パレート図の使用が最も適切である分析対象はどれか。
平成21年度 秋期 ITパスポート試験 問2
This question's explanation and your answers so far.
問 2
Incorrectア
The reliability of a production process is evaluated with metrics such as availability, MTBF, and failure rate, or with a control chart that shows over time whether the process is in a stable state. A Pareto chart shows the rank order and composition ratio of items; it is not a tool for measuring reliability itself.
Incorrectイ
A histogram is the right tool for looking at variation in product weight (the shape, center, and spread of the distribution). A histogram divides measured values into classes and shows the frequency distribution, whereas a Pareto chart arranges classified items in descending order, so it is not suited to grasping the variation of a continuous quantity.
- ウ
正解 A Pareto chart is one of the seven QC tools. It arranges classified items as bars in descending order of count or amount, and overlays a line showing the cumulative composition ratio. Its purpose is to visualize how large a share each item takes of the whole, so that countermeasures can be focused on the small number of top items (a priority-focused approach, the same idea as ABC analysis). Choice ウ, the composition ratio of quality defects by cause, means classifying defects by cause, ordering them from most to fewest, and identifying the leading causes with the cumulative ratio — exactly the typical use of a Pareto chart. Therefore ウ is the most appropriate.
Incorrectエ
A scatter diagram is used to examine the relationship between two varying factors (for example temperature and defect rate), that is, whether a correlation exists. A Pareto chart shows the composition ratio along a single classification axis and cannot express the relationship between two variables.
- Pareto chart = bars sorted in descending order plus a cumulative-ratio line. It visualizes composition ratios and identifies the small number of top-priority items (ABC analysis, priority focus); it is one of the seven QC tools.
- Choosing among the seven QC tools: use a histogram for the distribution of variation, a scatter diagram for the relationship between two variables, a control chart for process stability over time, and a cause-and-effect (fishbone) diagram for enumerating causes.