Analyzing Bird Strike Data using SAS Analytical Tools

This project focuses on analyzing bird strike data using SAS analytical tools. It covers tasks such as creating a data dictionary, analyzing different species involved, calculating the percentage of times pilots were warned of birds, and visualizing the data using geomaps and graphs. The goal is to identify bird strike records and develop analytical models with risk factors.

Statistics
Student Name:
Register Number:
Submission Date:
1

Table of Contents
Executive summary.........................................................................................................................3
Background......................................................................................................................................3
Tasks................................................................................................................................................4
Discussion of Findings..................................................................................................................21
Conclusion.....................................................................................................................................23
Reference.......................................................................................................................................24
Appendix........................................................................................................................................24
2

Table of Figures
Figure 1: Data Dictionary................................................................................................................5
Figure 2 :Different species...............................................................................................................6
Figure 3:Percentage of the time was the Pilot Warned of Birds......................................................8
Figure 4: Airline/Operators............................................................................................................10
Figure 5: Proportion (percentage) of bird strikes...........................................................................11
Figure 6: Aircraft...........................................................................................................................12
Figure 7: Whilst the Aircraft..........................................................................................................13
Figure 8: Bird Strike......................................................................................................................13
Figure 9: Precipitation Conditions.................................................................................................14
Figure 10: Species of Wildlife.......................................................................................................14
Figure 11: Most costly Species of Wildlife....................................................................................15
Figure 12: Species Bald Eagle and Canadian Goose.....................................................................16
Figure 13: Crosstab for the data values Pilot warned of bird strike, Altitude bin, and the total ...17
Figure 14: Geomap of the Bird Strike Data...................................................................................17
Figure 15: Sum of Records, Bird Strike Count by Airport............................................................17
Figure 16: Bird Strike Count by Airport........................................................................................18
Figure 17: Cluster Analysis...........................................................................................................19
3

Executive summary
In this project to analyse the bird strike data by using the SAS analytical tool. First task
is to create a data dictionary for the given dataset and analyse the different species that have been
involved in bird strikes. Then calculate the percentage of times for pilot being warned of birds,
large wildlife and medium wildlife (Liao and Lee, 2013).To calculate how many instances are
present in the dataset and to investigate particular bird’s data. Plot the geomap for the bird strikes
which used to find the place for birds and then calculate the sum of the total records. Compare
the bird strike cost per airport. The goal of the analysis was to identify the bird strike record and
develop analytical models with these risk factors. The analysis is based on all the categories and
measures the value dataset and an analysis of the particular data for category dataset. The
category dataset contains the aircraft data, airplane data, origin state, wildlife size and wild life
species data’s. The data containing the information about wildlife species dataset than measures
which is used for calculating the number of data present in this dataset. The analysis has used
data sources such as wildlife species, airline bird strikes, aircraft data, bald eagle, Canadian
goose, pilot warned of bird strike, altitude bin, total cost and time of service. To find the analysis
with a high prevalence of individual data in multiple records ID’s which would make it difficult
to obtain consistent for the individual affected. Compare the two variable data’s which used to
describe the cluster difference from one another in terms of the two variable’s used. Two models
are developed by using SAS analytical tools for analysis of the overall fatality risk. The
hierarchy approach required applications of the resolution process. Create a hierarchy process for
bird strikes dataset to analyse the dataset and use it to create the decision process.
4

Background
The objective of this project is to analyze the bird strikes data using SAS programming of
the visualization on the cloud application. Data Analysis using SAS Cloud Usage is considered
for the initial phase, the SAS cloud apparatus creates a group and the bird strikes are definitively
defined as a bird between a plane or aircraft or aircraft or ground roll (A Survey of SNP Data
Analysis, 2018). The terms are often spent in hunting of ground bird strikes are common and a
significant threat to aircraft safety. Analyze the data that can first create the group data for the
dataset. Analyze the aircraft of the Bird strikes dataset that contains the analyzed categories of
the data fields, which contains include the Airlines operators, aircraft flight number, aircraft
model, type of aircraft operation, aircraft name, flight date, freedom enroute, indicated derange,
impact to flight data field, analyzing the sky condition of the aircraft operation service,
precipitation, aircraft number, wide life number, altitude bin of the data analysis using cloud SAS
programming application. The birds can detect airplane landing lights and weather radar and can
avoid the airplane. The bird strikes are issues on the airplanes and engine operation working and
spinner making which affects the aircraft operation system.
Tasks
In this project to analyse the data for bird strikes dataset by using the SAS analytical tools
to visualize the data. First step is to create the data dictionary for the given dataset and visualize
that data set.
5

