What a Magnitude!

Algebra Level 2

a 2 + 1 + b 2 + 16 + c 2 + 49 \large \sqrt{a^2+1}+\sqrt{b^2+16}+\sqrt{c^2+49}

For a , b , c R a,b,c \in \mathbb R such that a + b + c = 5 a+b+c=5 , what is the minimum value of the expression above?


The answer is 13.

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.

10 solutions

We know that:

a + b + c a + b + c a = ( a , 1 ) ; b = ( b , 4 ) ; c = ( c , 7 ) a + b + c = ( 5 , 12 ) a + b + c = 13 ( a + b + c = 5 ) a + b + c 13 \left| \vec { a } +\vec { b } +\vec { c } \right| \le \left| \vec { a } \right| +\left| \vec { b } \right| +\left| \vec { c } \right| \\ \vec { a } =(a,1);\vec { b } =(b,4);\vec { c } =(c,7)\\ \Rightarrow \vec { a } +\vec { b } +\vec { c } =(5,12)\Rightarrow \left| \vec { a } +\vec { b } +\vec { c } \right| =13\quad (\because a+b+c=5)\\ \Rightarrow \left| \vec { a } \right| +\left| \vec { b } \right| +\left| \vec { c } \right| \ge 13

You know, when you have 5 5 and 12 12 , The answer is definitely 13 \boxed{13}

Nicely done :)

Shashwat Shukla - 6 years, 3 months ago

Too good !! Why didn't I think of Vectors ? !

+1

A Former Brilliant Member - 6 years, 3 months ago

Wonderful, now thats effing awesome!

Mvs Saketh - 6 years, 3 months ago

Nice One +1 , Finally I also Solved It in Similar manner by using Quadrilateral inequality concept , But instead of Using vector's notations , I used Complex number's notations. I'am bit handy in complex rather in vector!

Deepanshu Gupta - 6 years, 3 months ago

Good solution! We would need to clarify though that the equality takes place when the three vectors are in the same direction. That is, when b = 4 a \vec{b}=4\vec{a} and (\vec{c}=7\vec{a}},) and that is why the minimum value is reached.

Arturo Presa - 4 years ago

Absolutely BRILLIANT!!!

Prayas Rautray - 3 years, 11 months ago
Shashwat Shukla
Mar 5, 2015

This is such a beautiful question... I'll outline a physics/elementary geometry based approach.

The given quantities are obviously distances from A(a,1) , B(b,4) and C(c,7).

Let's consider a simpler case by fixing c c (for now).

That is, let's find the maximum value for P A + P B PA+PB subject to the constraint a + b = k a+b=k for some k.

Look at the figure below. (Note that the figure marked as A is actually the reflection of A about the y-axis, but that doesn't matter as far as distance is concerned)

First note that point A A is constrained to move along y = 1 y=1 and similarly B B has to move along y = 4 y=4 . Also, their horizontal separation is fixed(=k).

Forget point P for now and instead consider the question, what is the minimum value of D A + D B DA+DB for any point D D on y = 0 y=0 ? Well, that's pretty easy...

Reflect B B (or A) about y = 0 y=0 and join this new point to A. The point where this line meets y=0 is the required point. This is equivalent to using Fermat's principle of least time. Now, this point is such that the angle of 'incidence' equals the angle of 'refraction'.

Okay, so now if we move the point A A (and thus B also has to move the same distance) such that point P coincides with D D , then this configuration of A a n d B A \quad and \quad B is what we need.

From geometry, we will get, a b = 1 4 \frac{a}{b}=\frac{1}{4} .

But we did this after fixing the point C C . We could have fixed point A or B. Therefore, i f if , the conditions we get by fixing the three points separately yield a common solution, then we are done.

(P.S : This is also what happens if we use Langrange multipliers. i.e We get three equations which we solve to get a concurrent solution).

Fortunately the solution does exist. Based on our previous reasoning it is easy to see that for the given values, a : b : c = 1 : 4 : 7 a:b:c=1:4:7 and a + b + c = 5 a+b+c=5

This yields, a= 5 12 \frac{5}{12} , b= 5 3 \frac{5}{3} , c= 35 12 \frac{35}{12} .

Note that P, A, B, C lie along a straight line in this case.

Shukla ji! Wah wah! It just felt like that they had to lie on a straight line... I do not know why. It somehow always ends up being a straight line!

Raghav Vaidyanathan - 6 years, 3 months ago

Log in to reply

The master of intuition strikes again... wow :D

Shashwat Shukla - 6 years, 3 months ago

Do you speak Hindi ?

