Let x , y , z be real numbers satisfying x + y + z = 1 5 and x y + y z + z x = 7 2 . Let the minimum value of x be S , and the maximum value M .
Find the value of S + M .
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.
Another interesting way to arrive at the relation x + 2 y = 1 5 is to directly substitute z = 1 5 − x − y into x y + y z + x z = 7 2 , yielding − x 2 − x y + 1 5 x − y 2 + 1 5 y = 7 2 , the equation of a conic section.
By method of the discriminant, this section is identified as an ellipse, an ellipse containing only real pairs ( x , y ) that solve the given equations (recall that z can always be defined in terms of x and y ).
To find the minimum and maximum values of x , simply consider the ellipse: these values will occur at the horizontal extrema, where tangent lines are vertical, or more simply when ∂ x ∂ = 0 .
∂ x ∂ [ − x 2 − x y + 1 5 x − y 2 + 1 5 y = 7 2 ] ⇒ − 2 x − y + 1 5 = 0 ⇒ x + 2 y = 1 5 .
In the first solution, you say that " the solution set to the given set of equations will be the intersection of this sphere with the plane , which describes a circle." How do you know it is this and not, for example, the intersection of the sphere with xy+yz+zx?
Log in to reply
It will be that as well, but it is easier to visualize what the intersection of a sphere and a plane looks like than what the intersection of a sphere and x y + y z + z x = 7 2 looks like. (The cross-sections of this last curve are hyperbolas.)
Log in to reply
How does the symmetry condition impose a necessity for extremal points?
If you changed the 72 to 73, would the answer still be 10? I'm aware that you need to be given more than the value of x+y+z in the question (otherwise there would be no min or max), but at the moment it just seems like 72 is only giving nice integers for the min and max, and not actually affecting the final answer.
Log in to reply
Good point. If x y + y z + z x = k then
M = 5 + 1 0 0 − 3 4 k and S = 5 − 1 0 0 − 3 4 k ,
which will give us M + S = 1 0 for k ≤ 7 5 . (For k > 7 5 there wouldn't be any real solution points.) So we could have just stopped when we found the quadratic in x .
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 |
|
1 2 3 4 5 6 7 |
|
To find the domain, x, we will have to substitute a variable to make a 2-variable equation. We take the first equation and put it in terms of z: x + y + z = 1 5 ⟹ z = 1 5 − x − y Plugging this into the second equation, we get x y + y ( 1 5 − x − y ) + x ( 1 5 − x − y ) = 7 2 x y + 1 5 y − x y − y 2 + 1 5 x − x 2 − x y = 7 2 − y 2 + ( 1 5 − x ) y + 1 5 x − x 2 − 7 2 = 0 y 2 + ( x − 1 5 ) y + x 2 − 1 5 x + 7 2 = 0 Using the quadratic formula, we get y = 2 − ( x − 1 5 ) ± x 2 − 3 0 x + 2 2 5 − 4 ( x 2 − 1 5 x + 7 2 ) y = 2 1 5 − x ± − 3 x 2 + 3 0 x − 6 3 Knowing that y is a real number and not an imaginary number, we know − 3 x 2 + 3 0 x − 6 3 must be positive. Therefore, we can now set boundaries and find the maximum and minimum values. − 3 x 2 + 3 0 x − 6 3 > 0 3 x 2 − 3 0 x + 6 3 < 0 3 x 2 − 3 0 x + 6 3 = 0 x = 2 ⋅ 3 3 0 ± 9 0 0 − 4 ( 3 ) ( 3 6 ) S = 6 3 0 − 9 0 0 − 4 3 2 , M = 6 3 0 + 9 0 0 − 4 3 2 S + M = 5 + 5 + 6 4 6 8 − 6 4 6 8 = 1 0
I really liked your solution. It makes the problem really simple. Thanks for posting it!
Thanks for your solution. It is easy to understand by large number of people.
Simple solution.
No, knowing that y is a real number, we know that that quadratic expression must be nonnegative.
My method involves the use of Cauchy-Schwarz Inequality.
