Let be a regular 2006-gon. A diagonal of is called *good segment* if its endpoints divide the boundary of into two parts, each composed of an odd number of sides of . The sides of are also called *good segment*.
Suppose has been dissected into triangles by 2003 diagonals, no two of which have a common point in the interior of . Find the maximum number of isosceles triangles having two *good segments* that could appear in such a configuration.
Problem 1978
Official solution
First Solution: We start with the following lemma.
Lemma Let is a diagonal used in , and is non-major and contains segments of , then there are at most good triangles with vertices on . More precise there are at most
good triangles with vertices on
*Proof:* Without loss of generality, we may assume that . We induct on .
The bases cases for and are trivial. Assume the statement is true for with and . We consider the case .
Let be a triangle in with on . (Note that , and lie on non-major arc on in clockwise order. By the induction hypothesis, there are at most
good triangles with vertices on . Similar result holds for .
Because is a triangle in , we conclude that if a good triangles has its vertices on then either it is , or all its vertices are on exactly one of or . We can now apply the induction hypothesis and . We conclude that there are at most
good triangles with vertices on .
To finish our proof, we need to reduce the value of the right-hand side of (‡) by 1. We consider the following two cases.
In the first case, we assume that is not good. The summand 1 on the right-hand of (†) should be taken out, and we are done.
In the second case, we assume that is good. Since is non-major, and . We must have and must be the two equal good sides, and both must be the good sides. Hence both and are odd, and so we can improve (†) to
and similar result hold for . Then (‡) can be improved to
completing our induction.
■
Since is non-major, and . Since is good, we must have and be the good segments (with equal lengths). Thus and are both odd. By the induction hypothesis, there are at most
good triangles with vertices on . Similar result holds for .
Now we prove our main result. Let be the longest diagonal used in . Let be a non-obtuse triangle in . Without loss of generality, we may assume that . Since is non-obtuse, , , and are all non-major. By the lemma, there are at most
good triangles besides .
If is not good, we are done. If it is, then exactly two of , , and are odd, and so (*) is strict inequality. We still have at most good triangles in this case, completing our proof.
Second Solution: Let () be a good triangle, with and being good segments. This means that there are an odd number of sides of between and and also between and . We say and belong to triangle ABC.
At least one side in each of these groups does not belong to any other good triangle. This is so because any odd triangle whose vertices are among the points between and has two sides of equal length and therefore has an even number of sides belonging to it in total. Eliminating all sides belonging to any other good triangle in must therefore leave at least one side that belongs to no other good triangle. Same argument applies to . Let us assign these two sides (one in and one in ) to triangle .
To each good triangle we have thus assigned a pair of sides, with no two good triangles sharing an assigned side. It follows that at most 1003 good triangles can appear in the triangulation; that is, .