In this blog post, we will discuss the similarities and differences between run charts and control charts. This is a request from Umasankar Natarajan, a visitor to my blog.
These are essential tools for quality management that help identify trends and errors in the product or process. Control charts and run charts let you know:
- How the process is performing.
- How today’s performance compares to yesterday’s.
- Whether you are performing as planned.
These charts are as easy to draw as they are essential communication tools for a project manager.
Run Chart Vs Control Chart
The control chart is one of seven basic tools of quality, but run charts are not. However, the latter is equally useful in quality management.
Run Chart
A run chart is straightforward. It shows the variation for a single data group over time. It helps you analyze:
- Trends in the process.
- If the process has shifted.
- Whether or not the process has any patterns.
Small to medium-sized organizations use run charts to communicate project progress with the client.
You will need to draw the planned progress line. Once the project starts, you will update this chart with a new line showing the actual progress. This lets the client see status and variation.
Example of Run Chart
If you are a cricket fan, you will enjoy an example of a run chart in a live cricket match where the commentator shows you the first team’s score. Then, you’ll see the progress comparison of both teams as the second team starts batting.
Benefits of Run Charts
- It’s easy to draft.
- It’s easy to analyze and interpret.
- It does not require much technical skill.
- It is a straightforward representation of the data.
Limitations of Run Charts
- They don’t have any statistical control limits; they don’t show you the upper and lower tolerance and threshold limits.
- Run charts cannot show if the process is stable and controlled.
To understand a run chart, you must know the context around the data.
Without further information, you may think a trend is normal when it is a variation. Sometimes, you may think a trend is abnormal when it is not.
Control Chart
While working at Bell Labs, Mr. Walter A. Shewhart developed this chart; therefore, many experts call it the Shewhart chart. It helps you study changes in the process.
A control chart is one of the seven basic tools of control, a modified version of the run chart. If you add control limits to a run chart, it will become a control chart.
Elements of Control Charts
- Mean
- Limits
- Specification limits
A centerline (red) called the mean or goal is surrounded by two lines (dotted green) called limits. These are the upper and lower control limits (UCL and LCL). These are surrounded by two others (purple), known as the upper and lower specification limits.
These limits are provided in the contract and cannot be crossed. The project manager determines the upper and lower specification limits to set the process boundaries. You must take corrective action if the process crosses the set limits.
You can say that the process is under control if 99.73% (3-sigma) of the data points fall between the upper and lower control limits.
Control charts help you find answers to the following:
- Is the process under control?
- Is the project moving in the right direction?
- Are the deliverables within the specification limits?
The Rule of Seven
You might think that instant intervention is needed when an observation is outside the control or specification limits. This is incorrect. Under one condition, you will start an investigation even if the data points are within the control limits. This case is known as the “Rule of Seven.”
This rule says that if seven or more consecutive data points fall on one side of the mean, you should investigate, even if they fall within the control limits.
Types of Control Charts
- c Control Charts: A c control chart helps you find non-conforming elements in mass production.
- p Control Charts: These control charts are useful when elements are unequal.
- U Control Charts: These charts help you investigate when the parameters fall outside the upper or lower limit.
- X-Bar and R Charts: These charts use the X-Bar or the mean to find subgroups. The Range plots (R Chart) are the subgroups based on upper and lower control limits. These charts are useful when you need to analyze five or fewer subgroups.
- X-Bar and S Charts: These charts are useful for five or more subgroups, and the standard deviations are considered in upper and lower control limits based on the mean or X-Bar.
Example of Control Charts
You can use the c control chart to see if your production process produces the product to all specifications.
You can use a p control chart to find accidents in any particular place.
Refer to this page for more on the control chart.
Usage of Control Charts
- To find and correct errors in an ongoing process.
- To see if the process is stable.
- To analyze the pattern in the process or product output.
Limitations of Control Charts
- Control charts show common cause and special cause variations. Common cause variations are normal and usually do not require intervention, while special cause variations require attention. A control chart may show you a false special cause variation which wastes your time and resources.
- Although control charts are easy to understand, they require knowledge of mathematical concepts like mean and standard deviation to draw the diagram. This chart requires skilled and trained personnel.
- If the limits are placed incorrectly, you will get false indications.
Summary
Run charts and control charts are important tools in project management. A look at either can give you the information you may not get by reading reports. Both charts are handy; however, control charts provide more information than run charts. With the proper use of the former, you can eliminate errors in products and processes and focus on improving the process.
What is your experience with the run chart and control chart? Please share through the comments section.
