Optoelectronics, Optics, Lights and Lasers
May 17, 2012, 12:37:06 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News:
 
   Home   Help Search Login Register  
Pages: [1]
  Print  
Author Topic: Statistics Question Please Help.?  (Read 2330 times)
Chris_Boe
Newbie
*
Posts: 3


View Profile
« on: July 02, 2011, 06:00:12 AM »

I am working on a group assignment for a statistics class. One of the members completely dropped the ball and didn't do his part. Now I'm stuck with it and I don't know how to proceed. Please help...

Here's his question:

Why did you choose this figure as the warranty? What percentage of bulbs would need to be replaced if you chose a warranty of 7000 or 7500? Justify your answer statistically. How do you explain this to the board of directors, marketing people and financial people?

This is the preceeding question and answer that may be needed to complete it:

Your employer, Washington Electric Inc., wants to offer a warranty on the new compact fluorescent light bulb that they have produced and tested.  You are called into a meeting and operational experts provide the following data: mean bulb life = 8000 hours, standard deviation = 400 hours (assume a normal distribution).  The financial people tell you that the firm cannot afford to replace more than 2.5% of the bulbs under warranty.  Some members of the board of directors are pressuring you to come up with a warranty of 7000 hours. The marketing people are pressuring you to create a warranty of 7500 hours. Use the data and adhere to the 2.5% financial constraint above to make your calculations and recommend the highest warranty that you can. What do you recommend as a warranty?  

Bulbs replaced under 0.5000
warranty




2.5% or 0.0250
0.4750



                           X = ?               mean = 8000                scale of warranty

The area under the normal curve between X and the mean is equal to 0.4750 (0.5000 – 0.0250)
According to Appendix B.1 the closest area to this value is 0.4750 which gives a z value of 1.96 and because the value is on the left side of the mean it is actually -1.96.            

     Insert all values into formula to find z and solve for X
z = X – mean
    standard deviation

-1.96 = X – 8000
              400

-1.96 (400) = X- 8000

X = 8000 – 1.96 (400)

X = 7216


•I recommend to Washington Electric Inc. to have a warranty of 7216 hours which will meet the requirement to only replace 2.5% of the bulbs under warranty.

Anyone have any ideas on how to solve this?
Logged
mathsmanretired
Newbie
*
Posts: 2


View Profile
« Reply #1 on: July 02, 2011, 06:24:06 AM »

Standardize each of the figures 7000 and 7500 using the formula

z = (x - mean)/sd

z1 = (7000 - 8000)/400

z2 = (7500 - 8000)/400.

Use probability tables for P(z < z1) and P(z < z2). You may need to do this by using the symmetry of the Standard Normal curve because z1 and z2 will be negative and probabilities are often given only for positive z values.

EDIT. I think that you answer would need to include some explanation about the nature of normal distribution curves.
Logged
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.13 | SMF © 2006-2011, Simple Machines LLC | Privacy Policy Valid XHTML 1.0! Valid CSS!