頁籤選單縮合
| 題 名 | The Interaction betweeen Pilots and the Automated Systems during the Landing Phase=航機降落階段飛行員與自動駕駛系統之人機互動 |
|---|---|
| 作 者 | Carugati, Claudio; Li, Wen-chin; Brennan, Martin; Li, Wen-chin; Carugati, Claudio; Brennan, Martin; | 書刊名 | 航空安全及管理季刊 |
| 卷 期 | 2:3 2015.07[民104.07] |
| 頁 次 | 頁227-251 |
| 分類號 | 557.94 |
| 關鍵詞 | 人為失誤鑑別技術; 人為因素預測模組; 飛行安全; 階層任務分析; Automation; Human error template; Aviation; Autoland; Pilot error; |
| 語 文 | 英文(English) |
| 中文摘要 | 最近十年間,有關人為失誤之議題已成為國際上對飛機「適航性」認證當局關注的重點,尤其針對新一代高度自動化的飛機。人為失誤的原因是多方面的,這些原因與現代飛機的設計之間有著非常複雜的互動關係。本研究共有12位商用飛機之飛行員(平均飛行時數9130小時)參與,研究目的則是在探討在降落階段因座艙自動化所激發的人為失誤,研究方法則以「人為失誤預測模組(Human Errors Template, HET)」做為資料蒐集之工具。許多被認為是「飛行員人為失誤」的飛安事件,實際是「設計錯誤」所造成的。這種論述顯示出完善的自動化設計對降低人為失誤是極其重要的。人為失誤的發生與飛機各系統的設計有著非常複雜的互動關係,儘管新一代高度自動化的飛機具備眾多優點,但新型態的「人為失誤」亦相繼在座艙中被激發出來。「人為失誤預測模組」是以檢查表形態,包含十二項錯誤模式,做為預測人為失誤之方法,並應用「階層任務分析(HTA)」,將每一個任務所需執行的步驟分析到最底層,分析者必須辨認出每一任務步驟潛在錯誤模式發生之機率,與對飛行安全影響程度之嚴重性。階層任務分析法所提供的相關訊息,更被應用於人為失誤鑑別技術(Human Error Identification, HEI)。本研究發現在降落階段有關飛行員與航管間之溝通可能干擾對自動化系統之操控,而激發出程序失誤或對自動化系統之模式混淆的潛在人為失誤之模式。透過改善「軟體」及「硬體裝備」之設計,讓座艙操作環境降低對飛行員執行任務時之干擾,使整個系統在人-機器-環境的介面能整合在以人為主的人因設計理念,才能有效提升飛行安全。 |
| 英文摘要 | Advances in technology and automation have contributed to the steady reduction of aviation occurrences during the last decades. The evolution of the modern aircraft and the complexity of its systems, however, has introduced new forms of human errors. Twelve commercial pilots with an average of 9,130 flying hours took part in this research. The Human Error Template (HET) technique was used as the research tool for the data collection, while the analysis of the error modes was conducted through an innovative likelihood and criticality matrix. The objective of the study was the analysis of potential automation-induced pilot errors during the performance of an automatic landing. Furthermore, through the evaluation of the relationships between pilots and aircraft systems, it was proposed to ascertain the frequency of human errors occurring during the autoland sequence, and to provide recommendations for future interventions concerning design, procedures, and pilots training. The findings indicate an excellent level of interaction, co-operation and co-ordination between aircraft systems and crew members. Nevertheless, distractions, interruptions and forgetfulness have been highlighted by the participants as common triggers of automation related pilot errors. The 'too late' selection on the Multipurpose Control and Display Unit (MCDU) of the Approach (APPR) PHASE mode, along with the 'missed' selection of the APPR button, and disconnections of the autopilots were indicated by the highest number of participants (91.7%) as the most common procedural error modes. Communications and calls from Air Traffic Controllers (ATCs) have been stated as one of the major contributory factors for automation omission errors. Specific training and simulator sessions to address the most common pilot error modes, implementation of a modified landing checklists to include further cross-checks for these critical operations, and advancements in the datalink technology to reduce the communication between pilots and ATCs - thereby minimising disruptions, were proposed as possible solutions to mitigate or eliminate the problems. |
本系統中英文摘要資訊取自各篇刊載內容。