I am Mohammad Fahad Usmani, B.E. PMP, PMI-RMP. I have been blogging on project management topics since 2011. To date, thousands of professionals have passed the PMP exam using my resources.
We are running with run chart in our process, but sample size is 2, is it correct or not.
Please help me. On urgent basis.
I did not understand your question Naveen. Can you explain it in more detail?
Hi..usually corrective actions will be taken after defect repair..so that we make sure it wont get repeated…in this case if the point appears out of upper control limit,is it defect?
It is not a defect as it is still between specification limits but you will try to bring it between control limits.
Very nicely explained in a simple language . I was struggling to understand this .
I have one small question. What is the difference between run and a shift in run chart. What I get is that shift is also a type of run.( Run being defined as consecutive points on same side of median). where can I read more blogs from you.?
Thanks
You can read about the shift in run chart by following the given link:
http://www.sixsigmadaily.com/run-chart-shifts/
Other blog posts published on this blog can be find by clicking below given link:
https://pmstudycircle.com/study-notes/
Reaily good explain i search this topic and got useful things
Thanks Sanjay for your visit and leaving comment.
Hi Fahad,
Thank you for the explanation. What happens if the data points fall on specification limit? Can the product be still delivered to customer ?
Regards,
Pruthvi
Usually any data falls outside the line considered as a defect, otherwise not.
Hello,
Don’t you think the above diagram is incorrect? Specification limit should be within Control Limit? Can you please explain?
Specification limits are set by the client that cannot be breached, on the other hand control limits are set by the project manager to keep the process under control.
Nice to see that over the past few years, people have found this continuing post useful.
I would like to note that the LCL and UCL are set by the data that is being analyzed (specifically ¯x±3?) . The person performing the analytics should not set or adjust the control limits. The control chart also does not include specification limits (LSL and USL). These metrics would be included in a process capability analysis. Basically the control chart (also termed as a process behavior chart) is the Voice of Process (VOP) to quantitatively show how the process is functioning. Matched with the capability analysis, you are then able to take the data of process output and demonstrate how that relates to your customer specifications, or Voice of Customer (VOC). Altering the control chart from the original form dilutes the VOP signals you are looking for understanding the process behavior. Another very good resource for understanding control charts is Understanding Variation: The Key to Managing Chaos, by Donald Wheeler.
Hi Fahad,
I am having trouble trying to understand a Control Chart question from Oliver’s free 75 question, wonder if you or anyone could shine some light here.
A production process has been defined as part of an industrial equipment manufacturing project. The process is intended to produce steel bolts with a length of 20 cm. The control limits are 19.955cm and 20.045cm. The measurements made at the end of the process yielded the following results:
20.033cm, 19.982cm, 19,995cm, 20.006cm, 19.970cm, 19.968cm, 19.963cm, 19.958cm, 19.962cm, 19.979cm, 19.959cm.
What should be done?
(a) The process is under control. It should not be adjusted.
(b) A special cause should be investigated, the process should be adjusted.
(c) The control limits should be adjusted.
(d) The measuring equipment should be recalibrated.
I roughly plotted the chart and gathered these thoughts:
– the measurements are all within the control limits
– there are eight measurements closely aligned (Rule of 7) and two measurements away from them
I answered (d) however the answer is (b) special cause. However, you mentioned above that special cause applies when the measurements are out of control limits. Hence, I am confuse on how to answer such question. When we do apply (a), (b), (c) and (d)?
If you notice last seven readings, you will find that they are on one side of the mean. So the rule of seven will apply here.
Good. Useful for simple understanding.
Thanks Nagendra.
Great explanation! Thank you :)
You are welcome Veena.
Thanks again! you made the ‘control chart’ concept very clear to me!
You are welcome Icm.
Good illustrations for Control and Run Charts as well as explanations!!
Thank you Oliver for your comment.
1) Can specifications limit come in the control limit ? Who put control limit Project manager or member of project team or stackholder, customer, end-user etc.
2) Do specifications limit show the sigma value?
All these are defined in the quality management plan, and the planning required consultation with all concerned stakeholders.
Thank a lot. I’m learning Quality in PMBOK5. Your post is useful. :)
You are welcome Luu.
Does seven run rule apply for the quality control sample prepared for longer time (eg one month) and used each day in a routine lab?
The rule of seven clearly says that if seven or more consecutive data points fall on one side of the mean, then you must find the reason for it, it does not matter even if these points fall within the control limits.
