Machine-Learning Tool for Human Factors Evaluation – Application to Lion Air Boeing 737-8 Max Accident

Abstract

The capability of learning from accidents as quickly as possible allows preventing repeated mistakes to happen. This has been shown by the small time interval between two accidents with the same aircraft model: the Boeing 737-8 MAX. However, learning from major accidents and subsequently update the developed accident models has been proved to be a cumbersome process. This is because safety specialists use to take a long period of time to read and digest the information, as the accident reports are usually very detailed, long and sometimes with a difficult language and structure. A strategy to automatically extract relevant information from report accidents and update model parameters is investigated. A machine-learning tool has been developed and trained on previous expert opinion on several accident reports. The intention is that for each new accident report that is issued, the machine can quickly identify the more relevant features in secondsinstead of waiting for some days for the expert opinion. This way, the model can be more quickly and dynamically updated. An application to the preliminary accident report of the 2018 Lion Air accident is provided to show the feasibility of the machine-learning proposed approach.

An industry perspective on the use of machine learning in drug and vaccine safety

“…Frontiers in Drug Safety and Regulation frontiersin.org accident models has historically been a cumbersome process because of the long period required to review and digest the information contained in long, detailed and highly technical accident reports prepared by safety specialists (Morais et al, 2019). Morais and colleagues developed an ML tool that uses text recognition and text classification, combined with a support vector machine for classifying text according to a predefined taxonomy, to create a 'virtual risk expert' that automatically extracts relevant information from accident reports.…”

Identification of human errors and influencing factors: A machine learning approach

“…Differently from the first test of the tool performed on the preliminary accident report (Morais et al, 2019), this research tested the machine-learning tool on the final accident report of the Lion Air Aircraft flight, issued on October 2019 (one year after the accident) (KNKT, 2019). Although the final accident report of Ethiopian airlines was reportedly issued (Marks and Dahir, 2020), the link was not accessible for unknown reasons until the date this paper was submitted to reviewers, thus not included in this research (Google, 2018;Zhang et al, 2019).…”

“…In order to absorb lessons learnt from different industry sectors, the objective is to continually add to the dataset reports only from industries with the same level of complexity regarding the interaction of organisational structure, technology and humans (Moura et al, 2016). The work hereby presented is a substantial improvement and extension of the strategy proposed by some of the authors of this paper in a conference (Morais et al, 2019). Therefore, the aim of the present research is not only to expand MATA-D, but to do it faster and timely.…”

Automatic Incident Triage in Radiation Oncology Incident Learning System

“…To the best of our knowledge, there is no work reported in the field of radiotherapy to identify the severity of the incidents reported using incident description. However there have been well reported research in other industries such as aviation, and nuclear [ 8 , 9 , 10 , 11 , 12 ] to classify the incidents reported in the respective fields. In healthcare there has been successful work done in classifying the verbal autopsies [ 13 ].…”