Combinations problem..

This topic has expert replies
Master | Next Rank: 500 Posts
Posts: 103
Joined: Sat Jun 02, 2012 9:46 pm
Thanked: 1 times

Combinations problem..

by topspin360 » Mon Jul 02, 2012 5:42 pm
How many ways are there to arrange six people (A,B,C,D,E,F) if B should always be behind A?

is the answer to this question 6!/2? Since 6! and half the times B is in front of A?

If my answer is correct, anyone know any more advanced versions of this type of problem?

Thanks.
Source: — Problem Solving |

Master | Next Rank: 500 Posts
Posts: 363
Joined: Sun Oct 17, 2010 3:24 pm
Thanked: 115 times
Followed by:3 members

by theCEO » Mon Jul 02, 2012 9:13 pm
How many ways are there to arrange six people (A,B,C,D,E,F) if B should always be behind A?

Lets merge A and B into one person and call them X.

The question then becomes, how many ways can you rearrange 5 people (x,c,d,e,f)?
Answer = 5*4*3*2*1 = 5!

Since A is always infront of B - there is one conformation of AB
Final answer = 1 * 5!

If question had stated A should be next to B - there would be 2 conformation (AB or BA)
The answer would be 2 * 5!

Newbie | Next Rank: 10 Posts
Posts: 1
Joined: Sat Feb 20, 2010 7:40 am
Thanked: 1 times

by lazzybug » Mon Jul 02, 2012 9:25 pm
Hi CEO,

Your proposed solution is valid in the condition that B is right behind A always. I suppose the question does not mention that.I have seen this question in MGMAT test and the solution provided is in line with Topspin....

Any experts who can throw some light please.....

Master | Next Rank: 500 Posts
Posts: 126
Joined: Sun Jun 24, 2012 10:11 am
Location: Chicago, IL
Thanked: 36 times
Followed by:7 members

by tutorphd » Mon Jul 02, 2012 10:31 pm
Topspin solution is correct if the problem doesn't mean 'right behind A'.

A more advanced version is to ask for the number of ways in which B is behind A, and C is behind B.
Skype / Chicago quant tutor in GMAT / GRE
https://gmat.tutorchicago.org/

Master | Next Rank: 500 Posts
Posts: 103
Joined: Sat Jun 02, 2012 9:46 pm
Thanked: 1 times

by topspin360 » Tue Jul 03, 2012 7:06 am
thanks forthe clarification tutorphd.

to answer your new question: would it be (n-1)! * (n-2)!

if the scenario looks like this: A B C D E F, there are 5 ways for B to be behind A, if A was in the second spot there would be 4 ways for B to be behind A. Similar logic can be applied to C behind B. Hence we come out with 5! for B and 4! for C, so it's 5! * 4!?

is this the most optimal way to do this?

User avatar
Legendary Member
Posts: 520
Joined: Sat Apr 28, 2012 9:12 pm
Thanked: 339 times
Followed by:49 members
GMAT Score:770

by eagleeye » Tue Jul 03, 2012 10:37 am
topspin360 wrote:thanks forthe clarification tutorphd.

to answer your new question: would it be (n-1)! * (n-2)!

if the scenario looks like this: A B C D E F, there are 5 ways for B to be behind A, if A was in the second spot there would be 4 ways for B to be behind A. Similar logic can be applied to C behind B. Hence we come out with 5! for B and 4! for C, so it's 5! * 4!?

is this the most optimal way to do this?
Hi topspin360:

Unfortunately, it won't be 5!*4!.
The reason is that When A is in first place, there are only 4 places where B can be, since C has to be after B. So we need to count only the cases where A>B>C order is maintained.

The one way which can always be used is fundamental principal of counting (the so-called slot method). Fix A, Then Fix B at various places and count the cases. If I am not sure of doing a problem, I tend to default to that one.

Fortunately, for the problem at hand, there is an easier (read "formulaic") way of doing these kind of order maintenance problems. It is specific; however it is pretty fast. When there are n objects and you need r objects to be in a certain relative order, the number of ways of doing so is :

n!/r!.

For example, for the first one you did, it was 6!/2!. For this new problem where A > B > C, 3 objects A, B and C are in a fixed order. Hence number of ways for the favorable cases among the 6! arrangements = 6!/3!.

Try to test it for A>B>C>D>E case.

Now for the more important idea of how did I come up with this idea. Let's start with an example.

Let's say there were 4 items A,B, and C, D and we needed to have the cases where A was always in front of B, and B in front of C. Now we can arrange A,B,C among themselves in 3! ways. We can arrange A,B,C,D in 4! ways. Of these 4! ways, A,B,C among themselves maintain 3! different orders. Hence 4!/3! are the ways that 3 items out of 4 maintain a certain order. (If you need to convince yourself, write out the 24 cases, and highlight the cases in which A > B > C order is maintained).
So the overall formula is for:
Q: If out of n objects, r maintain a certain relative order, what is the number of ways of in which these n objects can be arranged?
A: n!/r!.


Let me know if this helps :)

Master | Next Rank: 500 Posts
Posts: 363
Joined: Sun Oct 17, 2010 3:24 pm
Thanked: 115 times
Followed by:3 members

by theCEO » Tue Jul 03, 2012 1:54 pm
lazzybug wrote:Hi CEO,

Your proposed solution is valid in the condition that B is right behind A always. I suppose the question does not mention that.I have seen this question in MGMAT test and the solution provided is in line with Topspin....

Any experts who can throw some light please.....

Thanks for the clarification lazzybug. I was under the impression that behind means right behind!

Master | Next Rank: 500 Posts
Posts: 103
Joined: Sat Jun 02, 2012 9:46 pm
Thanked: 1 times

by topspin360 » Tue Jul 03, 2012 2:22 pm
eagleeye, thanks a lot for such a detailed explanation. If BeattheGmat had an option, I would have paid for your effort. I think BeattheGmat needs to incorporate some kind of a "Pay for a post" system.

User avatar
Legendary Member
Posts: 520
Joined: Sat Apr 28, 2012 9:12 pm
Thanked: 339 times
Followed by:49 members
GMAT Score:770

by eagleeye » Wed Jul 04, 2012 5:04 am
topspin360 wrote:eagleeye, thanks a lot for such a detailed explanation. If BeattheGmat had an option, I would have paid for your effort. I think BeattheGmat needs to incorporate some kind of a "Pay for a post" system.
You are welcome topspin360. We can figure out some "alternative arrangement" ;).