Experiences in mining aviation safety data

The goal of data analysis in aviation safety is simple: improve safety. How ever, the path to this goal is hard to identify. What data mining methods a re most applicable to this task? What data are available and how should the y be analyzed? How do we focus on the most interesting results? Our answers to these questions are based on a recent research project we completed. Th e encouraging news is that we found a number of aviation safety offices doi ng commendable work to collect and analyze safety-related data. But we also found a number of areas where data mining techniques could provide new too ls that either perform analyses that were not considered before, or that ca n now be done more easily. Currently, Aviation Safety offices collect and analyze the incident reports by a combination of manual and automated methods. Data analysis is done by safety officers who are well familiar with the domain, but not with data m ining methods. Some Aviation Safety officers have tools to automate the dat abase query and report generation process. However, the actual analysis is done by the officer with only fairly rudimentary tools to help extract the useful information from the data. Our research project looked at the application of data mining techniques to aviation safety data to help Aviation Safety officers with their analysis task. This effort led to the creation of a tool called the "Aviation Safety Data Mining Workbench". This paper describes the research effort, the work bench, the experience with data mining of Aviation Safety data, and lessons learned.