You are on page 1of 10

1

Performance Lawn Equipment: Case Study

Student’s Name

Professor’s Name

Institute Name

Date
2

Performance Lawn Equipment: Case Study

Chapter 8

Introduction

The purpose of the case study is to highlight the impact of the quality problems on the
defects, received from suppliers, provide a statistical study to determine the effects of a year of
education, grade point average & age, and statistical study for the production of a new design of
lawn mower engines.

Part 1: Impact of Quality Problems on Defects

To check the effect of the quality problems on the defects, there is a need to implement
the regression analysis on the Number of Defects specifically received from suppliers before
taking an initiative in August 2015.

Defects before Initiative - August 2015


870
860
f(x) = 0.0382353621013 x³ − 1.2089707528529 x² + 11.981601197445 x + 795.37151702786
850R² = 0.694627173288867
840
830
820
810
800
790
780
0 5 10 15 20 25

Graph 1 – Defects before Initiative – August 2015

It can be seen from the graph and trend line before the initiative taken in August 2015,
which the trend of Defects is moving up which shows an increasing amount of defects received
over time. The value of R square for the order 3 polynomial is calculated as 0.69, which indicates
that only a 69 percent increase in Defects is received. The Defects received can be related to the
time which shows problems in quality over time. The result indicates that if no initiative is taken
3

for the rectification of quality problems, it is highly likely that the number of defects will
continue to arise. The equation of the trend line shows the relation between the numbers of
defects over time as follows,

Number (No.) of Defects = 795. 37 + 11.98*T – 1.209 * T ^ 2+ 0.038 * T ^3

The “T” in the trend line equation shows the increment of time in months, zero time starts on
January 1st, 2014. It is one month before the collection of the first data point for the number of
defects.

The result of the regression analysis is as follows,

Image 1 – Regression Analysis for Defects

Significance and Hypothesis Testing

The following hypothesis needs to be used for the above regression analysis,

Null Hypotheses N0 = β1 = β2 = β 3 (T, T ^2, T ^ 3 are not significant variables. Therefore,


Defects do not change.

Alternate Hypothesis H1 states that at least one βj is not equal to zero (Means that at least one of
the time variables cannot be equal to zero, which means Defects can change with time)

The results of the Regression analysis show that the value of significance is 0.00059 and the
value of the variable Time is 0.00059, for the variable T^2 is 0.0050 and for the variable T^3 is
4

0.004 respectively. All of these values are lower than the α = 0.05 which shows that the null
hypothesis can be rejected therefore it is concluded that the results are statistically significant,
and the number of Defects keeps changing with time (Jim, 2022).

Now, for the second part, to check the effect of the quality problems on the defects, there
is a need to implement the regression analysis on the Number of Defects specifically received
from suppliers from January 2014 till December 2018.

Graph 2 – Defects after Taking an Initiative

It can be seen from the graph and trend line after the initiative was taken in August 2015,
that the trend of Defects is moving down which shows a reduced amount of defects received over
time. The value of R square for the order 3 polynomial is calculated as 0.98, which indicates that
only 98 percent of the reduction in Defects is calculated. The Defects received can be related to
the time which shows problems in quality over time.

The equation of the trend line shows the relation between the numbers of defects over
time as follows,

Number (No.) of Defects = 803.7+ 5.63*T – 0.21 * T ^ 2


5

The “T” in the trend line equation shows the increment of time in months, zero time starts
on January 1st, 2014. It is one month before the collection of the first data point for the number of
defects.

The result of the regression analysis is as follows,

Image 3 – Regression Analysis after Taking an Initiative

Significance and Hypothesis Testing

The following hypothesis needs to be used for the above regression analysis,

Null Hypotheses N0 = β1 = β2 = β 3 (T, T ^2, are not significant variables. Therefore, Defects
Do not change.

