Bullmer Fabric Cutter Reliability - Refined
library(WeibullR)
library(readxl)
rm(list = ls()) # remove all environment
Blmr <- read_excel("Bullmer.xlsx", col_types = c("date", "skip", "text", "text", "text", "text",
"numeric", "numeric", "numeric", "skip",
"skip", "skip"))
#View(Blmr)
f <- function(time, title, pch) {
today = as.numeric(Sys.Date()-as.Date("2010-01-01"))
fday = as.numeric(as.Date("2018-06-01")-as.Date("2010-01-01"))
d <- wblr(time, pp = "median", pch = pch)
d$options$main <- c(title , d$options$main)
d2 <- wblr.conf(wblr.fit(d, dist = "weibull2p"))
d3 <- wblr.fit(d, dist = "weibull3p")
e2 <- wblr.conf(wblr.fit(d, dist = "lognormal"))
e3 <- wblr.fit(d, dist = "lognormal3p")
time<-c(sort(time),fday)
l<-length(time)
w <- c(d2$fit[[1]]$beta,d2$fit[[1]]$eta)
haz2 <- w["Beta"]/w["Eta"]*(time/w["Eta"])^(w["Beta"]-1)
w <- c(d3$fit[[1]]$beta,d3$fit[[1]]$eta, d3$fit[[1]]$t0)
haz3 <- w["Beta"]/w["Eta"]*((time-w["t0"])/w["Eta"])^(w["Beta"]-1)
w4 <- e2$fit[[1]]$fit_vec
w5 <- e3$fit[[1]]$fit_vec
haz4 <- dlnorm(time, w4["Mulog"], w4["Sigmalog"],log = F)/(plnorm(time,w4["Mulog"], w4["Sigmalog"],lower.tail = F, log.p = F))
haz5 <- dlnorm(time, w5["Mulog"], w5["Sigmalog"],log = F)/(plnorm(time,w5["Mulog"], w5["Sigmalog"],lower.tail = F, log.p = F))
LEGEND<-c(paste("Weibull 2p:",format(d2$fit[[1]]$fit_vec[3],digits = 3)),
paste("Weibull 3p:",format(d3$fit[[1]]$fit_vec[4],digits = 3)),
paste("Lognormal 2p:",format(w4[3], digits = 3)),
paste("Lognormal 3p:",format(w5[4], digits = 3)))
cat("Weibull 2-paprameters likelyhood")
plot.wblr(d2, in.legend.blives = F)
lines(c(today,today),c(-10,100),type = "l", col = "blue")
lines(c(fday,fday),c(-10,100),type = "l", col = "red")
cat("Weibull 3-paprameters likelyhood")
plot.wblr(d3)
lines(c(today,today),c(-10,100),type = "l", col = "blue")
lines(c(fday,fday),c(-10,100),type = "l", col = "red")
cat("Log-normal 2-paprameters likelyhood")
plot.wblr(e2, in.legend.blives = F)
lines(c(today,today),c(-10,100),type = "l", col = "blue")
lines(c(fday,fday),c(-10,100),type = "l", col = "red")
cat("Log-normal 3-paprameters likelyhood")
plot.wblr(e3)
lines(c(today,today),c(-10,100),type = "l", col = "blue")
lines(c(fday,fday),c(-10,100),type = "l", col = "red")
if (w["t0"]>0) {
YLIM<-c(min(haz3,haz2,haz4,haz5),max(haz3,haz2,haz4,haz5))
} else { YLIM<-c(min(haz2,haz4),max(haz2,haz4)) }
YLIM<-YLIM*100
cat("Failure prediction")
plot(time,haz2*100, type = "b", main = c(title,"Hazard rate"), xlab = "Time", ylab="Failure rate [%]",col="blue", ylim = YLIM, pch=0, family = "serif")
lines(time,haz4*100, type = "b", col="green", lty=3,pch=2)
if (w["t0"]>0) {
lines(time,haz3*100, type = "b", col="red", lty=2,pch=1 )
lines(time,haz5*100, type = "b" ,col="black" , lty=4,pch=3)
}
grid()
legend(x="topleft",legend = LEGEND, cex = 0.7, col=c("blue","red","green","black"),pch=c(0,1,2,3))
}
f2<-function(frame, title) {
mldt<-mean(frame$LDT , na.rm = T)
mttr<-mean(frame$TTR , na.rm = T)
mtbf<-f3(frame$Event)
cat(paste(title," Availability:\n"))
cat(paste("MTBF:",mtbf,"\n"))
cat(paste("MTTR:",mttr,"\n"))
cat(paste("MLDT:",mldt,"\n"))