A Former Brilliant Member - 6 years, 3 months ago

Log in to reply

Arey yaar! Dasvi kaksha thak Hindi hi pada tha meine!

Raghav Vaidyanathan - 6 years, 3 months ago

Log in to reply

@Raghav Vaidyanathan Kya baat hai yar...Hindi to koi tumse sikhe :D

Shashwat Shukla - 6 years, 3 months ago

@Raghav Vaidyanathan Haha , maaf kar do mujhe !!

And sorry if there are any spelling mistakes , I'm not good with the transliteration stuff !!

A Former Brilliant Member - 6 years, 3 months ago

Nice solution :) +1

A Former Brilliant Member - 6 years, 3 months ago

Really awesome Shashwat , This is the real Geometric Solution I was Looking For , Thanks For sharing it with us.

Deepanshu Gupta - 6 years, 3 months ago

Log in to reply

You're welcome :) ...Actually I also solved it using triangle inequality. But as you had started that discussion, I thought of posting this approach, just for variety.

@Deepanshu Gupta

Shashwat Shukla - 6 years, 3 months ago
Mvs Saketh
Mar 5, 2015

Without using Lagrange multipliers

let a= tan θ \tan {\theta} b= 4 tan ϕ 4\tan {\phi} c= 7 tan δ 7\tan {\delta}

Then we have

S = sec θ + 4 sec ϕ + 7 sec δ S = \sec {\theta} + 4\sec{\phi} + 7\sec{\delta}

where S is what we need to minimise given that tan θ + 4 tan ϕ + 7 tan δ = 5 \tan {\theta} + 4\tan {\phi} + 7\tan {\delta} =5

Now let us square S, to get

s e c 2 θ + 16 s e c 2 ϕ + 49 s e c 2 δ + 8 s e c θ s e c ϕ + 56 s e c ϕ s e c δ + 14 s e c δ s e c θ { sec }^{ 2 }\theta \quad +\quad 16\quad { sec }^{ 2 }\phi \quad +\quad 49\quad { sec }^{ 2 }\delta \quad \\ \\ +8sec\theta sec\phi +56sec\phi sec\delta \quad +\quad 14sec\delta \quad sec\theta \\

N o w \displaystyle Now

s e c θ s e c ϕ = 1 + t a n 2 θ 1 + t a n 2 ϕ = 1 + t a n θ t a n ϕ sec\theta sec\phi =\sqrt { 1+{ tan }^{ 2 }\theta } \sqrt { 1+{ tan }^{ 2 }\phi } \quad \ge \quad =1+tan\theta tan\phi

similarly doing for other two and then using

1 + t a n 2 θ = s e c 2 θ \\ 1+{ tan }^{ 2 }\theta ={ sec }^{ 2 }\theta

i get

s e c 2 θ + 16 s e c 2 ϕ + 49 s e c 2 δ + 8 s e c θ s e c ϕ + 56 s e c ϕ s e c δ + 14 s e c δ s e c θ 1 + 16 + 49 + 8 + 56 + 14 + t a n 2 θ + 16 t a n 2 ϕ + 49 t a n 2 δ + 8 t a n θ t a n ϕ + 56 t a n ϕ t a n δ + 14 t a n δ t a n θ = 144 + ( t a n θ + 4 t a n ϕ + 7 t a n δ ) 2 = 169 \\ { sec }^{ 2 }\theta \quad +\quad 16\quad { sec }^{ 2 }\phi \quad +\quad 49\quad { sec }^{ 2 }\delta \quad \\ \\ +8sec\theta sec\phi +56sec\phi sec\delta \quad +\quad 14sec\delta \quad sec\theta \\ \ge \quad 1+16+49+8+56+14+{ tan }^{ 2 }\theta \quad +\quad 16\quad { tan }^{ 2 }\phi \quad +\quad 49\quad { tan }^{ 2 }\delta \quad \\ \\ +8tan\theta tan\phi +56tan\phi tan\delta \quad +\quad 14tan\delta \quad tan\theta \quad =\quad 144+(tan\theta +4tan\phi +7tan\delta )^{ 2 }\quad =\quad 169

so S= 169 \sqrt {169} = 13

Nice solution :)

A Former Brilliant Member - 6 years, 3 months ago

+1 , Good Use of Inequalities :)

Deepanshu Gupta - 6 years, 3 months ago

with minkowski it s extremly easy

Omar El Mokhtar - 6 years, 3 months ago
Manuel Kahayon
Mar 4, 2016

DIRECT APPROACH USING MINKOWSKI'S INEQUALITY

a 2 + 1 2 + b 2 + 4 2 + c 2 + 7 2 ( a + b ) 2 + ( 1 + 4 ) 2 + c 2 + 7 2 ( a + b + c ) 2 + ( 1 + 4 + 7 ) 2 = 5 2 + 1 2 2 = 13 \sqrt{a^2+1^2} + \sqrt{b^2+4^2} + \sqrt{c^2+7^2} \geq \sqrt{(a+b)^2+(1+4)^2}+ \sqrt{c^2+7^2} \geq \sqrt{(a+b+c)^2+(1+4+7)^2} = \sqrt{5^2+12^2} = \boxed {13}

Well, that's cool :-)

Pulkit Gupta - 5 years, 3 months ago

Log in to reply

Well, I used caps lock because the length of all their solutions is driving me MAD!

