rClassMonteCarlo.R

library(qualityTools)

seedToSet<-1
set.seed(seedToSet)
#Number of items for simulation
nSim<-5000

#Process 1, Customer Inquiry, Lognormal Distribution, mean of 1 hour standard deviation of .25
set.seed(seedToSet)
process1<-rlnorm(nSim,meanlog=1,sdlog=.25)

#process2=Loan Application, Average 24hr, minimum 7 maximum 42, Beta distribution, Alpha=2.5, Beta=7
set.seed(seedToSet)
process2<-rbeta(nSim,2.5,7)
process2<-7+process2*(42-7)

#Process3, Document Variation, 16 to 32 hr, but 20% of time could take 32 to 48 hours
set.seed(seedToSet)
process3v1<-runif(nSim,16,32)
set.seed(seedToSet)
process3v2<-runif(nSim,32,48)
prob3V1<-.8
set.seed(seedToSet)
process3Event<-runif(nSim,0,1)
process3 <-runif(nSim,0,1)
#set the actual process 3 results based on which event was chosen
for (i in 1:nSim){
  if(process3Event[i]<=prob3V1) {
    process3[i]<-process3v1[i]
  } else {
    process3[i]<-process3v2[i]
  }
}

#Process 4, Loan underwriting, approves or disproves, uniform of 1-8 hours
process4<-runif(nSim, 1,8)

#Process5, Loan Closing, average 16 hr, normal Distribution, 16, sd=2
process5<-rnorm(nSim,16,2)

#Process6, Loan Disbursement, 66% of time fals within 12 and 20...normal Dist, avg=16, sd=4
process6<-rnorm(nSim, 16,4)

#Total Process
processTotal <- process1+process2+process3+process4+process5+process6

#Plot the total
hist(processTotal)
summary(processTotal)

#capability

pcr(processTotal,"normal",50,96,96, main="Process Total")