and tool design

5.     Selection of measuring instrument ( 1/10^th of tolerance)

6.     Design weakness ( selection of tolerance)

A perfect state of process control is not possible in a production process. The aim of process control is not a perfection, but a reasonable and economical state of control.

Process Stability is the attribute of a process to perform consistently within the control limits. And a process is said to be capable when it starts delivering as per the customer specifications. 

 

A stable and Capable process is suddenly rendered as incapable when there is a sudden change to the specifications. For example, Car Companies producing Cars on EURO 4 standard are suddenly without a lead time asked to adhere to EURO 5 Specifications. In such a scenario, their process capability exhibits a downward trend and process becomes incapable. 

Benchmark Six Sigma Expert View by Venugopal R

 

Most of us would have understood about the three types of Variability…. viz. Instability, Variation and Off target. A control chart is a tool that helps to represent these variabilities, both statistically and in an easily understandable way.

If all the points fall within the limits and the points are in compliance with the rules for all other ‘runs’ as applicable for a control chart, the process is considered stable. If the overall variation, represented by the distance between the UCL and the LCL is such that the variation of the population falls within the specified tolerances, the process is considered capable.

 

Well.. when can we have a capable and stable process be rendered incapable?

 

One such possibility is if we take an unwanted action of altering the settings of the process. But why should this happen?

 

Those of us who are familiar with he “Type-1” error, would understand that there is a small risk of being misled by a control chart, that the process is not stable, when in truly is. Of course, such type-1 errors are small and are expected to occur very rarely, but it is a possibility where one can adversely tamper a process that truly has been capable and stable. However, if we stick to the control chart methodology, we will quickly discover that there has been a shift and would promptly restore it back.

 

Imagine a situation where no control charting is done and the decision to alter the process is taken based on ad-hoc measurements. The chances of disturbing a process away from its stability and capability is certainly high in such situations.

 

Another incident that comes to my mind is where the concept of ‘fits’ and ‘tolerances’ had not been applied effectively for the dimensions of two components that need to be fitted to each other. As per the definition of stability and capability, the processes of both the components complied, but when an extreme match between the components came up there were failures, due to improper fit. Here the assembly has been rendered incapable. Hence, apart from the process capability and stability of individual mating components, the study on fits and tolerances needs to be considered as well.

A stable and capable process can be rendered incapable "suddenly" if there is a shift or change in the specification limits. The chosen best answer is that of Manjeet as he provides two such examples. Devesh has also listed an example. Read through rest of the answers for a very concise review of capability and stability concept.