OPPORTUNITIES OF DEFFECT

When we improve our business processes, we are reducing the number of defects in our products and services for our customers. If there is a high level of complexity in our processes and products and services, there are more opportunities for defect. Therefore, to achieve a high Sigma capability, our goal is to reduce the total number of opportunities for defect, and concurrently increase the capability of the remaining opportunities.

An opportunity is defined as a set of circumstances that are favorable to some end. Each characteristic of our products and services is an opportunity. In general, four factors must be present for an opportunity to exist: a characteristic, a scale, a standard, and density. For example, characteristics that are critical to the customer are measured according to performance limits or success criteria. When we measure these characteristics, we are creating density, which is historical distribution. If all of these factors are said to be present, the opportunity is said to be “active.” If the density is missing, the opportunity is said to be “passive.” 

There may be many opportunities for defect for each characteristic of our product and service. However, some opportunities are active and some are passive. When we are not measuring opportunities, they are passive. But, when we measure, test, or inspect opportunities for defect, they are considered active. Any CTQ (Critical to Quality), CTD (Critical to Delivery), or CTC (Critical to Cost) would, by definition, constitute an opportunity for nonconformance, so long as it is actively measured and reported. In fact, in order to evaluate the true capability of our processes, it is important to consider only active opportunities for metric calculation. 

We can establish density for both discrete and continuous data. When we have collected the number of categorical occurrences per observation, our density is based on discrete data. When we have collected measurements that are continuous in nature, our density, the historical distribution of continuous data, is based on the total area under the curve. Different statistical distributions, such as the Binomial, Poisson, Normal, and Lognormal are the most frequently used distributions for CTs to establish density. As organizations evolve in Six Sigma, they will value measurements more and density will be based on continuous data, allowing for estimates of process capability to become more accurate. 

Opportunities can exist at any level of a hierarchy (complex, system, subsystem, component, and element levels). The hierarchy level concept can be applied to every business situation. For example, there are many levels to a policy, service, project, assembly, design, etc. For each of these levels, we count the number of opportunities and defects for each CT. This information will eventually be used to calculate separate Sigma metrics in order to obtain true estimates of process capability. As organizations evolve in Six Sigma, it will be interesting to separate opportunities into the different classes of nonconformance for each level of the hierarchy. 

There are three classes of nonconformance to standard:  defect, error, and fault. A failure at any given level of the hierarchy can be attributed to the independent or joint occurrence of all three types of nonconformance. These different classes are caused by different factors. A defect at any level of the hierarchy can be attributed to the nonconformance of a characteristic to a standard. An error is caused by an action that is not compliant to standard. A fault is attributed to the performance of a characteristic that is not compliant to standard. As we evolve in our projects, separating Sigma metrics for these three classes of nonconformance will enable us to better estimate the capability of our processes.