Evaluation of the ESReDA Cube Method for the Aviation Sector

More Info


Nowadays accident analysis tools focus on finding the cause of an accident so that lessons can be learned. The ESReDA Cube method is designed to explore a new field in accident analyses. It focuses on the next step, namely analysing the lessons learned out of an accident. It is checked if they are sufficient and if they caused any change(s). With this project the ESReDA Cube method will be evaluated for the first time for the aviation sector. The ESReDA Cube analysis is performed in several steps. First the accident is described. Next a three dimensional overview of the lessons learned is created. This overview, in the shape of a cube, is called the ESReDA Cube. The three dimensions the method explores are: operation level, system level and depth of learning level. To make the overview three questions are asked for every lesson learned. Every question is related to a specific dimension: • Operation level: “What needs be learned?” (content, structure, culture, context) • System level: “Who should learn?” (micro, meso, macro) • Depth of learning level: “How to learn?” (optimize, adapt, innovate) Behind every question, the available answers are presented between parentheses. The three answers represent a lesson learned as a set of three dimensional coordinates. In this way every lesson learned can be located in a specific part of the ESReDA Cube and an overview is created. Using this overview it can be analysed where exactly lessons are learned. By checking the spreading of the items over the cube it is checked if there is still learning potential somewhere. It is also possible to combine the results of several cases into one cube to analyse multiple cases together. The next step is to describe the impact of the lessons learned to check if something changed or if any implementation problems are encountered. With the results obtained conclusions can be drawn and if necessary new learning suggestions can be issued. In this thesis the method is applied on three accident cases. The three analysed cases all have stall as a contributing factor, and due to that they are quite similar. The cases chosen are: • Air France flight 447 • Colgan Air, Continental Connection flight 3407 • Turkish Airlines flight 1951 The cases are analysed both individually as well as together. The results show that the part of innovation on the depth of learning level is quite empty, which can indicate learning potential. Although new concepts to prevent stall accidents exist, they are not recommended. All lessons on the depth of learning level are learned in the part of adaptation and optimization. Stall accidents still occurred after these three accidents, which confirms that maybe other lessons should be learned instead of always optimizing and adapting. It can be concluded that more research should be done for new innovative concepts to prevent stall accidents. By applying the method on three cases it is seen that in the current method’s state and usage, problems are encountered which prevent a complete analysis. Only one level can be completely analysed, namely the depth of learning level. At the operation level no complete analysis is possible due to lack of input with only public sources and due to an insufficient lay-out. At the system level also no complete analysis is possible due to an insufficient lay-out of the meso and macro part. These problems cause the method to almost always obtain the same results, namely an empty aspect of innovation. Due to this the method is in its current state and usage not applicable for the aviation sector. It is recommended to change the lay-out of the operation and system level. Another recommendation is to change the target group of the method to investigators/instances with access to more than only public sources, like companies or safety agencies. To check if the new lay-out and target group make the method applicable for aviation further evaluation is recommended.