Import the bird strikes data from local drive then run the program using SAS program (Allen,
Campbell and Hu, 2015).
Figure 1: Data Dictionary
6

It display the bird strikes table content and the data information.
Figure 2 :Different species
To analyse how many different species are involved in bird strikes using SAS analytical tools.
7

First step is to select the bird strikes dataset on SAS analytical page that is then used to analyse
the wildlife species data and plot the graph for that dataset. It show the visualization table of
wildlife species data from the bird strikes dataset (Patel, 2012).
It shows all the wildlife species names and the frequency size for that data.
8

It visualizes all the dataset value and it uses this to calculate the number of species in this dataset.
It shows 610 different species (distinct) have been involved in bird strikes (Buelens, 2016).
Figure 3: Percentage times when the Pilot was warned of the Birds
9

The dataset contains the times when the pilot was warned of the birds which is used for
calculating the percentage of that times (Buelens, 2016). The pilot warned data is then visualized
on the SAS analytical page which is then used for calculating the value for the wildlife date.
It shows the visualization table for pilot warned dataset and the frequency for that data. It
contains the missing data, yes and no dataset that calculates the total data present in the dataset.
10

Figure 4: Airline/Operators
It shows the graph for airline operator data which is used to find the top airline’s in this dataset.
The frequency value for the airline’s total value and that is display’s the value as a percentage.
It shows large and medium wild life data and that is visualized as a graph.
11

Figure 5: Proportion (percentage) of bird strikes
It display’s the number of airline operator’s in this record and this airline list is based on the
frequency list. The top five airlines are Military airline, United airlines, Southwest airlines,
Business and FedEx express airlines.
12

Figure 6: Aircraft
It shows the aircraft flight number and the total cost value graph for the bird strikes dataset.
Military aircraft is most likely to experience a bird strike because it has more than 9000 airlines
based on this record.
Figure 7: Whilst the Aircraft
13

It shows the aircraft data visualized as a table for calculating the number of instances that are
present in this dataset.
Figure 8: Bird Strike
14

Figure 9: Precipitation Conditions
It shows the perception condition data for each dataset. Rain condition has the largest dataset.
Figure 10: Wildlife Species
15

Combine the wildlife data and the total cost data that shows the total cost datset for each wildlife
datset. To calculate the largest total cost for that data. The calcornia spices has the largest data
and it has more than 3500 data.
Figure 11: Most costly Wildlife Species
Plot the graph for wildlife species data and the airlines data for analysing the most costly
wildlife species. The American wildlife species data is most costly species of wildlife and the
data value is 2.
16

Figure 12: Species - Bald Eagle and Canadian goose
To select the Canada goose and the Bald eagle data from the Wildlife species. Plot the graph for
this data. Compare both the data’s; the Canada goose data has the highest value of wildlife data.
17

Figure 13: Crosstab of the data values of Pilot warned of bird strike, Altitude bin and the
total cost
To select the pilot warned of bird strike, attitude bin and total cost. It shows the cross tab table
for all the data’s and calculates the total value for that dataset.
Figure 14: Geomap of the Bird Strike Data
To create the geomap for bird strike dataset that display’s the origin state based on category data.
18

Figure 15: Sum of Records, Bird Strike Count by Airport
To select the airport data and the total cost value after eliminating the missing data from the bird
strike dataset. Plot the graph and visualize the airport data table and calculate the sum of the
record.
19

Figure 16: Bird Strike Count by Airport
Select the total cost value from the category dataset and plot the graph for that data.
Compare the airport and wildlife dataset which is used to calculate the largest data that is used
for identifying which data is the highest data from the given dataset.
Figure 17: Cluster Analysis
The cluster analysis is applied for time out of the service and the total cost data is used to create
the cluster matrix. Cluster techniques are sued for finding subgroups of observations within a
dataset. The same group has similar data and different groups have dissimilar dataset that cluster
observes the similar and dissimilar data from given dataset. To identify which observations are
alike and potentially categorize which are used for clustering methods. Most commonly used k
means clustering data for splitting a dataset into set of k groups (Mane and Shinde, 2015). We
have used two dataset’s such as time out of service and total cost to split the similar and
dissimilar data’s.
20

