Interpretando Gráficos R

The centerline on an R chart represents the mean sample range, calculated by averaging all sample ranges from the process data. In the cookie thickness example, the mean of ten sample ranges is 0.252, which serves as the central reference line. This value indicates the average variability expected when the process operates under normal, stable conditions.

Upper and lower control limits are calculated by multiplying the mean sample range by control chart constants D4 and D3, respectively. These constants depend on sample size; for ten measurements per batch, D3 equals 0.223 and D4 equals 1.777. The resulting limits define the boundaries for acceptable process variability.

When all points fall within control limits, the process variability is statistically stable and in control. This indicates that the variation in the process—such as cookie thickness fluctuations—results from normal, random causes rather than special causes. However, being in control for variability does not guarantee the process mean is also centered correctly.

For sample sizes smaller than six, the D3 constant is typically zero because small samples provide insufficient data to reliably estimate lower control limits. This mathematical convention prevents artificially restrictive lower limits that could misidentify normal variation as out-of-control. As sample size increases, D3 becomes positive, allowing meaningful lower limit calculation.

Consistent patterns such as gradual increases or decreases in range values, or points falling outside control limits, indicate process shifts or trends requiring investigation. These patterns suggest special causes affecting process variability, such as equipment wear or environmental changes. Random distribution of points within limits indicates a stable, controlled process.

An R chart monitors process variability or dispersion, while complementary charts like the X-bar chart track the process mean. Together, they provide complete process control by detecting changes in both spread and central tendency. A process can be in control for variability but out of control for mean, or vice versa, making both charts essential for quality assurance.

Range represents the difference between maximum and minimum values within a sample, while individual measurements are single data points. R charts use range to assess overall process dispersion, which is more sensitive to detecting variability changes than individual values alone. This approach provides a clearer picture of process spread over time.