دانش و فناوری هوافضا

دانش و فناوری هوافضا

تحلیل آماری و پیش‌بینی شدت سوانح هوایی با استفاده از الگوریتم‌های یادگیری ماشین: مطالعه‌ای بر داده‌های سوانح آمریکا در پایگاه داده ICAO

نوع مقاله : مقاله پژوهشی

نویسندگان
پرهام حسن 1
حمیدرضا ضرغامی 2
1 کارشناسی ارشد آمار ریاضی، دانشکده ریاضی، آمار و علوم کامپیوتر، دانشگاه تهران، ایران.
2 دانشیار مهندسی صنایع، دانشکده تحصیلات تکمیلی، دانشگاه علوم و فنون هوایی شهید ستاری، تهران، ایران.
چکیده
ایمنی پرواز و کاهش سوانح هوایی همواره از مهم‌ترین دغدغه‌های صنعت هوانوردی و نهادهای مرتبط به شمار می‌رود. این پژوهش با هدف تحلیل آمار توصیفی، استنباطی و پیش‌بینی شدت سوانح هوایی با استفاده از مدل‌های یادگیری ماشین انجام شده است. داده‌های پژوهش شامل سوانح هوایی با متغیرهایی چون مرحله پرواز، نوع موتور، نوع ریسک، نوع وسیله پرنده و تعداد موتورها است. در مرحله نخست، تحلیل‌های توصیفی و استنباطی نشان دادند که بین شدت سانحه و عواملی فنی و‌ محیطی رابطه معناداری وجود دارد. سپس، برای پیش‌بینی سوانح منجر به مرگ و میر از مدل‌های رگرسیون لجستیک، درخت تصمیم، جنگل تصادفی و درخت تقویت یافته شدید استفاده شده است. نتایج ارزیابی عملکرد نشان داد مدل XGBoost با دقت ۰٫876 و حساسیت ۰٫676 بهترین نتایج را ارائه کرده است. این یافته بیانگر برتری مدل‌های یادگیری ماشین نسبت به مدل‌های آماری سنتی در تحلیل الگوهای پیچیده سوانح هوایی است. نتایج این پژوهش می‌تواند مبنایی برای ارتقای ایمنی پرواز و کاهش ریسک سوانح در سازمان‌های هوانوردی باشد.
کلیدواژه‌ها
سوانح هوایی
تحلیل‌آماری
یادگیری ماشین
ایمنی پرواز
موضوعات

عنوان مقاله English

Statistical analysis and prediction of aviation accident severity using machine learning algorithms: A study of US accident data in ICAO database

نویسندگان English

Parham Hassan 1
Hamid Reza Zarghami 2
1 M.Sc. in Mathematical Statistics, Faculty of Mathematics, Statistics and Computer Science, University of Tehran, Iran.
2 Associate Professor of Industrial Engineering, Faculty of Graduate Studies, Shahid Sattari University of Aerospace Science and Technology, Tehran, Iran
چکیده English

Flight safety and reducing air accidents have always been the most important concerns of the aviation industry and related institutions. This study aimed to analyze descriptive and inferential statistics and predict the severity of air accidents using machine learning models. The research data included air accidents with variables such as flight phase, engine type, risk type, aircraft type, and number of engines. In the first stage, descriptive and inferential analyses showed that there is a significant relationship between the severity of the accident and technical and environmental factors. Then, logistic regression, decision tree, random forest, and robust boosted tree models were used to predict fatal accidents. The performance evaluation results showed that the XGBoost model provided the best results with an accuracy of 0.876 and a sensitivity of 0.676. This finding indicates the superiority of machine learning models over traditional statistical models in analyzing complex patterns of air accidents. The results of this research can be a basis for improving flight safety and reducing the risk of accidents in aviation organizations.

