Coco Cola case study

Descriptive statistical Report on Coca Cola fill ( Case Study )

PROBLEM STATEMENT:
Reports show Variations while bottling the fluid Content of 200 milliliter in the newly introduced bottles at the Coca Cola factory in Russia
Using Descriptive Statistical Analysis we need to assess the probability of satisfaction for the customer and the company.

OBJECTIVE:
To find the variation and its effects from the random sample of 50 bottles of 200 ml filled at the Coca cola factory in Russia and state a conclusive report on the population.

DATA ANALYSIS:
Random sample of fifty bottle fills at the Russian Coca Cola factory taken from the population. The measurements of each bottles are mentioned in the table below:
200.1
199.9
200.2
200.2
200
200.1
200.9
200.1
200.3
200.5
199.7
200.4
200.3
199.8
199.3
200.1
199.4
199.6
199.2
200.2
200.4
199.8
199.9
200.2
199.6
199.6
200.4
200.4
200.6
200.6
200.1
200.8
199.9
200
199.9
200.3
200.5
199.9
201.1
199.7
200.2
200.5
200.2
199.7
200.9
200.2
199.5
200.6
200.3
199.8

Using the given Data we arrive at the following measures mentioned below:
Summary measures for selected variables Coke
Count
50.000
Mean
200.118
Median
200.150
Standard deviation
0.419
Minimum
199.200
Maximum
201.100
Range
1.900
First quartile
199.825
Third quartile
200.400
Interquartile range
0.575
Skewness
0.037
1st percentile
199.249
90th percentile
200.600
95th percentile
200.855

1. Amongst the fifty random samples taken from the population the minimum Milliliter bottled amongst the sample is 199.2 ml.
2. The maximum measurement of fluid bottled amongst the sample of 50 units is 201.1 ml.
3. The Mean or the average content filled in the sample of 50 units is 200.118 ml.
4. The median divides the Sample into two, in this sample it shows that the median is 200.15 ml hence more than half of the sample is above 200.15 ml.
5. The above data shows us that more than 25% of the samples fall above the first quartile, hence we see that majority of the data lies towards the customer satisfaction side.

Below Mentioned is the confidence level of the Random sample data
Results for one-sample analysis for Coke
Confidence interval for standard deviation
Confidence level
95.0%
Sample standard deviation
0.419
Degrees of freedom
49
Lower limit
0.350
Upper limit
0.522
This shows 95% true effect lies in between.

Shown below is the frequency at which the bottled content is bottled in the random sample taken,
Frequency table for Coke
Upper limit
Category
Frequency
199
201
1

From the distribution table above, we understand that 30 Bottles out of the total 50 bottles from the random sample are packed with 200 ml or more than 200 ml and 20 bottles are packed less than 200 ml, from this sample we can derive that more number of bottles are on the customer satisfaction scale than on the other side.

The histogram shows us that less than only three bottles out of the sample was filled below 199.4 ml.

Above shown is the Box Plot of the random sample taken for the Study it shows the spread of data between the first quartile and the third quartile. It also shows the majority of the data lies between Q2 and Q3. The red dot on the chart denotes the mean, the line dividing the box is the median. This plot shows us that the only few bottles were filled below 200ml and majority were above the 200 ml.

CONCLUSION:
The above mentioned report shows us that the variance amongst the filled bottles exist. But with a positive result towards the customer satisfaction scale showing us that Majority of the bottles were filled above 200 ml and the bottles with filled with less than 199.825 ml comprises less than 25% of the entire random sample data. Hence the overall variance in bottling or filling Coke bottles in their Company Russia is more beneficial to the customer.