Manuel Kahayon - 5 years, 3 months ago

Same solution, simple and convenient for the win

P C - 5 years, 3 months ago

Log in to reply

Is it just me or are you browsing through my profile...? XD

Manuel Kahayon - 5 years, 3 months ago
Pranjal Jain
Nov 3, 2014

By lagrange's multipliers,

  1. a+b+c=5
  2. λ \lambda = a a 2 + 1 \frac{a}{\sqrt{a^{2}+1}}
  3. λ \lambda = b b 2 + 1 \frac{b}{\sqrt{b^{2}+1}}
  4. λ \lambda = c c 2 + 49 \frac{c}{\sqrt{c^{2}+49}}

By using 2,3 and 4, b=4a and c=7a. Substituting b and c in 1, a= 5 12 \frac{5}{12} , b= 5 3 \frac{5}{3} , c= 35 12 \frac{35}{12}

By using these values one can easily calculate given expression to be 13 \boxed{13}

can we use lagrange's multipliers for finding max value?

Omar El Mokhtar - 6 years, 6 months ago

Lagrangian is just 2 {\nabla}^{2} , right?

Kartik Sharma - 6 years, 3 months ago
Tapas Pal
Mar 25, 2015

if a,b,c are real not taking a+b+c=5,then min value when a=b=c=0,then its value is 1+4+7=12,but when c=5,then sqrt of (c^2+49)<9,then min value is<14,so it is 13.

Aakash Sidhwani
Mar 5, 2015

what is lagrange multiplier???

Incredible Mind
Mar 7, 2015

i tried by cauchy schartz but i am getting min as 15..Y is that..but i ended up guessing 14 and 13 to get right ANS

Kartik Sharma
Mar 6, 2015

Well, the solutions given are quite "over the head"(that's because they are really quite amazing). But an alternative always goes fine(I guess).

By AM-GM,

a 2 + 1 2 a \sqrt{{a}^{2} + 1} \geq \sqrt{2a}

b 2 + 16 8 b \sqrt{{b}^{2} + 16} \geq \sqrt{8b}

c 2 + 49 14 c \sqrt{{c}^{2} + 49} \geq \sqrt{14c}

Adding them all up,

a 2 + 1 + b 2 + 16 + c 2 + 49 2 a + 8 b + 14 c \sqrt{{a}^{2} + 1} + \sqrt{{b}^{2} + 16} + \sqrt{{c}^{2} + 49} \geq \sqrt{2a} + \sqrt{8b} + \sqrt{14c}

Just arranging it a little bit,

14 c + 8 b + 2 a \geq \sqrt{14c} + \sqrt{8b} + \sqrt{2a}

Using Chebyshev,

14 + 8 + 2 ( a + b + c ) 3 \geq \frac{\sqrt{14} + \sqrt{8} + \sqrt{2}(\sqrt{a} + \sqrt{b} + \sqrt{c})}{3}

Here I assume that a , b , c a,b,c are positive reals because that will not change the equation. Hence, a , b , c < 5 a,b,c <5 , hence, their sum of square roots will be just less than their sum. Therefore,

14 + 8 + 2 ( 5 ) 3 13.307163456488711 \geq \frac{\sqrt{14} + \sqrt{8} + \sqrt{2}(5)}{3} \approx 13.307163456488711

Hence, our final answer becomes 13 13 .

This is not one of the solutions advised! Vectors solutions is awesome. And also, whenever you see 5 5 and 12 12 and squares, answer has to be 13 13 .

Kartik Sharma - 6 years, 3 months ago

i think that it s false

Omar El Mokhtar - 6 years, 3 months ago

What you said was 'sum of square roots will be just less than their sum' but what you used in equation was the sum of square roots is greater than sum.

Harkirat Dhanoa - 5 years, 3 months ago
Emmanuel Torres
Feb 9, 2017

I plugged in c = 5. I got 1 + 4 + √73 < 14. So it has to be an integer greater than 12 and less than 14, I wonder what it could be?

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...