Discussion of Findings
The initial stages of the birds strikes was to create data for the initial stages of the
operation for creating the data source on the SAS cloud application using the data dictionary and
to find by analyzing each birds strikes categories of data fields and constraints of the SAS
visualization on the aircraft system (Yang, 2012). The second stage was to analyse each data file
on the SAS cloud application, analyse different species of the data set on the bird strike list on
the data table and to check which contains the data that includes the overall categories in the 24
data fields required in the data analyzing table of SAS application system. Different species of
the bird strikes is identified and analysis of the field or constraints which it includes is carried
out. Involving different species of bird strikes that includes the aircraft operator of the distinct
value is denoted by 372. The flight number of the distinct value is 5000 records which have to be
analyzed on the dataset. Aircraft model of the distinct value is 526 records which is specified by
the designing of the aircraft visualization. Aircraft type of the distinct value is 4 and this is
specified as the type of visualization on the SAS analysis application. The name of the aircraft
which contains the distinct records is to be analyzed. For the 1,704 categories of the bird strikes
are to be used for easily finding out and predicting the issues on the impact flight affected list on
the SAS cloud analysis application.
The third task is to be specified as the analysis of the pilot being warned of bird strikes or
wildlife and analyse all the data, save and find the location of the stored records of the SAS data
analysis. The identification of the data and the plot has to be once verified for each of the record
and give the answer as yes or no which has to be specified on the data records.
The first task is to find the specified as each stage of bird strike by the airline operation, as that
contains the analyze of the data fields and the five constraints as, US Airways, United Airlines,
Delta air lines, Northwest Airlines, American airlines of the bird strikes on the SAS data
analysis.
21

Presenting the fifth task, this is analyzing the percentage proposition of the large and medium
wide life of the bird strikes. This includes the time of the day, number of strikes, and size for the
species categories of the data analyzation.
The SAS data analyzing of the six tasks is to be specified as which aircraft would be most likely
to experience a bird strike on the data set,
22

The seventh stage of the aircraft bird strike is the analysis of how many instances it occurs when
the aircraft is in flight.
The eight stage of the analysis task is to specify when there is no cloud as compared with
overcast conditions and this will be analyzed.
Reflection
Big data is collected a large data then it generated the information for that data. It has
larger number of data so that process is very complex to analysis the data. The data contains
market trends data, hidden patterns data, unknown correlations and customer preferences that can
help organization make informed business decision. The big data handle the business data for
marketing purpose (Majumdar, Naraseeyappa and Ankalaki, 2017). To analyze datasets and that
data to help organizations make informed business decisions. The technologies and techniques
are applied for big data analytics. It is a form of advanced analytical which involves complex
applications with elements. The big data analytics elements are predictive models, statically
algorithms and that has high performance analytical systems. Big data analytical mainly used for
business benefits that include revenue opportunities, operational efficiency and competitive
23

advantages (Ruan and Zhang, 2017). To analyses growing volume of structured transaction data
for statistical analysis purpose. The data has two type of data such as unstructured data and semi
structured data.
Challenges
The internal systems and external systems sources are using for analysis the big data. It is
real time analyses on data fed into Hadoop systems through processing engine (Wegman, 2012).
To produce software for big data analysis that easier to use for artificial intelligence and machine
learning technologies. We have used artificial intelligence and machine learning concept for
implement the big data analytical.
Learning and Contribution
We have learning the how to collect the data and analysis that data. In additionally learning the
reporting of data for purposes of understating and optimizing learning and the environment in
which it occurs. To determine the probable future outcomes and we can detect similarities in
behaviors or detect anomalous pattern.
Conclusion
In this project, we have successfully analysed the given data using SAS analytical tool.
We have used bird strike dataset for analysing the data and visualizing the data. First task was to
create a data dictionary for the given dataset and analyse the different species which have been
involved in bird strikes. And then calculate the percentage of times for pilot warned birds, large
wildlife and medium wildlife. To calculate how many instances are present in the dataset and to
investigate the particular bids data. Plot the geomap for bird strikes which are used to find the
place for birds and calculate the sum of total records for comparing the bird strike cost by airport.
The goal of the analysis was to identify the bird strikes record and develop analytical models
with risk factors. The analysis was based on all categories and measures the value dataset and
analysis the particular data for category dataset. The category dataset contains the aircraft data,
airplane data, origin state, wildlife size and wild life species data. The data containing
information about wildlife species dataset than measures, which is used for calculating the
24

