Research On Hierarchical Evaluation And Diagnosis Method For Human Factors In Civil Aircraft Cockpit

Human errors have become the primary contributor to modern aviation accidents.The evaluation of cockpit human factors(CHF)has great significance to cockpit interface design improvement as well as to airworthiness certification.It is also one of the key issues that have to be addressed in China’s large aircraft program.Challenges for the design of a comprehensive CHF evaluation method exist in the following aspects: firstly,since there is lack of universal and comprehensive understanding of CHF,it is difficult to cover the airworthiness certification demand.Secondly,because of the various task scenarios,countless uncertainties during flight and strict requirement for non-interference,it is extremely difficult to design a valid,flexible and applicable evaluation method.Thirdly,existent methods mainly focus on judging but overlook diagnosis.Their results could provide little guidance to design improvement.The purpose of this study was to propose a comprehensive,valid,diagnostic,flexible and applicable method to evaluate human factors in large civil aircraft cockpit.The research focused on CHF evaluation and diagnosis theory,construction of CHF evaluation indicator system,design of the indicator integration method and application in airworthiness certification.The study was supported by National Basic Research Program of China(973 Program,under grant: 2010CB734103).The main innovations of this study include:1)Aiming at the limitation of the “workload” concept in human factors evaluation,this paper proposed a multidimensional conceptual framework of CHF based on “human-machine interaction and human-machine state”.The framework summarized the effects of various safety factors into variations of pilot-cockpit interactions,pilot state and system state,and used “cognitive activity”,“control activity”,“workload” and “flight performance” to describe and evaluate CHF.Simulated flight experiment showed that evaluation results based on the proposed conceptual framework were more comprehensive and sensitive than traditional evaluation method which only examined pilot workload.2)In order to solve the problem of poor diagnostic ability of the existing human factors evaluation methods,this paper proposed to reveal the source of safety problem from design perspective by examine the correlations between human-machine interaction and human-machine state.Specifically,correlation between cognitive activity and workload is used to reflect the difficulty of perception and processing the information in cockpit interface.Correlation between control activity and workload is used to reflect the control difficulty of the cockpit components.Correlation between cognitive activity and flight performance is used to reflect the conspicuity of cockpit warnings.Correlation between control activity and flight performance is used to reflect the control efficiency of cockpit component.3)Since existing human factors evaluation indicator system is inflexible,not comprehensive and can hardly meet the requirement of airworthiness standard,this study built a “Hierarchical and multi-faceted” evaluation indicator system based on quantitative objective measures,under guidance of the proposed conceptual framework and the airworthiness demand.For the ambiguity of physiological indicators,this study examined the correlations between multiple physiological indicators during flight task and revealed the potential meanings of correlated indicators.The indicator system was revised and optimized according to these results.The indicator system also can be flexibly adjusted according to actual circumstances.4)Aiming at the low reliability of existing evaluation method,this study proposed a “hierarchically group integration method” based on “weighted principal component scores”.This method adaptively utilizes the common trends in indicators,and solves the problem of heavy information losses.Results of simulated flight experiment proved that the method can not only provide satisfying sensitivity but also reduce the distortion of individual indicator.Moreover,it has the ability to explain the evaluation result in detail from multiple aspects.In order to validate the proposed method in real world,this study participated in ARJ21-700’s airworthiness compliance test for “Minimum Flight Crew” regulation(CCAR-25.1523).In this process,differences between laboratory environment and real world were carefully analyzed.Specific solutions were presented to all encountered problems.Evaluation results showed that the proposed method can effectively distinguish the CHF in different types of cockpit.It can also provide certain diagnosis from design perspectives.This study provided an effective new method for compliance certification of the “minimum flightcrew” regulation.It also accumulated abundant valuable experiences and made significant contributions to China’s airworthiness certification,and promotes the development of China’s aviation human factors field.Above all,this study provides important theoretical foundation,technical support and experience guidance to the evaluation of cockpit human factors.