The vertices of a regular 10-gon are labeled V 1 , V 2 , … V n , which is a permutation of { 1 , 2 , … , 1 0 } . Define a neighboring sum to be the sum of 3 consecutive vertices V i , V i + 1 and V i + 2 [where V 1 1 = V 1 , V 1 2 = V 2 ]. For each permutation σ , let N σ denote the maximum neighboring sum. As σ ranges over all permutations, what is the minimum value of N σ ?
Details and assumptions
If the integers are written as 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 1 0 around the circle, then the neighboring sums are 6 , 9 , 1 2 , 1 5 , 1 8 , 2 1 , 2 4 , 2 7 , 2 0 , 1 3 , and the maximum neighboring sum is 27.
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Common mistakes:
Need to show that minimum value is 18, instead of saying that "I can't do 17".
Need to show that 18 can be achieved.
Note: Students who used the pigeonhole principle on all of the neighboring sums, found that they can only conclude that the minimum value is ⌊ 1 0 5 5 × 3 ⌋ = 1 7 . By removing the most useless information (namely V 1 = 1 ), we can improve the bound.
For all i , we define f ( i ) = V i − 1 + V i + V i + 1 , where the indices are read modulo 1 0 . Note that for all σ , we have i = 1 ∑ 1 0 f ( i ) = 3 ( 1 + 2 + ⋯ + 1 0 ) = 1 6 5 , since each of the numbers must appear thrice in the sum. It follows that max ( f ( i ) ) ≥ 1 0 1 6 5 = 1 6 . 5 , i.e. max ( f ( i ) ) ≥ 1 7 .
Before we proceed, we prove the following claim.
Claim: ∀ i , f ( i + 1 ) = f ( i )
Proof: On the contrary, assume there exists an i such that f ( i + 1 ) = f ( i ) . This is equivalent to V i + V i + 1 + V i + 2 = V i − 1 + V i + V i + 1 , which in turn implies V i + 2 = V i − 1 , which is a contradiction since all V i 's are distinct. ■
Now, suppose there exists a permutation σ for which max ( f ( i ) ) = 1 7 . Note that more than 5 f ( i ) 's cannot be 1 7 , since otherwise there would exist an i such that f ( i ) = f ( i + 1 ) . If less than 5 f ( i ) 's are equal to 1 7 , we have i = 1 ∑ 9 f ( i ) ≤ 4 × 1 7 + 6 × 1 6 < 1 6 5 , which is a contradiction. We thus conclude there exist precisely 5 i 's such that f ( i ) = 1 7 . Let S denote the sum of the other f ( i ) 's. Then, S + 5 × 1 7 = 1 6 5 ⟹ S = 8 0 . But since the other 5 f ( i ) 's are atmost 1 6 and 8 0 = 1 6 × 5 , all the other f ( i ) 's must be 1 6 .
By the claim, it follows that the sequence { f ( i ) } i = 1 1 0 alternates between 1 6 and 1 7 . In other words, if f ( i ) = 1 7 , f ( i + 1 ) = 1 6 , and vice-versa.
Let { K 1 , K 2 , ⋯ , K 6 , K 7 } be any seven consecutive vertices of the polygon where f ( K 1 ) = 1 7 , so f ( K 2 ) = 1 6 , f ( K 3 ) = 1 7 , f ( K 4 ) = 1 6 , f ( K 5 ) = 1 7 , f ( K 6 ) = 1 6 , f ( K 7 ) = 1 7 . Then, note that K 1 + K 2 + K 3 + K 4 + K 5 + K 6 = f ( K 2 ) + f ( K 3 ) = 3 3 K 2 + K 3 + K 4 + K 5 + K 6 + K 7 = f ( K 3 ) + f ( K 6 ) = 3 3 . Combining them, we find out K 1 = K 7 , a contradiction.
The minimum potential value of max ( f ( i ) ) , thus, is 1 8 . A quick hack shows that this can indeed be attained for the permutation σ = { 1 , 6 , 8 , 4 , 3 , 1 0 , 5 , 2 , 9 , 7 } . We thus conclude our desired answer is 1 8 .
Very good! Nearly the same as my method.
I like your solution very much!
Typo: K 1 + K 2 + K 3 + K 4 + K 5 + K 6 = f ( K 2 ) + f ( K 5 ) = 3 3 .
The total of all 10 neighbouring sums is always 3*(1+2+...+10) = 165, because each number is included thrice. The highest neighboring sum is therefore at most ceiling(165/10) = 17.
