Confusing Cuboids

Geometry Level 5

Two different cuboids with integer dimensions have the same total edge length E E and total surface area A A .

One has the minimum possible nonzero volume given E E and A A , and the other has the maximum possible volume given E E and A A . What is the minimum possible sum of the two volumes?


The answer is 36.

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.

1 solution

Michael Mendrin
Dec 28, 2015

The proof will be posted later. The dimensions of the cuboids are 4 × 4 × 1 4 \times 4 \times 1 and 2 × 2 × 5 2 \times 2 \times 5 , the volumes being 16 + 20 = 36 16+20 =36 , while E = 36 E=36 and A = 48 A=48 .

Edit:

The dimensions x , y , z x,y,z in the following is interchangeable.

Let x , y , z x,y,z be the dimensions of a cuboid, and

E = 4 ( x + y + z ) E=4\left( x+y+z \right)
A = 2 ( x y + y z + z x ) A=2\left( xy+yz+zx \right)

We solve for x , y x,y , the solutions being

x = 1 8 ( E 4 z E 2 32 A + 8 E z 48 z 2 ) x=\dfrac { 1 }{ 8 } \left( E-4z-\sqrt { { E }^{ 2 }-32A+8Ez-48{ z }^{ 2 } } \right)
y = 1 8 ( E 4 z + E 2 32 A + 8 E z 48 z 2 ) y=\dfrac { 1 }{ 8 } \left( E-4z+\sqrt { { E }^{ 2 }-32A+8Ez-48{ z }^{ 2 } } \right)

or

x = 1 8 ( E 4 z + E 2 32 A + 8 E z 48 z 2 ) x=\dfrac { 1 }{ 8 } \left( E-4z+\sqrt { { E }^{ 2 }-32A+8Ez-48{ z }^{ 2 } } \right)
y = 1 8 ( E 4 z E 2 32 A + 8 E z 48 z 2 ) y=\dfrac { 1 }{ 8 } \left( E-4z-\sqrt { { E }^{ 2 }-32A+8Ez-48{ z }^{ 2 } } \right)

The volume x y z xyz works out to

1 4 ( 2 A z E z 2 + 4 z 3 ) \dfrac{1}{4}\left(2Az-E{z}^{2}+4{z}^{3}\right)

which is a real value for all real E , A , z E,A,z . Maximum and minimum volumes occurs when

z = 1 12 ( E E 2 24 A ) z=\dfrac { 1 }{ 12 } \left( E-\sqrt { { E }^{ 2 }-24A } \right)
z = 1 12 ( E + E 2 24 A ) z=\dfrac { 1 }{ 12 } \left( E+\sqrt { { E }^{ 2 }-24A } \right)

which yields the same volumes when

z = 1 12 ( E + 2 E 2 24 A ) z=\dfrac { 1 }{ 12 } \left( E+2\sqrt { { E }^{ 2 }-24A } \right)
z = 1 12 ( E 2 E 2 24 A ) z=\dfrac { 1 }{ 12 } \left( E-2\sqrt { { E }^{ 2 }-24A } \right)

which is the condition that the radical in x , y x,y in terms of E , A , z E,A,z vanish, so that all the dimensions are real. Hence for any given real E , A E,A such that 32 A > E 2 > 24 A 32A>{ E }^{ 2 }>24A , the dimensions of the cuboids of maximum and minimum volumes are (in any order of x , y , z x,y,z )

1 12 ( E + 2 E 2 24 A ) \dfrac { 1 }{ 12 } \left( E+2\sqrt { { E }^{ 2 }-24A } \right)
1 12 ( E E 2 24 A ) \dfrac { 1 }{ 12 } \left( E-\sqrt { { E }^{ 2 }-24A } \right)
1 12 ( E E 2 24 A ) \dfrac { 1 }{ 12 } \left( E-\sqrt { { E }^{ 2 }-24A } \right)

and

1 12 ( E 2 E 2 24 A ) \dfrac { 1 }{ 12 } \left( E-2\sqrt { { E }^{ 2 }-24A } \right)
1 12 ( E + E 2 24 A ) \dfrac { 1 }{ 12 } \left( E+\sqrt { { E }^{ 2 }-24A } \right)
1 12 ( E + E 2 24 A ) \dfrac { 1 }{ 12 } \left( E+\sqrt { { E }^{ 2 }-24A } \right)

For all these to be integers, E E must at least be a multiple of 12 12 , or 12 e 12e , and A A must at least be a multiple of 6 6 , or 6 a 6a . We try different integer values for e , a = 1 , 2 , 3 , . . . e, a=1,2,3,... and the smallest dimensions and volumes possible occur where e = 3 e=3 and a = 8 a=8 , the dimensions being 2 × 2 × 5 = 20 2 \times 2 \times 5=20 and 1 × 4 × 4 = 16 1 \times 4 \times 4=16 .