Alternate Hypothesis H1 states that at least one βj is not equal to zero (Means that at least one of
the time variables cannot be equal to zero, which means Defects can change with time)

The results of Regression analysis shows that the value of significance is approximately
equal to zero and the p-value of the variable Time is 2.76E-22, for the variable T^2 is 4.30E-22. ,
which is smaller than α = 0.05 meaning that we reject the Null Hypothesis and conclude that it is
statistically significant to say that the number of defects is changing (increasing) with time.
6

Conclusion

It can be seen in the result, that if we break the trend line into two parts, it can be seen
that in the first section from January 2014 to August 2015, the R square equals 0.62, which
means that 62% of the reduction in the defects received from suppliers relates to the time.
However, in the second section, the value of R Square equals 0.88 which shows that a 98 % of
reduction exists in the number of defects related to time.

Part 2: Field Service Staff Turnover

To determine the significant impact on the field service, a regression study has been
conducted on field service technician retention. The impact of the years of education, college
GPA, and age of the field service technician is measured based on the inputs of HR
Management. By using all these factors together, it is calculated from the results that collective
data has not any significant statistical impact on retention, which can easily be interpreted from
the value of R square which is calculated as 0.15. The insignificance of the data can be best
explained by the low value of adjusted R square equals 0.079 in the model. Moreover, it is also
found that one of the factors is quite statistically significant and is directly related to the retention
of the field service technician.

Results and Analysis

Age has a p-value of 0.0376 and there exists a 96 percent correlation exists between the
date of hire, age, and retention factor for field service technicians. In an organization, it cannot
be discriminated against easily based on the value of the age of field technician applicants.
However, it is also possible, that number of years of experience can be reduced to increase the
average age of new technicians that can bring on board. The requirement of experience changes
with the recruitment techniques to get the older applicants and it has a direct impact on the
retention trend and can increase the stability of employees who are staying with PLE for quite a
long period.
7

Image 3 – Regression Analysis of Field Service Technician

Similarly, the results of the normal probability output show that normal distribution for the years
on the y –axis, against the sample percentile plotted on the x-axis (Nist, 2022).

12
Normal Probability Plot
10
8
6
Y

4
2
0
0 20 40 60 80 100 120
Sample Percentile

Graph 3 – Normal Probability Plot for YearsPLE


8

Part 3: Engine Production Times

During the examination of the trend line for the chart of the engine production times, it is
observed that the line with the highest R squared value is the Logarithmic Trend Line. By
understanding the relation of the trend line, the natural log of the independent variable was taken
for the sample, and then regression analysis is performed with the production time. The result of
the ANOVA analysis shows that significance F is zero, therefore null hypothesis is rejected, and
it is concluded that the slope is statistically significant. However, by measuring the independent
p-values for the independent variable and the sample of LN, the value is less than 0.05.
Therefore, it is assumed that the coefficient of regressions is statistically significant (James,
2022).

70.0 Engine Production Time


f(x) = − 5.19597836499563 ln(x) + 66.1517452952855
60.0 R² = 0.995861257785294

50.0

40.0

30.0

20.0

10.0

0.0
0 10 20 30 40 50 60

Graph 4 – Trend Line For engine production time

Results & Analysis

The interpretation of these two tests shows that samples number are statistically
significant and do have an effect on the dependent variable, which in our case, is the engine
product time. Using the expected values for the trend line, it can be figured out the learning rate,
and it is observed that further down the lines when engines are samples, the smaller we can
expect the production time. When production samples are doubled from sample 25 to sample 50,
the learning rate we can get is 92.71 %, which shows that every time production of an engine
9

doubles in numbers, the average production time is expected to increase by 8.29 percent. The
figure is quite useful to help us better estimate the prediction cost for future new products and
designing of products without extensive running of the prototype.

Image 4 – Regression Analysis for Engine production time


10

References

James Chen (2022). Trend Line. Investopedia.

https://www.investopedia.com/terms/t/trendline.asp

Jim Frost (2022). How to Interpret P-values and Coefficients in Regression Analysis

https://statisticsbyjim.com/regression/interpret-coefficients-p-values-regression/

NIST (2022). Normal Probability Plot

https://www.itl.nist.gov/div898/handbook/eda/section3/normprpl.htm

You might also like