Featured Product
This Week in Quality Digest Live
Six Sigma Features
Tom Taormina
Champion business success and avoid risk
Michael Popenas
Enables companies to develop, design, test, and release products and services more quickly than other waterfall methods
James J. Kline
A sound quality management foundation exists in local government
Greg Hutchins
Risk is becoming our lens for everything from managing, to working, and even going to the store.
Ryan Ayers
The right data and measurement scales help companies organize, identify, analyze, and use data to inform strategies

More Features

Six Sigma News
Collect measurements, visual defect information, simple Go/No-Go situations from any online device
Good quality is adding an average of 11 percent to organizations’ revenue growth
Floor symbols and decals create a SMART floor environment, adding visual organization to any environment
A guide for practitioners and managers
Making lean Six Sigma easier and adaptable to current workplaces
Gain visibility into real-time quality data to improve manufacturing process efficiency, quality, and profits
Makes it faster and easier to find and return tools to their proper places
Version 3.1 increases flexibility and ease of use with expanded data formatting features
Provides accurate visual representations of the plan-do-study-act cycle

More News

Jay Arthur—The KnowWare Man

Six Sigma

Control Charts for Services

Service professionals should know their way around p, u, and XmR charts

Published: Tuesday, April 3, 2012 - 12:28

Although Quality Digest often has in-depth articles about the nuances of control charts, I’ve found that many beginners are at a loss to figure out how to organize their data, especially in service industries such as health care, hotels, and food. They complain that the examples are all manufacturing-oriented. While it’s pretty simple to organize the data, this hurdle seems insurmountable to many.


Data for services: defects and turnaround times

Services are concerned with two things: defects (i.e., mistakes, errors, omissions, etc.) and turnaround times. So for most defect-related applications we need a date, the number of defects, the sample size (total number of opportunities for a defect), and the percentage of nonconforming units. These four columns give you the ability to create p charts, u charts, or XmR charts (aka the X and moving range chart or the Individuals chart)—the most common charts in service industries. For ease of charting, organize the data in columns, not rows. Figure 1 shows what the column headings would look like in Microsoft Excel.

Figure 1: Example of column headings in Microsoft Excel

It’s important to use a reasonable time interval like hours, days, weeks, or months. Hospitals often report by quarter, but you would need five years of data (20 quarters) to get enough data to draw a useful control chart. Two years of monthly data (24 months) would give a useful control chart, but may not be granular enough for some applications.

One of our customers have asked me, “I have a back room operation; how do I track the number of times people don’t follow established procedures?”

“Simple,” I replied. “Just adapt these columns to your data.” The result is shown in figure 2.

Figure 2: Adapting columns to your data

In health care, hospitals track hand washing, because hand washing is the key to preventing hospital acquired infections that, according to the Centers for Disease Control and Prevention, kill 99,000 people a year. Figure 3 shows how the data would appear.

Figure 3: Tracking hand washing in a hospital

This format can also be used for defect data concerning patient falls, infection rates, and other hospital events. See figure 4.

Figure 4: Patient falls as defect data

Once you understand the nonconforming/sample/ratio format for defect data, it’s easy to lay out data for charting. Using the Excel add-in, QI Macros, it’s easy to create p, u, or XmR charts from this data. Use nonconforming/sample size (columns A:C) for p or u charts; use the ratio (column A and D) for XmR charts.

Note: The difference between a p chart and a u chart (see figure 5) is that in a p chart the item is either bad or good. In a u chart, the item can have more than one defect. In this case a single patient can fall more than once.

Figure 5: U chart of total patient falls—1000 pt days. Click for larger image.

Figure 6: XmR chart—falls/1000 patient days. Click for larger image.

I often provide both types of charts because the variable limits of a p or u chart sometimes confuse the viewer. Having to explain why the limits vary sometimes takes detracts from what the data is telling us.

For those people who have anxiety about using the XmR chart for ratios, see the article, “When in Doubt, Get the X Chart Out,” by Thomas Pyzdek.

