Cpk: Indispensable Index or Misleading Measure?

What is capability analysis and why do it?

Published: Thursday, December 8, 2016 - 10:26

Capability analysis is a set of calculations used to assess whether a system is able to meet a set of requirements. Customers, engineers, or managers usually set the requirements, which can be specifications, goals, aims, or standards.

The primary reason for doing a capability analysis is to answer the question: “Can we meet customer requirements?” Or to be more specific: “Can our system produce consistently within tolerances required by the customer now and in the future?”

Capability analysis involves two entities: the producer and the consumer. The consumer sets the requirements and the producer must be able to meet those requirements.

Capability analysis can be used as structured communication that flows between producers and consumers. This communication happens up and down the supply chain and is useful in different contexts. For example, it can be useful when a manufacturer is considering the purchase of a new machine. It can also be useful when a producer is working through the production part approval process. And it can be useful during contract and pricing negotiations between a producer and a potential customer.

Because capability analysis plays this important role in the supply chain, the language of a capability analysis must be consistent and agreed to among interested parties. For this, guidance is provided in standards such as IATF 16949 and in publications such as those produced by the Automotive Industry Action Group (AIAG). The primary statistics resulting from capability analysis are the capability index (Cpk) and the performance index (Ppk).

Capability analysis can add value in the supply chain, if you proceed with caution. Following are a few suggestions to make its use more effective.

Understand capability analysis

As Mark Twain said, “Facts are stubborn things, but statistics are pliable.” Both when consuming and when producing a Cpk result, be aware that Cpk is easily manipulated.

Whether you are the producer or the consumer, if you are using capability analysis for decision making in your organization, you must fully understand its use. Computing Cpk is not difficult; understanding whether or not the value is a reasonable representation of the underlying process, however, is critical to using capability analysis effectively.

There are many resources for learning the mechanics and formulas for doing a capability analysis, including those offered by PQ Systems. Additional recommended resources for enhancing your understanding of capability analysis include:
• Process Evaluation Handbook (Donald J. Wheeler, SPC Press, 2000)
• Beyond Capability Confusion (Donald J. Wheeler, SPC Press, second edition, 2000)
• SPC Quickstart Guide (AIAG, 2000)
• Statistical Process Control (AIAG, 2005)

Once your understanding of capability analysis has been developed, ongoing practice is required to keep the knowledge current and to pass it on to others involved in quality management and monitoring.

Understand what your customers want/need

Once you have a working knowledge of capability analysis, you need to communicate with your customers about their requirements. Typically, the customer will have requirements for many different characteristics or metrics. However, not all of these will be equally important.

Once the characteristics requiring capability analysis are identified, the next questions are how often the analysis should be done and how the results will be communicated. There are many options here. The important result is communication—discussion and agreement between producer and customer.

Study what your current system can do (your capability)

For each metric where the customer has a specific requirement, you must conduct a capability analysis to see if you are currently meeting these needs. When doing this analysis, it is important to remember that control must come before capability. In other words, assess your process for statistical process control first. If the system is not predictable (in statistical control, reflected on a control chart) you must address problems before doing a capability analysis.

Compare the customer needs with your capability

You know what the customer needs and you know what you can do. In this step you compare these numbers.

Define ongoing monitoring for system changes

Once you are fluent in both performing capability analysis and communicating its results, it is important that your team implement ongoing procedures for paying attention to how well you are doing. This might include weekly, monthly, or quarterly reviews of capability analysis on all key customer metrics.

In a similar fashion, capability analysis should be seen as one of many tools that contribute to understanding and improving quality. A popular chart, known as Six-way analysis, is a good way to show not only capability but also statistical control, trends, and the distribution of your data. All of this is contained within a single-page visual that’s easy to consume, as seen in figure 1.

Figure 1: Six-way analysis (Click here for larger image)

Summary

Capability analysis plays an important role up and down the supply chain. Used wisely, it can be one measure to demonstrate a system’s ability to meet its numerical requirements. Used over time, it can be an excellent tool to demonstrate the extent of an improvement made to a process. But just as you would not run your entire business on your inventory turnover ratio alone, don’t run your entire manufacturing operation solely on capability analysis.

For more information on this topic, please register for the webinar “CPK: Indispensable Index or Misleading Measure?” which I will be presenting on Dec. 15, 2016, at 2 p.m. Eastern, 11 a.m. Pacific.

About The Author

Steve Daum

M. Stephen Daum is director of development for PQ Systems. Prior to assuming responsibility for development, Daum was the lead programmer on PQ’s statistical software products, a position he took in 1985. Daum has more than 20 years of experience with control charts and control charting software and has shared that experience through presentations, training, and educational sessions for organizations throughout the United States, England, and South Africa.