This project seeks to guide the company as it ventures into operation of commercial and private airplanes. The main focus is on comparing the risk levels of different aircraft makes and come up with aircraft models that present the lowest risk and highest potential for successful operations. Using Exploratory Data Analysis(EDA), the dataset from the National Transportation and Safety Board is used to profile the Aircraft models into either low risk or high risk depending on the number of accidents recorded.
- Determine the lowest-risk aircraft options for the company's expansion into the aviation industry.
- Evaluate various Aircraft Models in terms of reliability and market reputation by looking into the number of accidents the model was involved in over time.
- Use Exploratory Data Analysis(EDA) to clean the dataset, aggregating and formulating visualization tools for a clear presentation.
- Create an interactive dashboard so that the users can explore the analysis.
- A summary of the key insights from the Analysis and recommendations for the company.
In this project, we are going to use analytical skills to determine the lowest risk aircraft options for the company's expansion into the aviation industry. We will use Aviation_Data.csv from the National Transportation Safety Board that includes aviation accident data from 1962 to 2023. From the csv file used in this project, there are a total of 90348 entries and 32 columns.
python import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns %matplotlib inline df = pd.read_csv('Aviation_Data.csv', dtype={6: 'object', 7: 'object', 28: 'object'}) df.head()
python percent_missing = df.isnull().mean() * 100 missing_value_df = pd.DataFrame({'column_name': df.columns, 'percent_missing': percent_missing}) missing_value_df.sort_values('percent_missing', inplace=True) print(missing_value_df) df = df.loc[:, df.isnull().mean() < 0.3] df.head()
Treat the remaining Null values with mean for all numerical columns and mode for categorical columns
python numerical_cols = df.select_dtypes(include=['number']).columns categorical_cols = df.select_dtypes(exclude=['number']).columns numerical_means = df[numerical_cols].mean() categorical_modes = df[categorical_cols].mode().iloc[0] df[numerical_cols] = df[numerical_cols].fillna(numerical_means) df[categorical_cols] = df[categorical_cols].fillna(categorical_modes)
python duplicates = aviation_data.duplicated().sum() duplicates aviation_data= aviation_data.drop_duplicates()
`python
def treat_outliers(aviation_data, column): Q1 = aviation_data[column].quantile(0.25) Q3 = aviation_data[column].quantile(0.75) IQR = Q3 - Q1 lower_bound = Q1 - 1.5 * IQR upper_bound = Q3 + 1.5 * IQR median = aviation_data[column].median() aviation_data[column] = aviation_data[column].apply(lambda x: median if x < lower_bound or x > upper_bound else x)
for col in aviation_data.select_dtypes(include=['number']).columns: treat_outliers(aviation_data, col) `
A graph showing the number of accidents since 1942.
From the above graph, 1982 is the year that had most occurences of accidents. However, the number of accidents tends to go down two decades later. The reduction of the accidents may be due to production of new models of aeroplanes.
From the above graph, we can conclude that the Aircraft make with highest number of Accidents is Cessna. This make has over 25000 accidents since 1982 to 2022 which is a very high number compared other makes. Let's see which make has the least number of accidents since 1982 to 2022
The above graph shows top 10 makes and the number of accidents per year. Cessna Make is leading with the highest number of accidents, followed by Piper and Beech
Plotting a graph Aircraft Makes against the Number of accidents associated with the make.
There are 6133 makes with accident occurences less than two. Mckinnis, Aeronca-Bubeck Irving, Trike and 6130 more have incidents below 2. The above graph gives 10 Aircraft makes with number of accidents less than 2.
A graph showing number of accidents in different weather Conditions.
In aviation, instrument meteorological conditions (IMC) are weather conditions that require pilots to fly primarily by reference to flight instruments, and therefore under instrument flight rules (IFR), as opposed to flying by outside visual references under visual flight rules (VFR). Visual meteorological conditions (VMC) is an aviation flight category in which visual flight rules (VFR) flight is permitted—that is, conditions in which pilots have sufficient visibility to fly the aircraft maintaining visual separation from terrain and other aircraft. From the above graph, the number of accidents occuring from VMC are high compared to IMC weather conditions. This would mean that when there is sufficient visibility, that is under Visual Meteorological Conditions(VMC), the pilot is not required to apply the IFR(Instrument Flight Rules), the number of accidents go higher. However, when there is poor visibility due to poor weather conditions, that is under Instrument Meteorological Conditions(IMC), the pilots are required to apply the IFR and the number of accidents are less.
A count of Injuries in different categories
From the scatter plots, there is no relationship between Fatal Injuries and the Total Uninjured. There is also no relationship between Fatal Injuries and the Total Serious Injuries.
From the analysis the following conclusions were made;
- The Cessna Aircraft Model has a bad reputation from the high number of accidents it is involved in.
- There are at least 6000 Aircraft models the company can choose from. Further Analysis would be required to determine the operation cost and customer preferences either in personal or commercial flights.
- The number of accidents were higher when there was clear visibility and therefore poor weather conditions did not contribute to high number of accidents. The high number of accidents would be due to other reasons.
From the analysis of the aviation dataset, the following are the recommendations.
- The company need to watch out for the aircraft models with high number of accident occurrences since they reflect vulnerability. These include Cessna, Piper and Beech models which have recorded over 5000 number of accidents since 1948.
- The aviation industry improves with time as seen from the reduction of number of accidents from 1982. Therefore it is good to keep on researching for new model production with high level of efficiency and reliability.
- Operational cost was not covered by this analysis and therefore more analysis need to be undertaken to include the operation and maintenance costs of aircrafts for a better and informative decision to be made.
Thanks to Flatiron and Moringa School for providing the dataset and learning materials to guide me in this project. Special thanks to my Technical Mentors from Moringa School, Mark Tiba and Faith Rotich for their continued support in this journey.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent