Air transportation systems are designed to ensure that aircraft accidents are rare events. To minimize these accidents, factors causing or contributing to accidents must be understood and prevented. Previous research has studied accident data to determine these factors. The low rate of accidents however, makes it difficult to discover repeating patterns of these factors. In this research we employed a data mining technique to conduct a holistic analysis of aircraft incident data in relation to the accident data. The analysis identifies relationships between the accident and incident data and finds patterns of causal and contributory factors which are significantly associated with aircraft accidents. Keywordsaviation safety; aircraft accidents; aircraft incidents; data mining; contrast-set mining
[1]
工業講話会.
安全装置工業事故豫防法 = Industrial accident prevention
,
1918
.
[2]
B Strauss,et al.
EVERYDAY THREATS TO AIRCRAFT SAFETY
,
2003
.
[3]
Stephen D. Bay,et al.
Detecting change in categorical data: mining contrast sets
,
1999,
KDD '99.
[4]
C A Hart.
STUCK ON A PLATEAU: A COMMON PROBLEM. IN: ACCIDENT PRECURSOR ANALYSIS AND MANAGEMENT: REDUCING TECHNOLOGICAL RISK THROUGH DILIGENCE
,
2004
.
[5]
G.W.H. van Es.
Review of Air Traffic Management-related Accidents Worldwide: 1980-2001
,
2003
.
[6]
Yunlong Zhang,et al.
Safety of Efficiency: Link Between Operational Performance and Operational Errors in the National Airspace System
,
2004
.