Deming--working in Japan and else where in the mid-20th century--introduced very practical ideas of process control as a means to limit variations in product quality. Today, it is called defined process control.
Deming was a product guy; he came at quality from the point of view of the product: make the product the same way each time and make it work within limits that are acceptable to the customer. The modern poster child for defined process control is Six Sigma.
- Six Sigma is a problem solving methodology and defect control strategy with the purpose of identifying and mitigating error sources in defined process control.
- The control limits are established such that production yields less than approximately 3.4 errors in one million opportunities either above or below the control limits. This figure is derived from the error possibilities within six standard deviations of a bell-shaped curve, after allowing 1.5 standard deviations drift of the long-term average defect rate.
- The process derives its name from the Greek lower case s, called sigma and denoted σ; σ is the symbol used by statisticians for the standard deviation of a probability density function such as the bell curve.
Agile Thinking
Ken Schwaber—a leading SCRUM methodologist—objecting to defined process control, puts it this way, “[defined process control] is based on processes that work only because their degree of imprecision is acceptable… When defined process control cannot be achieved because of the complexity of the intermediate activities; something called empirical process control has to be employed.” In Schwaber’s view, software is too complex to expect defects to be contained within predefined error limits. Empirical control is the answer; empirical control is derived from observed facts, adapted to the situation, and not determined by preplanned limits from previous projects.
Deming contributes PDCA to agile projects
In spite of the fact that software projects offer little opportunity for statistical process control in the Six Sigma and Deming way, perhaps Deming's most noteworthy accomplishment from the perspective of project management and agile methodologies is his famous plan-do-check-act cycle that he originally adopted from Walter Shewhart. Plan-do-check-act (PDCA) envisions planning for what is to be done, then doing it—that is the plan-do. Next, measure results—measuring is the check activity—and then act on the measurement results. To act in the PDCA sense means to reflect upon lessons learned and provide feedback for corrective actions to the next iteration of the plan.
Deming was influenced by the work of the process statistician Walter A. Shewhart who is credited with identifying that processes have two variables: assignable cause and chance cause.
The former is systemic and capable of being corrected and maintained to an economical minimum; the latter is randomly occurring in frequency and intensity, not always present in the process, and is mitigated by establishing performance limits for a given process.
A project management tip: Deming introduced the PDCA cycle, which is wholly embraced by the EVO method.
• The cycle really applies to all agile iterations. The plan-do is equivalent to the planning session followed by development, test, and integration.
• Especially relevant is the check-act that provides measurement and feedback for continuous improvement.
• Deming focused on eliminating unsatisfactory results before they reached the customer. In agile parlance, every object must pass its unit, functional, and system test
Are you on LinkedIn? If so, share this article with your network
