Le 4 mai 2021, la plateforme Yahoo Questions/Réponses fermera. Elle est désormais accessible en mode lecture seule. Aucune modification ne sera apportée aux autres sites ou services Yahoo, ni à votre compte Yahoo. Vous trouverez plus d’informations sur l'arrêt de Yahoo Questions/Réponses et sur le téléchargement de vos données sur cette page d'aide.

Scythian1950 a posé la question dans Science & MathematicsMathematics · il y a 7 ans

Pizza puzzle, find maximum area of this piece?

Check graphic of a 12" diameter pizza with 3 straight cuts, leaving a right triangle piece in the center.

http://i254.photobucket.com/albums/hh120/Scythian1...

The arcs in degrees are to be one of each {15, 30, 45, 60, 90, 120}, totaling 360, in some order, not necessarily progressive. The graphic shown doesn't show correct arcs, do not rely on the graphic to assign arc degrees. What's the exact maximum area, in square inches, can this right triangle piece in the center can have? Please express your answer in this form:

A + B√2 + C√3 + D√6

where A, B, C, D are rational numbers, positive or negative.

An arc in degrees is the angle subtended from the center of the circle, not angles formed by intersection of cuts.

Mise à jour:

jibz, you can eliminate a lot of those cases just by inspection, i.e., the cases where the right triangle piece is obviously rather small. No need for calculations for those.

Mise à jour 2:

Congratulations, Falzoon, I think you've found one of the smallest possible right triangles, but at least it's dead on accurate. Er, one thumb up for producing a mathematically correct result, even if it's far from the maximum area? As of right now, you're ahead of everyone else, with the biggest piece to date.

Mise à jour 3:

Falzoon, there's a quite a few combinations that will result in a right triangle piece. I know of at least 7 that will get a right triangle piece with a greater area.

Mise à jour 4:

Falzoon, bingo. That's the maximum piece, and you've given its exact area. Congratulations.

Mise à jour 5:

Here's a graphic of this maximal piece:

http://i254.photobucket.com/albums/hh120/Scythian1...

Mise à jour 6:

Falzoon, I use Mathematica. However, I only need to compute the areas numerically first, and then the one with the most area can be worked out "exactly". Even then, Mathematica doesn't necessarily deliver the most concise "exact" answer. Actually, what I did was to plot out the different possibilities, which allows me to reject most of them offhand.

The key to simplifying this problem is using the fact that chords crossing at 90° divide the circle in such a way that opposite arcs add up to 180°. That results in about a dozen cases one has to actually look at.

3 réponses

Pertinence
  • il y a 7 ans
    Réponse favorite

    With an awful lot of work, (though it was exciting), I get -

    (-27) + (18)√2 + (-18)√3 + (27/2)√6

    With circle centre on (0, 0), begin at (0, 6) with the

    clockwise rotation order of 30, 120, 90, 45, 60, 15.

    EDIT:

    I guess you're not kidding, Scythian. I tested 6 hopeful diagrams, of which this one

    contained the only right-angled triangle, so I thought this must be it. Something must

    have gone astray with my initial programming to cut down the number of possibilities.

    EDIT:

    After correcting a few silly mistakes, I now get -

    Area = (-351) + (459/2)√2 + (-459/2)√3 + (351/2)√6 ≈ 5.9418

    I hope that's it because I'm getting dizzy and Wolfram's about to blow a gasket.

    That's with the order: 45, 15, 60, 30, 90, 120, clockwise from (0, 6).

    EDIT: Hold your horses for a while, because, even though I'm getting some answers,

    I think I'm getting wrong areas for the particular orders. Something is definitely fishy.

    EDIT: Ok, I think I have it now, once I fixed up a couple of minuses that should have

    been pluses and vice versa.

    The circular order of 15, 90, 60, 45, 30, 120 seems to give the largest area,

    so I'll get the answer to you asap.

    EDIT: Area = (18) + (-45/4)√2 + (15)√3 + (-33/4)√6 ≈ 7.862569

    BTW, for such an interesting question, I'm surprised at the number of stars so far.

