The process capability index and the related process performance index reflect the ability of a manufacturing process to meet specifications. When the process performance index is 2, there's a Six Sigma process. If the process is centered on the nominal, two non conformances per billion items are expected. If the process's mean shifts by 1.5 standard deviations, it implies 3.4 defects per million opportunities.
Well-established methods are available for calculation of traditional process performance metrics like Cp, Cpk, Pp, and Ppk. Commercially available software like Minitab will do this automatically, and even provide an estimate of the nonconforming fraction. The performance indices then translate into the language of Six Sigma, e.g. in terms of DPMO.
What happens when the process does not cooperate with the textbook assumption of a normal or bell curve distribution? This webinar by expert speaker William A. Levinson will provide you the answer to this dilemma while providing you these two key takeaways: (1) that the traditional methods and formulas can be off by several orders of magnitude with regards to the estimated nonconforming fraction, but (2) we can calculate accurate performance indices if we can identify the underlying statistical distribution. The method for doing this is, in fact, sanctioned by the Automotive Industry Action Group's statistical process control manual.
Who should attend?
Manufacturing managers, quality professionals, and purchasing professionals. The latter need to understand, for example, whether process capability reports from suppliers make sense. Quality engineers must meanwhile make sure the reports they provide to customers make sense!
William A. Levinson P.E.
William A. Levinson, P.E., is the principal of Levinson Productivity Systems, P.C. He is an ASQ Fellow, Certified Quality Engineer, Quality Auditor, Quality Manager, Reliability Engineer, and Six Sigma Black Belt. He is also the author of several books on quality, productivity, and management, of which the most recent ... More info