Investigating Airplane Crash Data with Watson Analytics and Cognos Analytics

Over the last century, the world has seen a number of airplane crashes both in the sky and on the ground. The development of technology aims to decrease technological difficulties and human errors. However, fatalities and aircraft damages caused by crashes do not cease to exist. Investigating Airplane Crash Data with Watson Analytics and Cognos Analytics is a big data project that involves discovering and examining airplane crash patterns using IBM's Watson Analytics and Cognos Analytics. This project aims to: (1) find some factors that contribute to crashes, (2) analyze patterns of the data collected from all over the world in the past decades, and (3) find replicable solutions for both aviation industry and customers. This paper looks for correlations between crashes and different variables and predicts when crashes are most likely to happen in a year. The analysis of over 73,000 data points, collected from a subsidiary of Google LLC Kaggle, displays a decrease in the number of airplane crashes and fatalities over the years, an increase in the number of passengers, predictive values for the future years, correlations between different variables, and strong correlations between the occurrence of crashes and the following time variable: yearly, monthly, and daily. The results found in this project will benefit the ongoing investigations into this important topic. Understanding what factors cause airplane crashes helps aviation industries make continuous improvement in flight safety, and help raise customer confidence with the use of statistical evidence.