servidor windows costo $ 100,000 hasta $ 200,000 actualizacion de software $ 275,000 a $ 500,000. capacitacion $9,000 a $10,000 Windows min $384000 max $710,000 estimado 435,000 servidor linux costo $ 80,000 hasta $ 210,000. actualizacion de software $ 250,000 a $ 525,000. capacitacion $ 8,000 a $ 17,500 Linux min 338000 max 752500 estimado 410000

get_montecarloest <- function(vmin,vmax,vest){
    mn <- (vmin + 4*vest + vmax)/6
  s <- abs((vmax - vmin)/6)
  
  valmontecarlo <- rnorm(1, mean=mn, sd=s)
  
  return(valmontecarlo)
}
get_sim<-function(x,vmin,vmax,vest){
    
    times <- get_montecarloest(vmin,vmax,vest)
    return (times)
}

hacemos test de la funcion de montecarlo

get_montecarloest(384000,710000,410000)

aplicamos la funcion para windows y linux con 10000 iteraciones

simulacion<-10000
sim_win<-sapply(1:simulacion, get_sim,vmin=384000,vmax=710000,vest=435000)
sim_lin<-sapply(1:simulacion, get_sim,vmin=338000,vmax=752500,vest=410000)
print(paste("windows: ",mean(sim_win)," Linux:", mean(sim_lin)))
[1] "windows:  471665.660191291  Linux: 454384.271605227"
print(paste("STD windows: ",sd(sim_win)," STD Linux:", sd(sim_lin)))
[1] "STD windows:  54245.3854852991  STD Linux: 68615.9216241588"

La mejor opcion es linux si no se evalua la desviacion estandar

plot( sim_win, type='l')


summary(sim_win)
   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
 272483  434773  471989  471666  508354  678870 
hist(sim_win)

hist(sim_lin)

library(tidyverse)
quantile(sim_win, probs = 0:10/10)
      0%      10%      20%      30%      40%      50%      60%      70%      80%      90%     100% 
272483.2 401101.2 426095.1 443531.2 458437.2 471988.7 485109.8 500351.3 517096.8 540895.8 678870.2 
print("probabilidaD windows:")
[1] "probabilidaD windows:"
length(sim_win[sim_win >472293])/length(sim_win)
[1] 0.4982
print("probabilidaD linux:")
[1] "probabilidaD linux:"
length(sim_lin[sim_lin >410000])/length(sim_lin)
[1] 0.7406

Basado en este resultado preferimos la opcion de windows por que la probabilidad de linux es mucho mayor

LS0tDQp0aXRsZTogIlNpbXVsYWNpb24gbW9udGVjYXJsbyINCm91dHB1dDogaHRtbF9ub3RlYm9vaw0KLS0tDQpzZXJ2aWRvciB3aW5kb3dzIGNvc3RvICQgMTAwLDAwMCBoYXN0YSAkIDIwMCwwMDANCmFjdHVhbGl6YWNpb24gZGUgc29mdHdhcmUgICQgMjc1LDAwMCBhICQgNTAwLDAwMC4gDQpjYXBhY2l0YWNpb24gJDksMDAwIGEgJDEwLDAwMA0KV2luZG93cyBtaW4gJDM4NDAwMCBtYXggJDcxMCwwMDAgZXN0aW1hZG8gIDQzNSwwMDAgDQpzZXJ2aWRvciBsaW51eCAgY29zdG8gJCAgODAsMDAwIGhhc3RhICQgIDIxMCwwMDAuDQphY3R1YWxpemFjaW9uIGRlIHNvZnR3YXJlICAkIDI1MCwwMDAgYSAkIDUyNSwwMDAuIA0KY2FwYWNpdGFjaW9uICQgOCwwMDAgYSAkIDE3LDUwMCANCkxpbnV4IG1pbiAzMzgwMDAgIG1heCA3NTI1MDAgZXN0aW1hZG8gNDEwMDAwDQpgYGB7cn0NCmBgYA0KYGBge3J9DQpnZXRfbW9udGVjYXJsb2VzdCA8LSBmdW5jdGlvbih2bWluLHZtYXgsdmVzdCl7DQogICAgbW4gPC0gKHZtaW4gKyA0KnZlc3QgKyB2bWF4KS82DQogIHMgPC0gYWJzKCh2bWF4IC0gdm1pbikvNikNCiAgDQogIHZhbG1vbnRlY2FybG8gPC0gcm5vcm0oMSwgbWVhbj1tbiwgc2Q9cykNCiAgDQogIHJldHVybih2YWxtb250ZWNhcmxvKQ0KfQ0KZ2V0X3NpbTwtZnVuY3Rpb24oeCx2bWluLHZtYXgsdmVzdCl7DQogICAgDQogICAgdGltZXMgPC0gZ2V0X21vbnRlY2FybG9lc3Qodm1pbix2bWF4LHZlc3QpDQogICAgcmV0dXJuICh0aW1lcykNCn0NCmBgYA0KaGFjZW1vcyB0ZXN0IGRlIGxhIGZ1bmNpb24gZGUgbW9udGVjYXJsbw0KYGBge3J9fA0KZ2V0X21vbnRlY2FybG9lc3QoMzg0MDAwLDcxMDAwMCw0MTAwMDApDQpgYGANCmFwbGljYW1vcyBsYSBmdW5jaW9uIHBhcmEgd2luZG93cyB5IGxpbnV4IGNvbiAxMDAwMCBpdGVyYWNpb25lcw0KYGBge3J9DQpzaW11bGFjaW9uPC0xMDAwMA0Kc2ltX3dpbjwtc2FwcGx5KDE6c2ltdWxhY2lvbiwgZ2V0X3NpbSx2bWluPTM4NDAwMCx2bWF4PTcxMDAwMCx2ZXN0PTQzNTAwMCkNCnNpbV9saW48LXNhcHBseSgxOnNpbXVsYWNpb24sIGdldF9zaW0sdm1pbj0zMzgwMDAsdm1heD03NTI1MDAsdmVzdD00MTAwMDApDQpwcmludChwYXN0ZSgid2luZG93czogIixtZWFuKHNpbV93aW4pLCIgTGludXg6IiwgbWVhbihzaW1fbGluKSkpDQpwcmludChwYXN0ZSgiU1REIHdpbmRvd3M6ICIsc2Qoc2ltX3dpbiksIiBTVEQgTGludXg6Iiwgc2Qoc2ltX2xpbikpKQ0KYGBgDQpMYSBtZWpvciBvcGNpb24gZXMgbGludXggc2kgbm8gc2UgZXZhbHVhIGxhIGRlc3ZpYWNpb24gZXN0YW5kYXINCmBgYHtyfQ0KcGxvdCggc2ltX3dpbiwgdHlwZT0nbCcpDQoNCmBgYA0KYGBge3J9DQoNCnN1bW1hcnkoc2ltX3dpbikNCmhpc3Qoc2ltX3dpbikNCmhpc3Qoc2ltX2xpbikNCmBgYA0KDQpgYGB7cn0NCmxpYnJhcnkodGlkeXZlcnNlKQ0KcXVhbnRpbGUoc2ltX3dpbiwgcHJvYnMgPSAwOjEwLzEwKQ0KcHJpbnQoInByb2JhYmlsaWRhRCB3aW5kb3dzOiIpDQpsZW5ndGgoc2ltX3dpbltzaW1fd2luID40NzIyOTNdKS9sZW5ndGgoc2ltX3dpbikNCg0KcHJpbnQoInByb2JhYmlsaWRhRCBsaW51eDoiKQ0KbGVuZ3RoKHNpbV9saW5bc2ltX2xpbiA+NDEwMDAwXSkvbGVuZ3RoKHNpbV9saW4pDQpgYGANCg0KQmFzYWRvIGVuIGVzdGUgcmVzdWx0YWRvIHByZWZlcmltb3MgbGEgb3BjaW9uIGRlIHdpbmRvd3MgcG9yIHF1ZSBsYSBwcm9iYWJpbGlkYWQgZGUgbGludXggZXMgbXVjaG8gbWF5b3I=