X-bar charts when the mean and standard deviation not known , Applied Statistics

2086_computation of covariance ungrouped data2.png Charts when the Mean and the Standard Deviation are not known

We consider the data corresponding to the example of Piston India Limited. Since we do not know population mean and the standard deviation, we employ the sample mean as a substitute. But for this data we have ten sample means and we do not know which one truly reflects the process mean. Therefore under these circumstances utilizing all the available data will be the best alternative. The calculation of mean from the sample means (this is referred to as grand mean and is denoted by   876_x charts.png  ), the calculation of the range and its mean are shown in the table 1.    

Figure 8

485_x charts1.png

 

The grand mean, that is 1204_x charts.png,  is calculated by employing either 1859_x charts3.png


Both gives us the same answer and it can be seen that the second formula is derived from the first one. Here 'n' denotes the number of samples that are randomly selected on a day and k denotes the number of days we sample the pistons.

Table 1

776_x charts2.png 

The value of the grand mean in our problem by the first formula is given by

1812_x charts4.png = 10.437 

and by the second version it is

1964_x charts5.png = 10.437.

As mentioned both the values are equal. This value of  1204_x charts.png is utilized in drawing the center line in the chart. Now we look at how we calculate the standard deviation.

As we have employed the sample mean as a substitute for population mean, do we substitute the sample standard deviation for the population standard deviation? No. Since the control charts are usually plotted on the factory floor and since it is easy for the personnel to calculate sample range as compared to the sample standard deviation, it became a norm to employ the mean of the sample range in the calculation of the standard deviation. However, the relationship between the process standard deviation and the mean of sample range is given by a factor denoted by "d2 ". This factor depends on the size of the sample "n". The value of d2 can be found from the tables. For computing the values of the upper and the lower limits we employ the following formulae.



1534_x charts6.png

Further these limits can be expressed as  1183_x charts7.png  A2 denoting  1545_x charts8.png . It is possible to look for these values directly from the tables.

From the tables, the value of d2 is equal to 2.704 for a sample size of 7. In our example the values of the lower and the upper control limits are given by

LCL  = 10.437 -  2497_x charts9.png  = 10.27
UCL   = 10.437 + 1936_x charts10.png = 10.60

Using these values we plot the center and the limit lines. Nevertheless, we employ  2086_computation of covariance ungrouped data2.png  values to get the  2086_computation of covariance ungrouped data2.png  chart and determine whether the process is 'in-control' or 'out-of-control'. In our problem the chart denotes that the process is in-control and there is enough scope for us to tighten the control limits.

Posted Date: 9/15/2012 7:08:49 AM | Location : United States







Related Discussions:- X-bar charts when the mean and standard deviation not known , Assignment Help, Ask Question on X-bar charts when the mean and standard deviation not known , Get Answer, Expert's Help, X-bar charts when the mean and standard deviation not known Discussions

Write discussion on X-bar charts when the mean and standard deviation not known
Your posts are moderated
Related Questions
Empirical Mode Where mode is ill-defined, its value may be ascertained by the following formula based upon the empirical relationship between Mean, Median and Mode: Mode = 3

Make a decision about the given claim. Use only the rare event rule, and make subjective estimates to determine whether events are likely. For example, if the claim is that a coi

Testing of Hypothesis One objective of sampling theory is Hypothesis Testing. Hypothesis testing begins by making an assumption about the population parameter. Then we gather

What is a null hypothesis? ..


The Null Hypothesis - H0:  The random errors will be normally distributed The Alternative Hypothesis - H1:  The random errors are not normally distributed Reject H0: when P-v

There are situations where none of the three averages is fully satisfactory. For example, if the number of items in a series is very small, none of these av

Question: (a) (i) Define the term multicollinearity. (ii) Explain why it is important to guard against multicollinearity. (b) (i) Sometimes we encounter missing values

TYPE I AND II Errors If a statistical hypothesis is tested, we may get the following four possible cases: The null hypothesis is true and it is accepted; The

Weighted Arithmetic Mean Another aspect to be considered is the importance we assign to each observation. The arithmetic mean as we calculated it so far gives equal