Peiwen Zhang

Mr. Peiwen Zhang

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 September 2023 in Cognition, Technology & Work
Verified
Journal Article
Open access
PDF file
HSHong Sun
Hong Sun
XZXin Zhou
Xin Zhou
PZPeiwen Zhang
Peiwen Zhang
XLXin Liu
Xin Liu
YLYunsong Lu
Yunsong Lu
WSWuyang Song
Wuyang Song
A pilot's control behavior has a direct impact on aviation safety. To adapt to new flight training reforms, a competency-based assessment method is proposed in this paper. First, assessment scenarios, key competencies and observable behaviours were determined. Then, observable behavior assessment indexes were developed based on standard operating procedures and practical examination standards. Flight training data were analyzed based on curve similarity theory and a time series fast retrieval algorithm, and the assessment indexes were scored. The competency assessment method was established by quantifying “HOW MANY” and “HOW OFTEN” observable behaviors according to the assessment indexes score results. Finally, the fuzzy C-means clustering algorithm was used to classify the assessment values, and assessment results were obtained. A teardrop pattern procedure was used to test the reliability of the assessment method. The results indicate that the correlation coefficient between the model evaluation and the expert evaluation is 0.947 with a significance level of 0.01, reliably and objectively quantifying manual flight performance and improving the scientific nature of manual flight performance assessment methods. Flight training institutions can use this method to establish diverse assessment scenarios and formulate corresponding assessment indexes based on the characteristics of each pilot to reasonably evaluate the manual flight performance of pilot. Targeted training methods can then be adopted for pilot with different performance levels to improve their competency.

28 June 2022 in Transportation Research Part E: Logistics and Transportation Review
Verified
Journal Article
Closed access
JWJiang Wu
Jiang Wu
PZPeiwen Zhang
Peiwen Zhang
YWYu Wang
Yu Wang
Jim ShiJim Shi
Jim Shi


22 February 2024 in Journal of Cleaner Production
Verified
Journal Article
Closed access
MWMinke Wang
Minke Wang
BYBeibei Ye
Beibei Ye
Shuang LinShuang Lin
Shuang Lin
CWChaofeng Wang
Chaofeng Wang
PZPeiwen Zhang
Peiwen Zhang

22 May 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
HSHong Sun
Hong Sun
方杨方铨 杨
方铨 杨
PZPeiwen Zhang
Peiwen Zhang
WZWenke Zhao
Wenke Zhao
Ensuring training safety is paramount to flight schools. In response to the inadequacy of traditional flight training assessment for comprehensive quantitative evaluation of cadet competency, an initial flight training competency assessment standard based on behavioral indicators was developed and optimized using the VENN model. Firstly, the Assessor Score Measurement Form (ASMF) was constructed according to the requirements of the Training Evaluation Worksheet specification, such as typical subjects, observations, and completion criteria. Secondly, based on the basic principles of the experience of the flight expert and the Competency-Based Training and Assessment (CBTA), a matrix of correlations between the observations and each competency-based behavioral indicator was created to construct a competency assessment matrix. In addition, a two-dimensional model for representing competency items characterized by behavioral indicators was established and an optimization model for competency assessment criteria was constructed. Finally, through combining actual flight training data, the proposed method was validated in the flight screening check phase. The results show that the optimized flight training competency assessment scheme can be well quantified and matched to real instructor ratings with an accuracy of 84%. The assessment worksheet, the assessment matrix, and the VENN competency rating model can be adapted to the different teaching requirements of each flight phase, achieving a perfect match between the behavioral indicators and the competency items, which is highly versatile. The proposed model can more accurately reflect the core competencies of flight trainees, enable quantitative assessment of behavioral indicators and competency items, and provide support for subsequent training of trainees.

1 January 2024 in Computer Modeling in Engineering & Sciences
Verified
Journal Article
Open access
PDF file
HSHong Sun
Hong Sun
FYFangquan Yang
Fangquan Yang
PZPeiwen Zhang
Peiwen Zhang
YJYang Jiao
Yang Jiao
YZYunxiang Zhao
Yunxiang Zhao
With the development of the integration of aviation safety and artificial intelligence, research on the combination of risk assessment and artificial intelligence is particularly important in the field of risk management, but searching for an efficient and accurate risk assessment algorithm has become a challenge for the civil aviation industry. Therefore, an improved risk assessment algorithm (PS-AE-LSTM) based on long short-term memory network (LSTM) with autoencoder (AE) is proposed for the various supervised deep learning algorithms in flight safety that cannot adequately address the problem of the quality on risk level labels. Firstly, based on the normal distribution characteristics of flight data, a probability severity (PS) model is established to enhance the quality of risk assessment labels. Secondly, autoencoder is introduced to reconstruct the flight parameter data to improve the data quality. Finally, utilizing the time-series nature of flight data, a long and short-term memory network is used to classify the risk level and improve the accuracy of risk assessment. Thus, a risk assessment experiment was conducted to analyze a fleet landing phase dataset using the PS-AE-LSTM algorithm to assess the risk level associated with aircraft hard landing events. The results show that the proposed algorithm achieves an accuracy of 86.45% compared with seven baseline models and has excellent risk assessment capability.