کلیدواژه‌ها English

Aviation Accidents
Statistical Analysis
Machine Learning
Flight Safety
[1] A. Nanyonga, U. Turhan, and G. Wild, "A Systematic Review of Machine Learning Analytic Methods for Aviation Accident Research," Sci, vol. 7, no. 3, p. 124, Sep. 2025.
[2] M. Caetano, “Aviation accident and incident forecasting combining occurrence investigation and meteorological data using machine learning”, Aviation, vol. 27, no. 1, pp. 47–56, Mar. 2023.
[3] R. A. Burnett and D. Si, "Prediction of injuries and fatalities in aviation accidents through machine learning," in Proc. Int. Conf. Compute and Data Analysis, May 2017, pp. 60–68.
[4] T. T. İnan and N. G. İnan, "Analysis of the primary factors affecting the most fatal aviation accidents: a machine learning approach," Reliability: Theory & Applications, vol. 17, no. 1 (67), pp. 164–177, 2022.
[5] T. T. İnan and N. G. İnan, "The analysis of fatal aviation accidents more than 100 dead passengers: an application of machine learning," Opsearch, vol. 59, no. 4, pp. 1377–1395, 2022.
[6] D. V. Silagyi II and D. Liu, "Prediction of severity of aviation landing accidents using support vector machine models," Accident Analysis & Prevention, vol. 187, p. 107043, 2023.
[7] F. Omrani, H. Etemadfard, and R. Shad, "Assessment of aviation accident datasets in severity prediction through machine learning," J. Air Transp. Manage., vol. 115, p. 102531, 2024.
[8] V. de Vries, "Classification of Aviation Safety Reports using Machine Learning," in Proc. 2020 Int. Conf. Artificial Intelligence and Data Analytics for Air Transportation (AIDA-AT), Singapore, pp. 1-6, Feb. 2020.
[9] A. Nanyonga, K. F. Joiner, U. Turhan, and G. Wild, "Deep Learning Approaches for Classifying Aviation Safety Incidents: Evidence from Australian Data," AI, vol. 6, no. 10, p. 251, 2025.
[10] X. Zhang and S. Mahadevan, "Ensemble machine learning models for aviation incident risk prediction," Decision Support Syst., vol. 116, pp. 48–63, 2019.
[11] A. O. Alkhamisi and R. Mehmood, “An Ensemble Machine and Deep Learning Model for Risk Prediction in Aviation Systems,” in Proc. 2020 6th Conf. Data Science and Machine Learning Applications (CDMA), Riyadh, Saudi Arabia, pp. 54-59, Mar. 2020.
[12] M. Monika, S. Verma, P. Kumar, and M. Kakran, "A stacking model for aviation high risk event decision making," in J. Phys.: Conf. Ser., vol. 1950, no. 1, p. 012069, IOP Publishing, 2021.
[13] A. Khattak, P. W. Chan, F. Chen, H. Peng, and C. Mongina Matara, “Missed Approach, a Safety-Critical Go-Around Procedure in Aviation: Prediction Based on Machine Learning-Ensemble Imbalance Learning,” Advances in Meteorology, vol. 2023, no. 1, p. 9119521, 2023.
[14] S. Wu, Q. Yuan, Z. Yan, and Q. Xu, "Analyzing accident injury severity via an extreme gradient boosting (XGBoost) model," J. Adv. Transp., vol. 2021, no. 1, p. 3771640, 2021.
[15] X. Shi, Y. D. Wong, M. Z. F. Li, C. Palanisamy, and C. Chai, “A feature learning approach based on XGBoost for driving assessment and risk prediction,” Accident Anal. Prev., vol. 129, pp. 170–179, 2019.
[16] X. Shen and S. Wei, “Application of XGBoost for Hazardous Material Road Transport Accident Severity Analysis,” IEEE Access, vol. 8, pp. 206806–206819, 2020.
[17] Y. Xing, Y. Wu, S. Zhang, L. Wang, H. Cui, B. Jia, and H. Wang, "Discovering latent themes in aviation safety reports using text mining and network analytics," Int. J. Transp. Sci. Technol., vol. 16, pp. 292–316, 2024.
[18] A. Ahadh, G. V. Binish, and R. Srinivasan, "Text mining of accident reports using semi-supervised keyword extraction and topic modeling," Process Saf. Environ. Prot., vol. 155, pp. 455–465, 2021.
[19] A. Nosrati Malekjahan, A. Husseinzadeh Kashan, and B. Ostadi, "Analysis of Iran’s civil aviation accidents and serious ‎incidents from 1979 to 2021 using failure mode and ‎effects analysis and multi-attribute decision making ‎methods," J. Aeronaut. Eng., vol. 25, no. 1, pp. 154-180, 2023 (in Persian).
[20] J. Raee, H. R. Zarghami, and A. A. Zand, "Analysis of I.R.I.A.A Aviation Accidents from 1989 to 2019 Using Data Mining Tools," Mil. Sci. Tactics, vol. 17, no. 56, pp. 29-50, 2021 (in Persian).
[21] M. Rahnama and S. Khansalari, "Analysis of Weather-Related Aviation Accidents in Iran Using ASN Information and ERA5 Reanalysis Data," Nivar, vol. 49, no. 130-131, pp. 100-121, 2025 (in Persian).