So using the same QC sample for a longer time could be one of the reasons for having
seven or more consecutive data points fall on one side of the mean? It is dificult to prepare fresh QC samples every day for a routine lab.
Once you get the sample and you find a cause for investigation, you will investigate and rectify the process. Then you will again take the sample to see if everything is fine.
Hello Fahad:
The exam questions, for example, include mean as 93.9887 and data point as 93.9845 or 93.9888. Should one round up or round down or simply compare. That is, should I assume that 93.9888 is above 93.9887?.
Regards
Akhtar
Usually after decimal, two digits are fine. You can also make it round, depends on the question.
I always like the way you put complex stuffs/methods in most simplest of the way possible! For example, while reading this, I noticed how superbly you gave the example of “cricket score line” to explain “Run Charts”. Keep the good work up!
Examples are always good to relate with the concepts, that is why I always try to include a real world example to make concept easier for visitors.
Dear Fahad
If UCL is 10 and point is at 10 and 11.
So 10 is considered out of control or in control. And for 11 the process is out of control or needs to be investigated for adjustment which one option is correct.
In fact you should specify it your quality management plan. However, if 10 is the UCL then it will not considered as out of control.
If it goes to 11, although it is under control, you need to take the corrective action.
Please note, it must not cross the upper and lower specification limits, in this case, you can say that it is out of control.
For a process to be out of control, it should have a data point outside control limits and not necessarily Specification limits.. therefore, this above example where process point is at 11 it is out of control.
Please correct my understanding if wrong.
It should below the upper specification limit. Specification limits are set by the customer and control limits by the project manager.
So if “11” is above control limit but below specification limit, most likely, it will be acceptable.
Dear Fahad
Nice explanations, Kindly explain about Assignable cause or Special Cause in Control Charts.
I think you are asking about the assignable cause and common cause because assignable cause and special cause are the same thing.
Common causes are normal causes that occur on your process. These are not considered as unusual. In control charts these are shown within the control limits.
Special cause are variance, not expected in the process. These are shown outside the control limits. You can also call them defects.
Thanks Fahad
Which of the following statements is true about control charts ?
a. Special causes are easier to predict and handle than common causes
b. Common causes are easier to predict and handle than Special causes
Can you explain.
I have already explained the common and special cause:
Common causes are normal causes that occur on your process. These are not considered as unusual. In control charts these are shown within the control limits.
Special cause are variance, not expected in the process. These are shown outside the control limits. You can also call them defects.
So as per my understanding common cause are easier to predict and manage. In fact these are not considered as abnormal as these falls within the control limits.
Thank a lot Fahad for great explanation.
You’re welcome Umasankar.
Brother,
My name is Royal I am working as an engineer in Saudi Arabia. I have cleared my PMP exam in my first attempt in Saudi on past 17th of this month. Really Your notes and guidance helped me a lots to my success & i have advised many of friends to take part of your blog during preparation .
Further, i need your suggestion to select coarse between PMI SP or PMI – RMP next to my PMP Bit i confused & i am basically from mechanical engineer, kindly suggest me to select the correct one to enhance my carrier.
Royal – PMP.
Congratulations Royal for passing the exam, and I am glad that my blog helped in your studies.
Regarding the choice, if it is between PMI-RMP and PMI-SP, I suggest you go for the PMI-RMP unless you have any other reasons to select the PMI-SP.
There is another course for scheduling professionals from from AACE (http://www.aacei.org/), which is known as PSP. You can also go for this course.
Thanks Brother, Can you share me your RMP study notes, if you can please send me the details to my following My mail id – [email protected] and also suggest me some good R.E.P for attend 30 PDU classes.
I don’t have anything in soft format to send you through email.
You can refer my lessons learned by clicking below given link. It will answers your rest of the questions.
https://pmstudycircle.com/2012/04/i-passed-my-pmi-rmp-risk-management-professional-exam-lesson-learned/
Hi Fahad,
Thanks for your beautiful explanation about Control Chart and Run Chart. I have one more suggestion. It would be great if you could define the lines meant for UCL and LCL with ‘text’. Even while descriping you can say the red lines are the Mean, Green lines are the UCL and LCL and the blue lines above/ below them is the specification limit.
I would also request you to explain the for Earned value management. I would like to know when you look at te chart marked with AC, EV and PV, how do you come to know exactl, by looking at the chart which is bigger? The PV or EV?
Thanks Vidya for your comment, I have updated these information into the text, picture will be updated soon as well.
Regarding your second request, I will write another blog post.
Hello Vidya,
control chart is updated with texts.