Unstable approach in aviation : causal factors and preventive interventions

• Main Author: Lai, Hsueh-Yi
• Other Authors: Chen Chun-Hsien
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Aeronautical engineering::Accidents and air safety
• Online Access: https://hdl.handle.net/10356/137147
• Institution: Nanyang Technological University

Landing safety has long been a concern in commercial airlines. In a landing system, a team with heterogeneous operators, including pilots and air traffic controllers (ATCOs), is required to collaboratively accomplish safe landing. The landing without satisfying stable approach criteria, known as unstable approach, may cause severe economic losses or even casualties. Therefore, the occurrence of an unstable approach has attracted research attention, which highlights the necessity of improving the quality of a landing system. To study the mechanism of such system failures, the quality of team interaction plays a critical role. Specifically, the synchronization of the team members’ respective shared mental models (SMMs) is compulsory for the development of holistic team cognition. The mental model disconnects (MMDs) among the team members cause interaction conflicts, jeopardizing a system’s reliability and performance. Accordingly, this research aims to study the contextual relationship among the occurring MMDs in order to pinpoint the root causes of the failures in a landing system that need to be mitigated. To fill the gaps, role-specific SMMs and the framework of evolving team cognition (FETC) are accordingly proposed. Three consecutive research efforts are performed in this thesis, namely, MMD mapping, implementation of FETC in an unstable approach, and investigation into better information transparency in a landing system. First, in MMD mapping, the exact items of the MMDs are extracted using a crew-based critical decision method. With the MMDs in the task-related SMM, a team may face difficulty in integrating the respective working results during an implicit interaction, which requires further negotiation to coordinate each other’s requirements. On the contrary, due to the possible MMDs in the team-related SMM, a team may fail to conduct the interaction by taking proper actions or critical information during an explicit interaction. This might result in the team missing the window to intercept landing risks from evolving. Furthermore, a new type of MMDs, called conflicting disconnects, are identified in contingent scenarios such as unexpected weather change. This MMD highlights the possible contradiction of the respective requirements between parties. Based on the above findings, the proposed FETC is successfully implemented in an unstable approach. This work applies agent-based modeling (ABM) to simulate the interaction of autonomous team members. The result shows that the occurrence of MMDs during interaction is relatively frequent around the terminal area. A strong relationship between team-related MMDs and landing instability is found, which explains 40% of the variance in exploratory factor analysis. Specifically, two major categories of team-related MMDs are classified, namely passive information dissemination and late initiation of corrective interaction. Considering both MMDs, a landing system’s sensitivity and reliability are further addressed to conduct satisfactory error detection and correction. Accordingly, further research towards better information transparency in a landing system is conducted to study the positive effects of information transparency on a landing system’s performance. To eliminate the information gaps that impede a system’s information transparency, the information required for parties to make decisions in level 1 situation awareness (SA) is shared for operators to mutually monitor possible goal changes in response to changing conditions. Dynamic goal sharing facilitates a team to conduct both implicit and explicit interactions in a proactive manner. To examine how a landing system is positively affected, two agent-based models are accordingly developed to simulate the current and improved landing system. Based on the simulation results, better information transparency greatly reduces the demands of refractive interaction, reflecting a reduction in system errors. Furthermore, the efficiency of resource utilization is further improved without causing significant delay. To facilitate the knowledge obtained further, future research aims to study how the mutual information in a team facilitates the operators’ role-specific SMMs, which is dedicated to identify the critical information needed to be shared in a team.