    EDIT: Here's a list of all 12 in descending area order -

    (15, 90, 60, 45, 30, 120): Area ≈ 7.862569

    (15, 45, 60, 90, 30, 120): Area ≈ 7.049387

    (15, 90, 60, 30, 45, 120): Area ≈ 3.630452

    (15, 30, 60, 90, 45, 120): Area ≈ 2.657675

    (15, 60, 30, 45, 120, 90): Area ≈ 1.869781

    (15, 45, 120, 90, 30, 60): Area ≈ 1.676400

    (15, 60, 45, 30, 120, 90): Area ≈ 0.474067

    (15, 30, 120, 90, 45, 60): Area ≈ 0.347041

    (15, 45, 60, 30, 90, 120): Area ≈ 0.319805

    (15, 30, 60, 45, 90, 120): Area ≈ 0.261120

    (15, 45, 120, 30, 90, 60): Area ≈ 0.188298

    (15, 30, 120, 45, 90, 60): Area ≈ 0.153745

    EDIT: To obtain an EXACT answer to a given set of arcs, I had to -

    1. Find the EXACT points at the ends of the arcs.

    2. Find the equations to the 3 triangle lines with EXACT slopes & EXACT y-intercepts.

    3. Find the EXACT (x,y) points of the 3 intersecting lines after equating to each other.

    4. Find the EXACT area using the area equation for 3 points.

    Scythian, each one takes quite a lot of time, even using WolframAlpha, so I was

    interested in knowing if you had a quicker method, such as software that can

    manipulate radicals and output radical answers (is there such a thing?), or do you

    have to do the hard slog too?

    EDIT: Thanks for the explanation, Scythian, In the end, once I had ironed out all

    the mistakes, the maximum popped out easily, and like you, all I had to do was

    calculate that one exactly. It's a pity I had to work through those others though,

    when I didn't really need to. "Opposite arcs adding to 180º" didn't occur to me.

    Instead, I used the perpendicular slopes formula. Having calculated 3 of the 12

    exactly, it looks like your answer format may be true for the other 9, but I'm not

    about to look into that soon.

  • ?
    Lv 6
    il y a 7 ans

    I'm still working on it, but I'm down to 12 cases. To be cont'd...

    • 6 different arcs can be arranged in (6-1)! = 120 different ways around a circle, but the arrangements come in pairs of mirror images. So really at most "only" 60 arrangements need looked at.

    • Let O be the center of the circle. For a given arrangement, label the arc endpoints P(1), ..., P(6) in, say, clockwise order, and for all n ∈ ℤ let P(n) = P(n mod 6). Then a chord must connect endpoints P(n) and P(n+3) for each n. Let A(n) := ∡(P(n),O,P(n+1)). So by hypothesis

    (1): { A(n) | n ∈ ℤ } = { τ/3, τ/4, τ/6, τ/8, τ/12, τ/24 },

    where τ := 2π.

    • The exterior (turning) angle between a pair of adjacent sides of a polygon is equal to the angle by which a line perpendicular to one of those sides must be turned in order to become perpendicular to the other. So in the case of a polygon formed by chords of a circle, this will be the same as the angle subtended by the chords' radial bisectors. Consequently, the exterior angle between chord (P(n),P(n+3)) and chord (P(n+1),P(n+4)) is equal to

    (2): ∡(n) := (1/2)(A(n) + A(n+3)).

    By hypothesis our polygon is a right triangle, so WLOG ∡(1) = τ/2 - τ/4 = τ/4. The only way this is possible, given (1) and (2), is if { A(1), A(4) } = { τ/3, τ/6 }. So WLOG let

    (A(1), A(4)) = (τ/3, τ/6).

    { A(1), A(3), A(5), A(6) } can be matched with {τ/4, τ/8, τ/12, τ/24} in 4! = 24 different ways. But half of them are mirror images of the other half. So just 12 cases remain.

  • il y a 7 ans

    We must realize that. in order to be left out piece after the three cuts has to be a right triangle. This makes it necessary that two cuts have to be perpendicular to each other. next the second perpendicular cut cannot be after 15 degree cut because the remaining will be 360 -15, a odd number so the other cuts will be fractional. For sane reason it cannot be 30 36- -30 is not divisible by 4. So let us try with 60.and let perpendicular cut be the diameter.

Vous avez d’autres questions ? Pour obtenir des réponses, posez vos questions dès maintenant.