FluffyGroCo Smart Pest Infestation Management: Report & Recommendations

Table of Contents
FLUFFYGROCO SMART PEST INFESTATION MANAGEMENT: REPORT &
RECOMMENDATIONS.................................................................................................. 1
1. Assessment of FluffyGroCo’s briefing note......................................................................1
2. Overview of the investigation.................................................................................... 2
3. Results................................................................................................................ 3
Investigating the impact of rain on infestation...................................................................4
General regression..................................................................................................... 4
The proportions of Crackety Crickling insect infestations.....................................................5
Temperature............................................................................................................. 6
Correlation.................................................................................................................. 6
Correlation analysis.................................................................................................... 6
4: Ethical and security considerations.................................................................................... 7
5. Data science in next steps and potential solutions..................................................................8
Predictive Causal Analytics............................................................................................. 8
Machine learning for predictions....................................................................................... 9
Prescriptive analytics..................................................................................................... 9
6. Report Appendix: Statistics and methodology......................................................................9
Temperature descriptive statistics.................................................................................... 10
List of reference............................................................................................................ 11
FLUFFYGROCO SMART PEST INFESTATION
MANAGEMENT: REPORT & RECOMMENDATIONS
1. Assessment of FluffyGroCo’s briefing note
Having done an in-depth review of FluffyGroco’s briefing note, the value of the Truffula tree to
the company and to the residents of Thneedville is very clear in my mind hence making the task
of issuing proper advice to FluffyGroco Company on the best method of managing Crackety
Crickling insect infestations to be of equal importance. That being said, the company briefing
note fell short in addressing different key concerns that are vital to the processes of coming up
1

with a data science-based elucidation to the problem of Crackety Crickling insect infestations.
Despite the fact that the methods discussed in the briefing note addressed issues pertaining to the
prediction of the infestation of the Truffula trees by Crackety Crickling insects, the dataset
provided by FluffyGroco is superficial when it comes to climate change in Randadoo, Uptagoo
and Nextafoo. More data is required in order to generate a more precise calculation of climate
change patterns in the region especially because the more accurate that the calculation of the
climate change patterns become, FluffyGroco and the farmers in the area can be able to deploy
crop protection mechanisms in a precise manner (Murata and Haraguchi, 2017).
Furthermore, even though the dataset provided by FluffyGroco contains the various
environmental conditions throughout the three plantations during different days, the info that can
be obtained from these data sets is still superficial at best. The lack of a proper analysis of the
dataset by FluffyGroco raises several questions that need to be addressed before any data
science-related methods of smart pest infestation management measures can be implemented.
For example, despite the fact that the briefing note indicates that the data was matched with sets
of voluntary reports by farmers on whether Crackety Crickling insect infestations were observed
in the two weeks following the initial date of observation, the dataset provided does not bring out
these information in a clear manner more so, because the accuracy of this information cannot be
ascertained. Instead of relying on information that was received from farmers on a random basis,
FluffyGroco Company should have used more real-time and more accurate models of obtaining
these data for example, by using in-field sensors, satellite radars, weather forecast reports and
climate scouting reports. In the past, farmers were presented with a several week window from
which to expect insect infestations on their land but currently, as a result of climate change, the
estimated time of arrival of these insects has most likely changed that is why, the usage of these
methods would have gone a long way in generating a more precise and accurate prediction of
Crackety Crickling insect infestations (Hassab, 2018).
That being said, we intend to carry out an investigation that will aid has in coming up with a
suitable proposal and the investigation and report shall be outlined as follows:
 The relationship between the availability of rain and the infestation of Truffula trees by
Crackety Crickling insect infestations shall be established using regression analysis.
 The relationship between the changes in temperature and the infestation of Truffula
trees by Crackety Crickling insects shall be established using correlation analysis.
 An in-depth report shall be written outlining the analysis and results, key ethical and
security considerations, how data science will be applied in the next steps and possible
solutions to the problem.
2

2. Overview of the investigation
Since the assessment of the FluffyGroco’s briefing note revealed a certain level of vagueness in
the dataset the initial step of the investigation was to establish whether the data was viable. The
application of ANOVA concept revealed that the level of significance for both F and P were
below 0.05 which meant that the data availed by FluffyGroco Company was fit; therefore, the
analysis of the data was commenced. Using the dataset provided by FluffyGroco Company, a
regression analysis was performed with an aim of establishing the statistical relationship between
the two initial variables which is the availability of rain versus the infestation of the Truffula
trees by the Crackety Crickling insects. In order to do this, null and alternative hypotheses were
generated as follows:
Hₒ: There is no significant association between rain and Crackety Crickling infestations.
Hı: There is a significant association between rain and Crackety Crickling infestations.
The application of regression analysis revealed that that there is a significant association between
the availability of rain and the infestation of the Truffula trees by Crackety Crickling insects.
Secondly, correlation analysis was used with an intention of establishing the association between
temperature and the likelihood of infestation of Truffula trees by Crackety Crickling insects.
Correlation analysis is a statistical analysis that is mostly made use of while measuring the power
and the relationship that exists amongst two sets of variables which in this case were temperature
versus the infestation of Truffula trees by Crackety Crickling insects. In order to be able to carry
out a comprehensive analysis of the two variables, temperature was divided into two sets which
were temperature below 15 degrees Celsius and temperature above 15 degrees Celsius. The
correlation analysis revealed that as the temperature increases, the rate of Crackety Crickling
insect infestations also rises.
In an attempt to present these findings in a clear and precise manner, tables and graphs were used
to present the analysis and results of the investigation. Furthermore, the investigation went a step
further and looked at various issues that are related to the matter of infestation of Truffula trees
by Crackety Crickling insects. These issues include the ethical and security considerations and
also, how FluffyGroco could apply the principles of data science in its subsequent steps
including potential data science based solutions to the problem of Crackety Crickling insect
infestations on Truffula trees (Eastman, 2018).
3. Results
To understand the statistical relationship which do exist within the two main variables (rain and
temperature) and the statistical impact they have on infestation, a regression and correlation
3

analysis are the best analysis techniques. According to Gunst 2018, regression analysis is used to
evaluate relationship between two or more variables.
Investigating the impact of rain on infestation
General regression
ANOVA
Df SS MS F
Significan
ce F
Regressio
n 1
9.87412
1
9.87412
1
52.0826
3 0.000625
Residual 5476
1038.17
1
0.18958
6
Total 5477
1048.04
5
Coefficint
s
Standar
d Err t Stat P-value
Interce
pt
0.238394
96
0.0123
23
19.345
92
5.4E-
76
X Var
1
0.084127
56
0.0223
52
3.7638
02
0.0001
73
The level of significance for both the F and P are below 0.05 which means that the data is fit,
should it had been above 0.05 the workings and analysis should have stopped.
The regression line is Y = 0.23839496 + 0.08412756X which means that there is a constant of
0.23839496 for the Crackety Crickling insect infestations to occur naturally and that this value is
be increased by 0.08412756 per unit rain.
Null and Alternative Hypothesis
H0: There is no significant association between rain and Crackety Crickling insect infestations
H1: There is significant association between rain and Crackety Crickling insect infestations
• Significance level, α; 5%.
T value at 3.76
Hence we reject the null hypothesis and concludes that, there is a significant association between
rain and Crackety Crickling insect infestations
4