We manipulate the equations such that we can do the inequality in terms of x .
x + y + z = 1 5 y + z = 1 5 − x
x y + y z + z x = 7 2 x y + x z = 7 2 − y z x ( y + z ) = 7 2 − y z x ( 1 5 − x ) = 7 2 − y z
⇒ y z = x 2 − 1 5 x + 7 2
y 2 + z 2 = ( 1 5 − x ) 2 − 2 ( x 2 − 1 5 x + 7 2 ) = 8 1 − x 2
( y 2 + z 2 ) ( 1 + 1 ) ≥ ( y + z ) 2 (Cauchy Schwarz Inequality)
2 ( 8 1 − x 2 ) ≥ ( 1 5 − x ) 2
3 x 2 − 3 0 x + 6 3 ≤ 0
3 ( x − 7 ) ( x − 3 ) ≤ 0
So 3 ≤ x ≤ 7
⇒ S = 3 , M = 7 , S + M = 1 0
Oh wow! I didn't expect a CS solution for a question like this. This is really surprising. Thanks for sharing!
Awesome solution buddy!
Let x , y and z be the solutions of a cubic equation. Then the equation p 3 − 1 5 p 2 + 7 2 p − c would have it's roots as x , y and z . Then p 3 − 1 5 p 2 + 7 2 p = c would have three solutions if c ∈ (108,112) as c would be a horizontal line parallel to the x-axis. All the three solutions are symmetrical so the smallest valued solution would be when c = 108, giving us two equal roots 6 and the other root 3 (Sum of the roots is 15). The largest valued solution would occur when c = 112, giving us two equal roots 4 and the other root 7.
So,
S
=
3
,
M
=
7
giving us
1
0
Oh nice, I tried a similar approach via cubic discriminant to get the answer.
First I transform to a new set of variables: a = x − 5 , b = y − 5 , c = z − 5 . Then { a + b + c = 0 a b + b c + a c = − 3 Suppose we have a solution ( a , b , c ) for this system. Then ( − a , − b , − c ) is obviously also a solution. In particular, if ( a ⋆ , b ⋆ , c ⋆ ) is a solution with minimum a , then ( − a ⋆ , − b ⋆ , − c ⋆ ) is a solution with maximum a and vice versa.
Therefore the answer to the problem is x min + x max = ( 5 + a min ) + ( 5 + a max ) = ( 5 + a ⋆ ) + ( 5 − a ⋆ ) = 5 + 5 = 1 0 .
In general, if { x + y + z = 3 p x y + y z + z x = 3 q we have x min + x max = 2 p .
With our transformed variables ( a , b , c ) = ( x − 5 , y − 5 , z − 5 ) it is also easier to find the minimum and maximum solutions.
Substitute c = − a − b : a b + b c + a c = − 3 ∴ a 2 + b 2 + a b = 3 . To find extreme values of a , differentiate and set d a = 0 : { ( 2 a + b ) d a + ( 2 b + a ) d b = 0 d a = 0 ∴ 2 b + a = 0 ∴ a = − 2 b . ∴ ( − 2 b ) 2 + b 2 + ( − 2 b ) b = 3 ∴ 3 b 2 = 3 ∴ b = ± 1 ∴ ⎩ ⎪ ⎨ ⎪ ⎧ ( a , b , c ) = ( − 2 , 1 , 1 ) or ( a , b , c ) = ( 2 , − 1 , − 1 ) ∴ ⎩ ⎪ ⎨ ⎪ ⎧ ( x , y , z ) = ( 3 , 6 , 6 ) or ( x , y , z ) = ( 7 , 4 , 4 ) . Thus we submit the solution x min + x max = 3 + 7 = 1 0 .
The general minimum and maximum solutions for { x + y + z = 3 p x y + y z + z x = 3 q are x min, max = p ∓ 2 p 2 − q .
Suppose we have a solution ( a , b , c ) for this system. Then ( − a , − b , − c ) is obviously also a solution. In particular, if ( a ⋆ , b ⋆ , c ⋆ ) is a solution with minimum a , then ( − a ⋆ , − b ⋆ , − c ⋆ ) is a solution with maximum a and vice versa.
Therefore the answer to the problem is...
Isn't it required to show that there exists reals x , y , z that satisfy these constraints FIRST?
Log in to reply
Your point is well-taken: my initial solution does not provide proof of existence.
I showed that if there is a solution (which implies that there is a maximum and a minimum solution), then their sum is equal to 10.
If the condition were false, there would be no way to submit a solution...
I used the method of Lagrangian multipliers.
We have x + y + z = 1 5 , and x y + y z + z x = 7 2 . The 2nd eq ⟹ x = y + z 7 2 − y z .
So maximizing/minimizing x means maximizing/minimizing x = y + z 7 2 − y z . The constraint being x + y + z = 1 5 .
