Aim: To analyze performance of 12 consecutive months and predict the performance for the following month.

##    Month Month Desc Received Resolved Reopen Backlog
## 1      1   Jan 2014      444      420     12      24
## 2      2   Feb 2014      452      433     13      19
## 3      3   Mar 2014      677      632      8      45
## 4      4   Apr 2014      612      592      8      20
## 5      5   May 2014      514      465      5      49
## 6      6   Jun 2014      512      491     19      93
## 7      7   Jul 2014      476      472     15       4
## 8      8   Aug 2014      381      399      7       0
## 9      9  Sept 2014      394      397     12       0
## 10    10   Oct 2014      419      408     10      11
## 11    11   Nov 2014      249      257      2       0
## 12    12   Dec 2014      374      353     10      21
## 13    13   Jan 2015      314      346      5       0
## 14    14   Feb 2015      344      331      6      13
## 15    15   Mar 2015      431      450      6       0

Plotting graph:

library(ggplot2)
gL1 <- ggplot(data, aes(x=Month, y=NumberOfCases)) + geom_line(aes(y=Received, colour="Received"))+labs(title="Performance Analysis") + geom_line(aes(y=Resolved, colour="Resolved")) + geom_line(aes(y=Reopen, colour="Reopen")) + scale_colour_manual("", breaks=c("Received","Resolved","Reopen"), values=c("purple","maroon","orange"))
gL1

plot of chunk performance

Analysis:

A sudden increase in inflow experienced in March 2014 caused an increase in the number of reopened cases in the following weeks. There was an unexpected rise in the number of reopened cases in December 2014. This has been classified as an anomaly.

Load Analysis

Plotting graph:

library(ggplot2)
gL1 <- ggplot(data, aes(x=Month, y=NumberOfCases)) + geom_line(aes(y=Received, colour="Received"))+labs(title="Load Analysis") + geom_line(aes(y=Resolved, colour="Resolved")) + geom_line(aes(y=Backlog, colour="Backlog")) + scale_colour_manual("", breaks=c("Received","Resolved","Backlog"), values=c("purple","maroon","orange"))
gL1

plot of chunk load

Analysis:

As the inflow increased early 2014, an increase in backlog was observed in the following weeks.

Predicting Performance for the Next Month

For Received cases:

library(lattice)
library(caret)
## Loading required package: ggplot2
meanReceived <- mean(data$Received)
meanReceived
## [1] 439.5
stdev <- sd (data$Received)
stdev
## [1] 110.2
n <- 18
meanReceived + c(-1,1) * qt(.975, n-1) * stdev / sqrt(n)
## [1] 384.7 494.3

Therefore, there is a 95% chance that the number of Received cases for the next month will be between 384.7 and 494.3.

For Resolved cases:

meanResolved <- mean(data$Resolved)
meanResolved
## [1] 429.7
stdevRes <- sd (data$Resolved)
stdevRes
## [1] 96.18
meanResolved + c(-1,1) * qt(.975, n-1) * stdevRes / sqrt(n)
## [1] 381.9 477.6

Therefore, there is a 95% chance that the number of Resolved cases for the next month will be between 381.9 and 477.6.

For Reopened cases:

meanReopen <- mean(data$Reopen)
meanReopen
## [1] 9.2
stdevReo <- sd (data$Reopen)
stdevReo
## [1] 4.443
meanReopen + c(-1,1) * qt(.975, n-1) * stdevReo / sqrt(n)
## [1]  6.99 11.41

Therefore, there is a 95% chance that the number of Reopened cases for the next month will be between 6.99 and 11.41.