Find the Maximum

Algebra Level 3

As a , b , c a,b,c range over all positive reals less than 1 1 satisfying 3 a b c = 1 1 a + 1 b + 1 c , 3abc = \dfrac{1}{\dfrac{1}{a} + \dfrac{1}{b} + \dfrac{1}{c}}, find the maximum value of 1 a 2 + 1 b c + 1 b 2 + 1 c a + 1 c 2 + 1 a b . \dfrac{1}{a^2+1-bc} + \dfrac{1}{b^2+1-ca} + \dfrac{1}{c^2+1-ab}.

Details and assumptions
- This problem is not original.


The answer is 3.

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.

2 solutions

The following solution isn't mine.


The condition is equivalent to a b + b c + c a = 1 3 ab+bc+ca= \dfrac{1}{3} .

I claim that the expression never exceeds 3 3 . This is equivalent to showing that ( a b + b c + c a ) ( 1 a 2 + 1 b c + 1 b 2 + 1 c a + 1 c 2 + 1 a b ) 1 cyc 2 ( a b + b c + c a ) a 2 b c + 1 2 3 cyc 2 a b + 2 b c + 2 c a a 2 b c + 1 1 cyc 1 2 ( a b + b c + c a ) a 2 b c + 1 1 cyc a 2 ( a + b + c ) a 3 a b c + a 1. (ab+bc+ca )\left(\dfrac{1}{a^2+1-bc} + \dfrac{1}{b^2+1-ca} + \dfrac{1}{c^2+1-ab} \right) \leq 1 \\ \iff \displaystyle \sum_{\text{cyc}} \dfrac{2(ab+bc+ca)}{a^2-bc+1} \leq 2 \\ \iff 3 - \displaystyle \sum_{\text{cyc}} \dfrac{2ab+2bc+2ca}{a^2-bc+1} \geq 1 \\ \iff \displaystyle \sum_{\text{cyc}} 1 - \dfrac{2(ab+bc+ca)}{a^2-bc+1} \geq 1 \\ \iff \displaystyle \sum_{\text{cyc}} \dfrac{a^2(a+b+c)}{a^3-abc+a} \geq 1. By Titu's lemma, cyc a 2 a 3 a b c + a ( a + b + c ) 2 a 3 + b 3 + c 3 3 a b c + a + b + c . \displaystyle \sum_{\text{cyc}} \dfrac{a^2}{a^3-abc+a} \geq \dfrac{(a+b+c)^2}{a^3+b^3+c^3 - 3abc + a+b+c}. Hence, it suffices to show that ( a + b + c ) 3 a 3 + b 3 + c 3 3 a b c + a + b + c , (a+b+c)^3 \geq a^3+b^3+c^3-3abc+a+b+c , which is, in fact, an identity. Hence, we're done.

Equality holds for a = b = c = 1 3 . a=b=c=\dfrac{1}{\sqrt{3}}.

Hence, our desired answer is 3 \boxed{3} .


This problem is from a recent China TST.

Aditya Raut
Jul 24, 2014

There are many elegant ways to solve this, but I am not at all in the moooooooooooood .

Equality case gives the extreme values! \color{#D61F06}{\text{Equality case gives the extreme values!}}

a = b = c a=b=c gives each of the fractions to be 1 1 , giving answer 3 3 .

Your second statement is not necessarily true.

Sreejato Bhattacharya - 6 years, 10 months ago

Log in to reply

Agreed, this is not there in many of the Cauchy Schwarz or other identity related question, but this one is symmetric in abc, so that was enough for bashing !

Aditya Raut - 6 years, 10 months ago

Log in to reply

First, what you did isn't called bashing, it's called guessing. And just because an expression is symmetric in its terms doesn't necessarily mean it's optimized when the terms are all equal. And even it were true, how would simply plugging the values show whether the optimum value you get is the maximum or the minimum?

Sreejato Bhattacharya - 6 years, 10 months ago

Log in to reply

@Sreejato Bhattacharya Why don't bunnies grow up :/

Aditya Raut - 6 years, 10 months ago

@sreehari vp @Sreejato Bhattacharya Can you post a solution and then help sort this into the corresponding Inequalities skill? Thanks!

Calvin Lin Staff - 6 years, 8 months ago

Log in to reply

Done. Thanks for reminding me, sir. :)

I'm not sure to which skill set this problem should go--this is more about algebraic manipulations than applying inequalities. Could you please help decide?

Sreejato Bhattacharya - 6 years, 8 months ago

I actually worked on this and showed that the expression was less than or equal to 2 + c y c l i c b c a 2 + 1 b c 2 + \displaystyle \sum_{cyclic} \dfrac{bc}{a^2+1-bc} . Then I guessed that that the answer was 3 3 . Seeing this then I tried working on a full solution by trying to show that c y c l i c b c a 2 + 1 b c 1 \displaystyle \sum_{cyclic} \dfrac{bc}{a^2+1-bc} \leq 1 . I tried that for sometime but failed. Can anyone help me from here?

@Sreejato Bhattacharya Can you help me?

Siam Habib - 6 years, 10 months ago

Actually, the equality case given is the GM-HM inequality, in which we know that the equality case occurs when a = b = c a=b=c .

Shaun Loong - 6 years, 10 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...