Time data

From a turnaround time perspective, service industries often track response times on a pass/fail (i.e., defect) basis: number of calls answered in 60 seconds (call center), number of cardiac patients given aspirin at arrival (emergency room), number of quotes delivered within 24 hours (insurance) and so on. This takes on the form of having met or missed a service commitment as shown in figure 7.

Figure 7: Tracking response time for services can be shown as a having met or missed a service commitment

While most managers like to know how well they are doing (% of commitments met), Six Sigma problem-solving methods work best when focused on the problem (% of commitments missed). Figure 8 shows managerial performance data.

Figure 8: P chart of missed commitment—total opportunities. Click for larger image.

Turnaround time data can also be shown in actual days, hours, or minutes. This could cover turnaround times in a hospital lab, time to resolve an appealed insurance claim, or days to pay claims, as shown in figure 9.


Figure 9: Showing turnaround time in actual days, hours, or minutes

When there are no headings and the data represents, for example, patient-by-patient or claim-by-claim turnaround times, the data can be easily charted as an XmR chart (figure 10) or histogram (figure 11) using the QI Macros.

Figure 10: XmR chart—days to pay insurance claims. Click for larger image.

Figure 11: Histogram—days to pay insurance claims. Click for larger image.

In this example, we can see that the process for paying claims is stable, but we might incur additional charges for exceeding 30 or 45 days before claims are paid, depending on the supplier’s requirements.

In summary

The most common control charts used in service industries are the p, u, and XmR charts. They can be used to monitor error rates, missed commitments, and turnaround times. Once the data is organized into columns, it’s easy to turn the data into a control chart. Just use these simple formats (shown in figure 12) as a guide to start collecting data in Excel.


Figure 12: Formats for turning the data that is organized into columns into a control chart

Use columns A:C for p or u charts. Use column D for the XmR chart, or you can use A in the time example for the XmR chart.


About The Author

Jay Arthur—The KnowWare Man’s picture

Jay Arthur—The KnowWare Man

Jay Arthur, speaker, trainer, founder of KnowWare International Inc., and developer of QI Macros for Excel, understands how to pinpoint areas for improvement in processes, people, and technology. He uses data to pinpoint broken processes and helps teams understand their communication styles and restore broken connections. Arthur is the author of Lean Six Sigma for Hospitals (McGraw-Hill, 2011), and Lean Six Sigma Demystified (McGraw-Hill, 2010), and QI Macros SPC Software for Excel. He has 30 years experience developing software. Located in Denver, KnowWare International helps service and manufacturing businesses use lean Six Sigma tools to drive dramatic performance improvements.


Systems thinking rather than tracking the data

As there is one example quoted in above article- one of the customers asked.."how to track the number of times people dont follow established procedures"?This question itself reveals the lack of profound knowledge and the Customer's organization health status.There is very good oppurtunity to improve more which would fix majority of the problems without plotting any control chart in the first instance itself

as its obvious,There is zero benefit of counting the number of people dont follow established procedures...all you have to do is..first define the problem clearly and do rootcause analysis..again, there is no scientific study required here to perform this.Just ask a common sense question- Why people are not following procedures ? is that all people ? or few?If its the case with all people, then your procedures are not easy to follow..revise your procedures..however, its not better to advise without first understanding- what are those procedures established for which its hard to follow?If its the case with only few people- then yes its the responsibility of management to apply holistic solution to fix any uneven things. Systems thinking approach helps here to understand the things in bigger picture. There is very excellent reference available "leaders handbook" by Scholtes to understand and fix the above type of problems.



For industries having difficulty with control charts, forget about p and u charts and use XmR instead.  They would also do well to forget about the Six Sigma nonsense and its ridiculous foundations.  Measure things and chart measurements wherever possible, rather than counting defects.

Control Charts for Services

The author of the subject article did not comment on the specific distributional requirements for using p and u charts. Data that do not follow the distributional requirements will not produce reliable charts. The author does not say if he tested the data and found them suitable.