Dr. Mikel Harry answers the question, “How can Six Sigma be applied to the Human Resources (HR) function?”
Listed are some simple examples for Human Resource projects:
1) Employee turnover rate.
2) Job satisfaction issues (surveys).
3) Management satisfaction.
4) Cafeteria food quality.
5) Policy deployment.
6) New hire process quality and cycle time.
7) Health care costs.
8) Safety and compliance issues.
9) Employee exit analysis.
10) University relations.
Remember, each “issue” is underpinned by at least one critical-to-quality characteristic (CTQ). For example, consider the issue referred to as “university relations.” Given this, we might ask: “What must go right during a typical encounter with a university official?” Also: “What could go wrong?” Such an internal examination of your past experience will help to reveal the CTQ’s. Without saying, it frequently helps to involve others. In additions, tools such as fault-tree analysis, failure modes effects analysis, fishbone diagrams, and other such methods can greatly assist.
Naturally, this discussion reflects the purpose and intent of the “Define” phase – as related to DMAIC. Following this, it will be necessary to establish a “scale of measure” for each CTQ. For example, let us again look at the issue of university relations. As would be known, it is important for corporations to maintain a good rapport with the university’s key staff. Consider the relations that come to play when attempting to recruit new college graduates. Often, executives that have previously cultivated good relations get the “first call” about a really good student. Of course, this constitutes value for the corporation, the university, as well as the student.
A metric that could be used to report on such a phenomenon might be related to “university staff satisfaction.” By design, such a scale could assume the numeric values 1 through 5. Of course, these quasi-measurements are really “assessments.” Nonetheless, this type of scale seeks to rate the “extent to which you agree (or disagree) with the following statements.”
On the surface, such an analysis appears to be nothing more than a simple “smiles test.” However, many things can be covertly gleamed from a statistical examination of the data. For example, it might be possible to surface the demographics that are most highly associated with student recruitment rates. Other “process level” factors can be cross-tabulated and subsequently correlated by way of such statistics as chi-square, the phi coefficient, lambda’s uncertainty coefficient, and so on. As one might surmise, such a level of investigation constitutes the “Analyze” phase.
The “Improve” phase typically involves the postulation of one or more new ways of doing things and then designing a confirmation study. For example, it might be postulated that “more follow-on by e-mail” will improve relations (on the 1-5 scale). In this case, the study is designed and then implemented. If this factor turns out to be among the “Vital Few” versus the “Trivial Many,” the Black Belt must then establish a control plan for the given contributory variable. Thus, the “Control” phase of DMAIC is executed. Of course, the ultimate gain in value must be validated over some rational period of time. Once this has been accomplished, the project is “closed out.”