Do you know if the process within your organization has what it needs to be effectuated repeatedly? A critical process to determine this is the Process Capability Analysis or Studies.

Understanding the core principles of Process Capability Analysis

Overall, this study uses data from an initial run of parts to predict the outcome of the process in terms of repeatability, to check whether a manufacturing process can produce parts that meet determined specifications over and over or not.

The thing is, that this is a critical study to be delivered to the organization’s customer, as part of a PPAP; and on more occasions than you think, it’s not done. In a nutshell, a Capability Analysis or Study measures consistency within the part production, according to the upper and lower limits. The idea is for it to be centered over the nominal indicated.

Cp and Cpk – how to measure them

Given the circumstances, two specific measures ought to be performed:

• The Cp (Process Capability): which is obtained through the next formula: Specification Width / Process Width. It accounts only for the spread or variation of the process.
• The Cpk (Process Capability Index): which is obtained through the next formula: Distance from Mean to Nearest Spec Limit / Distance from Mean to Process Edge. It accounts for both the spread and location of the process.

In general terms, when Cpk is higher it’s better. When Cpk is less than 1.0, it’s considered a poor result, and the analyzed process is tagged as not capable; a result between 1.0 and 1.33 is considered “barely capable”. In other words, a high Cpk value means that the organization and its process are expected to produce fewer defective parts, ultimately delivering better performance and fewer warranty complaints.

Seeking process stability overall is an inherent part of this goal. It’s considered as such when, on the different analysis runs, behavior remains equal in short and long terms and average and standard deviation are constant over time.

The difference between Cp and Cpk with Pp and Ppk

The Process Capability Analysis or Studies use data to predict the outcome of the part production process – however, it has to be differentiated from a Process Performance Analysis (Pp) and Process Performance Analysis Index (Ppk). These are employed to evaluate the manufacturing process and overall give a response to how the process is performed over a time interval (in other words, it’s a historical analysis rather than a predictive one).

The real insight here is that, if the process is stable enough, Cpk and Ppk will be equal, indicating that the actual performance matched the predicted potential performance; if it’s unstable, Cpk will be greater than Ppk, implying shifts or drifts over time.

Quick recommendations when data measuring:

Lastly, keep the next recommendations in mind. When recollecting data to perform this analysis, you should consider the following:

• Gages should be calibrated and gage resolution should be at 10% of the specification.
• Production order is critical for part measurement and recording; if they are not recorded that way, trends can be missed.
• Data that parts that pass and for parts that fail should be measured equally.
• Check that all measurements are traceable according to the 5Ms-1E Principle (Man, Machine, Method,  Material Measurement System, and Environmental Conditions).
• Do not mix measurements made with two different types of equipment into single data-sets; everything must be homogeneous.