Summary: | 碩士 === 國立成功大學 === 民航研究所 === 104 === Big data analysis has become a very popular topic recently. In various fields are trying to bring the new insight into their enterprise for increasing revenue or finding potential patterns. Furthermore the airline can seek new way to serve the marketplace and increase their profit, balancing both safety and revenue. Basis for big data analysis is on the volume, variety, velocity, and veracity of data. It needs to be well collected, analyzed, and applied that can reveal its value. For aviation, there is a strict regulation for collecting, preserving data, but most of the data are not well used.
In this study, based on CRISP-DM, a robust and well-proven data mining methodology, as the fundamental concept to create a data analytic processing which is suitable for aviation. A generic data thinking can help analysts to handle with diversified data type, a variety data source, and different analyzing software. We focus on the procedure that gives aviation analysts a logical and data thinking. The purpose of this study is to create a preliminary approach of aviation data analytic, because there is not a standard operation procedure yet. However, data analytic is a case by case study, it is impossible to have an all-powerful tool which is able to analyze every database. An engine sensor database is used in this procedure to give example for data thinking and investigating. Try to find hidden information to reveal its value.
Bringing big data analytic into aviation safety system does not mean to exclude experts' opinion. Human work with data can generate great insights.
|