Session: 14-04-01: Reliability and Safety in Transportation Systems I

Paper Number: 144773

144773 - The Modified Cognitive Reliability and Error Analysis in Aviation 

Human error is one of the most critical factors leading to aviation unsafe accidents． This paper demonstrates a modified cognitive reliability and error analysis (CREAM) to achieve more accurate human error probability (HEP) for flight safety. The traditional approach lacks the failure data and does not consider the common performance conditions (CPCs) weights in the scenario environment of aircraft operation processes. The modified CREAM adjusts the original CPCs to make it more reasonable considering the working situation of pilots in the cockpit. The modified CREAM uses cloud models to obtain evaluation indicators for the correlation degree of CPCs. It uses evidence theory to calculate the corelative weight of CPCs. The modified approach uses DANP (DEMATEL-Analytic Network Process) to construct a CPC network model and get the important weight of CPCs. Finally, we can obtain the comprehensive weight and then calculate the HEP. It is necessary to verify that the modified extended CREAM has correctness and sensitivity. The correctness analysis compares the results between the basic CREAM and the modified CREAM and changes any CPCs into positive and negative changes respectively. It takes the results of HEP being effective. The modified CREAM can provide references for aircraft operational reliability analysis.

 

In recent years, there has been increased interest in human reliability analysis (HRA). More and more scholars are paying attention to HRA. With the development of technology, the advanced aircraft cockpit has modern digital human-system interfaces (HSI), and they change the tasks of operators in aviation. Human reliability analysis can play an essential role in addressing the issue of safety in aviation and it gives a new challenge for HRA.

 The CREAM method can analyze the context and cognitive functions, and then apply in many fields. Liu uses the interval 2-tuple and cluster analysis to assess dependence in human reliability analysis. Zhou proposed a new approach that consists of a Bayesian network and fuzzy CREAM, and the conclusion that the method is consistent with CREAM. Ahn constructed a new framework that uses the fuzzy multiple attributive group decision-making method, Bayesian networks, and evident reasoning to modify CREAM, and then to solve human errors. The past decade has seen the rapid development of weight analysis in CREAM. It combines a hesitant fuzzy matrix into CREAM that handles the scores by experts and ensures the HFM-CREAM is right.  TIM calculated HEP combined CPC weights and analysis sensitivity using the modified CREAM. Kim adopted a profiling technique to quantify the weightings of performance shaping factors when performing HRA during LPSD operation.  The fuzzy analytic network process was used to ensure the weight of CPCs in the urban railway Considering the relationship of CPCs and the weight of CPCs, Zhang gave a modified fuzzy CREAM to tackle them. Sun uses DEMATEL and FAHP methods to improve the weight and verify its accuracy. It is taking the DANP to CREAM that the approach handles the correlations among the CPCs.

This paper gives a modified CREAM in order to calculate the human error probability more accurately. The modified CREAM has three parts, one is that we use cloud models to tackle the evaluation, and the others are to calculate the corelative weight using evidence theory and the important weight of CPCs using DANP. The comprehensive weight of CPCs has correlative weight and important weight. We use the modified CREAM to calculate the HEP of taking off and get the HEP is 0.06809. Finally, we take the correctness analysis and sensitivity analysis to verify the modified CREAM. The HEP ranged from 0.001 to 0.1 by the basic CREAM, and the result is right by the modified CREAM. The sensitivity analysis that the HEP is changing with the CPCs in different states. We can use the modified extended CREAM to calculate the HEP in the aircraft cockpit.

Presenting Author: Xiaodan Zhang Harbin Engineering University

Presenting Author Biography: Xiaodan Zhang received the B.S. from Liaoning University of Technology in 2015, and received
the M.S. from Jiaotong University in 2018. She is currently working toward the Ph.D. degree in mechanical
engineering with the Laboratory of Ship Aesthetics and Human Factors Engineering, College of Mechanical and Electrical Engineering, Harbin
Engineering University, Harbin, China.
Her research interests include human reliability for aviation.

Authors:

Xiaodan Zhang Harbin Engineering University
Shengyuan Yan Harbin Engineering University
Xin Liu Harbin Engineering University
Jian Song Harbin Engineering University
Waqas Ahmed Harbin Engineering University