Each laptop of 400 owned by a particular company has an 8% probability of not working. 20 are randomly selected to be used by the sales department.
What should the overall probability of not working be reduced to so that the likelihood of 5 being broken is reduced to 0.001?
p ≈ 0.0414
$${\left({\frac{{\mathtt{20}}{!}}{{\mathtt{5}}{!}{\mathtt{\,\times\,}}({\mathtt{20}}{\mathtt{\,-\,}}{\mathtt{5}}){!}}}\right)}{\mathtt{\,\times\,}}{{\mathtt{0.041\: \!4}}}^{{\mathtt{5}}}{\mathtt{\,\times\,}}{\left({\mathtt{1}}{\mathtt{\,-\,}}{\mathtt{0.041\: \!4}}\right)}^{{\mathtt{15}}} = {\mathtt{0.001\: \!000\: \!015\: \!738\: \!633\: \!1}}$$
So the overall probability should be about 4%.
.
I am not at all sure that I understand the question but the probability of exactly five being broken is given by this (I think) Where p is the prob of an individual one being broken.
In my interpretation of the question (which could easily be totally incorrect) the 400 and the 8% are irrelevant.
The web 2 calc won't solve this. Maybe the wolfram|alpha calc will give you an answer.
$${\left({\frac{{\mathtt{20}}{!}}{{\mathtt{5}}{!}{\mathtt{\,\times\,}}({\mathtt{20}}{\mathtt{\,-\,}}{\mathtt{5}}){!}}}\right)}{\mathtt{\,\times\,}}{{\mathtt{p}}}^{{\mathtt{5}}}{\mathtt{\,\times\,}}{\left({\mathtt{1}}{\mathtt{\,-\,}}{\mathtt{p}}\right)}^{{\mathtt{15}}} = {\mathtt{0.001}} = \tiny\text{Error: }$$
If you get the proper answer I'd really like you to post it.
Any one else here want to give it a shot?
p ≈ 0.0414
$${\left({\frac{{\mathtt{20}}{!}}{{\mathtt{5}}{!}{\mathtt{\,\times\,}}({\mathtt{20}}{\mathtt{\,-\,}}{\mathtt{5}}){!}}}\right)}{\mathtt{\,\times\,}}{{\mathtt{0.041\: \!4}}}^{{\mathtt{5}}}{\mathtt{\,\times\,}}{\left({\mathtt{1}}{\mathtt{\,-\,}}{\mathtt{0.041\: \!4}}\right)}^{{\mathtt{15}}} = {\mathtt{0.001\: \!000\: \!015\: \!738\: \!633\: \!1}}$$
So the overall probability should be about 4%.
.
Melody, WolframAlpha will solve this.............http://www.wolframalpha.com/input/?i=solve+C%2820%2C5%29+*p^5*%281-p%29^15+%3D+.001
thanks Chris, i thought wolfram would give the answer. I just had not gone there. :)
What I did was to divide both sides by ncr(20,5) and then take the fifth root of both sides. This leaves a quartic in p. There are four solutions to this quartic; two real and two complex. The 0.0414 was one of the real solutions (the other was around 0.6, so I doubt this was the desired result!).
(I used Mathcad to do the calculations).