Place each neighbouring sum in front of the physically centre of the three numbers it contains, like this:
Numbers Neighbouring sums a m p ; a a m p ; a m p ; a m p ; a m p ; b a m p ; a + b + c a m p ; a m p ; a m p ; c a m p ; b + c + d a m p ; a m p ; a m p ; d a m p ; a m p ; a m p ; a m p ; … a m p ; …
Suppose two adjacent neighbouring sums are equal. Then a+b+c = b+c+d; therefore, a = d. Contradiction. Therefore the assumption is false: no two neighbouring sums are equal.
Suppose the highest neighbouring sum is 17. There must be at least 5 17s among the neighbouring sums. Because no two adjacent neighbouring sums are equal, they must alternate between 17 and 16.
Numbers Neighbouring sums a m p ; a a m p ; 1 7 a m p ; a m p ; a m p ; b a m p ; 1 6 a m p ; a m p ; a m p ; c a m p ; 1 7 a m p ; a m p ; a m p ; d a m p ; 1 6 a m p ; a m p ; a m p ; e a m p ; 1 7 a m p ; a m p ; a m p ; f a m p ; a m p ; a m p ; a m p ; g a m p ; a m p ; a m p ; a m p ; … a m p ; … Then a + b + c + d + e + f = 17 + 16 = 33, and b + c + d + e + f + g = 16 + 17 = 33; therefore, a - g = 33 - 33 = 0, therefore a = g. Contradiction. Therefore, the assumption is false: the highest neighbouring sum cannot be 17.
The highest neighbouring sum can be 18, as shown here:
Numbers N’g sums a m p ; 1 0 a m p ; 1 8 a m p ; a m p ; a m p ; 5 a m p ; 1 7 a m p ; a m p ; a m p ; 2 a m p ; 1 6 a m p ; a m p ; a m p ; 9 a m p ; 1 8 a m p ; a m p ; a m p ; 7 a m p ; 1 7 a m p ; a m p ; a m p ; 1 a m p ; 1 4 a m p ; a m p ; a m p ; 6 a m p ; 1 5 a m p ; a m p ; a m p ; 8 a m p ; 1 8 a m p ; a m p ; a m p ; 4 a m p ; 1 5 a m p ; a m p ; a m p ; 3 a m p ; 1 7
Therefore, the answer is 18.
[Calvin - Slight edits to align the 'diagrams' as per latter posts]
Comments and replies:
Calvin:
Can you clarify why "Because no two adjacent neighbouring sums are equal, they must alternate between 17 and 16."? Why can't the neighboring sums be 17 15 17 16? This would avoid the issue raised in the next paragraph.
Otherwise, this is a good solution. Can you think of a way to show more directly that 18 is the minimum possible value?
C:
They all add to 165. There must be at most 5 17's. otherwise two of them will be adjacent. There must be at least 5 17's, otherwise the total would be at most 4 * 17 + 6 * 16 = 164. Therefore, there are exactly 5 17's, which must be placed in every other position to prevent them from meeting. The total of the other 5 is then 165 - 5 * 17 = 80. Each of them must be at most 16, and 16 * 5 = 80, so each must be exactly 16.
The tabs did not work. Rewriting the 'diagrams':
Numbers a b c d … Neighbouring sums a+b+c b+c+d …
Numbers 10 5 2 9 7 1 6 8 4 3 N’g sums 18 17 16 18 17 14 15 18 15 17
C:
The spaces did not work. Rewriting the ‘diagrams’:
Numbers .................a............b............c............d… Neighbouring sums...........a+b+c....b+c+d..........…
Numbers 10...5...2...9...7...1...6...8...4...3 N’g sums 18 17 16 18 17 14 15 18 15 17
C:
Also, the second diagram that I forgot:
Numbers ……………a………b………c………d………e………f………g… Neighbouring sums…………17…….16……..17…….16……..17………
Problem Loading...
Note Loading...
Set Loading...
WLOG let V 1 = 1 . Observe that the 3 neighbouring sums ( V 2 + V 3 + V 4 ) , ( V 5 + V 6 + V 7 ) , ( V 8 + V 9 + V 1 0 ) sum to 5 5 − V 1 = 5 4 . Thus by pigeonhole principle, at least one is larger than or equal to 5 4 / 3 = 1 8 .
We now only need to construct s.t. the maximum sum is 18. Arrange as follows:
1 , 7 , 3 , 8 , 5 , 4 , 9 , 2 , 6 , 1 0 .
The resulting neighbouring sums are:
1 1 , 1 8 , 1 6 , 1 7 , 1 8 , 1 5 , 1 7 , 1 8 , 1 7 , 1 8 .
Hence, we see that 18 is indeed the minimum value.