number of data present in this dataset. Data Analysis SAS Cloud Usage considers the initial
phases, the SAS cloud apparatus creates a group and the bird strikes are definitively defined as a
bird between a plane or aircraft or aircraft or ground roll. Aanalyzing the aircraft for the Bird
strikes dataset that contains the analyzed categories of the data fields that includes the airlines
operators, aircraft flight number, aircraft model, type of aircraft operation, aircraft name, flight
date, freedom route, indicated range, impact to flight data field, analyzing the sky condition of
the aircraft operation service, precipitation, aircraft number, wide life number and altitude bin of
the data analysis using cloud SAS programming application. Finally, we have successfully
analysed the give dataset by using SAS programming application.
Reference
Liao, S. and Lee, W. (2013). Sufficient-Cause Modeling with Matched Data Using
SAS. Epidemiology, 24(6), pp.936-937.
Mane, A. and Shinde, P. (2015). Document Clustering and Automatic Labeling for Forensic
Analysis Using High Performance Clustering Algorithm. International Journal Of
Engineering And Computer Science.
Patel, K. (2012). Analysis of Various Database Using Clustering Techniques. Global Journal
For Research Analysis, 3(7), pp.59-60.
Yang, J. (2012). Using SAS for data management, statistical analysis and
graphics. Pharmaceutical Statistics, 11(4), pp.346-346.
A Survey of SNP Data Analysis. (2018). Big Data Mining and Analytics, 1(3), pp.173-190.
Allen, G., Campbell, F. and Hu, Y. (2015). Comments on “visualizing statistical models”:
Visualizing modern statistical methods for Big Data. Statistical Analysis and Data Mining:
The ASA Data Science Journal, 8(4), pp.226-228.
Buelens, B. (2016). Visual Circular Analysis of 266 Years of Sunspot Counts. Big Data, 4(2),
pp.89-96.
25

Buelens, B. (2016). Visual Circular Analysis of 266 Years of Sunspot Counts. Big Data, 4(2),
pp.89-96.
Majumdar, J., Naraseeyappa, S. and Ankalaki, S. (2017). Analysis of agriculture data using data
mining techniques: application of big data. Journal of Big Data, 4(1).
Ruan, G. and Zhang, H. (2017). Closed-loop Big Data Analysis with Visualization and Scalable
Computing. Big Data Research, 8, pp.12-26.
Wegman, E. (2012). Special issue of statistical analysis and data mining. Statistical Analysis and
Data Mining, 5(3), pp.177-177.
Appendix
data BirdStrikes;
set sashelp.BirdStrikes; run;
proc datasets library=work;
modify BirdStrikes;
xattr set var
Height (Unit = "Inches")
Weight (Unit = "Pounds")
Smoking (Unit = "wildlife")
DeathCause (ValidValues = "wildlife species,aircraft,pilot warned of bird strike,total cost,altitude
bin")
26

Chol_Status (ValidValues = "<200 = Desirable; 200-239 = Borderline; >240 =
High")
run;
quit;
ods output variables=varlist;
ods output ExtendedAttributesVar=varlist2;
proc contents data=BirdStrikes; run;
proc sort data=varlist2; by attributevariable; run;
proc transpose data=varlist2 out=varlist2B;
by attributevariable;
id extendedattribute;
var attributecharvalue;
run;
proc sql;
create table datadictionaryB as
select v.num label='#',
v.variable label='Variable',
v.type label = 'Type',
v.len label = 'Length',
v.label label='Label',
27

x.validvalues label='Valid values',
x.unit label='Unit of measurement'
from varlist as v left join varlist2b as x
on upcase(v.variable) = upcase(x.attributevariable);
quit;
ods tagsets.excelxp file="BirdStrikes.xlsx" style=statistical;
proc print data=datadictionaryB noobs label; run;
ods tagsets.excelxp close;
28