# Question

The data in Table 9E.1 represent individual observations on molecular weight taken hourly from a chemical process. The target value of molecular weight is 1,050 and the process standard deviation is thought to be about σ = 25.

(a) Set up a tabular CUSUM for the mean of this process. Design the CUSUM to quickly detect a shift of about 1.0in the process mean.

(b) Is the estimate of σ used in part (a) of this problem reasonable?

(a) Set up a tabular CUSUM for the mean of this process. Design the CUSUM to quickly detect a shift of about 1.0in the process mean.

(b) Is the estimate of σ used in part (a) of this problem reasonable?

## Answer to relevant Questions

Set up a tabular CUSUM scheme for the flow width data used in Example 6.1 (see Tables 6.1 and 6.2). When the procedure is applied to all 45 samples, does the CUSUM react more quickly than the x chart to the shift in the ...Consider the “minute clinic” waiting time data in Exercise 6.66. These data may not be normally distributed. Set up a CUSUM chart for monitoring this process. Does the process seem to be in statistical ...Consider a standardized two-sided CUSUM with k = 0.2 and h=8. Use Siegmund’s procedure to evaluate the in-control ARL performance of this scheme. Find ARL1 for * = 0.5. In control ARL performance: Reconsider the data in Exercise 9.7. Apply an EWMA control chart to these data using = 0.1 and L = 2.7. 12.16 , CL = 0 = 950, UCL = 957.53, LCL = 942.47. Show that if the process is in control at the level µ, the exponentially weighted moving average is an unbiased estimator of the process mean.Post your question

0