∴ using Lagrangian multiplier, ∇ ( x + y + z − 1 5 ) = λ ∇ ( y + z 7 2 − y z ) , where ∇ = ( ∂ x ∂ , ∂ y ∂ , ∂ z ∂ ) .
Differentiating and comparing both sides, we get λ = 0 , and so y = z . Substituting this into the given equations and solving for x, we get a quadratic. Turns out that x = 3 or x = 7 .
But how do you know if the turning point that you've found isn't a saddle point (instead of a max/min point)?
Log in to reply
Here I kinda took the clue from the question to cut-short the process. Since the question said that there is a min and max for x , and I got exactly 2 turning points. So I skipped to check.
But it's not difficult to check. You can find the discriminant value ∇ at the turning point and classify it as max or min. You'll get ∇ < 0 , showing that it's either a max or min. I hope you got it. I'll be elaborate if you want.
Log in to reply
Nah, I got it. Thanks for your detailed response! +1
Considering P(q) = q³ + bq² +cq +d with roots x, y and z.
x+y+z = 15 implies that b = -15 and xy +yz +xz = 72 implies that c= 72.
So P(q) = q³ -15q² +72q +d.
Deriving this polynomial we will find the maximum and minimum values.
P'(q) = 3q² -30q +72 =0
Solving this equation, we can find a minimum value of 4 and a maximum value of 6
if these values are x values the sum is 10, if they're not, we have
z= 4 => x+y = 11 and xy = 28 => x = 7 and y =4 => (x,y,z) = (7,4,4)
z= 6 => x+y = 9 and xy = 18 => x=3 and y =6 => (x,y,z) = (3,6,6)
Derivatives certainly makes it simpler, doesn't it?
1 2 3 4 5 6 |
|
1 2 3 4 5 6 |
|
Min+Max = 3 + 7 = 1 0
The way i thought of is as:
since x, y ,z are real numbers. once of them can be zero, say y = 0, then i will be left with x + z = 15 and xz = 72
possible values for xz=72 can be 1 x 72 2 x 36 3 x 24 4 x 18 6 x 12 8 x 9 and for x+z = 15, the closest possible values can be: 8+9 = 17 or 6 + 12 = 18 which are still greater then 15, so, 5 <= x <= 8. so with kind of a wild guess, since i had 3 attempts, i started with 10 as an option and it was correct,
this is was just a thought process not a solution ;)
Problem Loading...
Note Loading...
Set Loading...
Note that x 2 + y 2 + z 2 = ( x + y + z ) 2 − 2 ( x y + y z + x z ) = 1 5 2 − 2 × 7 2 = 8 1 , which describes a sphere of radius 9 centered at the origin. So the solution set to the given set of equations will be the intersection of this sphere with the plane x + y + z = 1 5 , which describes a circle. This circle is symmetric about the plane y = z , (as y and z are interchangeable in the original system of equations), so both the maximum and minimum of x will occur when y = z . This transforms the given set of equations to
x + 2 y = 1 5 ⟹ 2 y = 1 5 − x and
2 x y + y 2 = y ( 2 x + y ) = 7 2 ⟹ 2 y ( 4 x + 2 y ) = 2 8 8 ⟹ ( 1 5 − x ) ( 4 x + ( 1 5 − x ) ) = 2 8 8 ⟹
( 1 5 − x ) ( 3 x + 1 5 ) = 2 8 8 ⟹ ( 1 5 − x ) ( x + 5 ) = 9 6 ⟹ x 2 − 1 0 x + 2 1 = ( x − 3 ) ( x − 7 ) = 0 .
So S = 3 and M = 7 , and thus S + M = 1 0 .
The minimum for x is achieved when ( x , y , z ) = ( 3 , 6 , 6 ) and then maximum when ( x , y , z ) = ( 7 , 4 , 4 ) .
Note that y + z = 1 5 − x and y z = 7 2 − x ( y + z ) = 7 2 − x ( 1 5 − x ) = 7 2 − 1 5 x + x 2 . Then
0 ≤ ( y − z ) 2 = ( y + z ) 2 − 4 y z = ( 1 5 − x ) 2 − 4 ( 7 2 − 1 5 x + x 2 ) ⟹
0 ≤ 2 2 5 − 3 0 x + x 2 − 2 8 8 + 6 0 x − 4 x 2 = − 3 x 2 + 3 0 x − 6 3 = − 3 ( x 2 − 1 0 x + 2 1 ) ⟹
0 ≤ − ( x − 3 ) ( x − 7 ) ⟹ 3 ≤ x ≤ 7 ,
which is in agreement with the previous result.