cat(paste("Technical:",format(mtbf/(mtbf+mttr),digits = 3),"\n"))
cat(paste("Operational:",format(mtbf/(mtbf+mldt+mttr),digits = 3),"\n"))
}
f3<-function(frame) {
frame<-sort(frame)
l<-length(frame)
x<-0.0
for (i in 2:l) {
x<-x+frame[i]-frame[i-1]
}
return(x/(l-1))
}
q <- wblr(Blmr$Event, pch = 0)
q1 <- wblr.conf(wblr.fit(q, dist = "weibull2p"))
plot.wblr(q1, in.legend.blives = F)
q2 <- wblr.conf(wblr.fit(q, dist = "lognormal"))
plot.wblr(q2, in.legend.blives = F)
q3 <- wblr.fit(q, dist = "weibull3p")
plot.wblr(q3)
q4 <- wblr.fit(q, dist = "lognormal3p")
plot.wblr(q4)
f2(Blmr,"Bulmer fabric cutter")
## Bulmer fabric cutter Availability:
## MTBF: 43.1833333333333
## MTTR: 0.211805555555554
## MLDT: 3.5
## Technical: 0.995
## Operational: 0.921
f(time = Blmr$Event[23:28], title = "Control events", pch = 2)
## Weibull 2-paprameters likelyhood
## Weibull 3-paprameters likelyhood
## Log-normal 2-paprameters likelyhood
## Log-normal 3-paprameters likelyhood
## Failure prediction
f2(Blmr[23:28,],"Control")
## Control Availability:
## MTBF: 54.4
## MTTR: 0.204166666665697
## MLDT: 6.2
## Technical: 0.996
## Operational: 0.895
f(time = Blmr$Event[29:38], title = "Cutter System failures", pch = 3)
## Weibull 2-paprameters likelyhood
## Weibull 3-paprameters likelyhood
## Log-normal 2-paprameters likelyhood
## Log-normal 3-paprameters likelyhood
## Failure prediction
f2(Blmr[29:38,],"Cutter System")
## Cutter System Availability:
## MTBF: 266.888888888889
## MTTR: 0.406250000001819
## MLDT: 0
## Technical: 0.998
## Operational: 0.998
f(time = Blmr$Event[39:45], title = "General Electric events", pch = 4)
## Weibull 2-paprameters likelyhood
## Weibull 3-paprameters likelyhood
## Log-normal 2-paprameters likelyhood
## Log-normal 3-paprameters likelyhood
## Failure prediction
f2(Blmr[39:45,],"Electric")
## Electric Availability:
## MTBF: 379.166666666667
## MTTR: 0.0625
## MLDT: 0
## Technical: 1
## Operational: 1
f(time = Blmr$Event[46:52], pch = 6, title = "Software related events")
## Weibull 2-paprameters likelyhood
## Weibull 3-paprameters likelyhood
## Log-normal 2-paprameters likelyhood
## Log-normal 3-paprameters likelyhood
## Failure prediction
f2(Blmr[46:52,],"Software")
## Software Availability:
## MTBF: 291.833333333333
## MTTR: 0.194444444445253
## MLDT: 0.333333333333333
## Technical: 0.999
## Operational: 0.998
f(time = Blmr$Event[53:55], pch = 8, title = "Stopper related events")
## Weibull 2-paprameters likelyhood
## Weibull 3-paprameters likelyhood
## Log-normal 2-paprameters likelyhood
## Log-normal 3-paprameters likelyhood
## Failure prediction
f2(Blmr[53:55,],"Stopper")
## Stopper Availability:
## MTBF: 387.5
## MTTR: 0.12499999999635
## MLDT: 15
## Technical: 1
## Operational: 0.962
f(time = Blmr$Event[56:61], pch = 9, title = "Vacuum System events")
## Weibull 2-paprameters likelyhood
## Weibull 3-paprameters likelyhood
## Log-normal 2-paprameters likelyhood
## Log-normal 3-paprameters likelyhood
## Failure prediction
f2(Blmr[56:61,],"Vacuum")
## Vacuum Availability:
## MTBF: 423.2
## MTTR: NaN
## MLDT: NaN
## Technical: NaN
## Operational: NaN