It’s combinatorics. Basically, whenever you have a pool of stuff, and you want to know how many different combinations you can have, you use combinatorics.
If the order doesn’t matter (e.g. how many different ways can you pick 7 letters/numbers, where ABC1234 is the same as 1A2B3C4), it’s a combination, if the order DOES matter (e.g. how many different license plates can there be, since ABC1234 is different from 1A2B3C4), it’s a permutation.
The number of possible PERMUTATIONS of r objects from a collection of m objects is: [sub]m[/sub]P[sub]r[/sub]=(m!)/(m-r)!
Example: how many 3 letter words are there, using the 26 letters of the alphabet?
[sub]26[/sub]P[sub]3[/sub]=26!/23!=156,000
NOTE: this does not include words in which letters appear more than once, I think.
The number of possible COMBINATIONS of r objects from a collection of m objects is: [sub]m[/sub]C[sub]r[/sub]=(m!)/(r!((m-r)!))
Example: how many different combinations of three letters are there (abc is the same as cba)?
[sub]26[/sub]C[sub]3[/sub]=(26!)/(3!((26-3)!))=2600.
Aside from that, it’s just applying logic and basic probability to problems. For example, if I have a selection of 100 engines, and I know 5 are defective, if I pick 5 engines, I can calculate the probability that I selected no defective engines, exactly 3 defective engines, or even all five defective engines.
Example: What is the probability of selecting more than three defective engines?
From the pool of defective engines, I have selected three, four, or five. From the pool of non-defective engines, I have selected 2, 1, or 0, respectively.
There are [sub]5[/sub]C[sub]5[/sub] ways in which I can select 5 defective engines from a pool of five, and [sub]95[/sub]C[sub]0[/sub] ways in which I can select 0 defective engines from a pool of 95 non-defective engines. You multiply these answers together to get the total number of ways to select 5 defective engines, and you get 1.
There are [sub]5[/sub]C[sub]4[/sub] times [sub]95[/sub]C[sub]1[/sub] ways to pick exactly four defective engines and one good engine. This comes out to 475.
There are [sub]5[/sub]C[sub]3[/sub] times [sub]95[/sub]C[sub]2[/sub] ways to pick exactly three defective engines and two good engines. This comes out to 44,650.
Adding the three together, we get 45126 ways to pick three or more defective engines. There is a total of [sub]100[/sub]C[sub]5[/sub] choices, or 75,287,520 possible ways of picking five engines from a pool of 100, and 45126 of those involve picking three or more defective engines. Therefore, the probability of picking three or more defective engines from a pool of 100 engines, of which five are defective is 45126/75287520, or 5.99*10[sup]-4[/sup], or .0006, or .06%.
Additionally, the likelyhood of picking NONE of the five defective engines is ([sub]5[/sub]C[sub]0[/sub]*[sub]95[/sub]C[sub]5[/sub])/[sub]100[/sub]C[sub]5[/sub], or 57940519/75287520, or .76959, or 77%.