DQM Common Units

Statistical Quality Control flashcards

Read the question, say the answer to yourself, then turn the card to check. The first 10 of 49 cards are free; the rest come with the subject.

Card 1 of 49

Jump to a card

Complete Flashcard Reference

Review every card in this deck. Open any card to reveal its answer. The first 10 are free; the other 39 come with the subject.

Distinguish a population from a sample, and a parameter from a statistic.

Population: the complete set of items or observations a study is interested in. Sample: a subset of the population actually observed or measured. Parameter: a numerical characteristic of the population, such as the true population mean. Statistic: a numerical characteristic calculated from a sample, used to estimate a parameter.

What is Yamane's formula for sample size, and what does it give for N = 500 at e = 0.05?

n = N / (1 + N·e²), where N is the population size and e the acceptable margin of error. n = 500 / (1 + 500 × 0.0025) = 500 / 2.25, about 222 units.

What is data coding?

Assigning a consistent numeric or short symbolic value to each category of a qualitative variable, for example 1 = surface scratch, 2 = dimensional error, 3 = material flaw. It lets the data be entered, sorted and counted consistently.

Which chart suits categories, continuous data in intervals, and a value over time?

Categories: a bar chart (comparing them) or a pie chart (each one's share of a whole). Continuous data grouped into intervals: a histogram. A value changing over time: a line graph.

What is the strength and the weakness of the mean, the median and the mode?

Mean: uses every value, but is pulled toward outliers. Median: unaffected by extreme outliers, so better for skewed data. Mode: the only measure meaningful for purely categorical data, but can be undefined or have more than one value.

What is the coefficient of variation, and why is it used?

CV = (standard deviation / mean) × 100%. It is a relative measure of dispersion: it expresses spread as a proportion of the mean rather than in the data's own units, so the variability of datasets on different scales can be compared. Standard deviation 2 and mean 5 give CV = 40%.

Which scenario suits the binomial, the Poisson and the normal distribution?

Binomial: a fixed number of pass/fail trials. Poisson: counting rare events, such as defects, over an interval. Normal: a continuous measurement clustering around an average, such as weight or fill volume.

What does a control chart plot, and what does it signal?

A process characteristic over time against a centre line (the process average) and upper and lower control limits, conventionally three standard deviations from the centre line. A point outside the limits, or an unusual pattern within them, signals special-cause variation worth investigating, as distinct from the common-cause variation every process has.

How are c chart control limits set, and what are they when c̄ = 4?

UCL = c̄ + 3√c̄ and LCL = c̄ − 3√c̄. With c̄ = 4: UCL = 4 + 3 × 2 = 10; LCL = 4 - 6 = -2, treated as 0 because a defect count cannot be negative.

What does Six Sigma aim for, and what are the steps of DMAIC?

A defect rate of no more than 3.4 defects per million opportunities, by reducing process variation. DMAIC: Define the problem, Measure current performance, Analyze root causes, Improve the process, Control it to sustain the gain.

What is sampling?