Apriori algorithm applied to the analysis of alert messages sent by executive aircrafts

• Grant number: 14/15770-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: October 01, 2014
• End date: March 31, 2015
• Field of knowledge: Engineering - Aerospace Engineering - Aerospace Systems
• Principal Investigator: Ricardo Augusto Souza Fernandes
• Grantee: Osmar Mendes da Silva Junior
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

Over the years, the operation of aircraft has obtained a large number of messages, which correspond to data that characterize the performance of the aircraft. Therefore, some aspects or indicators related to component failures and aircraft health can be obtained by analyzing these messages. However, such messages accumulate so faster than reliability engineers can process it. However, these messages contain information that can assist in aircraft failure management. So, the use of Data Mining techniques is important for obtaining patterns that identify sequences of alerts and prevent possible accidents. Following this context, we notice the possibility of the Data Mining contribute effectively to the prevention of aircraft accidents and/or to reduce the maintenance time. Thus, this research project aims to carry out an analysis of these messages through Apriori algorithm, which is designed to generate association rules, i.e., rules that show a trend or pattern of sequential messages. In order to validate the proposal of this work, the algorithm will run on databases of real messages provided by a Brazilian manufacturer executive aircraft. Furthermore, a comparative analysis between the obtained association rules for aircraft operating in South America and North America will be done in order to highlight features that are due to distinct regions of operation.