Fraction defective p chart - control charts for attributes, Operation Management

Fraction Defective P Chart - Control Charts for Attributes

This type of chart is used when it is impractical or inconvenient to maintain a constant sample size. Instead of plotting the number of defectives in each sample, the proportion (p) or fraction of defectives is calculated and then plotted. Clearly, to ensure that meaningful plots are made, the sample sizes must be large enough to ensure that a number of defectives will typically be found. This may seem to imply that some level of non-conformance is being accepted, but this is not necessarily so. For example, consider the Go/Not-go gauging example shown in the figure on page 249. Whilst the design limits of size on the component diameter might be, say, 24.98 mm and 25.02 mm; for process control purposes the gaps in the gauge might be set at, say, 24.99 and 25.01 respectively. Remembering that, for attributes, only two states are possible (defective/non defective) the relevant probability distribution is the binomial distribution, the mean of which (¯p) can be estimated by inspecting a large number of items:

¯p = total number of defectives/ total number inspected

An estimate of the standard deviation of the binomial distribution (σp), for samples of size n is given by:

σp = √(¯p (1-¯p) / n)

For small values of n the profiles of the binomial distributions are distinctly skewed, but, for large values of n and, particularly, when the product p n is greater than about 5, the distributions are sufficiently symmetrical to be well-approximated by the normal distribution. In this latter situation the control chart limits can be determined in the same manner as for variables charts.

2206_Fraction Defective P Chart - Control Charts for Attributes.png

It will be realised that the implied constant sample size (n) in these expressions presents some difficulty in a procedure when the sample size is allowed to vary. This can be surmounted by using an average value for n, and limiting variation in the size of subsequent samples to, say, ±20 per cent of this value: the results will still be sufficiently accurate for most process control purposes.

Posted Date: 3/16/2013 1:41:42 AM | Location : United States







Related Discussions:- Fraction defective p chart - control charts for attributes, Assignment Help, Ask Question on Fraction defective p chart - control charts for attributes, Get Answer, Expert's Help, Fraction defective p chart - control charts for attributes Discussions

Write discussion on Fraction defective p chart - control charts for attributes
Your posts are moderated
Related Questions
The Legal environment of HRM includes EEO, Affirmative Action, and Americans with Disabilities Act. Discuss each explaining significance.

What are the different network perspectives in innovative research? Why do authors differ in how they define innovation networks? How can networks influence actions of its members?

1 In Johnson v. Transportation Agency, Santa Clara County, a female employee was promoted to the position of road dispatcher, despite the fact that a male candidate had scored two

The number of allocated cells a balanced transportation model with 3 sources and 3 destinations should have in order NOT to be degenerate is

When you look at the pay of unionized employees vs non-union in similar job classifications, what are the differences in terms of amount of pay, how the pay levels are determined,

Which of the following rules maximizes the number of orders processed in a center and minimizes the number orders waiting in the queue? Shortest operation next First-come/fir

Why mary parker follett ideas tended to be popular

Wally and Beaver were general partners in a car dealership. Using the partnership name, Wally entered into a contract to purchase 30 used cars from Car Lots. Car Lots was not aware

Discuss why organizations need to have the correct performance metrics to drive quality. Think of examples in the news and their outcomes.

Explain how personal selling can help solve the problem of information overload.