Page 337 - Six Sigma Demystified

Part 3 S i x S i g m a To o l S 317

due to common or special causes of variation, which is determined using a
control chart. This distinction is important to understand our response to the
high cost or occurrence: If it is due to common causes, then we must change
the underlying nature of the process, whereas if it is due to a special cause of
variation, we can respond to that particular issue.
The Pareto chart in Figure F.24 displays the number of errors of each type of
error associated with a painting process. “Color Wrong” represents the largest
source of errors (30 percent), followed by “Color Off” (an additional 24 per-
cent), and “Chips” (another 19 percent), so these categories represent the best
opportunities for improvement (if all errors have the same costs).

PerT Analysis

Program evaluation and review techniques (PERT) analysis allows probabilistic
estimates of activity times, such as in project scheduling or cycle time analysis.

When to Use

Define Stage
• To identify project activities that determine the total project duration

Analyze Stage
• To identify the critical path for cycle time reduction

Improve Stage

• To verify the reduction in process critical-path cycle time

Methodology

Begin with activity times on the critical path, such as those determined using
an activity network diagram. We assume that each activity time follows a β
distribution, where a is an estimate of the optimistic time, b is an estimate of
the pessimistic time, and m is an estimate of the most likely time (usually the
deterministic time applied to the activity network diagram).
The estimated duration for each activity is then calculated as

μ = (a + 4m + b)/6