Very interesting question.

I guess the last thing to prove would be for the cases when maximum/minimum solution is not integer but there is integer solution like for these: E = 48 , A = 82 E=48, A= 82 , edges: 1 × 5 × 6 1 \times 5 \times 6 and 2 × 3 × 7 2 \times 3 \times 7 . Is there a proof that there are no more than two integer edge cuboids for each pair of E , A E, A ?

Maria Kozlowska - 5 years, 5 months ago

Log in to reply

The solutions for x , y x,y in terms of E , A , z E,A,z is such that x = y x=y in the extrema, i.e, when the radical vanishes. In other words, given E E and A A , there could be range of possible values for x , y , z x,y,z , which are usually all different, except for maximum and minimum volumes for real x , y , z x,y,z , in which case at least two of the dimensions are the same.

Given a sufficiently large E E and A A there could be a number of cuboids with integer dimensions that have that edge length and surface area, but there can only be one maximum volume and one minimum volume, and I don't see how there could be more than one cuboid with such maximum or minimum volume, particularly when you have the restriction that at least two of the dimensions must be the same.

This problem does have a number of curious mathematical properties, and I wonder if I can come up with a number theory problem based on some of such properties.

Michael Mendrin - 5 years, 5 months ago

Log in to reply

If you say so. I just looked at values for edge lengths < 20 <20 and they do not have more two cuboids for each pair of E, A values.

Happy New Year!

Maria Kozlowska - 5 years, 5 months ago

Log in to reply

@Maria Kozlowska Oh yes, Happy New Year for you as well, and congrats for being the first to solve this.

Michael Mendrin - 5 years, 5 months ago

Log in to reply

@Michael Mendrin You were gracious enough to make it not too difficult this time.

I checked the results and there are two cases for small edge lengths where there are three cuboids for each E, A. Thanks for your an answer, as I assumed there are two solutions for each set only.

You asked about pasta in our cuisine. I just figured that one now. We actually serve uszka with barszcz on Wigilia (Christmas Eve supper). Uszka are tortellini sized pasta filled with mushrooms, or best wild dried mushrooms (boletus are the best). For desert in some regions there are łazanki z makiem, pasta with ground popyseeds with honey, nuts, raisins. In other regions people serve kutia instead, where pasta is replaced with wheat whole seeds.

I still have few more questions you posted to solve. It might take me awhile but I am not loosing hope.

Maria Kozlowska - 5 years, 5 months ago

Log in to reply

@Maria Kozlowska Oh, borscht is one of my favorite soups. But I prefer it without the beets, i.e., "poor southern Russian folk" borscht.

I think I better hurry up and start posting more Level 5 problems.

Michael Mendrin - 5 years, 5 months ago

Log in to reply

@Michael Mendrin Do you have a good recipe for borscht?

Maria Kozlowska - 5 years, 5 months ago

Log in to reply

@Maria Kozlowska This is fairly representative of the different ways of making beet-less borscht

BORSCH

5 Lbs beef
6 Qts water
3 Whole potatoes, peeled
½ Cup Lima beans, dry
½ Cup Garbanzos
2 Tbs Salt
3 Carrots, diced
3 Stalks celery, diced
1 Bell pepper, diced
½ Small head cabbage, diced
1 Qt Tomatoes, cooked
1 Small can tomato sauce
1 Pkg Lima beans, frozen
1 Pkg Peas, frozen
Fresh dill

Soak garbanzos overnight
Boil water
Put in meat, potatoes, beans, garbanzos
Boil until meat is done
Take out meat
Put in carrots, celery
Boil 20-30 minutes
Put rest of ingredients
Boil 20-30 minutes until done

Repeated boiling over time improves flavor

Alternatively, boil meat & bones for
good soup stock before putting in
any ingredients

Michael Mendrin - 5 years, 5 months ago

Log in to reply

@Michael Mendrin Thank you. It sounds like a tasty and very filling meal.

Maria Kozlowska - 5 years, 5 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...