Statistics help. Please A company manufactures a brand of light bulb with a lifetimes in months that is normally distributed with mean 3 and variance 1. A consumer buys a number of these bulbs with the intention of replacing them successively as they burn out. The light bulbs have independent lifetimes. What is the smallest number of bulbs to be purchased so that the succession of light bulbs produces light for at least 40 months with probability at least 0.9772?
\(X_1, X_2,\cdots, X_n\)~ \(N(3,1)\) \(X=\sum_{i=1}^n X_i\)~ \(N(3n,n)\)
\(P(X \geq 40)\geq 0.97772\)
\(=1-P(X\leq 40)\geq 0.97772\) Am I Ok up to now?
Yeah
okay..
\[1-P (\dfrac{X-its~mean}{its~standard~deviation}\leq \dfrac{40-its~mean}{its~standard~deviation})\geq 0.9772\]
That gives me. \[P (\dfrac{X-its~mean}{its~standard~deviation}\leq \dfrac{40-its~mean}{its~standard~deviation})\leq 1- 0.9772=0.22228\]
right?
There is a calculation mistake in RHS
0.0228
where?
oh yes.
then, \(P(Z\leq \dfrac{40-3n}{\sqrt n})\leq 0.02228\)
Okay
use calculator, the invnorm, I have the ( ) =-2
Okays...
and solve for n, I got n = 16 and n =11.1
Perfect! I got 16
wait, let me double check,
(y) Take your time
Yes, but we have 2 roots, 16 and 11.1, why do you pick 16?
I actually used a different method... I got 16 and a negative root which I eliminated
P(T >= 40) = \(\large{P(Z\ge \frac{40 - 3n}{\sqrt{n}})}\) = \(\large{ 1 - \phi (\frac{40 - 3n}{\sqrt{n}})}\) = \(\large{\phi(\frac{-40 + 3n}{\sqrt{n}}) }\) = 0.9722 So, I got.. \(\large{\frac{-40 + 3n}{\sqrt{n}}}\) = 2 => \(3n - 2\sqrt{n} - 40 = 0\) => \(\sqrt{n} = -10/3\) or 4
I ignored the other one.. and got n = 16
Also, n must be a whole number... we can't have fractional units of bulbs
The question is "at least", hence if we have fraction, then, take floor function. why not? One more \(\sqrt n= -10/3\rightarrow n = 100/9\) . Either ways, you cannot get rid it off, right?
Hmm.. there is that too.. why don't we just plug in n = 100/9 and find out whether it satisfied our original condition or not
Yes, that is a good idea, Show me, please
One sec...
So, let's assume 12 bulbs
So, expected life = 36 months variance = 12 => S.D. = sqrt(12)
Now, we need a lifetime of at least 40 months... i.e. 1.155 s.d. away from our expected life
What are the chances for that?
GGGGGGot you. thank you so much. I need completely understanding why I can get rid of the other root. Now, I know why. Again, thank you very much.
Glad I could help.. it's been a whole year since I last read probability distributions :P
btw, I got solution for my PDE problem also :)
Join our real-time social learning platform and learn together with your friends!