# Le Monde puzzle [#1002]

April 3, 2017
By

(This article was first published on R – Xi'an's Og, and kindly contributed to R-bloggers)

For once and only because it is part of this competition, a geometric Le Monde mathematical puzzle:

Given both diagonals of lengths p=105 and q=116, what is the parallelogram with the largest area? and when the perimeter is furthermore constrained to be L=290?

This made me jump right away to the quadrilateral page on Wikipedia, which reminds us that the largest area occurs when the diagonals are orthogonal, in which case it is A=½pq. Only the angle between the diagonals matters. Imposing the perimeter in addition is not solved there, so I wrote an R code looking at all the integer solutions, based on one of the numerous formulae for the area, like

$\sqrt{(S-a)(s-b)(s-c)(s-d)-(ac+bd+pq)(ac+bd-pq)/4}$

where s is the half-perimeter and a,b,c,d are the lengths of the four sides:

p=105
q=116
s=145
for (alpha in (1:500)/1000){
ap=alpha*p;ap2=ap^2;omap=p-ap;omap2=omap^2
for (beta in (1:999)/1000){
bq=beta*q;bq2=bq^2;ombq=q-bq;ombq2=ombq^2
for (teta in (1:9999)*pi/10000){
d=sqrt(ap2+bq2-2*ap*bq*cos(teta))
a=sqrt(ap2+ombq2+2*ap*ombq*cos(teta))
b=sqrt(omap2+ombq2-2*omap*ombq*cos(teta))
c=sqrt(omap2+bq2+2*omap*bq*cos(teta))
if (abs(a+b+c+d-2*s)2*maxur){
maxur=p*q*sin(teta)/2
sole=c(a,b,c,d,alpha,beta,teta)}}}}


This code returned an area of 4350, to compare with the optimal 6090 (which is recovered by the above R code when the diagonal lengths are identical and the perimeter is the one of the associated square). (As Jean-Louis Foulley pointed out to me, this area can be found directly by assuming the quadrilateral is a parallelogram.)

Filed under: Kids, R Tagged: geometry, Le Monde, mathematical puzzle, optimisation, R, Rmpfr

R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...