Scribd
Upload
44 views

Statistical Methods For Decision Making

1) The document analyzes temperature data from a cold storage facility over different seasons to calculate averages and probabilities. 2) The overall average temperature for the full year was 3.002466 degrees C with a standard deviation of 0.4658 degrees C. 3) The probability of temperature falling below 2 degrees C was 1.57% and the probability of exceeding 4 degrees C was 1.61%, resulting in a potential penalty of 10% of the annual maintenance contract fee.

Uploaded by

surajkadam
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
44 views

Statistical Methods For Decision Making

1) The document analyzes temperature data from a cold storage facility over different seasons to calculate averages and probabilities. 2) The overall average temperature for the full year was 3.002466 degrees C with a standard deviation of 0.4658 degrees C. 3) The probability of temperature falling below 2 degrees C was 1.57% and the probability of exceeding 4 degrees C was 1.61%, resulting in a potential penalty of 10% of the annual maintenance contract fee.

Uploaded by

surajkadam
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Group Assignment(Cold

Storage)
-Group 11(Raj,Navnath,Suraj,Shantanu)
Mean cold storage temperature for
Summer, Winter and Rainy Season
 Code :
library(ggplot2)
df<-read.csv("Cold_Storage_Temp_Data.csv") #read the file and store in df
head(df)
summary(df)
Seas_Avg<-aggregate(df$Temperature,by=list(df$Season),FUN=mean)
colnames(Seas_Avg)<-c("Season","Average_temperature")
ggplot(Seas_Avg,aes(x=Season,y=Average_temperature))+geom_col(aes(fill=Seaso
n))+geom_text(aes(label=round(Average_temperature,2),hjust=0, vjust=-0.5))
Result
Overall mean for the full year
 Code:
ggplot(df,aes(1:length(df$Temp
erature),df$Temperature))+geo
m_point(aes(col=Season))+geom
_line(aes(y=mean(df$Temperatu
re),col="Overall Average"))
+xlab("Index")+ylab("temperatur
e")
mean(df$Temperature)

Mean temp : 3.002466


Std dev of Temperature

 Code :
ggplot(data =
normal,aes(x=Temperature,y=He
ight))+geom_line()+geom_line(ae
s(x=mean(df$Temperature),col="
Mean"))+geom_line(aes(x=mean(
df$Temperature)+sd(df$Tempera
ture),col="sd1"))+geom_line(aes(
x=mean(df$Temperature)-
sd(df$Temperature),col="sd-1"))
sd(df$Temperature)

Standard 0.4658
deviation temp :
What is the probability of temperature
having fallen below 2 deg C?
 Code:pnorm(2,mean=mean(df$Temperature),sd(d
f$Temperature))
 Code 2: create function to shade
shade <- function(x, bound) {
Height <- dnorm(x, mean =
mean(df$Temperature),
sd = sd(df$Temperature))
Height[x > bound] <- NA
return(Height)
}

 Code 3: ggplot(data =
normal,aes(x=Temperature,y=Height))+geom_line
()+stat_function(fun = shade,args = 2.0,geom =
"area", alpha =
.5,fill="red")+ggtitle(paste0("Probability: ",
round(pnorm(2,mean=mean(df$Temperature),sd(
df$Temperature),lower.tail = T),4)*100,"%"))
What is the probability of temperature
having gone above 4 deg C?
 Code :ggplot(data =
normal,aes(x=Temperature,y=Height))+geo
m_line()+stat_function(fun = shade,args =
4.0,geom = "area", alpha =
.5,fill="red")+ggtitle(paste0("Probability: ",
round(pnorm(4.0,mean=mean(df$Temperat
ure),sd(df$Temperature),lower.tail =
F),4)*100,"%"))

 Code :
shade <- function(x, bound) {
Height <- dnorm(x, mean =
mean(df$Temperature), sd = sd(df$Temperature))
Height[x < bound] <- NA
return(Height)
}
What will be the penalty for the AMC
Company?
 If the probability of temperature going outside the 2 - 4 C during the one-year contract was above 2.5%
and less than 5% then the penalty would be 10% of AMC (annual maintenance case). In case it exceeded
5% then the penalty would be 25% of the AMC fee.
 Probability of temperature going outside the 2 - 4 C = Probability of temp having gone above 4 and
below 2 => 1.61% + 1.57% = 3.18%
 Thus penalty will be 10% of AMC.
Which Hypothesis test shall be performed to
check if corrective action is needed at the
cold storage plant?

 We need to perform a T-test. Population Standard deviation is unknown


 We have to check do we have enough evidence from the sample to prove that
the temperature of the plant has increased by 3.9 C.
 If we find enough statistical evidence to prove that the temperature is above
3.9 C. We can conclude that there is definitely some issue from our end.
State the Hypothesis, perform
hypothesis test and determine p-value
 Null Hypothesis : mu <=3.9
 Alternative Hypothesis : mu>3.9
 Use code : t.test(x=dfmar18$Temperature,mu=3.9,conf.level =
0.99,alternative = "greater")
 Result :
Inference

 P value is 0.0047 which is less than 0.01 .


 Thus we reject Null Hypothesis(Saying that mean is less than or equal to 3.9)
 We accept alternative hypothesis (Saying that mean is greater than 3.9)
 At a confidence level of 99 %

 Conclusion : Since we reject that the mean is less than or equal to 3.9 it can
be said that there is some issue from the dairy’s end. The manager needs to
extra vigilant through out the day control temperature of the storage. Since it
is summer season a natural increase in temperature is likely to increase.
 There is enough evidence to conclude that mean is greater than 3.9.

You might also like